From 608cf5b7935e1d03d1a8854a928e9bf8b30835be Mon Sep 17 00:00:00 2001 From: Patrick Steinhardt Date: Tue, 2 Sep 2025 12:47:41 +0200 Subject: [PATCH 001/248] gitlab-ci: disable realtime monitoring to unbreak Windows jobs The GitLab CI runners using Windows machines have realtime monitoring via Windows Defender enabled by default. This has just now started to cause issues in our CI jobs using Microsoft Visual Studio: Program 'meson.exe' failed to run: Operation did not complete successfully because the file contains a virus or potentially unwanted softwareAt line:356 char:1 + meson setup build --vsenv -Dperl=disabled -Dbackend_max_links=1 -Dcre ... + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. At line:356 char:1 + meson setup build --vsenv -Dperl=disabled -Dbackend_max_links=1 -Dcre ... + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + CategoryInfo : ResourceUnavailable: (:) [], ApplicationFailedException + FullyQualifiedErrorId : NativeCommandFailed The detected issue is more likely than not completely bogus, but it breaks the jobs. Fix the issue by disabling realtime monitoring. Besides unbreaking CI, it also improves our build times a bit: - Building Git goes from 26 to 22 minutes. - Executing tests goes from ~1h for one slice of tests to ~30 minutes. This is still painfully slow, but the issue here is that the Windows runners on GitLab CI are quite underwhelming overall. Signed-off-by: Patrick Steinhardt Signed-off-by: Junio C Hamano --- .gitlab-ci.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index bb6d5b976cdc36..ef57798e2b0afb 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -119,6 +119,7 @@ build:mingw64: variables: NO_PERL: 1 before_script: + - Set-MpPreference -DisableRealtimeMonitoring $true - ./ci/install-sdk.ps1 -directory "git-sdk" script: - git-sdk/usr/bin/bash.exe -l -c 'ci/make-test-artifacts.sh artifacts' @@ -135,6 +136,7 @@ test:mingw64: - job: "build:mingw64" artifacts: true before_script: + - Set-MpPreference -DisableRealtimeMonitoring $true - git-sdk/usr/bin/bash.exe -l -c 'tar xf artifacts/artifacts.tar.gz' - New-Item -Path .git/info -ItemType Directory - New-Item .git/info/exclude -ItemType File -Value "/git-sdk" @@ -148,6 +150,7 @@ test:mingw64: tags: - saas-windows-medium-amd64 before_script: + - Set-MpPreference -DisableRealtimeMonitoring $true - choco install -y git meson ninja openssl - Import-Module $env:ChocolateyInstall\helpers\chocolateyProfile.psm1 - refreshenv From c0bec06cfedb930ecb3cd8c8cdb0f3e14e5e9308 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Wed, 24 Sep 2025 13:57:15 -0700 Subject: [PATCH 002/248] diff --no-index: fix logic for paths ending in '/' If one of the two provided paths for git diff --no-index ends in a '/', a failure similar to the following occurs: $ git diff --no-index -- /tmp/ /tmp/ ':!' fatal: `pos + len' is too far after the end of the buffer This occurs because of an incorrect calculation of the skip lengths in diff_no_index(). The code wants to calculate the length of the string, but add one in case the string doesn't end with a slash. The method it uses is incorrect, as it always checks the trailing NUL character of the string. This will never be a '/', so we always add one. In the event that we *do* have a trailing slash, this will create an off-by-one length error later when using the skip value. The most straightforward fix would be to correct the skip1 and skip2 lengths by using ends_with(). However, Johannes made a good point that the existing logic is wasting a lot of computation. We generate the match string by copying the path in and then skipping almost all of it immediately with a potentially expensive memmove() from the strbuf_remove() call. We also re-initialize the match stringbuf each time we call read_directory_contents. The read_directory_contents really wants a path that is rooted at the start of the directory scan. We're currently building this by taking the full path and stripping out the start portion. Instead, replace this logic by building up the portion of the match as we go. Start by initializing two strbuf in diff_no_index containing the empty string. Pass these into queue_diff, which in turn passes the appropriate left or right side into read_directory_contents. As before, we build up the matches by appending elements to the match path and then clearing them using strbuf_setlen. In the recursive portion of the queue_diff algorithm, we build up new match paths the same way that we build up new buffer paths, by appending the elements and then clearing them with strbuf_setlen after each iteration. This is cheaper as it avoids repeated allocations, and is a bit simpler to track what is going on. Add a couple of test cases that pass in paths already ending in '/', to ensure the tests cover this regression. Reported-by: Johannes Schindelin Closes: https://lore.kernel.org/git/c75ec5f9-407a-6555-d4fb-bb629d54ec61@gmx.de/ Signed-off-by: Jacob Keller [jc: small leakfixes at the end of diff_no_index()] Signed-off-by: Junio C Hamano --- diff-no-index.c | 63 ++++++++++++++++++++++------------------ t/t4053-diff-no-index.sh | 16 ++++++++++ 2 files changed, 51 insertions(+), 28 deletions(-) diff --git a/diff-no-index.c b/diff-no-index.c index 88ae4cee56ba41..f320424f05fe0f 100644 --- a/diff-no-index.c +++ b/diff-no-index.c @@ -21,30 +21,21 @@ static int read_directory_contents(const char *path, struct string_list *list, const struct pathspec *pathspec, - int skip) + struct strbuf *match) { - struct strbuf match = STRBUF_INIT; - int len; + int len = match->len; DIR *dir; struct dirent *e; if (!(dir = opendir(path))) return error("Could not open directory %s", path); - if (pathspec) { - strbuf_addstr(&match, path); - strbuf_complete(&match, '/'); - strbuf_remove(&match, 0, skip); - - len = match.len; - } - while ((e = readdir_skip_dot_and_dotdot(dir))) { if (pathspec) { int is_dir = 0; - strbuf_setlen(&match, len); - strbuf_addstr(&match, e->d_name); + strbuf_setlen(match, len); + strbuf_addstr(match, e->d_name); if (NOT_CONSTANT(DTYPE(e)) != DT_UNKNOWN) { is_dir = (DTYPE(e) == DT_DIR); } else { @@ -57,7 +48,7 @@ static int read_directory_contents(const char *path, struct string_list *list, } if (!match_leading_pathspec(NULL, pathspec, - match.buf, match.len, + match->buf, match->len, 0, NULL, is_dir)) continue; } @@ -65,7 +56,7 @@ static int read_directory_contents(const char *path, struct string_list *list, string_list_insert(list, e->d_name); } - strbuf_release(&match); + strbuf_setlen(match, len); closedir(dir); return 0; } @@ -169,7 +160,8 @@ static struct diff_filespec *noindex_filespec(const struct git_hash_algo *algop, static int queue_diff(struct diff_options *o, const struct git_hash_algo *algop, const char *name1, const char *name2, int recursing, - const struct pathspec *ps, int skip1, int skip2) + const struct pathspec *ps, + struct strbuf *ps_match1, struct strbuf *ps_match2) { int mode1 = 0, mode2 = 0; enum special special1 = SPECIAL_NONE, special2 = SPECIAL_NONE; @@ -208,10 +200,12 @@ static int queue_diff(struct diff_options *o, const struct git_hash_algo *algop, struct string_list p2 = STRING_LIST_INIT_DUP; int i1, i2, ret = 0; size_t len1 = 0, len2 = 0; + size_t match1_len = ps_match1->len; + size_t match2_len = ps_match2->len; - if (name1 && read_directory_contents(name1, &p1, ps, skip1)) + if (name1 && read_directory_contents(name1, &p1, ps, ps_match1)) return -1; - if (name2 && read_directory_contents(name2, &p2, ps, skip2)) { + if (name2 && read_directory_contents(name2, &p2, ps, ps_match2)) { string_list_clear(&p1, 0); return -1; } @@ -235,6 +229,11 @@ static int queue_diff(struct diff_options *o, const struct git_hash_algo *algop, strbuf_setlen(&buffer1, len1); strbuf_setlen(&buffer2, len2); + if (ps) { + strbuf_setlen(ps_match1, match1_len); + strbuf_setlen(ps_match2, match2_len); + } + if (i1 == p1.nr) comp = 1; else if (i2 == p2.nr) @@ -245,18 +244,28 @@ static int queue_diff(struct diff_options *o, const struct git_hash_algo *algop, if (comp > 0) n1 = NULL; else { - strbuf_addstr(&buffer1, p1.items[i1++].string); + strbuf_addstr(&buffer1, p1.items[i1].string); + if (ps) { + strbuf_addstr(ps_match1, p1.items[i1].string); + strbuf_complete(ps_match1, '/'); + } n1 = buffer1.buf; + i1++; } if (comp < 0) n2 = NULL; else { - strbuf_addstr(&buffer2, p2.items[i2++].string); + strbuf_addstr(&buffer2, p2.items[i2].string); + if (ps) { + strbuf_addstr(ps_match2, p2.items[i2].string); + strbuf_complete(ps_match2, '/'); + } n2 = buffer2.buf; + i2++; } - ret = queue_diff(o, algop, n1, n2, 1, ps, skip1, skip2); + ret = queue_diff(o, algop, n1, n2, 1, ps, ps_match1, ps_match2); } string_list_clear(&p1, 0); string_list_clear(&p2, 0); @@ -346,7 +355,8 @@ int diff_no_index(struct rev_info *revs, const struct git_hash_algo *algop, int implicit_no_index, int argc, const char **argv) { struct pathspec pathspec, *ps = NULL; - int i, no_index, skip1 = 0, skip2 = 0; + struct strbuf ps_match1 = STRBUF_INIT, ps_match2 = STRBUF_INIT; + int i, no_index; int ret = 1; const char *paths[2]; char *to_free[ARRAY_SIZE(paths)] = { 0 }; @@ -387,11 +397,6 @@ int diff_no_index(struct rev_info *revs, const struct git_hash_algo *algop, NULL, &argv[2]); if (pathspec.nr) ps = &pathspec; - - skip1 = strlen(paths[0]); - skip1 += paths[0][skip1] == '/' ? 0 : 1; - skip2 = strlen(paths[1]); - skip2 += paths[1][skip2] == '/' ? 0 : 1; } else if (argc > 2) { warning(_("Limiting comparison with pathspecs is only " "supported if both paths are directories.")); @@ -415,7 +420,7 @@ int diff_no_index(struct rev_info *revs, const struct git_hash_algo *algop, revs->diffopt.flags.exit_with_status = 1; if (queue_diff(&revs->diffopt, algop, paths[0], paths[1], 0, ps, - skip1, skip2)) + &ps_match1, &ps_match2)) goto out; diff_set_mnemonic_prefix(&revs->diffopt, "1/", "2/"); diffcore_std(&revs->diffopt); @@ -431,6 +436,8 @@ int diff_no_index(struct rev_info *revs, const struct git_hash_algo *algop, for (i = 0; i < ARRAY_SIZE(to_free); i++) free(to_free[i]); strbuf_release(&replacement); + strbuf_release(&ps_match1); + strbuf_release(&ps_match2); if (ps) clear_pathspec(ps); return ret; diff --git a/t/t4053-diff-no-index.sh b/t/t4053-diff-no-index.sh index 01db9243abfe4f..e0ea437685b0ab 100755 --- a/t/t4053-diff-no-index.sh +++ b/t/t4053-diff-no-index.sh @@ -322,6 +322,22 @@ test_expect_success 'diff --no-index with pathspec' ' test_cmp expect actual ' +test_expect_success 'diff --no-index first path ending in slash with pathspec' ' + test_expect_code 1 git diff --name-status --no-index a/ b 1 >actual && + cat >expect <<-EOF && + D a/1 + EOF + test_cmp expect actual +' + +test_expect_success 'diff --no-index second path ending in slash with pathspec' ' + test_expect_code 1 git diff --name-status --no-index a b/ 1 >actual && + cat >expect <<-EOF && + D a/1 + EOF + test_cmp expect actual +' + test_expect_success 'diff --no-index with pathspec no matches' ' test_expect_code 0 git diff --name-status --no-index a b missing ' From 3860985105a40f425effc49642693e90e84e1863 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 9 Oct 2025 07:46:22 +0000 Subject: [PATCH 003/248] refs: forbid clang to complain about unreachable code When `NO_SYMLINK_HEAD` is defined, `create_ref_symlink()` is hard-coded as `(-1)`, and as a consequence the condition `!create_ref_symlink()` always evaluates to false, rendering any code guarded by that condition unreachable. Therefore, clang is _technically_ correct when it complains about unreachable code. It does completely miss the fact that this is okay because on _other_ platforms, where `NO_SYMLINK_HEAD` is not defined, the code isn't unreachable at all. Let's use the same trick as in 82e79c63642c (git-compat-util: add NOT_CONSTANT macro and use it in atfork_prepare(), 2025-03-17) to appease clang while at the same time keeping the `-Wunreachable` flag to potentially find _actually_ unreachable code. Signed-off-by: Johannes Schindelin Signed-off-by: Junio C Hamano --- refs/files-backend.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/refs/files-backend.c b/refs/files-backend.c index 088b52c740b9ff..814decf323e14d 100644 --- a/refs/files-backend.c +++ b/refs/files-backend.c @@ -3186,7 +3186,13 @@ static int files_transaction_finish(struct ref_store *ref_store, * next update. If not, we try and create a regular symref. */ if (update->new_target && refs->prefer_symlink_refs) - if (!create_ref_symlink(lock, update->new_target)) + /* + * By using the `NOT_CONSTANT()` trick, we can avoid + * errors by `clang`'s `-Wunreachable` logic that would + * report that the `continue` statement is not reachable + * when `NO_SYMLINK_HEAD` is `#define`d. + */ + if (NOT_CONSTANT(!create_ref_symlink(lock, update->new_target))) continue; if (update->flags & REF_NEEDS_COMMIT) { From b3ac6e737db8635aaed0c355ebaf291b63fb0461 Mon Sep 17 00:00:00 2001 From: Kristoffer Haugsbakk Date: Wed, 8 Oct 2025 13:48:46 +0200 Subject: [PATCH 004/248] doc: fix accidental literal blocks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Make sure that normal paragraphs in most user-facing docs[1] don’t use literal blocks. This can easily happen if you try to maintain indentation in order to continue a block; that might work in e.g. Markdown variants, but not in AsciiDoc. The fixes are straightforward, i.e. just deindent the block and maybe add line continuations. The only exception is git-sparse-checkout(1) where we also replace indentation used for *intended* literal blocks with `----`. † 1: These have not been considered: • `Documentation/howto/` • `Documentation/technical/` • `Documentation/gitprotocol*` Signed-off-by: Kristoffer Haugsbakk Acked-by: Jeff King Signed-off-by: Junio C Hamano --- Documentation/config/core.adoc | 4 +- Documentation/git-config.adoc | 18 +++---- Documentation/git-rev-parse.adoc | 14 ++--- Documentation/git-shortlog.adoc | 4 +- Documentation/git-sparse-checkout.adoc | 72 ++++++++++++++++---------- 5 files changed, 64 insertions(+), 48 deletions(-) diff --git a/Documentation/config/core.adoc b/Documentation/config/core.adoc index 3fbe83eef1612c..8866ed27714b4a 100644 --- a/Documentation/config/core.adoc +++ b/Documentation/config/core.adoc @@ -75,8 +75,8 @@ The built-in file system monitor is currently available only on a limited set of supported platforms. Currently, this includes Windows and MacOS. + - Otherwise, this variable contains the pathname of the "fsmonitor" - hook command. +Otherwise, this variable contains the pathname of the "fsmonitor" +hook command. + This hook command is used to identify all files that may have changed since the requested date/time. This information is used to speed up diff --git a/Documentation/git-config.adoc b/Documentation/git-config.adoc index 511b2e26bfb00f..a633ab8ec3a9e1 100644 --- a/Documentation/git-config.adoc +++ b/Documentation/git-config.adoc @@ -117,15 +117,15 @@ OPTIONS --comment :: Append a comment at the end of new or modified lines. - - If __ begins with one or more whitespaces followed - by "#", it is used as-is. If it begins with "#", a space is - prepended before it is used. Otherwise, a string " # " (a - space followed by a hash followed by a space) is prepended - to it. And the resulting string is placed immediately after - the value defined for the variable. The __ must - not contain linefeed characters (no multi-line comments are - permitted). ++ +If __ begins with one or more whitespaces followed +by "#", it is used as-is. If it begins with "#", a space is +prepended before it is used. Otherwise, a string " # " (a +space followed by a hash followed by a space) is prepended +to it. And the resulting string is placed immediately after +the value defined for the variable. The __ must +not contain linefeed characters (no multi-line comments are +permitted). --all:: With `get`, return all values for a multi-valued key. diff --git a/Documentation/git-rev-parse.adoc b/Documentation/git-rev-parse.adoc index cc32b4b4f0f999..18383e52af72ad 100644 --- a/Documentation/git-rev-parse.adoc +++ b/Documentation/git-rev-parse.adoc @@ -174,13 +174,13 @@ for another option. Allow oids to be input from any object format that the current repository supports. - - Specifying "sha1" translates if necessary and returns a sha1 oid. - - Specifying "sha256" translates if necessary and returns a sha256 oid. - - Specifying "storage" translates if necessary and returns an oid in - encoded in the storage hash algorithm. ++ +Specifying "sha1" translates if necessary and returns a sha1 oid. ++ +Specifying "sha256" translates if necessary and returns a sha256 oid. ++ +Specifying "storage" translates if necessary and returns an oid in +encoded in the storage hash algorithm. Options for Objects ~~~~~~~~~~~~~~~~~~~ diff --git a/Documentation/git-shortlog.adoc b/Documentation/git-shortlog.adoc index d8ab38dcc1f1a6..aa92800c69ccc3 100644 --- a/Documentation/git-shortlog.adoc +++ b/Documentation/git-shortlog.adoc @@ -44,8 +44,8 @@ OPTIONS describe each commit. '' can be any string accepted by the `--format` option of 'git log', such as '* [%h] %s'. (See the "PRETTY FORMATS" section of linkgit:git-log[1].) - - Each pretty-printed commit will be rewrapped before it is shown. ++ +Each pretty-printed commit will be rewrapped before it is shown. --date=:: Show dates formatted according to the given date string. (See diff --git a/Documentation/git-sparse-checkout.adoc b/Documentation/git-sparse-checkout.adoc index 529a8edd9c1ed8..b5fe5da041676c 100644 --- a/Documentation/git-sparse-checkout.adoc +++ b/Documentation/git-sparse-checkout.adoc @@ -264,34 +264,50 @@ patterns in non-cone mode has a number of shortcomings: inconsistent. * It has edge cases where the "right" behavior is unclear. Two examples: - - First, two users are in a subdirectory, and the first runs - git sparse-checkout set '/toplevel-dir/*.c' - while the second runs - git sparse-checkout set relative-dir - Should those arguments be transliterated into - current/subdirectory/toplevel-dir/*.c - and - current/subdirectory/relative-dir - before inserting into the sparse-checkout file? The user who typed - the first command is probably aware that arguments to set/add are - supposed to be patterns in non-cone mode, and probably would not be - happy with such a transliteration. However, many gitignore-style - patterns are just paths, which might be what the user who typed the - second command was thinking, and they'd be upset if their argument - wasn't transliterated. - - Second, what should bash-completion complete on for set/add commands - for non-cone users? If it suggests paths, is it exacerbating the - problem above? Also, if it suggests paths, what if the user has a - file or directory that begins with either a '!' or '#' or has a '*', - '\', '?', '[', or ']' in its name? And if it suggests paths, will - it complete "/pro" to "/proc" (in the root filesystem) rather than to - "/progress.txt" in the current directory? (Note that users are - likely to want to start paths with a leading '/' in non-cone mode, - for the same reason that .gitignore files often have one.) - Completing on files or directories might give nasty surprises in - all these cases. ++ +First, two users are in a subdirectory, and the first runs ++ +---- +git sparse-checkout set '/toplevel-dir/*.c' +---- ++ +while the second runs ++ +---- +git sparse-checkout set relative-dir +---- ++ +Should those arguments be transliterated into ++ +---- +current/subdirectory/toplevel-dir/*.c +---- ++ +and ++ +---- +current/subdirectory/relative-dir +---- ++ +before inserting into the sparse-checkout file? The user who typed +the first command is probably aware that arguments to set/add are +supposed to be patterns in non-cone mode, and probably would not be +happy with such a transliteration. However, many gitignore-style +patterns are just paths, which might be what the user who typed the +second command was thinking, and they'd be upset if their argument +wasn't transliterated. ++ +Second, what should bash-completion complete on for set/add commands +for non-cone users? If it suggests paths, is it exacerbating the +problem above? Also, if it suggests paths, what if the user has a +file or directory that begins with either a '!' or '#' or has a '*', +'\', '?', '[', or ']' in its name? And if it suggests paths, will +it complete "/pro" to "/proc" (in the root filesystem) rather than to +"/progress.txt" in the current directory? (Note that users are +likely to want to start paths with a leading '/' in non-cone mode, +for the same reason that .gitignore files often have one.) +Completing on files or directories might give nasty surprises in +all these cases. * The excessive flexibility made other extensions essentially impractical. `--sparse-index` is likely impossible in non-cone From 0c4f1346ca57814a988be0846f9ec1560dea87dc Mon Sep 17 00:00:00 2001 From: Solly Date: Wed, 15 Oct 2025 15:03:29 +0100 Subject: [PATCH 005/248] t2401: update path checks using test_path helpers Update old-style shell path checks to use the modern test helpers 'test_path_is_file' and 'test_path_is_dir' for improved runtime diagnosis. Signed-off-by: Solly Signed-off-by: Junio C Hamano --- t/t2401-worktree-prune.sh | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/t/t2401-worktree-prune.sh b/t/t2401-worktree-prune.sh index fe671d41974281..f8f28c76ee56c7 100755 --- a/t/t2401-worktree-prune.sh +++ b/t/t2401-worktree-prune.sh @@ -24,8 +24,8 @@ test_expect_success 'prune files inside $GIT_DIR/worktrees' ' Removing worktrees/abc: not a valid directory EOF test_cmp expect actual && - ! test -f .git/worktrees/abc && - ! test -d .git/worktrees + test_path_is_missing .git/worktrees/abc && + test_path_is_missing .git/worktrees ' test_expect_success 'prune directories without gitdir' ' @@ -36,8 +36,8 @@ Removing worktrees/def: gitdir file does not exist EOF git worktree prune --verbose 2>actual && test_cmp expect actual && - ! test -d .git/worktrees/def && - ! test -d .git/worktrees + test_path_is_missing .git/worktrees/def && + test_path_is_missing .git/worktrees ' test_expect_success SANITY 'prune directories with unreadable gitdir' ' @@ -47,8 +47,8 @@ test_expect_success SANITY 'prune directories with unreadable gitdir' ' chmod u-r .git/worktrees/def/gitdir && git worktree prune --verbose 2>actual && test_grep "Removing worktrees/def: unable to read gitdir file" actual && - ! test -d .git/worktrees/def && - ! test -d .git/worktrees + test_path_is_missing .git/worktrees/def && + test_path_is_missing .git/worktrees ' test_expect_success 'prune directories with invalid gitdir' ' @@ -57,8 +57,8 @@ test_expect_success 'prune directories with invalid gitdir' ' : >.git/worktrees/def/gitdir && git worktree prune --verbose 2>actual && test_grep "Removing worktrees/def: invalid gitdir file" actual && - ! test -d .git/worktrees/def && - ! test -d .git/worktrees + test_path_is_missing .git/worktrees/def && + test_path_is_missing .git/worktrees ' test_expect_success 'prune directories with gitdir pointing to nowhere' ' @@ -67,8 +67,8 @@ test_expect_success 'prune directories with gitdir pointing to nowhere' ' echo "$(pwd)"/nowhere >.git/worktrees/def/gitdir && git worktree prune --verbose 2>actual && test_grep "Removing worktrees/def: gitdir file points to non-existent location" actual && - ! test -d .git/worktrees/def && - ! test -d .git/worktrees + test_path_is_missing .git/worktrees/def && + test_path_is_missing .git/worktrees ' test_expect_success 'not prune locked checkout' ' @@ -76,23 +76,23 @@ test_expect_success 'not prune locked checkout' ' mkdir -p .git/worktrees/ghi && : >.git/worktrees/ghi/locked && git worktree prune && - test -d .git/worktrees/ghi + test_path_is_dir .git/worktrees/ghi ' test_expect_success 'not prune recent checkouts' ' test_when_finished rm -r .git/worktrees && git worktree add jlm HEAD && - test -d .git/worktrees/jlm && + test_path_is_dir .git/worktrees/jlm && rm -rf jlm && git worktree prune --verbose --expire=2.days.ago && - test -d .git/worktrees/jlm + test_path_is_dir .git/worktrees/jlm ' test_expect_success 'not prune proper checkouts' ' test_when_finished rm -r .git/worktrees && git worktree add --detach "$PWD/nop" main && git worktree prune && - test -d .git/worktrees/nop + test_path_is_dir .git/worktrees/nop ' test_expect_success 'prune duplicate (linked/linked)' ' @@ -103,8 +103,8 @@ test_expect_success 'prune duplicate (linked/linked)' ' mv .git/worktrees/w2/gitdir.new .git/worktrees/w2/gitdir && git worktree prune --verbose 2>actual && test_grep "duplicate entry" actual && - test -d .git/worktrees/w1 && - ! test -d .git/worktrees/w2 + test_path_is_dir .git/worktrees/w1 && + test_path_is_missing .git/worktrees/w2 ' test_expect_success 'prune duplicate (main/linked)' ' @@ -116,7 +116,7 @@ test_expect_success 'prune duplicate (main/linked)' ' mv repo wt && git -C wt worktree prune --verbose 2>actual && test_grep "duplicate entry" actual && - ! test -d .git/worktrees/wt + test_path_is_missing .git/worktrees/wt ' test_expect_success 'not prune proper worktrees inside linked worktree with relative paths' ' From d014fb2914e3c318b72cd1f4efb777677aa15805 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 16 Oct 2025 15:48:09 +0000 Subject: [PATCH 006/248] build(deps): bump actions/download-artifact from 4 to 5 Bumps [actions/download-artifact](https://github.com/actions/download-artifact) from 4 to 5. - [Release notes](https://github.com/actions/download-artifact/releases) - [Commits](https://github.com/actions/download-artifact/compare/v4...v5) Signed-off-by: Junio C Hamano --- .github/workflows/main.yml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index d122e79415a13e..ea192f6ff7a0a4 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -140,7 +140,7 @@ jobs: cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - name: download tracked files and build artifacts - uses: actions/download-artifact@v4 + uses: actions/download-artifact@v5 with: name: windows-artifacts path: ${{github.workspace}} @@ -226,7 +226,7 @@ jobs: steps: - uses: git-for-windows/setup-git-for-windows-sdk@v1 - name: download tracked files and build artifacts - uses: actions/download-artifact@v4 + uses: actions/download-artifact@v5 with: name: vs-artifacts path: ${{github.workspace}} @@ -292,7 +292,7 @@ jobs: shell: pwsh run: pip install meson ninja - name: Download build artifacts - uses: actions/download-artifact@v4 + uses: actions/download-artifact@v5 with: name: windows-meson-artifacts path: build From 63541ed9bc9213349f8957ecb362a0e8a97ad1f7 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 16 Oct 2025 15:48:10 +0000 Subject: [PATCH 007/248] build(deps): bump actions/checkout from 4 to 5 Bumps [actions/checkout](https://github.com/actions/checkout) from 4 to 5. - [Release notes](https://github.com/actions/checkout/releases) - [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md) - [Commits](https://github.com/actions/checkout/compare/v4...v5) Signed-off-by: Junio C Hamano --- .github/workflows/check-style.yml | 2 +- .github/workflows/check-whitespace.yml | 2 +- .github/workflows/coverity.yml | 2 +- .github/workflows/main.yml | 22 +++++++++++----------- 4 files changed, 14 insertions(+), 14 deletions(-) diff --git a/.github/workflows/check-style.yml b/.github/workflows/check-style.yml index c052a5df2381e9..19a145d4ad0c5a 100644 --- a/.github/workflows/check-style.yml +++ b/.github/workflows/check-style.yml @@ -20,7 +20,7 @@ jobs: jobname: ClangFormat runs-on: ubuntu-latest steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 with: fetch-depth: 0 diff --git a/.github/workflows/check-whitespace.yml b/.github/workflows/check-whitespace.yml index d0a78fc426f9e4..928fd4cfe2456d 100644 --- a/.github/workflows/check-whitespace.yml +++ b/.github/workflows/check-whitespace.yml @@ -19,7 +19,7 @@ jobs: check-whitespace: runs-on: ubuntu-latest steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 with: fetch-depth: 0 diff --git a/.github/workflows/coverity.yml b/.github/workflows/coverity.yml index 01a0437b2f2634..cfa17d394a7dbc 100644 --- a/.github/workflows/coverity.yml +++ b/.github/workflows/coverity.yml @@ -38,7 +38,7 @@ jobs: COVERITY_LANGUAGE: cxx COVERITY_PLATFORM: overridden-below steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - name: install minimal Git for Windows SDK if: contains(matrix.os, 'windows') uses: git-for-windows/setup-git-for-windows-sdk@v1 diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index ea192f6ff7a0a4..ea1d9591cf6a39 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -112,7 +112,7 @@ jobs: group: windows-build-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - uses: git-for-windows/setup-git-for-windows-sdk@v1 - name: build shell: bash @@ -173,10 +173,10 @@ jobs: group: vs-build-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - uses: git-for-windows/setup-git-for-windows-sdk@v1 - name: initialize vcpkg - uses: actions/checkout@v4 + uses: actions/checkout@v5 with: repository: 'microsoft/vcpkg' path: 'compat/vcbuild/vcpkg' @@ -258,7 +258,7 @@ jobs: group: windows-meson-build-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - uses: actions/setup-python@v5 - name: Set up dependencies shell: pwsh @@ -286,7 +286,7 @@ jobs: group: windows-meson-test-${{ matrix.nr }}-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - uses: actions/setup-python@v5 - name: Set up dependencies shell: pwsh @@ -331,7 +331,7 @@ jobs: TEST_OUTPUT_DIRECTORY: ${{github.workspace}}/t runs-on: ${{matrix.vector.pool}} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - run: ci/install-dependencies.sh - run: ci/run-build-and-tests.sh - name: print test failures @@ -352,7 +352,7 @@ jobs: CI_JOB_IMAGE: ubuntu-latest runs-on: ubuntu-latest steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - run: ci/install-dependencies.sh - run: ci/run-build-and-minimal-fuzzers.sh dockerized: @@ -429,7 +429,7 @@ jobs: else apt-get -q update && apt-get -q -y install git fi - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - run: ci/install-dependencies.sh - run: useradd builder --create-home - run: chown -R builder . @@ -454,7 +454,7 @@ jobs: group: static-analysis-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - run: ci/install-dependencies.sh - run: ci/run-static-analysis.sh - run: ci/check-directional-formatting.bash @@ -469,7 +469,7 @@ jobs: group: sparse-${{ github.ref }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - name: Install other dependencies run: ci/install-dependencies.sh - run: make sparse @@ -485,6 +485,6 @@ jobs: CI_JOB_IMAGE: ubuntu-latest runs-on: ubuntu-latest steps: - - uses: actions/checkout@v4 + - uses: actions/checkout@v5 - run: ci/install-dependencies.sh - run: ci/test-documentation.sh From b195b9526bc7fa642b9ff9dba56753acac165ea7 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 16 Oct 2025 15:48:11 +0000 Subject: [PATCH 008/248] build(deps): bump actions/setup-python from 5 to 6 Bumps [actions/setup-python](https://github.com/actions/setup-python) from 5 to 6. - [Release notes](https://github.com/actions/setup-python/releases) - [Commits](https://github.com/actions/setup-python/compare/v5...v6) Signed-off-by: Junio C Hamano --- .github/workflows/main.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index ea1d9591cf6a39..fa7c60f95fdca0 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -259,7 +259,7 @@ jobs: cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - uses: actions/checkout@v5 - - uses: actions/setup-python@v5 + - uses: actions/setup-python@v6 - name: Set up dependencies shell: pwsh run: pip install meson ninja @@ -287,7 +287,7 @@ jobs: cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - uses: actions/checkout@v5 - - uses: actions/setup-python@v5 + - uses: actions/setup-python@v6 - name: Set up dependencies shell: pwsh run: pip install meson ninja From 96978d7545a00dbffe99252a4749a7554ce124bb Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 16 Oct 2025 15:48:12 +0000 Subject: [PATCH 009/248] build(deps): bump actions/github-script from 7 to 8 Bumps [actions/github-script](https://github.com/actions/github-script) from 7 to 8. - [Release notes](https://github.com/actions/github-script/releases) - [Commits](https://github.com/actions/github-script/compare/v7...v8) Signed-off-by: Junio C Hamano --- .github/workflows/main.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index fa7c60f95fdca0..aa6bce673b4e99 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -63,7 +63,7 @@ jobs: echo "skip_concurrent=$skip_concurrent" >>$GITHUB_OUTPUT - name: skip if the commit or tree was already tested id: skip-if-redundant - uses: actions/github-script@v7 + uses: actions/github-script@v8 if: steps.check-ref.outputs.enabled == 'yes' with: github-token: ${{secrets.GITHUB_TOKEN}} From 623f7af28417805d941b1618157b9d93c02347af Mon Sep 17 00:00:00 2001 From: Jeff King Date: Fri, 17 Oct 2025 04:36:41 -0400 Subject: [PATCH 010/248] diff: restore redirection to /dev/null for diff_from_contents In --quiet mode, since we produce only an exit code for "something was changed" and no actual output, we can often get by with just a tree-level diff. However, certain options require us to actually look at the file contents (e.g., if we are ignoring whitespace changes). We have a flag "diff_from_contents" for that, and if it is set we call diff_flush() on each path. To avoid producing any output (since we were asked to be --quiet), we traditionally just redirected the output to /dev/null. That changed in b55e6d36eb (diff: ensure consistent diff behavior with ignore options, 2025-08-08), which replaced that with a "dry_run" flag. In theory, with dry_run set, we should produce no output. But it carries a risk of regression: if we forget to respect dry_run in any of the output paths, we'll accidentally produce output. And indeed, there is at least one such regression in that commit, as it covered only the case where we actually call into xdiff, and not creation or deletion diffs, where we manually generate the headers. We even test this case in t4035, but only with diff-tree, which does not show the bug by default because it does not require diff_from_contents. But git-diff does, because it allows external diff programs by default (so we must dig into each diff filepair to decide if it requires running an external diff that may declare two distinct blobs to actually be the same). We should fix all of those code paths to respect dry_run correctly, but in the meantime we can protect ourselves more fully by restoring the redirection to /dev/null. This gives us an extra layer of protection against regressions dues to other code paths we've missed. Though the original issue was reported with "git diff" (and due to its default of --ext-diff), I've used "diff-tree -w" in the new test. It triggers the same issue, but I think the fact that "-w" implies diff_from_contents is a bit more obvious, and fits in with the rest of t4035. Reported-by: Jake Zimmerman Signed-off-by: Jeff King Signed-off-by: Junio C Hamano --- diff.c | 9 +++++++++ t/t4035-diff-quiet.sh | 4 ++++ 2 files changed, 13 insertions(+) diff --git a/diff.c b/diff.c index cb04a9a6f26f5c..9b8d658b9ebb8e 100644 --- a/diff.c +++ b/diff.c @@ -6876,6 +6876,15 @@ void diff_flush(struct diff_options *options) if (output_format & DIFF_FORMAT_NO_OUTPUT && options->flags.exit_with_status && options->flags.diff_from_contents) { + /* + * run diff_flush_patch for the exit status. setting + * options->file to /dev/null should be safe, because we + * aren't supposed to produce any output anyway. + */ + diff_free_file(options); + options->file = xfopen("/dev/null", "w"); + options->close_file = 1; + options->color_moved = 0; for (i = 0; i < q->nr; i++) { struct diff_filepair *p = q->queue[i]; if (check_pair_status(p)) diff --git a/t/t4035-diff-quiet.sh b/t/t4035-diff-quiet.sh index 0352bf81a90a38..35eaf0855f86a5 100755 --- a/t/t4035-diff-quiet.sh +++ b/t/t4035-diff-quiet.sh @@ -50,6 +50,10 @@ test_expect_success 'git diff-tree HEAD HEAD' ' test_expect_code 0 git diff-tree --quiet HEAD HEAD >cnt && test_line_count = 0 cnt ' +test_expect_success 'git diff-tree -w HEAD^ HEAD' ' + test_expect_code 1 git diff-tree --quiet -w HEAD^ HEAD >cnt && + test_line_count = 0 cnt +' test_expect_success 'git diff-files' ' test_expect_code 0 git diff-files --quiet >cnt && test_line_count = 0 cnt From 50927f4f683a35ad1c76c8a02a1759a076d3f8f8 Mon Sep 17 00:00:00 2001 From: Jeff King Date: Fri, 17 Oct 2025 04:44:55 -0400 Subject: [PATCH 011/248] status: make coloring of "-z --short" consistent When running "git status -z --short", the marker on modified index entries (e.g., "M") is colorized, but the "??" marker for untracked entries is not. Let's fix the "??" entries to show color here. At first glance you might think that neither should be colorized, as usually one would use "-z" to get machine-readable output. But this is a tricky and unusual case. We have two output formats, "--short" and "--porcelain" which are substantially similar, but differ in that "--short" is for humans who want something short and "--porcelain" is for machines. And "-z" by itself, without any other output option, does default to "--porcelain", so "git status -z" will not colorize anything. But if you explicitly ask for "-z" and "--short" together, then that is asking for the human-readable output, but separated by NULs. This is unlikely to be useful directly, but could for example be used if the output will be shown to a human outside of the terminal. At any rate, the current behavior is clearly wrong (since we colorize some things but not others), and I think colorizing everything is the least-surprising thing we can do here. Reported-by: Langbart Signed-off-by: Jeff King Signed-off-by: Junio C Hamano --- t/t7508-status.sh | 11 +++++++++++ wt-status.c | 4 ++-- 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/t/t7508-status.sh b/t/t7508-status.sh index cdc1d6fcc78f49..abad229e9d9eda 100755 --- a/t/t7508-status.sh +++ b/t/t7508-status.sh @@ -717,6 +717,17 @@ test_expect_success TTY 'status -s with color.status' ' ' +test_expect_success TTY 'status -s keeps colors with -z' ' + test_when_finished "rm -f output.*" && + test_terminal git status -s -z >output.raw && + # convert back to newlines to avoid portability issues with + # test_decode_color and test_cmp, and to let us use the same expected + # output as earlier tests + tr "\0" "\n" output.nl && + test_decode_color output && + test_cmp expect output +' + cat >expect <<\EOF ## main...upstream [ahead 1, behind 2] M dir1/modified diff --git a/wt-status.c b/wt-status.c index 454601afa15a95..d6917f0a832d80 100644 --- a/wt-status.c +++ b/wt-status.c @@ -2051,13 +2051,13 @@ static void wt_shortstatus_status(struct string_list_item *it, static void wt_shortstatus_other(struct string_list_item *it, struct wt_status *s, const char *sign) { + color_fprintf(s->fp, color(WT_STATUS_UNTRACKED, s), "%s", sign); if (s->null_termination) { - fprintf(s->fp, "%s %s%c", sign, it->string, 0); + fprintf(s->fp, " %s%c", it->string, 0); } else { struct strbuf onebuf = STRBUF_INIT; const char *one; one = quote_path(it->string, s->prefix, &onebuf, QUOTE_PATH_QUOTE_SP); - color_fprintf(s->fp, color(WT_STATUS_UNTRACKED, s), "%s", sign); fprintf(s->fp, " %s\n", one); strbuf_release(&onebuf); } From 330a54099ecebf930ebe73797025b02d2dc9e506 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Torsten=20B=C3=B6gershausen?= Date: Tue, 21 Oct 2025 11:48:49 +0200 Subject: [PATCH 012/248] unicode: update the width tables to Unicode 17 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Unicode 17 is out. Update the unicode with table. https://blog.unicode.org/2025/09/unicode-170-release-announcement.html Signed-off-by: Torsten Bögershausen Signed-off-by: Junio C Hamano --- unicode-width.h | 33 +++++++++++++++++++++------------ 1 file changed, 21 insertions(+), 12 deletions(-) diff --git a/unicode-width.h b/unicode-width.h index 3ffee123a08366..b701129515ebc6 100644 --- a/unicode-width.h +++ b/unicode-width.h @@ -143,7 +143,8 @@ static const struct interval zero_width[] = { { 0x1A65, 0x1A6C }, { 0x1A73, 0x1A7C }, { 0x1A7F, 0x1A7F }, -{ 0x1AB0, 0x1ACE }, +{ 0x1AB0, 0x1ADD }, +{ 0x1AE0, 0x1AEB }, { 0x1B00, 0x1B03 }, { 0x1B34, 0x1B34 }, { 0x1B36, 0x1B3A }, @@ -229,7 +230,7 @@ static const struct interval zero_width[] = { { 0x10D24, 0x10D27 }, { 0x10D69, 0x10D6D }, { 0x10EAB, 0x10EAC }, -{ 0x10EFC, 0x10EFF }, +{ 0x10EFA, 0x10EFF }, { 0x10F46, 0x10F50 }, { 0x10F82, 0x10F85 }, { 0x11001, 0x11001 }, @@ -306,6 +307,9 @@ static const struct interval zero_width[] = { { 0x11A59, 0x11A5B }, { 0x11A8A, 0x11A96 }, { 0x11A98, 0x11A99 }, +{ 0x11B60, 0x11B60 }, +{ 0x11B62, 0x11B64 }, +{ 0x11B66, 0x11B66 }, { 0x11C30, 0x11C36 }, { 0x11C38, 0x11C3D }, { 0x11C3F, 0x11C3F }, @@ -362,6 +366,10 @@ static const struct interval zero_width[] = { { 0x1E2EC, 0x1E2EF }, { 0x1E4EC, 0x1E4EF }, { 0x1E5EE, 0x1E5EF }, +{ 0x1E6E3, 0x1E6E3 }, +{ 0x1E6E6, 0x1E6E6 }, +{ 0x1E6EE, 0x1E6EF }, +{ 0x1E6F5, 0x1E6F5 }, { 0x1E8D0, 0x1E8D6 }, { 0x1E944, 0x1E94A }, { 0xE0001, 0xE0001 }, @@ -429,10 +437,10 @@ static const struct interval double_width[] = { { 0xFF01, 0xFF60 }, { 0xFFE0, 0xFFE6 }, { 0x16FE0, 0x16FE4 }, -{ 0x16FF0, 0x16FF1 }, -{ 0x17000, 0x187F7 }, -{ 0x18800, 0x18CD5 }, -{ 0x18CFF, 0x18D08 }, +{ 0x16FF0, 0x16FF6 }, +{ 0x17000, 0x18CD5 }, +{ 0x18CFF, 0x18D1E }, +{ 0x18D80, 0x18DF2 }, { 0x1AFF0, 0x1AFF3 }, { 0x1AFF5, 0x1AFFB }, { 0x1AFFD, 0x1AFFE }, @@ -474,7 +482,7 @@ static const struct interval double_width[] = { { 0x1F680, 0x1F6C5 }, { 0x1F6CC, 0x1F6CC }, { 0x1F6D0, 0x1F6D2 }, -{ 0x1F6D5, 0x1F6D7 }, +{ 0x1F6D5, 0x1F6D8 }, { 0x1F6DC, 0x1F6DF }, { 0x1F6EB, 0x1F6EC }, { 0x1F6F4, 0x1F6FC }, @@ -484,11 +492,12 @@ static const struct interval double_width[] = { { 0x1F93C, 0x1F945 }, { 0x1F947, 0x1F9FF }, { 0x1FA70, 0x1FA7C }, -{ 0x1FA80, 0x1FA89 }, -{ 0x1FA8F, 0x1FAC6 }, -{ 0x1FACE, 0x1FADC }, -{ 0x1FADF, 0x1FAE9 }, -{ 0x1FAF0, 0x1FAF8 }, +{ 0x1FA80, 0x1FA8A }, +{ 0x1FA8E, 0x1FAC6 }, +{ 0x1FAC8, 0x1FAC8 }, +{ 0x1FACD, 0x1FADC }, +{ 0x1FADF, 0x1FAEA }, +{ 0x1FAEF, 0x1FAF8 }, { 0x20000, 0x2FFFD }, { 0x30000, 0x3FFFD } }; From 3da4413dbcdb8df021739ca6f20fe4f0bcd1fd3c Mon Sep 17 00:00:00 2001 From: Junio C Hamano Date: Wed, 22 Oct 2025 10:39:12 -0700 Subject: [PATCH 013/248] diff: make sure the other caller of diff_flush_patch_quietly() is silent Earlier, we added is a protection for the loop that computes "git diff --quiet -w" to ensure calls to the diff_flush_patch_quietly() helper stays quiet. Do the same for another loop that deals with options like "--name-status" to make calls to the same helper. Helped-by: Jeff King Signed-off-by: Junio C Hamano --- diff.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) diff --git a/diff.c b/diff.c index 9b8d658b9ebb8e..7b5601de2f3c80 100644 --- a/diff.c +++ b/diff.c @@ -6814,18 +6814,38 @@ void diff_flush(struct diff_options *options) DIFF_FORMAT_NAME | DIFF_FORMAT_NAME_STATUS | DIFF_FORMAT_CHECKDIFF)) { + /* + * make sure diff_Flush_patch_quietly() to be silent. + */ + FILE *dev_null = NULL; + int saved_color_moved = options->color_moved; + + if (options->flags.diff_from_contents) { + dev_null = xfopen("/dev/null", "w"); + options->color_moved = 0; + } for (i = 0; i < q->nr; i++) { struct diff_filepair *p = q->queue[i]; if (!check_pair_status(p)) continue; - if (options->flags.diff_from_contents && - !diff_flush_patch_quietly(p, options)) - continue; + if (options->flags.diff_from_contents) { + FILE *saved_file = options->file; + int found_changes; + options->file = dev_null; + found_changes = diff_flush_patch_quietly(p, options); + options->file = saved_file; + if (!found_changes) + continue; + } flush_one_pair(p, options); } + if (options->flags.diff_from_contents) { + fclose(dev_null); + options->color_moved = saved_color_moved; + } separator++; } From b7fb2194b96604898ff1ddd8dd1a394aa71532a9 Mon Sep 17 00:00:00 2001 From: Patrick Steinhardt Date: Thu, 23 Oct 2025 09:14:59 +0200 Subject: [PATCH 014/248] t7528: work around ETOOMANY in OpenSSH 10.1 and newer In t7528 we spawn an SSH agent to verify that we can sign a commit via it. This test has started to fail on some machines: +++ ssh-agent unix_listener_tmp: path "/home/pks/Development/git/build/test-output/trash directory.t7528-signed-commit-ssh/.ssh/agent/s.UTulegefEg.agent.UrPHumMXPq" too long for Unix domain socket main: Couldn't prepare agent socket As it turns out this is caused by a change in OpenSSH 10.1 [1]: * ssh-agent(1), sshd(8): move agent listener sockets from /tmp to under ~/.ssh/agent for both ssh-agent(1) and forwarded sockets in sshd(8). Instead of creating the socket in "/tmp", OpenSSH now creates the socket in our home directory. And as the home directory gets modified to be located in our test output directory we end up with paths that are somewhat long. But Linux has a rather short limit of 108 characters for socket paths, and other systems have even lower limits, so it is very easy now to exceed the limit and run into the above error. Work around the issue by using `ssh-agent -T`, which instructs it to use the old behaviour and create the socket in "/tmp" again. This switch has only been introduced with 10.1 though, so for older versions we have to fall back to not using it. That's fine though, as older versions know to put the socket into "/tmp" already. An alternative approach would be to abbreviate the socket name itself so that we create it as e.g. "sshsock" in the trash directory. But taking the above example we'd still end up with a path that is 91 characters long. So we wouldn't really have a lot of headroom, and it is quite likely that some developers would see the issue on their machines. [1]: https://www.openssh.com/txt/release-10.1 Reported-by: Xi Ruoyao Suggested-by: brian m. carlson Helped-by: Jeff King Helped-by: Lauri Tirkkonen Signed-off-by: Patrick Steinhardt Signed-off-by: Junio C Hamano --- t/t7528-signed-commit-ssh.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/t7528-signed-commit-ssh.sh b/t/t7528-signed-commit-ssh.sh index 0f887a3ebee19b..b50306b9b3952d 100755 --- a/t/t7528-signed-commit-ssh.sh +++ b/t/t7528-signed-commit-ssh.sh @@ -82,7 +82,7 @@ test_expect_success GPGSSH 'create signed commits' ' test_expect_success GPGSSH 'sign commits using literal public keys with ssh-agent' ' test_when_finished "test_unconfig commit.gpgsign" && test_config gpg.format ssh && - eval $(ssh-agent) && + eval $(ssh-agent -T || ssh-agent) && test_when_finished "kill ${SSH_AGENT_PID}" && test_when_finished "test_unconfig user.signingkey" && mkdir tmpdir && From bb5c624209fcaebd60b9572b2cc8c61086e39b57 Mon Sep 17 00:00:00 2001 From: Junio C Hamano Date: Sun, 26 Oct 2025 17:47:52 -0700 Subject: [PATCH 015/248] Git 2.51.2 Signed-off-by: Junio C Hamano --- Documentation/RelNotes/2.51.2.adoc | 45 ++++++++++++++++++++++++++++++ GIT-VERSION-GEN | 2 +- RelNotes | 2 +- 3 files changed, 47 insertions(+), 2 deletions(-) create mode 100644 Documentation/RelNotes/2.51.2.adoc diff --git a/Documentation/RelNotes/2.51.2.adoc b/Documentation/RelNotes/2.51.2.adoc new file mode 100644 index 00000000000000..f0be60333abb37 --- /dev/null +++ b/Documentation/RelNotes/2.51.2.adoc @@ -0,0 +1,45 @@ +Git 2.51.2 Release Notes +======================== + +In addition to fixes for an unfortunate regression introduced in Git +2.51.1 that caused "git diff --quiet -w" to be not so quiet when there +are additions, deletions and conflicts, this maintenance release merges +more fixes/improvements that have landed on the master front, primarily +to make the CI part of the system a bit more robust. + + +Fixes since Git 2.51.1 +---------------------- + + * Recently we attempted to improve "git diff -w --quiet" and friends + to handle cases where patch output would be suppressed, but it + introduced a bug that emits unnecessary output, which has been + corrected. + + * The code to squelch output from "git diff -w --name-status" + etc. for paths that "git diff -w -p" would have stayed silent + leaked output from dry-run patch generation, which has been + corrected. + + * Windows "real-time monitoring" interferes with the execution of + tests and affects negatively in both correctness and performance, + which has been disabled in Gitlab CI. + + * An earlier addition to "git diff --no-index A B" to limit the + output with pathspec after the two directories misbehaved when + these directories were given with a trailing slash, which has been + corrected. + + * The "--short" option of "git status" that meant output for humans + and "-z" option to show NUL delimited output format did not mix + well, and colored some but not all things. The command has been + updated to color all elements consistently in such a case. + + * Unicode width table update. + + * Recent OpenSSH creates the Unix domain socket to communicate with + ssh-agent under $HOME instead of /tmp, which causes our test to + fail doe to overly long pathname in our test environment, which has + been worked around by using "ssh-agent -T". + +Also contains various documentation updates, code cleanups and minor fixups. diff --git a/GIT-VERSION-GEN b/GIT-VERSION-GEN index 13d355766645fa..309bccb019a6bd 100755 --- a/GIT-VERSION-GEN +++ b/GIT-VERSION-GEN @@ -1,6 +1,6 @@ #!/bin/sh -DEF_VER=v2.51.1 +DEF_VER=v2.51.2 LF=' ' diff --git a/RelNotes b/RelNotes index 624baf8c5eb711..8e03a738f66bfc 120000 --- a/RelNotes +++ b/RelNotes @@ -1 +1 @@ -Documentation/RelNotes/2.51.1.adoc \ No newline at end of file +Documentation/RelNotes/2.51.2.adoc \ No newline at end of file From a4a7ff108ee47aafb161f4d7a06a7aee5137a5c9 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:05 +0000 Subject: [PATCH 016/248] sparse-checkout: remove use of the_repository The logic for the 'git sparse-checkout' builtin uses the_repository all over the place, despite some use of a repository struct in different method parameters. Complete this removal of the_repository by using 'repo' when possible. In one place, there was already a local variable 'r' that was set to the_repository, so move that to a method parameter. We cannot remove the USE_THE_REPOSITORY_VARIABLE declaration as we are still using global constants for the state of the sparse-checkout. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- builtin/sparse-checkout.c | 117 ++++++++++++++++++++------------------ 1 file changed, 62 insertions(+), 55 deletions(-) diff --git a/builtin/sparse-checkout.c b/builtin/sparse-checkout.c index 8c333b3e2e145b..06de61bd9d0d33 100644 --- a/builtin/sparse-checkout.c +++ b/builtin/sparse-checkout.c @@ -204,12 +204,12 @@ static void clean_tracked_sparse_directories(struct repository *r) ensure_full_index(r->index); } -static int update_working_directory(struct pattern_list *pl) +static int update_working_directory(struct repository *r, + struct pattern_list *pl) { enum update_sparsity_result result; struct unpack_trees_options o; struct lock_file lock_file = LOCK_INIT; - struct repository *r = the_repository; struct pattern_list *old_pl; /* If no branch has been checked out, there are no updates to make. */ @@ -327,7 +327,8 @@ static void write_cone_to_file(FILE *fp, struct pattern_list *pl) string_list_clear(&sl, 0); } -static int write_patterns_and_update(struct pattern_list *pl) +static int write_patterns_and_update(struct repository *repo, + struct pattern_list *pl) { char *sparse_filename; FILE *fp; @@ -336,15 +337,15 @@ static int write_patterns_and_update(struct pattern_list *pl) sparse_filename = get_sparse_checkout_filename(); - if (safe_create_leading_directories(the_repository, sparse_filename)) + if (safe_create_leading_directories(repo, sparse_filename)) die(_("failed to create directory for sparse-checkout file")); hold_lock_file_for_update(&lk, sparse_filename, LOCK_DIE_ON_ERROR); - result = update_working_directory(pl); + result = update_working_directory(repo, pl); if (result) { rollback_lock_file(&lk); - update_working_directory(NULL); + update_working_directory(repo, NULL); goto out; } @@ -372,25 +373,26 @@ enum sparse_checkout_mode { MODE_CONE_PATTERNS = 2, }; -static int set_config(enum sparse_checkout_mode mode) +static int set_config(struct repository *repo, + enum sparse_checkout_mode mode) { /* Update to use worktree config, if not already. */ - if (init_worktree_config(the_repository)) { + if (init_worktree_config(repo)) { error(_("failed to initialize worktree config")); return 1; } - if (repo_config_set_worktree_gently(the_repository, + if (repo_config_set_worktree_gently(repo, "core.sparseCheckout", mode ? "true" : "false") || - repo_config_set_worktree_gently(the_repository, + repo_config_set_worktree_gently(repo, "core.sparseCheckoutCone", mode == MODE_CONE_PATTERNS ? "true" : "false")) return 1; if (mode == MODE_NO_PATTERNS) - return set_sparse_index_config(the_repository, 0); + return set_sparse_index_config(repo, 0); return 0; } @@ -410,7 +412,7 @@ static enum sparse_checkout_mode update_cone_mode(int *cone_mode) { return MODE_ALL_PATTERNS; } -static int update_modes(int *cone_mode, int *sparse_index) +static int update_modes(struct repository *repo, int *cone_mode, int *sparse_index) { int mode, record_mode; @@ -418,20 +420,20 @@ static int update_modes(int *cone_mode, int *sparse_index) record_mode = (*cone_mode != -1) || !core_apply_sparse_checkout; mode = update_cone_mode(cone_mode); - if (record_mode && set_config(mode)) + if (record_mode && set_config(repo, mode)) return 1; /* Set sparse-index/non-sparse-index mode if specified */ if (*sparse_index >= 0) { - if (set_sparse_index_config(the_repository, *sparse_index) < 0) + if (set_sparse_index_config(repo, *sparse_index) < 0) die(_("failed to modify sparse-index config")); /* force an index rewrite */ - repo_read_index(the_repository); - the_repository->index->updated_workdir = 1; + repo_read_index(repo); + repo->index->updated_workdir = 1; if (!*sparse_index) - ensure_full_index(the_repository->index); + ensure_full_index(repo->index); } return 0; @@ -448,7 +450,7 @@ static struct sparse_checkout_init_opts { } init_opts; static int sparse_checkout_init(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { struct pattern_list pl; char *sparse_filename; @@ -464,7 +466,7 @@ static int sparse_checkout_init(int argc, const char **argv, const char *prefix, }; setup_work_tree(); - repo_read_index(the_repository); + repo_read_index(repo); init_opts.cone_mode = -1; init_opts.sparse_index = -1; @@ -473,7 +475,7 @@ static int sparse_checkout_init(int argc, const char **argv, const char *prefix, builtin_sparse_checkout_init_options, builtin_sparse_checkout_init_usage, 0); - if (update_modes(&init_opts.cone_mode, &init_opts.sparse_index)) + if (update_modes(repo, &init_opts.cone_mode, &init_opts.sparse_index)) return 1; memset(&pl, 0, sizeof(pl)); @@ -485,14 +487,14 @@ static int sparse_checkout_init(int argc, const char **argv, const char *prefix, if (res >= 0) { free(sparse_filename); clear_pattern_list(&pl); - return update_working_directory(NULL); + return update_working_directory(repo, NULL); } - if (repo_get_oid(the_repository, "HEAD", &oid)) { + if (repo_get_oid(repo, "HEAD", &oid)) { FILE *fp; /* assume we are in a fresh repo, but update the sparse-checkout file */ - if (safe_create_leading_directories(the_repository, sparse_filename)) + if (safe_create_leading_directories(repo, sparse_filename)) die(_("unable to create leading directories of %s"), sparse_filename); fp = xfopen(sparse_filename, "w"); @@ -511,7 +513,7 @@ static int sparse_checkout_init(int argc, const char **argv, const char *prefix, add_pattern("!/*/", empty_base, 0, &pl, 0); pl.use_cone_patterns = init_opts.cone_mode; - return write_patterns_and_update(&pl); + return write_patterns_and_update(repo, &pl); } static void insert_recursive_pattern(struct pattern_list *pl, struct strbuf *path) @@ -674,7 +676,8 @@ static void add_patterns_literal(int argc, const char **argv, add_patterns_from_input(pl, argc, argv, use_stdin ? stdin : NULL); } -static int modify_pattern_list(struct strvec *args, int use_stdin, +static int modify_pattern_list(struct repository *repo, + struct strvec *args, int use_stdin, enum modify_type m) { int result; @@ -696,22 +699,23 @@ static int modify_pattern_list(struct strvec *args, int use_stdin, } if (!core_apply_sparse_checkout) { - set_config(MODE_ALL_PATTERNS); + set_config(repo, MODE_ALL_PATTERNS); core_apply_sparse_checkout = 1; changed_config = 1; } - result = write_patterns_and_update(pl); + result = write_patterns_and_update(repo, pl); if (result && changed_config) - set_config(MODE_NO_PATTERNS); + set_config(repo, MODE_NO_PATTERNS); clear_pattern_list(pl); free(pl); return result; } -static void sanitize_paths(struct strvec *args, +static void sanitize_paths(struct repository *repo, + struct strvec *args, const char *prefix, int skip_checks) { int i; @@ -752,7 +756,7 @@ static void sanitize_paths(struct strvec *args, for (i = 0; i < args->nr; i++) { struct cache_entry *ce; - struct index_state *index = the_repository->index; + struct index_state *index = repo->index; int pos = index_name_pos(index, args->v[i], strlen(args->v[i])); if (pos < 0) @@ -779,7 +783,7 @@ static struct sparse_checkout_add_opts { } add_opts; static int sparse_checkout_add(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { static struct option builtin_sparse_checkout_add_options[] = { OPT_BOOL_F(0, "skip-checks", &add_opts.skip_checks, @@ -796,7 +800,7 @@ static int sparse_checkout_add(int argc, const char **argv, const char *prefix, if (!core_apply_sparse_checkout) die(_("no sparse-checkout to add to")); - repo_read_index(the_repository); + repo_read_index(repo); argc = parse_options(argc, argv, prefix, builtin_sparse_checkout_add_options, @@ -804,9 +808,9 @@ static int sparse_checkout_add(int argc, const char **argv, const char *prefix, for (int i = 0; i < argc; i++) strvec_push(&patterns, argv[i]); - sanitize_paths(&patterns, prefix, add_opts.skip_checks); + sanitize_paths(repo, &patterns, prefix, add_opts.skip_checks); - ret = modify_pattern_list(&patterns, add_opts.use_stdin, ADD); + ret = modify_pattern_list(repo, &patterns, add_opts.use_stdin, ADD); strvec_clear(&patterns); return ret; @@ -825,7 +829,7 @@ static struct sparse_checkout_set_opts { } set_opts; static int sparse_checkout_set(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { int default_patterns_nr = 2; const char *default_patterns[] = {"/*", "!/*/", NULL}; @@ -847,7 +851,7 @@ static int sparse_checkout_set(int argc, const char **argv, const char *prefix, int ret; setup_work_tree(); - repo_read_index(the_repository); + repo_read_index(repo); set_opts.cone_mode = -1; set_opts.sparse_index = -1; @@ -856,7 +860,7 @@ static int sparse_checkout_set(int argc, const char **argv, const char *prefix, builtin_sparse_checkout_set_options, builtin_sparse_checkout_set_usage, 0); - if (update_modes(&set_opts.cone_mode, &set_opts.sparse_index)) + if (update_modes(repo, &set_opts.cone_mode, &set_opts.sparse_index)) return 1; /* @@ -870,10 +874,10 @@ static int sparse_checkout_set(int argc, const char **argv, const char *prefix, } else { for (int i = 0; i < argc; i++) strvec_push(&patterns, argv[i]); - sanitize_paths(&patterns, prefix, set_opts.skip_checks); + sanitize_paths(repo, &patterns, prefix, set_opts.skip_checks); } - ret = modify_pattern_list(&patterns, set_opts.use_stdin, REPLACE); + ret = modify_pattern_list(repo, &patterns, set_opts.use_stdin, REPLACE); strvec_clear(&patterns); return ret; @@ -891,7 +895,7 @@ static struct sparse_checkout_reapply_opts { static int sparse_checkout_reapply(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { static struct option builtin_sparse_checkout_reapply_options[] = { OPT_BOOL(0, "cone", &reapply_opts.cone_mode, @@ -912,12 +916,12 @@ static int sparse_checkout_reapply(int argc, const char **argv, builtin_sparse_checkout_reapply_options, builtin_sparse_checkout_reapply_usage, 0); - repo_read_index(the_repository); + repo_read_index(repo); - if (update_modes(&reapply_opts.cone_mode, &reapply_opts.sparse_index)) + if (update_modes(repo, &reapply_opts.cone_mode, &reapply_opts.sparse_index)) return 1; - return update_working_directory(NULL); + return update_working_directory(repo, NULL); } static char const * const builtin_sparse_checkout_disable_usage[] = { @@ -927,7 +931,7 @@ static char const * const builtin_sparse_checkout_disable_usage[] = { static int sparse_checkout_disable(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { static struct option builtin_sparse_checkout_disable_options[] = { OPT_END(), @@ -955,7 +959,7 @@ static int sparse_checkout_disable(int argc, const char **argv, * are expecting to do that when disabling sparse-checkout. */ give_advice_on_expansion = 0; - repo_read_index(the_repository); + repo_read_index(repo); memset(&pl, 0, sizeof(pl)); hashmap_init(&pl.recursive_hashmap, pl_hashmap_cmp, NULL, 0); @@ -966,13 +970,13 @@ static int sparse_checkout_disable(int argc, const char **argv, add_pattern("/*", empty_base, 0, &pl, 0); prepare_repo_settings(the_repository); - the_repository->settings.sparse_index = 0; + repo->settings.sparse_index = 0; - if (update_working_directory(&pl)) + if (update_working_directory(repo, &pl)) die(_("error while refreshing working directory")); clear_pattern_list(&pl); - return set_config(MODE_NO_PATTERNS); + return set_config(repo, MODE_NO_PATTERNS); } static char const * const builtin_sparse_checkout_check_rules_usage[] = { @@ -987,14 +991,17 @@ static struct sparse_checkout_check_rules_opts { char *rules_file; } check_rules_opts; -static int check_rules(struct pattern_list *pl, int null_terminated) { +static int check_rules(struct repository *repo, + struct pattern_list *pl, + int null_terminated) +{ struct strbuf line = STRBUF_INIT; struct strbuf unquoted = STRBUF_INIT; char *path; int line_terminator = null_terminated ? 0 : '\n'; strbuf_getline_fn getline_fn = null_terminated ? strbuf_getline_nul : strbuf_getline; - the_repository->index->sparse_checkout_patterns = pl; + repo->index->sparse_checkout_patterns = pl; while (!getline_fn(&line, stdin)) { path = line.buf; if (!null_terminated && line.buf[0] == '"') { @@ -1006,7 +1013,7 @@ static int check_rules(struct pattern_list *pl, int null_terminated) { path = unquoted.buf; } - if (path_in_sparse_checkout(path, the_repository->index)) + if (path_in_sparse_checkout(path, repo->index)) write_name_quoted(path, stdout, line_terminator); } strbuf_release(&line); @@ -1016,7 +1023,7 @@ static int check_rules(struct pattern_list *pl, int null_terminated) { } static int sparse_checkout_check_rules(int argc, const char **argv, const char *prefix, - struct repository *repo UNUSED) + struct repository *repo) { static struct option builtin_sparse_checkout_check_rules_options[] = { OPT_BOOL('z', NULL, &check_rules_opts.null_termination, @@ -1055,7 +1062,7 @@ static int sparse_checkout_check_rules(int argc, const char **argv, const char * free(sparse_filename); } - ret = check_rules(&pl, check_rules_opts.null_termination); + ret = check_rules(repo, &pl, check_rules_opts.null_termination); clear_pattern_list(&pl); free(check_rules_opts.rules_file); return ret; @@ -1084,8 +1091,8 @@ int cmd_sparse_checkout(int argc, repo_config(the_repository, git_default_config, NULL); - prepare_repo_settings(the_repository); - the_repository->settings.command_requires_full_index = 0; + prepare_repo_settings(repo); + repo->settings.command_requires_full_index = 0; return fn(argc, argv, prefix, repo); } From 71948f02e4858c8b0d7472ada1e73db67b9cd578 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:06 +0000 Subject: [PATCH 017/248] sparse-checkout: add basics of 'clean' command When users change their sparse-checkout definitions to add new directories and remove old ones, there may be a few reasons why directories no longer in scope remain (ignored or excluded files still exist, Windows handles are still open, etc.). When these files still exist, the sparse index feature notices that a tracked, but sparse, directory still exists on disk and thus the index expands. This causes a performance hit _and_ the advice printed isn't very helpful. Using 'git clean' isn't enough (generally '-dfx' may be needed) but also this may not be sufficient. Add a new subcommand to 'git sparse-checkout' that removes these tracked-but-sparse directories. The implementation details provide a clear definition of what is happening, but it is difficult to describe this without including the internal implementation details. The core operation converts the index to a sparse index (in memory if not already on disk) and then deletes any directories in the worktree that correspond with a sparse directory entry in that sparse index. In the most common case, this means that a file will be removed if it is contained within a directory that is both tracked and outside of the sparse-checkout definition. However, there can be exceptions depending on the current state of the index: * If the worktree has a modification to a tracked, sparse file, then that file's parent directories will be expanded instead of represented as sparse directories. Siblings of those parent directories may be considered sparse. * If the user staged a sparse file with "git add --sparse", then that file loses the SKIP_WORKTREE bit until the sparse-checkout is reapplied. Until then, that file's parent directories are not represented as sparse directory entries and thus will not be removed. Siblings of those parent directories may be considered sparse. (There may be other reasons why the SKIP_WORKTREE bit was removed for a file and this impact on the sparse directories will apply to those as well.) * If the user has a merge conflict outside of the sparse-checkout definition, then those conflict entries prevent the parent directories from being represented as sparse directory entries and thus are not removed. * The cases above present reasons why certain _file conditions_ will impact which _directories_ are considered sparse. The list of tracked directories that are outside of the sparse-checkout definition but not represented as a sparse directory further reduces the list of files that will be removed. For these complicated reasons, the documentation details a potential list of files that will be "considered for removal" instead of defining the list concretely. The special cases can be handled by resolving conflicts, committing staged changes, and running 'git sparse-checkout reapply' to update the SKIP_WORKTREE bits as expected by the sparse-checkout definition. It is important to make clear that this operation will remove ignored and excluded files which would normally be ignored even by 'git clean -f' unless the '-x' or '-X' option is provided. This is the most extreme method for doing this, but it works when the sparse-checkout is in cone mode and is expected to rescope based on directories, not files. The current implementation always deletes these sparse directories without warning. This is unacceptable for a released version, but those features will be added in changes coming immediately after this one. Note that this will not remove an untracked directory (or any of its contents) if its parent is a tracked directory within the sparse-checkout definition. This is required to prevent removing data created by tools that perform caching operations for editors or build tools. Thus, 'git sparse-checkout clean' is both more aggressive and more careful than 'git clean -fx': * It is more aggressive because it will remove _tracked_ files within the sparse directories. * It is less aggressive because it will leave _untracked_ files that are not contained in sparse directories. These special cases will be handled more explicitly in a future change that expands tests for the 'git sparse-checkout clean' command. We handle some of the modified, staged, and committed states including some impact on 'git status' after cleaning. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- Documentation/git-sparse-checkout.adoc | 19 ++++- builtin/sparse-checkout.c | 64 ++++++++++++++- t/t1091-sparse-checkout-builtin.sh | 103 +++++++++++++++++++++++++ 3 files changed, 184 insertions(+), 2 deletions(-) diff --git a/Documentation/git-sparse-checkout.adoc b/Documentation/git-sparse-checkout.adoc index b5fe5da041676c..a765424e0776d7 100644 --- a/Documentation/git-sparse-checkout.adoc +++ b/Documentation/git-sparse-checkout.adoc @@ -9,7 +9,7 @@ git-sparse-checkout - Reduce your working tree to a subset of tracked files SYNOPSIS -------- [verse] -'git sparse-checkout' (init | list | set | add | reapply | disable | check-rules) [] +'git sparse-checkout' (init | list | set | add | reapply | disable | check-rules | clean) [] DESCRIPTION @@ -111,6 +111,23 @@ flags, with the same meaning as the flags from the `set` command, in order to change which sparsity mode you are using without needing to also respecify all sparsity paths. +'clean':: + Opportunistically remove files outside of the sparse-checkout + definition. This command requires cone mode to use recursive + directory matches to determine which files should be removed. A + file is considered for removal if it is contained within a tracked + directory that is outside of the sparse-checkout definition. ++ +Some special cases, such as merge conflicts or modified files outside of +the sparse-checkout definition could lead to keeping files that would +otherwise be removed. Resolve conflicts, stage modifications, and use +`git sparse-checkout reapply` in conjunction with `git sparse-checkout +clean` to resolve these cases. ++ +This command can be used to be sure the sparse index works efficiently, +though it does not require enabling the sparse index feature via the +`index.sparse=true` configuration. + 'disable':: Disable the `core.sparseCheckout` config setting, and restore the working directory to include all files. diff --git a/builtin/sparse-checkout.c b/builtin/sparse-checkout.c index 06de61bd9d0d33..f7caa28f3f00aa 100644 --- a/builtin/sparse-checkout.c +++ b/builtin/sparse-checkout.c @@ -2,6 +2,7 @@ #define DISABLE_SIGN_COMPARE_WARNINGS #include "builtin.h" +#include "abspath.h" #include "config.h" #include "dir.h" #include "environment.h" @@ -23,7 +24,7 @@ static const char *empty_base = ""; static char const * const builtin_sparse_checkout_usage[] = { - N_("git sparse-checkout (init | list | set | add | reapply | disable | check-rules) []"), + N_("git sparse-checkout (init | list | set | add | reapply | disable | check-rules | clean) []"), NULL }; @@ -924,6 +925,66 @@ static int sparse_checkout_reapply(int argc, const char **argv, return update_working_directory(repo, NULL); } +static char const * const builtin_sparse_checkout_clean_usage[] = { + "git sparse-checkout clean [-n|--dry-run]", + NULL +}; + +static const char *msg_remove = N_("Removing %s\n"); + +static int sparse_checkout_clean(int argc, const char **argv, + const char *prefix, + struct repository *repo) +{ + struct strbuf full_path = STRBUF_INIT; + const char *msg = msg_remove; + size_t worktree_len; + + struct option builtin_sparse_checkout_clean_options[] = { + OPT_END(), + }; + + setup_work_tree(); + if (!core_apply_sparse_checkout) + die(_("must be in a sparse-checkout to clean directories")); + if (!core_sparse_checkout_cone) + die(_("must be in a cone-mode sparse-checkout to clean directories")); + + argc = parse_options(argc, argv, prefix, + builtin_sparse_checkout_clean_options, + builtin_sparse_checkout_clean_usage, 0); + + if (repo_read_index(repo) < 0) + die(_("failed to read index")); + + if (convert_to_sparse(repo->index, SPARSE_INDEX_MEMORY_ONLY) || + repo->index->sparse_index == INDEX_EXPANDED) + die(_("failed to convert index to a sparse index; resolve merge conflicts and try again")); + + strbuf_addstr(&full_path, repo->worktree); + strbuf_addch(&full_path, '/'); + worktree_len = full_path.len; + + for (size_t i = 0; i < repo->index->cache_nr; i++) { + struct cache_entry *ce = repo->index->cache[i]; + if (!S_ISSPARSEDIR(ce->ce_mode)) + continue; + strbuf_setlen(&full_path, worktree_len); + strbuf_add(&full_path, ce->name, ce->ce_namelen); + + if (!is_directory(full_path.buf)) + continue; + + printf(msg, ce->name); + + if (remove_dir_recursively(&full_path, 0)) + warning_errno(_("failed to remove '%s'"), ce->name); + } + + strbuf_release(&full_path); + return 0; +} + static char const * const builtin_sparse_checkout_disable_usage[] = { "git sparse-checkout disable", NULL @@ -1080,6 +1141,7 @@ int cmd_sparse_checkout(int argc, OPT_SUBCOMMAND("set", &fn, sparse_checkout_set), OPT_SUBCOMMAND("add", &fn, sparse_checkout_add), OPT_SUBCOMMAND("reapply", &fn, sparse_checkout_reapply), + OPT_SUBCOMMAND("clean", &fn, sparse_checkout_clean), OPT_SUBCOMMAND("disable", &fn, sparse_checkout_disable), OPT_SUBCOMMAND("check-rules", &fn, sparse_checkout_check_rules), OPT_END(), diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh index ab3a105ffff253..bdb7b21e327b2a 100755 --- a/t/t1091-sparse-checkout-builtin.sh +++ b/t/t1091-sparse-checkout-builtin.sh @@ -1050,5 +1050,108 @@ test_expect_success 'check-rules null termination' ' test_cmp expect actual ' +test_expect_success 'clean' ' + git -C repo sparse-checkout set --cone deep/deeper1 && + git -C repo sparse-checkout reapply && + mkdir repo/deep/deeper2 repo/folder1 && + + # Add untracked files + touch repo/deep/deeper2/file && + touch repo/folder1/file && + + cat >expect <<-\EOF && + Removing deep/deeper2/ + Removing folder1/ + EOF + + git -C repo sparse-checkout clean >out && + test_cmp expect out && + + test_path_is_missing repo/deep/deeper2 && + test_path_is_missing repo/folder1 +' + +test_expect_success 'clean with sparse file states' ' + test_when_finished git reset --hard && + git -C repo sparse-checkout set --cone deep/deeper1 && + mkdir repo/folder2 && + + # create an untracked file and a modified file + touch repo/folder2/file && + echo dirty >repo/folder2/a && + + # First clean/reapply pass will do nothing. + git -C repo sparse-checkout clean >out && + test_must_be_empty out && + test_path_exists repo/folder2/a && + test_path_exists repo/folder2/file && + + git -C repo sparse-checkout reapply 2>err && + test_grep folder2 err && + test_path_exists repo/folder2/a && + test_path_exists repo/folder2/file && + + # Now, stage the change to the tracked file. + git -C repo add --sparse folder2/a && + + # Clean will continue not doing anything. + git -C repo sparse-checkout clean >out && + test_line_count = 0 out && + test_path_exists repo/folder2/a && + test_path_exists repo/folder2/file && + + # But we can reapply to remove the staged change. + git -C repo sparse-checkout reapply 2>err && + test_grep folder2 err && + test_path_is_missing repo/folder2/a && + test_path_exists repo/folder2/file && + + # We can clean now. + cat >expect <<-\EOF && + Removing folder2/ + EOF + git -C repo sparse-checkout clean >out && + test_cmp expect out && + test_path_is_missing repo/folder2 && + + # At the moment, the file is staged. + cat >expect <<-\EOF && + M folder2/a + EOF + + git -C repo status -s >out && + test_cmp expect out && + + # Reapply persists the modified state. + git -C repo sparse-checkout reapply && + cat >expect <<-\EOF && + M folder2/a + EOF + git -C repo status -s >out && + test_cmp expect out && + + # Committing the change leads to resolved status. + git -C repo commit -m "modified" && + git -C repo status -s >out && + test_must_be_empty out && + + # Repeat, but this time commit before reapplying. + mkdir repo/folder2/ && + echo dirtier >repo/folder2/a && + git -C repo add --sparse folder2/a && + git -C repo sparse-checkout clean >out && + test_must_be_empty out && + test_path_exists repo/folder2/a && + + # Committing without reapplying makes it look like a deletion + # due to no skip-worktree bit. + git -C repo commit -m "dirtier" && + git -C repo status -s >out && + test_must_be_empty out && + + git -C repo sparse-checkout reapply && + git -C repo status -s >out && + test_must_be_empty out +' test_done From 17e6e2c31a4bb60e16e872876587ae4bf5cb3f3e Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:07 +0000 Subject: [PATCH 018/248] sparse-checkout: match some 'clean' behavior The 'git sparse-checkout clean' subcommand is somewhat similar to 'git clean' in that it will delete files that should not be in the worktree. The big difference is that it focuses on the directories that should not be in the worktree due to cone-mode sparse-checkout. It also does not discriminate in the kinds of files and focuses on deleting entire directories. However, there are some restrictions that would be good to bring over from 'git clean', specifically how it refuses to do anything without the '-f'/'--force' or '-n'/'--dry-run' arguments. The 'clean.requireForce' config can be set to 'false' to imply '--force'. Add this behavior to avoid accidental deletion of files that cannot be recovered from Git. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- Documentation/git-sparse-checkout.adoc | 9 +++++ builtin/sparse-checkout.c | 15 ++++++- t/t1091-sparse-checkout-builtin.sh | 54 +++++++++++++++++++++++++- 3 files changed, 76 insertions(+), 2 deletions(-) diff --git a/Documentation/git-sparse-checkout.adoc b/Documentation/git-sparse-checkout.adoc index a765424e0776d7..d0fcc27ddb2349 100644 --- a/Documentation/git-sparse-checkout.adoc +++ b/Documentation/git-sparse-checkout.adoc @@ -127,6 +127,15 @@ clean` to resolve these cases. This command can be used to be sure the sparse index works efficiently, though it does not require enabling the sparse index feature via the `index.sparse=true` configuration. ++ +To prevent accidental deletion of worktree files, the `clean` subcommand +will not delete any files without the `-f` or `--force` option, unless +the `clean.requireForce` config option is set to `false`. ++ +The `--dry-run` option will list the directories that would be removed +without deleting them. Running in this mode can be helpful to predict the +behavior of the clean comand or to determine which kinds of files are left +in the sparse directories. 'disable':: Disable the `core.sparseCheckout` config setting, and restore the diff --git a/builtin/sparse-checkout.c b/builtin/sparse-checkout.c index f7caa28f3f00aa..d777b64960668d 100644 --- a/builtin/sparse-checkout.c +++ b/builtin/sparse-checkout.c @@ -931,6 +931,7 @@ static char const * const builtin_sparse_checkout_clean_usage[] = { }; static const char *msg_remove = N_("Removing %s\n"); +static const char *msg_would_remove = N_("Would remove %s\n"); static int sparse_checkout_clean(int argc, const char **argv, const char *prefix, @@ -939,8 +940,12 @@ static int sparse_checkout_clean(int argc, const char **argv, struct strbuf full_path = STRBUF_INIT; const char *msg = msg_remove; size_t worktree_len; + int force = 0, dry_run = 0; + int require_force = 1; struct option builtin_sparse_checkout_clean_options[] = { + OPT__DRY_RUN(&dry_run, N_("dry run")), + OPT__FORCE(&force, N_("force"), PARSE_OPT_NOCOMPLETE), OPT_END(), }; @@ -954,6 +959,13 @@ static int sparse_checkout_clean(int argc, const char **argv, builtin_sparse_checkout_clean_options, builtin_sparse_checkout_clean_usage, 0); + repo_config_get_bool(repo, "clean.requireforce", &require_force); + if (require_force && !force && !dry_run) + die(_("for safety, refusing to clean without one of --force or --dry-run")); + + if (dry_run) + msg = msg_would_remove; + if (repo_read_index(repo) < 0) die(_("failed to read index")); @@ -977,7 +989,8 @@ static int sparse_checkout_clean(int argc, const char **argv, printf(msg, ce->name); - if (remove_dir_recursively(&full_path, 0)) + if (dry_run <= 0 && + remove_dir_recursively(&full_path, 0)) warning_errno(_("failed to remove '%s'"), ce->name); } diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh index bdb7b21e327b2a..e6b768a8da959a 100755 --- a/t/t1091-sparse-checkout-builtin.sh +++ b/t/t1091-sparse-checkout-builtin.sh @@ -1059,12 +1059,29 @@ test_expect_success 'clean' ' touch repo/deep/deeper2/file && touch repo/folder1/file && + test_must_fail git -C repo sparse-checkout clean 2>err && + grep "refusing to clean" err && + + git -C repo config clean.requireForce true && + test_must_fail git -C repo sparse-checkout clean 2>err && + grep "refusing to clean" err && + + cat >expect <<-\EOF && + Would remove deep/deeper2/ + Would remove folder1/ + EOF + + git -C repo sparse-checkout clean --dry-run >out && + test_cmp expect out && + test_path_exists repo/deep/deeper2 && + test_path_exists repo/folder1 && + cat >expect <<-\EOF && Removing deep/deeper2/ Removing folder1/ EOF - git -C repo sparse-checkout clean >out && + git -C repo sparse-checkout clean -f >out && test_cmp expect out && test_path_is_missing repo/deep/deeper2 && @@ -1076,6 +1093,10 @@ test_expect_success 'clean with sparse file states' ' git -C repo sparse-checkout set --cone deep/deeper1 && mkdir repo/folder2 && + # The previous test case checked the -f option, so + # test the config option in this one. + git -C repo config clean.requireForce false && + # create an untracked file and a modified file touch repo/folder2/file && echo dirty >repo/folder2/a && @@ -1154,4 +1175,35 @@ test_expect_success 'clean with sparse file states' ' test_must_be_empty out ' +test_expect_success 'clean with merge conflict status' ' + git clone repo clean-merge && + + echo dirty >clean-merge/deep/deeper2/a && + touch clean-merge/folder2/extra && + + cat >input <<-EOF && + 0 $ZERO_OID folder1/a + 100644 $(git -C clean-merge rev-parse HEAD:folder1/a) 1 folder1/a + EOF + git -C clean-merge update-index --index-info err && + grep "failed to convert index to a sparse index" err && + + mkdir -p clean-merge/folder1/ && + echo merged >clean-merge/folder1/a && + git -C clean-merge add --sparse folder1/a && + + # deletes folder2/ but leaves staged change in folder1 + # and dirty change in deep/deeper2/ + cat >expect <<-\EOF && + Removing folder2/ + EOF + + git -C clean-merge sparse-checkout clean -f >out && + test_cmp expect out +' + test_done From af8b38c09857e409098e63f659c8f1dfe9e5ef55 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:08 +0000 Subject: [PATCH 019/248] dir: add generic "walk all files" helper There is sometimes a need to visit every file within a directory, recursively. The main example is remove_dir_recursively(), though it has some extra flags that make it want to iterate over paths in a custom way. There is also the fill_directory() approach but that involves an index and a pathspec. This change adds a new for_each_file_in_dir() method that will be helpful in the next change. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- dir.c | 28 ++++++++++++++++++++++++++++ dir.h | 14 ++++++++++++++ 2 files changed, 42 insertions(+) diff --git a/dir.c b/dir.c index dfb4d40103fb4a..7731f7b9d305b7 100644 --- a/dir.c +++ b/dir.c @@ -30,6 +30,7 @@ #include "read-cache-ll.h" #include "setup.h" #include "sparse-index.h" +#include "strbuf.h" #include "submodule-config.h" #include "symlinks.h" #include "trace2.h" @@ -87,6 +88,33 @@ struct dirent *readdir_skip_dot_and_dotdot(DIR *dirp) return e; } +int for_each_file_in_dir(struct strbuf *path, file_iterator fn, const void *data) +{ + struct dirent *e; + int res = 0; + size_t baselen = path->len; + DIR *dir = opendir(path->buf); + + if (!dir) + return 0; + + while (!res && (e = readdir_skip_dot_and_dotdot(dir)) != NULL) { + unsigned char dtype = get_dtype(e, path, 0); + strbuf_setlen(path, baselen); + strbuf_addstr(path, e->d_name); + + if (dtype == DT_REG) { + res = fn(path->buf, data); + } else if (dtype == DT_DIR) { + strbuf_addch(path, '/'); + res = for_each_file_in_dir(path, fn, data); + } + } + + closedir(dir); + return res; +} + int count_slashes(const char *s) { int cnt = 0; diff --git a/dir.h b/dir.h index fc9be7b427a134..20d4a078d61ef8 100644 --- a/dir.h +++ b/dir.h @@ -536,6 +536,20 @@ int get_sparse_checkout_patterns(struct pattern_list *pl); */ int remove_dir_recursively(struct strbuf *path, int flag); +/* + * This function pointer type is called on each file discovered in + * for_each_file_in_dir. The iteration stops if this method returns + * non-zero. + */ +typedef int (*file_iterator)(const char *path, const void *data); + +struct strbuf; +/* + * Given a directory path, recursively visit each file within, including + * within subdirectories. + */ +int for_each_file_in_dir(struct strbuf *path, file_iterator fn, const void *data); + /* * Tries to remove the path, along with leading empty directories so long as * those empty directories are not startup_info->original_cwd. Ignores From 6394cadc24d93be90febf24d41c885cc3d32bd96 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:09 +0000 Subject: [PATCH 020/248] sparse-checkout: add --verbose option to 'clean' The 'git sparse-checkout clean' subcommand is focused on directories, deleting any tracked sparse directories to clean up the worktree and make the sparse index feature work optimally. However, this directory-focused approach can leave users wondering why those directories exist at all. In my experience, these files are left over due to ignore or exclude patterns, Windows file handles, or possibly merge conflict resolutions. Add a new '--verbose' option for users to see all the files that are being deleted (with '--force') or would be deleted (with '--dry-run'). Based on usage, users may request further context on this list of files for states such as tracked/untracked, unstaged/staged/conflicted, etc. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- Documentation/git-sparse-checkout.adoc | 5 +++++ builtin/sparse-checkout.c | 26 ++++++++++++++++++++++++-- t/t1091-sparse-checkout-builtin.sh | 14 +++++++++++--- 3 files changed, 40 insertions(+), 5 deletions(-) diff --git a/Documentation/git-sparse-checkout.adoc b/Documentation/git-sparse-checkout.adoc index d0fcc27ddb2349..0d1618f161ed63 100644 --- a/Documentation/git-sparse-checkout.adoc +++ b/Documentation/git-sparse-checkout.adoc @@ -136,6 +136,11 @@ The `--dry-run` option will list the directories that would be removed without deleting them. Running in this mode can be helpful to predict the behavior of the clean comand or to determine which kinds of files are left in the sparse directories. ++ +The `--verbose` option will list every file within the directories that +are considered for removal. This option is helpful to determine if those +files are actually important or perhaps to explain why the directory is +still present despite the current sparse-checkout. 'disable':: Disable the `core.sparseCheckout` config setting, and restore the diff --git a/builtin/sparse-checkout.c b/builtin/sparse-checkout.c index d777b64960668d..15d51e60a86533 100644 --- a/builtin/sparse-checkout.c +++ b/builtin/sparse-checkout.c @@ -930,6 +930,24 @@ static char const * const builtin_sparse_checkout_clean_usage[] = { NULL }; +static int list_file_iterator(const char *path, const void *data) +{ + const char *msg = data; + + printf(msg, path); + return 0; +} + +static void list_every_file_in_dir(const char *msg, + const char *directory) +{ + struct strbuf path = STRBUF_INIT; + + strbuf_addstr(&path, directory); + for_each_file_in_dir(&path, list_file_iterator, msg); + strbuf_release(&path); +} + static const char *msg_remove = N_("Removing %s\n"); static const char *msg_would_remove = N_("Would remove %s\n"); @@ -940,12 +958,13 @@ static int sparse_checkout_clean(int argc, const char **argv, struct strbuf full_path = STRBUF_INIT; const char *msg = msg_remove; size_t worktree_len; - int force = 0, dry_run = 0; + int force = 0, dry_run = 0, verbose = 0; int require_force = 1; struct option builtin_sparse_checkout_clean_options[] = { OPT__DRY_RUN(&dry_run, N_("dry run")), OPT__FORCE(&force, N_("force"), PARSE_OPT_NOCOMPLETE), + OPT__VERBOSE(&verbose, N_("report each affected file, not just directories")), OPT_END(), }; @@ -987,7 +1006,10 @@ static int sparse_checkout_clean(int argc, const char **argv, if (!is_directory(full_path.buf)) continue; - printf(msg, ce->name); + if (verbose) + list_every_file_in_dir(msg, ce->name); + else + printf(msg, ce->name); if (dry_run <= 0 && remove_dir_recursively(&full_path, 0)) diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh index e6b768a8da959a..7b15fa669c4662 100755 --- a/t/t1091-sparse-checkout-builtin.sh +++ b/t/t1091-sparse-checkout-builtin.sh @@ -1053,11 +1053,11 @@ test_expect_success 'check-rules null termination' ' test_expect_success 'clean' ' git -C repo sparse-checkout set --cone deep/deeper1 && git -C repo sparse-checkout reapply && - mkdir repo/deep/deeper2 repo/folder1 && + mkdir -p repo/deep/deeper2 repo/folder1/extra/inside && # Add untracked files touch repo/deep/deeper2/file && - touch repo/folder1/file && + touch repo/folder1/extra/inside/file && test_must_fail git -C repo sparse-checkout clean 2>err && grep "refusing to clean" err && @@ -1074,7 +1074,15 @@ test_expect_success 'clean' ' git -C repo sparse-checkout clean --dry-run >out && test_cmp expect out && test_path_exists repo/deep/deeper2 && - test_path_exists repo/folder1 && + test_path_exists repo/folder1/extra/inside/file && + + cat >expect <<-\EOF && + Would remove deep/deeper2/file + Would remove folder1/extra/inside/file + EOF + + git -C repo sparse-checkout clean --dry-run --verbose >out && + test_cmp expect out && cat >expect <<-\EOF && Removing deep/deeper2/ From 76a3806582e3d7ba43967a63d1f0bac46a379010 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 6 Nov 2024 20:34:50 +0100 Subject: [PATCH 021/248] sideband: mask control characters The output of `git clone` is a vital component for understanding what has happened when things go wrong. However, these logs are partially under the control of the remote server (via the "sideband", which typically contains what the remote `git pack-objects` process sends to `stderr`), and is currently not sanitized by Git. This makes Git susceptible to ANSI escape sequence injection (see CWE-150, https://cwe.mitre.org/data/definitions/150.html), which allows attackers to corrupt terminal state, to hide information, and even to insert characters into the input buffer (i.e. as if the user had typed those characters). To plug this vulnerability, disallow any control character in the sideband, replacing them instead with the common `^` (e.g. `^[` for `\x1b`, `^A` for `\x01`). There is likely a need for more fine-grained controls instead of using a "heavy hammer" like this, which will be introduced subsequently. Signed-off-by: Johannes Schindelin --- sideband.c | 17 +++++++++++++++-- t/t5409-colorize-remote-messages.sh | 12 ++++++++++++ 2 files changed, 27 insertions(+), 2 deletions(-) diff --git a/sideband.c b/sideband.c index 8f15b98a65444f..5425c91722183b 100644 --- a/sideband.c +++ b/sideband.c @@ -66,6 +66,19 @@ void list_config_color_sideband_slots(struct string_list *list, const char *pref list_config_item(list, prefix, keywords[i].keyword); } +static void strbuf_add_sanitized(struct strbuf *dest, const char *src, int n) +{ + strbuf_grow(dest, n); + for (; n && *src; src++, n--) { + if (!iscntrl(*src) || *src == '\t' || *src == '\n') + strbuf_addch(dest, *src); + else { + strbuf_addch(dest, '^'); + strbuf_addch(dest, 0x40 + *src); + } + } +} + /* * Optionally highlight one keyword in remote output if it appears at the start * of the line. This should be called for a single line only, which is @@ -81,7 +94,7 @@ static void maybe_colorize_sideband(struct strbuf *dest, const char *src, int n) int i; if (!want_color_stderr(use_sideband_colors())) { - strbuf_add(dest, src, n); + strbuf_add_sanitized(dest, src, n); return; } @@ -114,7 +127,7 @@ static void maybe_colorize_sideband(struct strbuf *dest, const char *src, int n) } } - strbuf_add(dest, src, n); + strbuf_add_sanitized(dest, src, n); } diff --git a/t/t5409-colorize-remote-messages.sh b/t/t5409-colorize-remote-messages.sh index fa5de4500a4f50..d0745c391b2625 100755 --- a/t/t5409-colorize-remote-messages.sh +++ b/t/t5409-colorize-remote-messages.sh @@ -98,4 +98,16 @@ test_expect_success 'fallback to color.ui' ' grep "error: error" decoded ' +test_expect_success 'disallow (color) control sequences in sideband' ' + write_script .git/color-me-surprised <<-\EOF && + printf "error: Have you \\033[31mread\\033[m this?\\n" >&2 + exec "$@" + EOF + test_config_global uploadPack.packObjectshook ./color-me-surprised && + test_commit need-at-least-one-commit && + git clone --no-local . throw-away 2>stderr && + test_decode_color decoded && + test_grep ! RED decoded +' + test_done From d8881bad5813652fff5a557d68604766fd4f0a77 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:10 +0000 Subject: [PATCH 022/248] sparse-index: point users to new 'clean' action In my experience, the most-common reason that the sparse index must expand to a full one is because there is some leftover file in a tracked directory that is now outside of the sparse-checkout. The new 'git sparse-checkout clean' command will find and delete these directories, so point users to it when they hit the sparse index expansion advice. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- sparse-index.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sparse-index.c b/sparse-index.c index 5634abafaa07ed..5d14795063b578 100644 --- a/sparse-index.c +++ b/sparse-index.c @@ -32,7 +32,8 @@ int give_advice_on_expansion = 1; "Your working directory likely has contents that are outside of\n" \ "your sparse-checkout patterns. Use 'git sparse-checkout list' to\n" \ "see your sparse-checkout definition and compare it to your working\n" \ - "directory contents. Running 'git clean' may assist in this cleanup." + "directory contents. Running 'git sparse-checkout clean' may assist\n" \ + "in this cleanup." struct modify_index_context { struct index_state *write; From 8b431bc8dccd1bdd3d629620bc7f3394a774638b Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 6 Nov 2024 21:07:51 +0100 Subject: [PATCH 023/248] sideband: introduce an "escape hatch" to allow control characters The preceding commit fixed the vulnerability whereas sideband messages (that are under the control of the remote server) could contain ANSI escape sequences that would be sent to the terminal verbatim. However, this fix may not be desirable under all circumstances, e.g. when remote servers deliberately add coloring to their messages to increase their urgency. To help with those use cases, give users a way to opt-out of the protections: `sideband.allowControlCharacters`. Signed-off-by: Johannes Schindelin --- Documentation/config.adoc | 2 ++ Documentation/config/sideband.adoc | 5 +++++ sideband.c | 10 ++++++++++ t/t5409-colorize-remote-messages.sh | 8 +++++++- 4 files changed, 24 insertions(+), 1 deletion(-) create mode 100644 Documentation/config/sideband.adoc diff --git a/Documentation/config.adoc b/Documentation/config.adoc index 05f1ca7293f717..ae6dcb6a6c0607 100644 --- a/Documentation/config.adoc +++ b/Documentation/config.adoc @@ -521,6 +521,8 @@ include::config/sequencer.adoc[] include::config/showbranch.adoc[] +include::config/sideband.adoc[] + include::config/sparse.adoc[] include::config/splitindex.adoc[] diff --git a/Documentation/config/sideband.adoc b/Documentation/config/sideband.adoc new file mode 100644 index 00000000000000..3fb5045cd79581 --- /dev/null +++ b/Documentation/config/sideband.adoc @@ -0,0 +1,5 @@ +sideband.allowControlCharacters:: + By default, control characters that are delivered via the sideband + are masked, to prevent potentially unwanted ANSI escape sequences + from being sent to the terminal. Use this config setting to override + this behavior. diff --git a/sideband.c b/sideband.c index 5425c91722183b..da0f9644559735 100644 --- a/sideband.c +++ b/sideband.c @@ -26,6 +26,8 @@ static struct keyword_entry keywords[] = { { "error", GIT_COLOR_BOLD_RED }, }; +static int allow_control_characters; + /* Returns a color setting (GIT_COLOR_NEVER, etc). */ static int use_sideband_colors(void) { @@ -39,6 +41,9 @@ static int use_sideband_colors(void) if (use_sideband_colors_cached >= 0) return use_sideband_colors_cached; + repo_config_get_bool(the_repository, "sideband.allowcontrolcharacters", + &allow_control_characters); + if (!repo_config_get_string_tmp(the_repository, key, &value)) use_sideband_colors_cached = git_config_colorbool(key, value); else if (!repo_config_get_string_tmp(the_repository, "color.ui", &value)) @@ -68,6 +73,11 @@ void list_config_color_sideband_slots(struct string_list *list, const char *pref static void strbuf_add_sanitized(struct strbuf *dest, const char *src, int n) { + if (allow_control_characters) { + strbuf_add(dest, src, n); + return; + } + strbuf_grow(dest, n); for (; n && *src; src++, n--) { if (!iscntrl(*src) || *src == '\t' || *src == '\n') diff --git a/t/t5409-colorize-remote-messages.sh b/t/t5409-colorize-remote-messages.sh index d0745c391b2625..fb31e8525418a1 100755 --- a/t/t5409-colorize-remote-messages.sh +++ b/t/t5409-colorize-remote-messages.sh @@ -105,9 +105,15 @@ test_expect_success 'disallow (color) control sequences in sideband' ' EOF test_config_global uploadPack.packObjectshook ./color-me-surprised && test_commit need-at-least-one-commit && + git clone --no-local . throw-away 2>stderr && test_decode_color decoded && - test_grep ! RED decoded + test_grep ! RED decoded && + + rm -rf throw-away && + git -c sideband.allowControlCharacters clone --no-local . throw-away 2>stderr && + test_decode_color decoded && + test_grep RED decoded ' test_done From 067d53f72d16f9cbdfc5cacd6d17d1fc7d3c0069 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 12 Sep 2025 10:30:11 +0000 Subject: [PATCH 024/248] t: expand tests around sparse merges and clean With the current implementation of 'git sparse-checkout clean', we notice that a file that was in a conflicted state does not get cleaned up because of some internal details around the SKIP_WORKTREE bit. This test is documenting the current behavior before we update it in the following change. Signed-off-by: Derrick Stolee Signed-off-by: Junio C Hamano --- t/t1091-sparse-checkout-builtin.sh | 56 ++++++++++++++++++------------ 1 file changed, 34 insertions(+), 22 deletions(-) diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh index 7b15fa669c4662..b2da4feaeff9ec 100755 --- a/t/t1091-sparse-checkout-builtin.sh +++ b/t/t1091-sparse-checkout-builtin.sh @@ -1183,35 +1183,47 @@ test_expect_success 'clean with sparse file states' ' test_must_be_empty out ' -test_expect_success 'clean with merge conflict status' ' - git clone repo clean-merge && +test_expect_success 'sparse-checkout operations with merge conflicts' ' + git clone repo merge && - echo dirty >clean-merge/deep/deeper2/a && - touch clean-merge/folder2/extra && + ( + cd merge && + mkdir -p folder1/even/more/dirs && + echo base >folder1/even/more/dirs/file && + git add folder1 && + git commit -m "base" && - cat >input <<-EOF && - 0 $ZERO_OID folder1/a - 100644 $(git -C clean-merge rev-parse HEAD:folder1/a) 1 folder1/a - EOF - git -C clean-merge update-index --index-info folder1/even/more/dirs/file && + git commit -a -m "right" && - git -C clean-merge sparse-checkout set deep/deeper1 && + git checkout -b left HEAD~1 && + echo left >folder1/even/more/dirs/file && + git commit -a -m "left" && - test_must_fail git -C clean-merge sparse-checkout clean -f 2>err && - grep "failed to convert index to a sparse index" err && + git checkout -b merge && + git sparse-checkout set deep/deeper1 && - mkdir -p clean-merge/folder1/ && - echo merged >clean-merge/folder1/a && - git -C clean-merge add --sparse folder1/a && + test_must_fail git merge -m "will-conflict" right && - # deletes folder2/ but leaves staged change in folder1 - # and dirty change in deep/deeper2/ - cat >expect <<-\EOF && - Removing folder2/ - EOF + test_must_fail git sparse-checkout clean -f 2>err && + grep "failed to convert index to a sparse index" err && - git -C clean-merge sparse-checkout clean -f >out && - test_cmp expect out + echo merged >folder1/even/more/dirs/file && + git add --sparse folder1 && + git merge --continue && + + test_path_exists folder1/even/more/dirs/file && + + # clean does not remove the file, because the + # SKIP_WORKTREE bit was not cleared by the merge command. + git sparse-checkout clean -f >out && + test_line_count = 0 out && + test_path_exists folder1/even/more/dirs/file && + + git sparse-checkout reapply && + test_path_is_missing folder1 + ) ' test_done From 1e81481d444d8aec20fff0a9d9c8aa5e38243ef1 Mon Sep 17 00:00:00 2001 From: Colin Stagner Date: Tue, 9 Sep 2025 22:11:24 -0500 Subject: [PATCH 025/248] contrib/subtree: fix split with squashed subtrees 98ba49ccc2 (subtree: fix split processing with multiple subtrees present, 2023-12-01) increases the performance of git subtree split --prefix=subA by ignoring subtree merges which are outside of `subA/`. It also introduces a regression. Subtree merges that should be retained are incorrectly ignored if they: 1. are nested under `subA/`; and 2. are merged with `--squash`. For example, a subtree merged like: git subtree merge --squash --prefix=subA/subB "$rev" # ^^^^^^^^ ^^^^ is erroneously ignored during a split of `subA`. This causes missing tree files and different commit hashes starting in git v2.44.0-rc0. The method: should_ignore_subtree_split_commit REV should test only a single commit REV, but the combination of git log -1 --grep=... actually searches all *parent* commits until a `--grep` match is discovered. Rewrite this method to test only one REV at a time. Extract commit information with a single `git` call as opposed to three. The `test` conditions for rejecting a commit remain unchanged. Unit tests now cover nested subtrees. Signed-off-by: Colin Stagner Acked-by: Phillip Wood Signed-off-by: Junio C Hamano --- contrib/subtree/git-subtree.sh | 36 +++++++++++---- contrib/subtree/t/t7900-subtree.sh | 71 ++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+), 8 deletions(-) diff --git a/contrib/subtree/git-subtree.sh b/contrib/subtree/git-subtree.sh index 3fddba797cb92c..17106d1a721519 100755 --- a/contrib/subtree/git-subtree.sh +++ b/contrib/subtree/git-subtree.sh @@ -785,20 +785,40 @@ ensure_valid_ref_format () { die "fatal: '$1' does not look like a ref" } -# Usage: check if a commit from another subtree should be +# Usage: should_ignore_subtree_split_commit REV +# +# Check if REV is a commit from another subtree and should be # ignored from processing for splits should_ignore_subtree_split_commit () { assert test $# = 1 - local rev="$1" - if test -n "$(git log -1 --grep="git-subtree-dir:" $rev)" + + git show \ + --no-patch \ + --no-show-signature \ + --format='%(trailers:key=git-subtree-dir,key=git-subtree-mainline)' \ + "$1" | + ( + have_mainline= + subtree_dir= + + while read -r trailer val + do + case "$trailer" in + git-subtree-dir:) + subtree_dir="${val%/}" ;; + git-subtree-mainline:) + have_mainline=y ;; + esac + done + + if test -n "${subtree_dir}" && + test -z "${have_mainline}" && + test "${subtree_dir}" != "$arg_prefix" then - if test -z "$(git log -1 --grep="git-subtree-mainline:" $rev)" && - test -z "$(git log -1 --grep="git-subtree-dir: $arg_prefix$" $rev)" - then - return 0 - fi + return 0 fi return 1 + ) } # Usage: process_split_commit REV PARENTS diff --git a/contrib/subtree/t/t7900-subtree.sh b/contrib/subtree/t/t7900-subtree.sh index 3edbb33af46971..316dc5269e2b6f 100755 --- a/contrib/subtree/t/t7900-subtree.sh +++ b/contrib/subtree/t/t7900-subtree.sh @@ -9,6 +9,9 @@ This test verifies the basic operation of the add, merge, split, pull, and push subcommands of git subtree. ' +GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME=main +export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME + TEST_DIRECTORY=$(pwd)/../../../t . "$TEST_DIRECTORY"/test-lib.sh . "$TEST_DIRECTORY"/lib-gpg.sh @@ -68,6 +71,33 @@ test_create_pre2_32_repo () { git -C "$1-clone" replace HEAD^2 $new_commit } +# test_create_subtree_add REPO ORPHAN PREFIX FILENAME ... +# +# Create a simple subtree on a new branch named ORPHAN in REPO. +# The subtree is then merged into the current branch of REPO, +# under PREFIX. The generated subtree has has one commit +# with subject and tag FILENAME with a single file "FILENAME.t" +# +# When this method returns: +# - the current branch of REPO will have file PREFIX/FILENAME.t +# - REPO will have a branch named ORPHAN with subtree history +# +# additional arguments are forwarded to "subtree add" +test_create_subtree_add () { + ( + cd "$1" && + orphan="$2" && + prefix="$3" && + filename="$4" && + shift 4 && + last="$(git branch --show-current)" && + git switch --orphan "$orphan" && + test_commit "$filename" && + git checkout "$last" && + git subtree add --prefix="$prefix" "$@" "$orphan" + ) +} + test_expect_success 'shows short help text for -h' ' test_expect_code 129 git subtree -h >out 2>err && test_must_be_empty err && @@ -426,6 +456,47 @@ test_expect_success 'split with multiple subtrees' ' --squash --rejoin -d -m "Sub B Split 1" 2>&1 | grep -w "\[1\]")" = "" ' +# When subtree split-ing a directory that has other subtree +# *merges* underneath it, the split must include those subtrees. +# This test creates a nested subtree, `subA/subB`, and tests +# that the tree is correct after a subtree split of `subA/`. +# The test covers: +# - An initial `subtree add`; and +# - A follow-up `subtree merge` +# both with and without `--squashed`. +for is_squashed in '' 'y' +do + test_expect_success "split keeps nested ${is_squashed:+--squash }subtrees that are part of the split" ' + subtree_test_create_repo "$test_count" && + ( + cd "$test_count" && + mkdir subA && + test_commit subA/file1 && + test_create_subtree_add \ + . mksubtree subA/subB file2 ${is_squashed:+--squash} && + test_path_is_file subA/file1.t && + test_path_is_file subA/subB/file2.t && + git subtree split --prefix=subA --branch=bsplit && + git checkout bsplit && + test_path_is_file file1.t && + test_path_is_file subB/file2.t && + git checkout mksubtree && + git branch -D bsplit && + test_commit file3 && + git checkout main && + git subtree merge \ + ${is_squashed:+--squash} \ + --prefix=subA/subB mksubtree && + test_path_is_file subA/subB/file3.t && + git subtree split --prefix=subA --branch=bsplit && + git checkout bsplit && + test_path_is_file file1.t && + test_path_is_file subB/file2.t && + test_path_is_file subB/file3.t + ) + ' +done + test_expect_success 'split sub dir/ with --rejoin from scratch' ' subtree_test_create_repo "$test_count" && test_create_commit "$test_count" main1 && From 348357b8e7a279d6a211e0bd74c67b07bf7ad25c Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 18 Nov 2024 21:42:57 +0100 Subject: [PATCH 026/248] sideband: do allow ANSI color sequences by default The preceding two commits introduced special handling of the sideband channel to neutralize ANSI escape sequences before sending the payload to the terminal, and `sideband.allowControlCharacters` to override that behavior. However, some `pre-receive` hooks that are actively used in practice want to color their messages and therefore rely on the fact that Git passes them through to the terminal. In contrast to other ANSI escape sequences, it is highly unlikely that coloring sequences can be essential tools in attack vectors that mislead Git users e.g. by hiding crucial information. Therefore we can have both: Continue to allow ANSI coloring sequences to be passed to the terminal, and neutralize all other ANSI escape sequences. Signed-off-by: Johannes Schindelin --- Documentation/config/sideband.adoc | 17 ++++++-- sideband.c | 61 ++++++++++++++++++++++++++--- t/t5409-colorize-remote-messages.sh | 16 +++++++- 3 files changed, 84 insertions(+), 10 deletions(-) diff --git a/Documentation/config/sideband.adoc b/Documentation/config/sideband.adoc index 3fb5045cd79581..f347fd6b33004a 100644 --- a/Documentation/config/sideband.adoc +++ b/Documentation/config/sideband.adoc @@ -1,5 +1,16 @@ sideband.allowControlCharacters:: By default, control characters that are delivered via the sideband - are masked, to prevent potentially unwanted ANSI escape sequences - from being sent to the terminal. Use this config setting to override - this behavior. + are masked, except ANSI color sequences. This prevents potentially + unwanted ANSI escape sequences from being sent to the terminal. Use + this config setting to override this behavior: ++ +-- + color:: + Allow ANSI color sequences, line feeds and horizontal tabs, + but mask all other control characters. This is the default. + false:: + Mask all control characters other than line feeds and + horizontal tabs. + true:: + Allow all control characters to be sent to the terminal. +-- diff --git a/sideband.c b/sideband.c index da0f9644559735..bd452b7ee69421 100644 --- a/sideband.c +++ b/sideband.c @@ -26,7 +26,11 @@ static struct keyword_entry keywords[] = { { "error", GIT_COLOR_BOLD_RED }, }; -static int allow_control_characters; +static enum { + ALLOW_NO_CONTROL_CHARACTERS = 0, + ALLOW_ALL_CONTROL_CHARACTERS = 1, + ALLOW_ANSI_COLOR_SEQUENCES = 2 +} allow_control_characters = ALLOW_ANSI_COLOR_SEQUENCES; /* Returns a color setting (GIT_COLOR_NEVER, etc). */ static int use_sideband_colors(void) @@ -41,8 +45,24 @@ static int use_sideband_colors(void) if (use_sideband_colors_cached >= 0) return use_sideband_colors_cached; - repo_config_get_bool(the_repository, "sideband.allowcontrolcharacters", - &allow_control_characters); + switch (repo_config_get_maybe_bool(the_repository, "sideband.allowcontrolcharacters", &i)) { + case 0: /* Boolean value */ + allow_control_characters = i ? ALLOW_ALL_CONTROL_CHARACTERS : + ALLOW_NO_CONTROL_CHARACTERS; + break; + case -1: /* non-Boolean value */ + if (repo_config_get_string_tmp(the_repository, "sideband.allowcontrolcharacters", + &value)) + ; /* huh? `get_maybe_bool()` returned -1 */ + else if (!strcmp(value, "color")) + allow_control_characters = ALLOW_ANSI_COLOR_SEQUENCES; + else + warning(_("unrecognized value for `sideband." + "allowControlCharacters`: '%s'"), value); + break; + default: + break; /* not configured */ + } if (!repo_config_get_string_tmp(the_repository, key, &value)) use_sideband_colors_cached = git_config_colorbool(key, value); @@ -71,9 +91,37 @@ void list_config_color_sideband_slots(struct string_list *list, const char *pref list_config_item(list, prefix, keywords[i].keyword); } +static int handle_ansi_color_sequence(struct strbuf *dest, const char *src, int n) +{ + int i; + + /* + * Valid ANSI color sequences are of the form + * + * ESC [ [ [; ]*] m + */ + + if (allow_control_characters != ALLOW_ANSI_COLOR_SEQUENCES || + n < 3 || src[0] != '\x1b' || src[1] != '[') + return 0; + + for (i = 2; i < n; i++) { + if (src[i] == 'm') { + strbuf_add(dest, src, i + 1); + return i; + } + if (!isdigit(src[i]) && src[i] != ';') + break; + } + + return 0; +} + static void strbuf_add_sanitized(struct strbuf *dest, const char *src, int n) { - if (allow_control_characters) { + int i; + + if (allow_control_characters == ALLOW_ALL_CONTROL_CHARACTERS) { strbuf_add(dest, src, n); return; } @@ -82,7 +130,10 @@ static void strbuf_add_sanitized(struct strbuf *dest, const char *src, int n) for (; n && *src; src++, n--) { if (!iscntrl(*src) || *src == '\t' || *src == '\n') strbuf_addch(dest, *src); - else { + else if ((i = handle_ansi_color_sequence(dest, src, n))) { + src += i; + n -= i; + } else { strbuf_addch(dest, '^'); strbuf_addch(dest, 0x40 + *src); } diff --git a/t/t5409-colorize-remote-messages.sh b/t/t5409-colorize-remote-messages.sh index fb31e8525418a1..a755c49a74e634 100755 --- a/t/t5409-colorize-remote-messages.sh +++ b/t/t5409-colorize-remote-messages.sh @@ -100,7 +100,7 @@ test_expect_success 'fallback to color.ui' ' test_expect_success 'disallow (color) control sequences in sideband' ' write_script .git/color-me-surprised <<-\EOF && - printf "error: Have you \\033[31mread\\033[m this?\\n" >&2 + printf "error: Have you \\033[31mread\\033[m this?\\a\\n" >&2 exec "$@" EOF test_config_global uploadPack.packObjectshook ./color-me-surprised && @@ -108,12 +108,24 @@ test_expect_success 'disallow (color) control sequences in sideband' ' git clone --no-local . throw-away 2>stderr && test_decode_color decoded && + test_grep RED decoded && + test_grep "\\^G" stderr && + tr -dc "\\007" actual && + test_must_be_empty actual && + + rm -rf throw-away && + git -c sideband.allowControlCharacters=false \ + clone --no-local . throw-away 2>stderr && + test_decode_color decoded && test_grep ! RED decoded && + test_grep "\\^G" stderr && rm -rf throw-away && git -c sideband.allowControlCharacters clone --no-local . throw-away 2>stderr && test_decode_color decoded && - test_grep RED decoded + test_grep RED decoded && + tr -dc "\\007" actual && + test_file_not_empty actual ' test_done From db9d7ca394f1096bc5c13147c5880fd1015a6971 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 30 Oct 2024 19:48:46 +0100 Subject: [PATCH 027/248] unix-socket: avoid leak when initialization fails When a Unix socket is initialized, the current directory's path is stored so that the cleanup code can `chdir()` back to where it was before exit. If the path that needs to be stored exceeds the default size of the `sun_path` attribute of `struct sockaddr_un` (which is defined as a 108-sized byte array on Linux), a larger buffer needs to be allocated so that it can hold the path, and it is the responsibility of the `unix_sockaddr_cleanup()` function to release that allocated memory. In Git's CI, this stack allocation is not necessary because the code is checked out to `/home/runner/work/git/git`. Concatenate the path `t/trash directory.t0301-credential-cache/.cache/git/credential/socket` and a terminating NUL, and you end up with 96 bytes, 12 shy of the default `sun_path` size. However, I use worktrees with slightly longer paths: `/home/me/projects/git/yes/i/nest/worktrees/to/organize/them/` is more in line with what I have. When I recently tried to locally reproduce a failure of the `linux-leaks` CI job, this t0301 test failed (where it had not failed in CI). The reason: When `credential-cache` tries to reach its daemon initially by calling `unix_sockaddr_init()`, it is expected that the daemon cannot be reached (the idea is to spin up the daemon in that case and try again). However, when this first call to `unix_sockaddr_init()` fails, the code returns early from the `unix_stream_connect()` function _without_ giving the cleanup code a chance to run, skipping the deallocation of above-mentioned path. The fix is easy: do not return early but instead go directly to the cleanup code. Signed-off-by: Johannes Schindelin --- unix-socket.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/unix-socket.c b/unix-socket.c index 8860203c3f46dc..1fa0cf6c15c721 100644 --- a/unix-socket.c +++ b/unix-socket.c @@ -84,7 +84,7 @@ int unix_stream_connect(const char *path, int disallow_chdir) struct unix_sockaddr_context ctx; if (unix_sockaddr_init(&sa, path, &ctx, disallow_chdir) < 0) - return -1; + goto fail; fd = socket(AF_UNIX, SOCK_STREAM, 0); if (fd < 0) goto fail; From 89013364f06c0b256052dcaafe22f1820fd82628 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 2 Jun 2025 14:45:24 +0200 Subject: [PATCH 028/248] mingw: avoid relative `#include`s We want to make them relative to the top-level directory. Signed-off-by: Johannes Schindelin --- compat/mingw.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..da99473f56d451 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1,22 +1,22 @@ #define USE_THE_REPOSITORY_VARIABLE #define DISABLE_SIGN_COMPARE_WARNINGS -#include "../git-compat-util.h" +#include "git-compat-util.h" #include "win32.h" #include #include #include #include -#include "../strbuf.h" -#include "../run-command.h" -#include "../abspath.h" -#include "../alloc.h" +#include "strbuf.h" +#include "run-command.h" +#include "abspath.h" +#include "alloc.h" #include "win32/lazyload.h" -#include "../config.h" -#include "../environment.h" -#include "../trace2.h" -#include "../symlinks.h" -#include "../wrapper.h" +#include "config.h" +#include "environment.h" +#include "trace2.h" +#include "symlinks.h" +#include "wrapper.h" #include "dir.h" #include "gettext.h" #define SECURITY_WIN32 From 78b8fb030d6ec4ca863819ca8a6fa3a2cc9240e4 Mon Sep 17 00:00:00 2001 From: Jeff King Date: Mon, 13 Jan 2025 01:26:01 -0500 Subject: [PATCH 029/248] grep: prevent `^$` false match at end of file In some implementations, `regexec_buf()` assumes that it is fed lines; Without `REG_NOTEOL` it thinks the end of the buffer is the end of a line. Which makes sense, but trips up this case because we are not feeding lines, but rather a whole buffer. So the final newline is not the start of an empty line, but the true end of the buffer. This causes an interesting bug: $ echo content >file.txt $ git grep --no-index -n '^$' file.txt file.txt:2: This bug is fixed by making the end of the buffer consistently the end of the final line. The patch was applied from https://lore.kernel.org/git/20250113062601.GD767856@coredump.intra.peff.net/ Reported-by: Olly Betts Signed-off-by: Johannes Schindelin --- grep.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/grep.c b/grep.c index 932647e4a6580b..79166934a989be 100644 --- a/grep.c +++ b/grep.c @@ -1646,6 +1646,8 @@ static int grep_source_1(struct grep_opt *opt, struct grep_source *gs, int colle bol = gs->buf; left = gs->size; + if (left && gs->buf[left-1] == '\n') + left--; while (left) { const char *eol; int hit; From 4c7bcd214739300aa884257a2e94bba160edd815 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 2 Jun 2025 14:46:43 +0200 Subject: [PATCH 030/248] mingw: order `#include`s alphabetically It allows for more consistent patches that way. Signed-off-by: Johannes Schindelin --- compat/mingw.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index da99473f56d451..736a07a028ab4d 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2,25 +2,25 @@ #define DISABLE_SIGN_COMPARE_WARNINGS #include "git-compat-util.h" -#include "win32.h" -#include -#include -#include -#include -#include "strbuf.h" -#include "run-command.h" #include "abspath.h" #include "alloc.h" -#include "win32/lazyload.h" #include "config.h" +#include "dir.h" #include "environment.h" -#include "trace2.h" +#include "gettext.h" +#include "run-command.h" +#include "strbuf.h" #include "symlinks.h" +#include "trace2.h" +#include "win32.h" +#include "win32/lazyload.h" #include "wrapper.h" -#include "dir.h" -#include "gettext.h" +#include +#include +#include #define SECURITY_WIN32 #include +#include #include #define STATUS_DELETE_PENDING ((NTSTATUS) 0xC0000056) From 9fa03f2d367fc1b3c46f448d362ac44c53eaa890 Mon Sep 17 00:00:00 2001 From: Sverre Rabbelier Date: Sun, 24 Jul 2011 15:54:04 +0200 Subject: [PATCH 031/248] t9350: point out that refs are not updated correctly This happens only when the corresponding commits are not exported in the current fast-export run. This can happen either when the relevant commit is already marked, or when the commit is explicitly marked as UNINTERESTING with a negative ref by another argument. This breaks fast-export basec remote helpers. Signed-off-by: Sverre Rabbelier --- t/t9350-fast-export.sh | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/t/t9350-fast-export.sh b/t/t9350-fast-export.sh index 8f85c69d62f17a..d8ecc4705fb180 100755 --- a/t/t9350-fast-export.sh +++ b/t/t9350-fast-export.sh @@ -970,4 +970,15 @@ test_expect_success GPG 'export and import of doubly signed commit' ' fi ' +cat > expected << EOF +reset refs/heads/master +from $(git rev-parse master) + +EOF + +test_expect_failure 'refs are updated even if no commits need to be exported' ' + git fast-export master..master > actual && + test_cmp expected actual +' + test_done From 271b1fa1d713d3d7aed2626c2ff8ed77d5071690 Mon Sep 17 00:00:00 2001 From: Sverre Rabbelier Date: Sat, 28 Aug 2010 20:49:01 -0500 Subject: [PATCH 032/248] transport-helper: add trailing -- [PT: ensure we add an additional element to the argv array] Signed-off-by: Sverre Rabbelier Signed-off-by: Johannes Schindelin --- transport-helper.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/transport-helper.c b/transport-helper.c index 0789e5bca53282..3e3f9e66c39f02 100644 --- a/transport-helper.c +++ b/transport-helper.c @@ -499,6 +499,8 @@ static int get_exporter(struct transport *transport, for (size_t i = 0; i < revlist_args->nr; i++) strvec_push(&fastexport->args, revlist_args->items[i].string); + strvec_push(&fastexport->args, "--"); + fastexport->git_cmd = 1; return start_command(fastexport); } From 1856befe9543212357292df6ef74b2eb6491b860 Mon Sep 17 00:00:00 2001 From: Sverre Rabbelier Date: Sun, 24 Jul 2011 00:06:00 +0200 Subject: [PATCH 033/248] remote-helper: check helper status after import/export Signed-off-by: Johannes Schindelin Signed-off-by: Sverre Rabbelier --- t/t5801-remote-helpers.sh | 2 +- transport-helper.c | 15 +++++++++++++++ 2 files changed, 16 insertions(+), 1 deletion(-) diff --git a/t/t5801-remote-helpers.sh b/t/t5801-remote-helpers.sh index d21877150ed82e..3917da47276825 100755 --- a/t/t5801-remote-helpers.sh +++ b/t/t5801-remote-helpers.sh @@ -262,7 +262,7 @@ test_expect_success 'push update refs failure' ' echo "update fail" >>file && git commit -a -m "update fail" && git rev-parse --verify testgit/origin/heads/update >expect && - test_expect_code 1 env GIT_REMOTE_TESTGIT_FAILURE="non-fast forward" \ + test_must_fail env GIT_REMOTE_TESTGIT_FAILURE="non-fast forward" \ git push origin update && git rev-parse --verify testgit/origin/heads/update >actual && test_cmp expect actual diff --git a/transport-helper.c b/transport-helper.c index 3e3f9e66c39f02..7545e50d27ed63 100644 --- a/transport-helper.c +++ b/transport-helper.c @@ -505,6 +505,19 @@ static int get_exporter(struct transport *transport, return start_command(fastexport); } +static void check_helper_status(struct helper_data *data) +{ + int pid, status; + + pid = waitpid(data->helper->pid, &status, WNOHANG); + if (pid < 0) + die("Could not retrieve status of remote helper '%s'", + data->name); + if (pid > 0 && WIFEXITED(status)) + die("Remote helper '%s' died with %d", + data->name, WEXITSTATUS(status)); +} + static int fetch_with_import(struct transport *transport, int nr_heads, struct ref **to_fetch) { @@ -541,6 +554,7 @@ static int fetch_with_import(struct transport *transport, if (finish_command(&fastimport)) die(_("error while running fast-import")); + check_helper_status(data); /* * The fast-import stream of a remote helper that advertises @@ -1160,6 +1174,7 @@ static int push_refs_with_export(struct transport *transport, if (finish_command(&exporter)) die(_("error while running fast-export")); + check_helper_status(data); if (push_update_refs_status(data, remote_refs, flags)) return 1; From fe8ef44015667884166b66ff4ea225f4ce70afa0 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 18 Apr 2017 12:09:08 +0200 Subject: [PATCH 034/248] mingw: demonstrate a problem with certain absolute paths On Windows, there are several categories of absolute paths. One such category starts with a backslash and is implicitly relative to the drive associated with the current working directory. Example: c: git clone https://github.com/git-for-windows/git \G4W should clone into C:\G4W. There is currently a problem with that, in that mingw_mktemp() does not expect the _wmktemp() function to prefix the absolute path with the drive prefix, and as a consequence, the resulting path does not fit into the originally-passed string buffer. The symptom is a "Result too large" error. Reported by Juan Carlos Arevalo Baeza. Signed-off-by: Johannes Schindelin --- t/t5580-unc-paths.sh | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/t/t5580-unc-paths.sh b/t/t5580-unc-paths.sh index 65ef1a3628ee94..2c30f3b51d3183 100755 --- a/t/t5580-unc-paths.sh +++ b/t/t5580-unc-paths.sh @@ -20,14 +20,11 @@ fi UNCPATH="$(winpwd)" case "$UNCPATH" in [A-Z]:*) + WITHOUTDRIVE="${UNCPATH#?:}" # Use administrative share e.g. \\localhost\C$\git-sdk-64\usr\src\git # (we use forward slashes here because MSYS2 and Git accept them, and # they are easier on the eyes) - UNCPATH="//localhost/${UNCPATH%%:*}\$/${UNCPATH#?:}" - test -d "$UNCPATH" || { - skip_all='could not access administrative share; skipping' - test_done - } + UNCPATH="//localhost/${UNCPATH%%:*}\$$WITHOUTDRIVE" ;; *) skip_all='skipping UNC path tests, cannot determine current path as UNC' @@ -35,6 +32,18 @@ case "$UNCPATH" in ;; esac +test_expect_failure 'clone into absolute path lacking a drive prefix' ' + USINGBACKSLASHES="$(echo "$WITHOUTDRIVE"/without-drive-prefix | + tr / \\\\)" && + git clone . "$USINGBACKSLASHES" && + test -f without-drive-prefix/.git/HEAD +' + +test -d "$UNCPATH" || { + skip_all='could not access administrative share; skipping' + test_done +} + test_expect_success setup ' test_commit initial ' From 9abf52cb05ba78247ad3788b8ddd0cbcd45a875e Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 7 Dec 2018 13:39:30 +0100 Subject: [PATCH 035/248] clean: do not traverse mount points It seems to be not exactly rare on Windows to install NTFS junction points (the equivalent of "bind mounts" on Linux/Unix) in worktrees, e.g. to map some development tools into a subdirectory. In such a scenario, it is pretty horrible if `git clean -dfx` traverses into the mapped directory and starts to "clean up". Let's just not do that. Let's make sure before we traverse into a directory that it is not a mount point (or junction). This addresses https://github.com/git-for-windows/git/issues/607 Signed-off-by: Johannes Schindelin --- builtin/clean.c | 14 ++++++++++++++ compat/mingw.c | 22 ++++++++++++++++++++++ compat/mingw.h | 3 +++ git-compat-util.h | 4 ++++ path.c | 39 +++++++++++++++++++++++++++++++++++++++ path.h | 1 + t/t7300-clean.sh | 9 +++++++++ 7 files changed, 92 insertions(+) diff --git a/builtin/clean.c b/builtin/clean.c index a1977b92dc89dd..55643c5b766e54 100644 --- a/builtin/clean.c +++ b/builtin/clean.c @@ -41,6 +41,8 @@ static const char *msg_remove = N_("Removing %s\n"); static const char *msg_would_remove = N_("Would remove %s\n"); static const char *msg_skip_git_dir = N_("Skipping repository %s\n"); static const char *msg_would_skip_git_dir = N_("Would skip repository %s\n"); +static const char *msg_skip_mount_point = N_("Skipping mount point %s\n"); +static const char *msg_would_skip_mount_point = N_("Would skip mount point %s\n"); static const char *msg_warn_remove_failed = N_("failed to remove %s"); static const char *msg_warn_lstat_failed = N_("could not lstat %s\n"); static const char *msg_skip_cwd = N_("Refusing to remove current working directory\n"); @@ -185,6 +187,18 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, goto out; } + if (is_mount_point(path)) { + if (!quiet) { + quote_path(path->buf, prefix, "ed, 0); + printf(dry_run ? + _(msg_would_skip_mount_point) : + _(msg_skip_mount_point), quoted.buf); + } + *dir_gone = 0; + + goto out; + } + dir = opendir(path->buf); if (!dir) { /* an empty dir could be removed even if it is unreadble */ diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..1da8a999833c5f 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2666,6 +2666,28 @@ pid_t waitpid(pid_t pid, int *status, int options) return -1; } +int mingw_is_mount_point(struct strbuf *path) +{ + WIN32_FIND_DATAW findbuf = { 0 }; + HANDLE handle; + wchar_t wfilename[MAX_PATH]; + int wlen = xutftowcs_path(wfilename, path->buf); + if (wlen < 0) + die(_("could not get long path for '%s'"), path->buf); + + /* remove trailing slash, if any */ + if (wlen > 0 && wfilename[wlen - 1] == L'/') + wfilename[--wlen] = L'\0'; + + handle = FindFirstFileW(wfilename, &findbuf); + if (handle == INVALID_HANDLE_VALUE) + return 0; + FindClose(handle); + + return (findbuf.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) && + (findbuf.dwReserved0 == IO_REPARSE_TAG_MOUNT_POINT); +} + int xutftowcsn(wchar_t *wcs, const char *utfs, size_t wcslen, int utflen) { int upos = 0, wpos = 0; diff --git a/compat/mingw.h b/compat/mingw.h index 444daedfa52469..af6fc3f12970bf 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -36,6 +36,9 @@ static inline void convert_slashes(char *path) if (*path == '\\') *path = '/'; } +struct strbuf; +int mingw_is_mount_point(struct strbuf *path); +#define is_mount_point mingw_is_mount_point #define PATH_SEP ';' char *mingw_query_user_email(void); #define query_user_email mingw_query_user_email diff --git a/git-compat-util.h b/git-compat-util.h index 398e0fac4fab60..7ae64af46345cb 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -383,6 +383,10 @@ static inline int git_has_dir_sep(const char *path) #define has_dir_sep(path) git_has_dir_sep(path) #endif +#ifndef is_mount_point +#define is_mount_point is_mount_point_via_stat +#endif + #ifndef query_user_email #define query_user_email() NULL #endif diff --git a/path.c b/path.c index 7f56eaf9930374..ca78007f9c7691 100644 --- a/path.c +++ b/path.c @@ -1423,6 +1423,45 @@ char *strip_path_suffix(const char *path, const char *suffix) return offset == -1 ? NULL : xstrndup(path, offset); } +int is_mount_point_via_stat(struct strbuf *path) +{ + size_t len = path->len; + dev_t current_dev; + struct stat st; + + if (!strcmp("/", path->buf)) + return 1; + + strbuf_addstr(path, "/."); + if (lstat(path->buf, &st)) { + /* + * If we cannot access the current directory, we cannot say + * that it is a bind mount. + */ + strbuf_setlen(path, len); + return 0; + } + current_dev = st.st_dev; + + /* Now look at the parent directory */ + strbuf_addch(path, '.'); + if (lstat(path->buf, &st)) { + /* + * If we cannot access the parent directory, we cannot say + * that it is a bind mount. + */ + strbuf_setlen(path, len); + return 0; + } + strbuf_setlen(path, len); + + /* + * If the device ID differs between current and parent directory, + * then it is a bind mount. + */ + return current_dev != st.st_dev; +} + int daemon_avoid_alias(const char *p) { int sl, ndot; diff --git a/path.h b/path.h index e67348f25397cc..7a46488f8d706d 100644 --- a/path.h +++ b/path.h @@ -172,6 +172,7 @@ int normalize_path_copy(char *dst, const char *src); int strbuf_normalize_path(struct strbuf *src); int longest_ancestor_length(const char *path, struct string_list *prefixes); char *strip_path_suffix(const char *path, const char *suffix); +int is_mount_point_via_stat(struct strbuf *path); int daemon_avoid_alias(const char *path); /* diff --git a/t/t7300-clean.sh b/t/t7300-clean.sh index 00d4070156243b..7c3a1ca91df534 100755 --- a/t/t7300-clean.sh +++ b/t/t7300-clean.sh @@ -800,4 +800,13 @@ test_expect_success 'traverse into directories that may have ignored entries' ' ) ' +test_expect_success MINGW 'clean does not traverse mount points' ' + mkdir target && + >target/dont-clean-me && + git init with-mountpoint && + cmd //c "mklink /j with-mountpoint\\mountpoint target" && + git -C with-mountpoint clean -dfx && + test_path_is_file target/dont-clean-me +' + test_done From 8aeb35851d530e3edb8717aadee4ecf99b7967ef Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 9 Apr 2012 13:04:35 -0500 Subject: [PATCH 036/248] Always auto-gc after calling a fast-import transport After importing anything with fast-import, we should always let the garbage collector do its job, since the objects are written to disk inefficiently. This brings down an initial import of http://selenic.com/hg from about 230 megabytes to about 14. In the future, we may want to make this configurable on a per-remote basis, or maybe teach fast-import about it in the first place. Signed-off-by: Johannes Schindelin --- transport-helper.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/transport-helper.c b/transport-helper.c index 7545e50d27ed63..af4a717ad538d2 100644 --- a/transport-helper.c +++ b/transport-helper.c @@ -22,6 +22,8 @@ #include "packfile.h" static int debug; +/* TODO: put somewhere sensible, e.g. git_transport_options? */ +static int auto_gc = 1; struct helper_data { char *name; @@ -588,6 +590,13 @@ static int fetch_with_import(struct transport *transport, } } strbuf_release(&buf); + if (auto_gc) { + struct child_process cmd = CHILD_PROCESS_INIT; + + cmd.git_cmd = 1; + strvec_pushl(&cmd.args, "gc", "--auto", "--quiet", NULL); + run_command(&cmd); + } return 0; } From 0a86d81e44bd07c4f1db4b5a43c591b3aad154bf Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 18 Apr 2017 12:38:30 +0200 Subject: [PATCH 037/248] mingw: allow absolute paths without drive prefix When specifying an absolute path without a drive prefix, we convert that path internally. Let's make sure that we handle that case properly, too ;-) This fixes the command git clone https://github.com/git-for-windows/git \G4W Signed-off-by: Johannes Schindelin --- compat/mingw.c | 10 +++++++++- t/t5580-unc-paths.sh | 2 +- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..cad9ba54642d00 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1165,11 +1165,19 @@ unsigned int sleep (unsigned int seconds) char *mingw_mktemp(char *template) { wchar_t wtemplate[MAX_PATH]; + int offset = 0; + if (xutftowcs_path(wtemplate, template) < 0) return NULL; + + if (is_dir_sep(template[0]) && !is_dir_sep(template[1]) && + iswalpha(wtemplate[0]) && wtemplate[1] == L':') { + /* We have an absolute path missing the drive prefix */ + offset = 2; + } if (!_wmktemp(wtemplate)) return NULL; - if (xwcstoutf(template, wtemplate, strlen(template) + 1) < 0) + if (xwcstoutf(template, wtemplate + offset, strlen(template) + 1) < 0) return NULL; return template; } diff --git a/t/t5580-unc-paths.sh b/t/t5580-unc-paths.sh index 2c30f3b51d3183..e9df367d5777fd 100755 --- a/t/t5580-unc-paths.sh +++ b/t/t5580-unc-paths.sh @@ -32,7 +32,7 @@ case "$UNCPATH" in ;; esac -test_expect_failure 'clone into absolute path lacking a drive prefix' ' +test_expect_success 'clone into absolute path lacking a drive prefix' ' USINGBACKSLASHES="$(echo "$WITHOUTDRIVE"/without-drive-prefix | tr / \\\\)" && git clone . "$USINGBACKSLASHES" && From efc59a2bc8c4378d20551127a31fa4c0f7dcbdf9 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 11 Dec 2018 12:55:26 +0100 Subject: [PATCH 038/248] clean: remove mount points when possible Windows' equivalent to "bind mounts", NTFS junction points, can be unlinked without affecting the mount target. This is clearly what users expect to happen when they call `git clean -dfx` in a worktree that contains NTFS junction points: the junction should be removed, and the target directory of said junction should be left alone (unless it is inside the worktree). Signed-off-by: Johannes Schindelin --- builtin/clean.c | 13 +++++++++++++ compat/mingw.h | 1 + t/t7300-clean.sh | 1 + 3 files changed, 15 insertions(+) diff --git a/builtin/clean.c b/builtin/clean.c index 55643c5b766e54..b447155ab2cf2a 100644 --- a/builtin/clean.c +++ b/builtin/clean.c @@ -41,8 +41,10 @@ static const char *msg_remove = N_("Removing %s\n"); static const char *msg_would_remove = N_("Would remove %s\n"); static const char *msg_skip_git_dir = N_("Skipping repository %s\n"); static const char *msg_would_skip_git_dir = N_("Would skip repository %s\n"); +#ifndef CAN_UNLINK_MOUNT_POINTS static const char *msg_skip_mount_point = N_("Skipping mount point %s\n"); static const char *msg_would_skip_mount_point = N_("Would skip mount point %s\n"); +#endif static const char *msg_warn_remove_failed = N_("failed to remove %s"); static const char *msg_warn_lstat_failed = N_("could not lstat %s\n"); static const char *msg_skip_cwd = N_("Refusing to remove current working directory\n"); @@ -188,6 +190,7 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, } if (is_mount_point(path)) { +#ifndef CAN_UNLINK_MOUNT_POINTS if (!quiet) { quote_path(path->buf, prefix, "ed, 0); printf(dry_run ? @@ -195,6 +198,16 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, _(msg_skip_mount_point), quoted.buf); } *dir_gone = 0; +#else + if (!dry_run && unlink(path->buf)) { + int saved_errno = errno; + quote_path(path->buf, prefix, "ed, 0); + errno = saved_errno; + warning_errno(_(msg_warn_remove_failed), quoted.buf); + *dir_gone = 0; + ret = -1; + } +#endif goto out; } diff --git a/compat/mingw.h b/compat/mingw.h index af6fc3f12970bf..fb83cdaf4e982c 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -39,6 +39,7 @@ static inline void convert_slashes(char *path) struct strbuf; int mingw_is_mount_point(struct strbuf *path); #define is_mount_point mingw_is_mount_point +#define CAN_UNLINK_MOUNT_POINTS 1 #define PATH_SEP ';' char *mingw_query_user_email(void); #define query_user_email mingw_query_user_email diff --git a/t/t7300-clean.sh b/t/t7300-clean.sh index 7c3a1ca91df534..6f16f3893191e7 100755 --- a/t/t7300-clean.sh +++ b/t/t7300-clean.sh @@ -806,6 +806,7 @@ test_expect_success MINGW 'clean does not traverse mount points' ' git init with-mountpoint && cmd //c "mklink /j with-mountpoint\\mountpoint target" && git -C with-mountpoint clean -dfx && + test_path_is_missing with-mountpoint/mountpoint && test_path_is_file target/dont-clean-me ' From ef29ee39d22825f9184881a3691c6f4829f26063 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sun, 20 Oct 2019 22:08:58 +0200 Subject: [PATCH 039/248] win32/pthread: avoid name clashes with winpthread When asking the mingw-w64 variant of GCC to compile C11 code, it seems to link implicitly to libwinpthread, which does implement a pthread emulation (that is more complete than Git's). In preparation for vendoring in mimalloc (which requires C11 support), let's keep preferring Git's own pthread emulation. To avoid linker errors where it thinks that the `pthread_self` and the `pthread_create` symbols are defined twice, let's give our version a `win32_` prefix, just like we already do for `pthread_join()`. Signed-off-by: Johannes Schindelin --- compat/win32/pthread.c | 6 +++--- compat/win32/pthread.h | 8 +++++--- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/compat/win32/pthread.c b/compat/win32/pthread.c index 58980a529c3eb9..4cc6356f757d80 100644 --- a/compat/win32/pthread.c +++ b/compat/win32/pthread.c @@ -21,8 +21,8 @@ static unsigned __stdcall win32_start_routine(void *arg) return 0; } -int pthread_create(pthread_t *thread, const void *attr UNUSED, - void *(*start_routine)(void *), void *arg) +int win32_pthread_create(pthread_t *thread, const void *attr UNUSED, + void *(*start_routine)(void *), void *arg) { thread->arg = arg; thread->start_routine = start_routine; @@ -53,7 +53,7 @@ int win32_pthread_join(pthread_t *thread, void **value_ptr) } } -pthread_t pthread_self(void) +pthread_t win32_pthread_self(void) { pthread_t t = { NULL }; t.tid = GetCurrentThreadId(); diff --git a/compat/win32/pthread.h b/compat/win32/pthread.h index e2b5c4f64c9b91..476e78b911b857 100644 --- a/compat/win32/pthread.h +++ b/compat/win32/pthread.h @@ -50,8 +50,9 @@ typedef struct { DWORD tid; } pthread_t; -int pthread_create(pthread_t *thread, const void *unused, - void *(*start_routine)(void*), void *arg); +int win32_pthread_create(pthread_t *thread, const void *unused, + void *(*start_routine)(void*), void *arg); +#define pthread_create win32_pthread_create /* * To avoid the need of copying a struct, we use small macro wrapper to pass @@ -62,7 +63,8 @@ int pthread_create(pthread_t *thread, const void *unused, int win32_pthread_join(pthread_t *thread, void **value_ptr); #define pthread_equal(t1, t2) ((t1).tid == (t2).tid) -pthread_t pthread_self(void); +pthread_t win32_pthread_self(void); +#define pthread_self win32_pthread_self static inline void NORETURN pthread_exit(void *ret) { From 452bb2ec1c6caf9fde37699a5902af5e716571b0 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 12 Aug 2022 12:44:15 +0200 Subject: [PATCH 040/248] git-compat-util: avoid redeclaring _DEFAULT_SOURCE We are about to vendor in `mimalloc`'s source code which we will want to include `compat/posix.h` after defining that constant. Signed-off-by: Johannes Schindelin --- compat/posix.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/compat/posix.h b/compat/posix.h index 067a00f33b83f3..626d71256f8428 100644 --- a/compat/posix.h +++ b/compat/posix.h @@ -70,7 +70,9 @@ #define _ALL_SOURCE 1 #define _GNU_SOURCE 1 #define _BSD_SOURCE 1 +#ifndef _DEFAULT_SOURCE #define _DEFAULT_SOURCE 1 +#endif #define _NETBSD_SOURCE 1 #define _SGI_SOURCE 1 From 9ab117a9d06bf43b9a3d92a4ef6951ea31e4875c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sat, 5 Jul 2025 09:01:01 +0200 Subject: [PATCH 041/248] Import the source code of mimalloc v2.2.4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Update to newer mimalloc versions like this: update_mimalloc () { test $# = 1 || { echo "Need a mimalloc version" 1>&2; return 1 }; for oneline in 'mimalloc: adjust for building inside Git' 'Import the source code of mimalloc'; do git revert -n HEAD^{/^"$oneline"} && git checkout HEAD -- Makefile && git commit -sm "Temporarily revert \"$oneline\"" -m 'In preparation for upgrading to a newer mimalloc version.' || return 1; done; for file in $(git show --format='%n' --name-only --diff-filter=A HEAD^{/^"Import the source code of mimalloc "}) compat/mimalloc/arena-abandon.c compat/mimalloc/free.c compat/mimalloc/libc.c compat/mimalloc/prim/prim.c compat/mimalloc/mimalloc-stats.h; do file2=${file#compat/mimalloc/}; case "$file2" in segment-cache.c) : no longer needed; continue ;; bitmap.h | *.c) file2=src/$file2 ;; *.h) file2=include/$file2 ;; esac; mkdir -p "${file%/*}" && git -C /usr/src/mimalloc/ show "$1":$file2 > "$file" && git add "$file" || { echo "Failed: $file2 -> $file" 1>&2; return 1 }; done; conv_sed='sed -n "/^ *eval/d;/ /p"' && git commit -sm "Import the source code of mimalloc $1" -m "Update to newer mimalloc versions like this:" -m "$(set | sed -n '/^update_mimalloc *() *$/,/^}/{s/^./ &/;p}')" -m ' update_mimalloc $MIMALLOC_VERSION' -m 'For convenience, you can set `MIMALLOC_VERSION` and then run:' -m ' eval "$(git show -s | '"$conv_sed"')"' || return 1; git cherry-pick HEAD^{/^'mimalloc: adjust for building inside Git'} || return 1 } update_mimalloc $MIMALLOC_VERSION For convenience, you can set `MIMALLOC_VERSION` and then run: eval "$(git show -s | sed -n "/^ *eval/d;/ /p")" Signed-off-by: Matthias Aßhauer --- compat/mimalloc/LICENSE | 21 + compat/mimalloc/alloc-aligned.c | 360 ++++++ compat/mimalloc/alloc.c | 692 +++++++++++ compat/mimalloc/arena-abandon.c | 346 ++++++ compat/mimalloc/arena.c | 1043 ++++++++++++++++ compat/mimalloc/bitmap.c | 441 +++++++ compat/mimalloc/bitmap.h | 119 ++ compat/mimalloc/free.c | 572 +++++++++ compat/mimalloc/heap.c | 733 ++++++++++++ compat/mimalloc/init.c | 714 +++++++++++ compat/mimalloc/libc.c | 334 ++++++ compat/mimalloc/mimalloc-stats.h | 103 ++ compat/mimalloc/mimalloc.h | 612 ++++++++++ compat/mimalloc/mimalloc/atomic.h | 557 +++++++++ compat/mimalloc/mimalloc/internal.h | 1153 ++++++++++++++++++ compat/mimalloc/mimalloc/prim.h | 421 +++++++ compat/mimalloc/mimalloc/track.h | 145 +++ compat/mimalloc/mimalloc/types.h | 685 +++++++++++ compat/mimalloc/options.c | 663 +++++++++++ compat/mimalloc/os.c | 770 ++++++++++++ compat/mimalloc/page-queue.c | 390 ++++++ compat/mimalloc/page.c | 1049 +++++++++++++++++ compat/mimalloc/prim/osx/prim.c | 9 + compat/mimalloc/prim/prim.c | 76 ++ compat/mimalloc/prim/unix/prim.c | 934 +++++++++++++++ compat/mimalloc/prim/windows/prim.c | 878 ++++++++++++++ compat/mimalloc/random.c | 258 ++++ compat/mimalloc/segment-map.c | 142 +++ compat/mimalloc/segment.c | 1702 +++++++++++++++++++++++++++ compat/mimalloc/stats.c | 635 ++++++++++ 30 files changed, 16557 insertions(+) create mode 100644 compat/mimalloc/LICENSE create mode 100644 compat/mimalloc/alloc-aligned.c create mode 100644 compat/mimalloc/alloc.c create mode 100644 compat/mimalloc/arena-abandon.c create mode 100644 compat/mimalloc/arena.c create mode 100644 compat/mimalloc/bitmap.c create mode 100644 compat/mimalloc/bitmap.h create mode 100644 compat/mimalloc/free.c create mode 100644 compat/mimalloc/heap.c create mode 100644 compat/mimalloc/init.c create mode 100644 compat/mimalloc/libc.c create mode 100644 compat/mimalloc/mimalloc-stats.h create mode 100644 compat/mimalloc/mimalloc.h create mode 100644 compat/mimalloc/mimalloc/atomic.h create mode 100644 compat/mimalloc/mimalloc/internal.h create mode 100644 compat/mimalloc/mimalloc/prim.h create mode 100644 compat/mimalloc/mimalloc/track.h create mode 100644 compat/mimalloc/mimalloc/types.h create mode 100644 compat/mimalloc/options.c create mode 100644 compat/mimalloc/os.c create mode 100644 compat/mimalloc/page-queue.c create mode 100644 compat/mimalloc/page.c create mode 100644 compat/mimalloc/prim/osx/prim.c create mode 100644 compat/mimalloc/prim/prim.c create mode 100644 compat/mimalloc/prim/unix/prim.c create mode 100644 compat/mimalloc/prim/windows/prim.c create mode 100644 compat/mimalloc/random.c create mode 100644 compat/mimalloc/segment-map.c create mode 100644 compat/mimalloc/segment.c create mode 100644 compat/mimalloc/stats.c diff --git a/compat/mimalloc/LICENSE b/compat/mimalloc/LICENSE new file mode 100644 index 00000000000000..53315ebee557ac --- /dev/null +++ b/compat/mimalloc/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2018-2025 Microsoft Corporation, Daan Leijen + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/compat/mimalloc/alloc-aligned.c b/compat/mimalloc/alloc-aligned.c new file mode 100644 index 00000000000000..3d3202eb574971 --- /dev/null +++ b/compat/mimalloc/alloc-aligned.c @@ -0,0 +1,360 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2021, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" // mi_prim_get_default_heap + +#include // memset + +// ------------------------------------------------------ +// Aligned Allocation +// ------------------------------------------------------ + +static bool mi_malloc_is_naturally_aligned( size_t size, size_t alignment ) { + // objects up to `MI_MAX_ALIGN_GUARANTEE` are allocated aligned to their size (see `segment.c:_mi_segment_page_start`). + mi_assert_internal(_mi_is_power_of_two(alignment) && (alignment > 0)); + if (alignment > size) return false; + if (alignment <= MI_MAX_ALIGN_SIZE) return true; + const size_t bsize = mi_good_size(size); + return (bsize <= MI_MAX_ALIGN_GUARANTEE && (bsize & (alignment-1)) == 0); +} + +#if MI_GUARDED +static mi_decl_restrict void* mi_heap_malloc_guarded_aligned(mi_heap_t* heap, size_t size, size_t alignment, bool zero) mi_attr_noexcept { + // use over allocation for guarded blocksl + mi_assert_internal(alignment > 0 && alignment < MI_BLOCK_ALIGNMENT_MAX); + const size_t oversize = size + alignment - 1; + void* base = _mi_heap_malloc_guarded(heap, oversize, zero); + void* p = mi_align_up_ptr(base, alignment); + mi_track_align(base, p, (uint8_t*)p - (uint8_t*)base, size); + mi_assert_internal(mi_usable_size(p) >= size); + mi_assert_internal(_mi_is_aligned(p, alignment)); + return p; +} + +static void* mi_heap_malloc_zero_no_guarded(mi_heap_t* heap, size_t size, bool zero) { + const size_t rate = heap->guarded_sample_rate; + // only write if `rate!=0` so we don't write to the constant `_mi_heap_empty` + if (rate != 0) { heap->guarded_sample_rate = 0; } + void* p = _mi_heap_malloc_zero(heap, size, zero); + if (rate != 0) { heap->guarded_sample_rate = rate; } + return p; +} +#else +static void* mi_heap_malloc_zero_no_guarded(mi_heap_t* heap, size_t size, bool zero) { + return _mi_heap_malloc_zero(heap, size, zero); +} +#endif + +// Fallback aligned allocation that over-allocates -- split out for better codegen +static mi_decl_noinline void* mi_heap_malloc_zero_aligned_at_overalloc(mi_heap_t* const heap, const size_t size, const size_t alignment, const size_t offset, const bool zero) mi_attr_noexcept +{ + mi_assert_internal(size <= (MI_MAX_ALLOC_SIZE - MI_PADDING_SIZE)); + mi_assert_internal(alignment != 0 && _mi_is_power_of_two(alignment)); + + void* p; + size_t oversize; + if mi_unlikely(alignment > MI_BLOCK_ALIGNMENT_MAX) { + // use OS allocation for very large alignment and allocate inside a huge page (dedicated segment with 1 page) + // This can support alignments >= MI_SEGMENT_SIZE by ensuring the object can be aligned at a point in the + // first (and single) page such that the segment info is `MI_SEGMENT_SIZE` bytes before it (so it can be found by aligning the pointer down) + if mi_unlikely(offset != 0) { + // todo: cannot support offset alignment for very large alignments yet +#if MI_DEBUG > 0 + _mi_error_message(EOVERFLOW, "aligned allocation with a very large alignment cannot be used with an alignment offset (size %zu, alignment %zu, offset %zu)\n", size, alignment, offset); +#endif + return NULL; + } + oversize = (size <= MI_SMALL_SIZE_MAX ? MI_SMALL_SIZE_MAX + 1 /* ensure we use generic malloc path */ : size); + // note: no guarded as alignment > 0 + p = _mi_heap_malloc_zero_ex(heap, oversize, false, alignment); // the page block size should be large enough to align in the single huge page block + // zero afterwards as only the area from the aligned_p may be committed! + if (p == NULL) return NULL; + } + else { + // otherwise over-allocate + oversize = (size < MI_MAX_ALIGN_SIZE ? MI_MAX_ALIGN_SIZE : size) + alignment - 1; // adjust for size <= 16; with size 0 and aligment 64k, we would allocate a 64k block and pointing just beyond that. + p = mi_heap_malloc_zero_no_guarded(heap, oversize, zero); + if (p == NULL) return NULL; + } + mi_page_t* page = _mi_ptr_page(p); + + // .. and align within the allocation + const uintptr_t align_mask = alignment - 1; // for any x, `(x & align_mask) == (x % alignment)` + const uintptr_t poffset = ((uintptr_t)p + offset) & align_mask; + const uintptr_t adjust = (poffset == 0 ? 0 : alignment - poffset); + mi_assert_internal(adjust < alignment); + void* aligned_p = (void*)((uintptr_t)p + adjust); + if (aligned_p != p) { + mi_page_set_has_aligned(page, true); + #if MI_GUARDED + // set tag to aligned so mi_usable_size works with guard pages + if (adjust >= sizeof(mi_block_t)) { + mi_block_t* const block = (mi_block_t*)p; + block->next = MI_BLOCK_TAG_ALIGNED; + } + #endif + _mi_padding_shrink(page, (mi_block_t*)p, adjust + size); + } + // todo: expand padding if overallocated ? + + mi_assert_internal(mi_page_usable_block_size(page) >= adjust + size); + mi_assert_internal(((uintptr_t)aligned_p + offset) % alignment == 0); + mi_assert_internal(mi_usable_size(aligned_p)>=size); + mi_assert_internal(mi_usable_size(p) == mi_usable_size(aligned_p)+adjust); + #if MI_DEBUG > 1 + mi_page_t* const apage = _mi_ptr_page(aligned_p); + void* unalign_p = _mi_page_ptr_unalign(apage, aligned_p); + mi_assert_internal(p == unalign_p); + #endif + + // now zero the block if needed + if (alignment > MI_BLOCK_ALIGNMENT_MAX) { + // for the tracker, on huge aligned allocations only the memory from the start of the large block is defined + mi_track_mem_undefined(aligned_p, size); + if (zero) { + _mi_memzero_aligned(aligned_p, mi_usable_size(aligned_p)); + } + } + + if (p != aligned_p) { + mi_track_align(p,aligned_p,adjust,mi_usable_size(aligned_p)); + #if MI_GUARDED + mi_track_mem_defined(p, sizeof(mi_block_t)); + #endif + } + return aligned_p; +} + +// Generic primitive aligned allocation -- split out for better codegen +static mi_decl_noinline void* mi_heap_malloc_zero_aligned_at_generic(mi_heap_t* const heap, const size_t size, const size_t alignment, const size_t offset, const bool zero) mi_attr_noexcept +{ + mi_assert_internal(alignment != 0 && _mi_is_power_of_two(alignment)); + // we don't allocate more than MI_MAX_ALLOC_SIZE (see ) + if mi_unlikely(size > (MI_MAX_ALLOC_SIZE - MI_PADDING_SIZE)) { + #if MI_DEBUG > 0 + _mi_error_message(EOVERFLOW, "aligned allocation request is too large (size %zu, alignment %zu)\n", size, alignment); + #endif + return NULL; + } + + // use regular allocation if it is guaranteed to fit the alignment constraints. + // this is important to try as the fast path in `mi_heap_malloc_zero_aligned` only works when there exist + // a page with the right block size, and if we always use the over-alloc fallback that would never happen. + if (offset == 0 && mi_malloc_is_naturally_aligned(size,alignment)) { + void* p = mi_heap_malloc_zero_no_guarded(heap, size, zero); + mi_assert_internal(p == NULL || ((uintptr_t)p % alignment) == 0); + const bool is_aligned_or_null = (((uintptr_t)p) & (alignment-1))==0; + if mi_likely(is_aligned_or_null) { + return p; + } + else { + // this should never happen if the `mi_malloc_is_naturally_aligned` check is correct.. + mi_assert(false); + mi_free(p); + } + } + + // fall back to over-allocation + return mi_heap_malloc_zero_aligned_at_overalloc(heap,size,alignment,offset,zero); +} + + +// Primitive aligned allocation +static void* mi_heap_malloc_zero_aligned_at(mi_heap_t* const heap, const size_t size, const size_t alignment, const size_t offset, const bool zero) mi_attr_noexcept +{ + // note: we don't require `size > offset`, we just guarantee that the address at offset is aligned regardless of the allocated size. + if mi_unlikely(alignment == 0 || !_mi_is_power_of_two(alignment)) { // require power-of-two (see ) + #if MI_DEBUG > 0 + _mi_error_message(EOVERFLOW, "aligned allocation requires the alignment to be a power-of-two (size %zu, alignment %zu)\n", size, alignment); + #endif + return NULL; + } + + #if MI_GUARDED + if (offset==0 && alignment < MI_BLOCK_ALIGNMENT_MAX && mi_heap_malloc_use_guarded(heap,size)) { + return mi_heap_malloc_guarded_aligned(heap, size, alignment, zero); + } + #endif + + // try first if there happens to be a small block available with just the right alignment + if mi_likely(size <= MI_SMALL_SIZE_MAX && alignment <= size) { + const uintptr_t align_mask = alignment-1; // for any x, `(x & align_mask) == (x % alignment)` + const size_t padsize = size + MI_PADDING_SIZE; + mi_page_t* page = _mi_heap_get_free_small_page(heap, padsize); + if mi_likely(page->free != NULL) { + const bool is_aligned = (((uintptr_t)page->free + offset) & align_mask)==0; + if mi_likely(is_aligned) + { + void* p = (zero ? _mi_page_malloc_zeroed(heap,page,padsize) : _mi_page_malloc(heap,page,padsize)); // call specific page malloc for better codegen + mi_assert_internal(p != NULL); + mi_assert_internal(((uintptr_t)p + offset) % alignment == 0); + mi_track_malloc(p,size,zero); + return p; + } + } + } + + // fallback to generic aligned allocation + return mi_heap_malloc_zero_aligned_at_generic(heap, size, alignment, offset, zero); +} + + +// ------------------------------------------------------ +// Optimized mi_heap_malloc_aligned / mi_malloc_aligned +// ------------------------------------------------------ + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_malloc_aligned_at(mi_heap_t* heap, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_malloc_zero_aligned_at(heap, size, alignment, offset, false); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_malloc_aligned(mi_heap_t* heap, size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_malloc_aligned_at(heap, size, alignment, 0); +} + +// ensure a definition is emitted +#if defined(__cplusplus) +void* _mi_extern_heap_malloc_aligned = (void*)&mi_heap_malloc_aligned; +#endif + +// ------------------------------------------------------ +// Aligned Allocation +// ------------------------------------------------------ + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_zalloc_aligned_at(mi_heap_t* heap, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_malloc_zero_aligned_at(heap, size, alignment, offset, true); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_zalloc_aligned(mi_heap_t* heap, size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_zalloc_aligned_at(heap, size, alignment, 0); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_calloc_aligned_at(mi_heap_t* heap, size_t count, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(count, size, &total)) return NULL; + return mi_heap_zalloc_aligned_at(heap, total, alignment, offset); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_calloc_aligned(mi_heap_t* heap, size_t count, size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_calloc_aligned_at(heap,count,size,alignment,0); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_malloc_aligned_at(size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_malloc_aligned_at(mi_prim_get_default_heap(), size, alignment, offset); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_malloc_aligned(size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_malloc_aligned(mi_prim_get_default_heap(), size, alignment); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_zalloc_aligned_at(size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_zalloc_aligned_at(mi_prim_get_default_heap(), size, alignment, offset); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_zalloc_aligned(size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_zalloc_aligned(mi_prim_get_default_heap(), size, alignment); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_calloc_aligned_at(size_t count, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_calloc_aligned_at(mi_prim_get_default_heap(), count, size, alignment, offset); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_calloc_aligned(size_t count, size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_calloc_aligned(mi_prim_get_default_heap(), count, size, alignment); +} + + +// ------------------------------------------------------ +// Aligned re-allocation +// ------------------------------------------------------ + +static void* mi_heap_realloc_zero_aligned_at(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, size_t offset, bool zero) mi_attr_noexcept { + mi_assert(alignment > 0); + if (alignment <= sizeof(uintptr_t)) return _mi_heap_realloc_zero(heap,p,newsize,zero); + if (p == NULL) return mi_heap_malloc_zero_aligned_at(heap,newsize,alignment,offset,zero); + size_t size = mi_usable_size(p); + if (newsize <= size && newsize >= (size - (size / 2)) + && (((uintptr_t)p + offset) % alignment) == 0) { + return p; // reallocation still fits, is aligned and not more than 50% waste + } + else { + // note: we don't zero allocate upfront so we only zero initialize the expanded part + void* newp = mi_heap_malloc_aligned_at(heap,newsize,alignment,offset); + if (newp != NULL) { + if (zero && newsize > size) { + // also set last word in the previous allocation to zero to ensure any padding is zero-initialized + size_t start = (size >= sizeof(intptr_t) ? size - sizeof(intptr_t) : 0); + _mi_memzero((uint8_t*)newp + start, newsize - start); + } + _mi_memcpy_aligned(newp, p, (newsize > size ? size : newsize)); + mi_free(p); // only free if successful + } + return newp; + } +} + +static void* mi_heap_realloc_zero_aligned(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, bool zero) mi_attr_noexcept { + mi_assert(alignment > 0); + if (alignment <= sizeof(uintptr_t)) return _mi_heap_realloc_zero(heap,p,newsize,zero); + size_t offset = ((uintptr_t)p % alignment); // use offset of previous allocation (p can be NULL) + return mi_heap_realloc_zero_aligned_at(heap,p,newsize,alignment,offset,zero); +} + +mi_decl_nodiscard void* mi_heap_realloc_aligned_at(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_realloc_zero_aligned_at(heap,p,newsize,alignment,offset,false); +} + +mi_decl_nodiscard void* mi_heap_realloc_aligned(mi_heap_t* heap, void* p, size_t newsize, size_t alignment) mi_attr_noexcept { + return mi_heap_realloc_zero_aligned(heap,p,newsize,alignment,false); +} + +mi_decl_nodiscard void* mi_heap_rezalloc_aligned_at(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_realloc_zero_aligned_at(heap, p, newsize, alignment, offset, true); +} + +mi_decl_nodiscard void* mi_heap_rezalloc_aligned(mi_heap_t* heap, void* p, size_t newsize, size_t alignment) mi_attr_noexcept { + return mi_heap_realloc_zero_aligned(heap, p, newsize, alignment, true); +} + +mi_decl_nodiscard void* mi_heap_recalloc_aligned_at(mi_heap_t* heap, void* p, size_t newcount, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(newcount, size, &total)) return NULL; + return mi_heap_rezalloc_aligned_at(heap, p, total, alignment, offset); +} + +mi_decl_nodiscard void* mi_heap_recalloc_aligned(mi_heap_t* heap, void* p, size_t newcount, size_t size, size_t alignment) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(newcount, size, &total)) return NULL; + return mi_heap_rezalloc_aligned(heap, p, total, alignment); +} + +mi_decl_nodiscard void* mi_realloc_aligned_at(void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_realloc_aligned_at(mi_prim_get_default_heap(), p, newsize, alignment, offset); +} + +mi_decl_nodiscard void* mi_realloc_aligned(void* p, size_t newsize, size_t alignment) mi_attr_noexcept { + return mi_heap_realloc_aligned(mi_prim_get_default_heap(), p, newsize, alignment); +} + +mi_decl_nodiscard void* mi_rezalloc_aligned_at(void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_rezalloc_aligned_at(mi_prim_get_default_heap(), p, newsize, alignment, offset); +} + +mi_decl_nodiscard void* mi_rezalloc_aligned(void* p, size_t newsize, size_t alignment) mi_attr_noexcept { + return mi_heap_rezalloc_aligned(mi_prim_get_default_heap(), p, newsize, alignment); +} + +mi_decl_nodiscard void* mi_recalloc_aligned_at(void* p, size_t newcount, size_t size, size_t alignment, size_t offset) mi_attr_noexcept { + return mi_heap_recalloc_aligned_at(mi_prim_get_default_heap(), p, newcount, size, alignment, offset); +} + +mi_decl_nodiscard void* mi_recalloc_aligned(void* p, size_t newcount, size_t size, size_t alignment) mi_attr_noexcept { + return mi_heap_recalloc_aligned(mi_prim_get_default_heap(), p, newcount, size, alignment); +} + + diff --git a/compat/mimalloc/alloc.c b/compat/mimalloc/alloc.c new file mode 100644 index 00000000000000..0fed5e754c1918 --- /dev/null +++ b/compat/mimalloc/alloc.c @@ -0,0 +1,692 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#ifndef _DEFAULT_SOURCE +#define _DEFAULT_SOURCE // for realpath() on Linux +#endif + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "mimalloc/prim.h" // _mi_prim_thread_id() + +#include // memset, strlen (for mi_strdup) +#include // malloc, abort + +#define MI_IN_ALLOC_C +#include "alloc-override.c" +#include "free.c" +#undef MI_IN_ALLOC_C + +// ------------------------------------------------------ +// Allocation +// ------------------------------------------------------ + +// Fast allocation in a page: just pop from the free list. +// Fall back to generic allocation only if the list is empty. +// Note: in release mode the (inlined) routine is about 7 instructions with a single test. +extern inline void* _mi_page_malloc_zero(mi_heap_t* heap, mi_page_t* page, size_t size, bool zero) mi_attr_noexcept +{ + mi_assert_internal(size >= MI_PADDING_SIZE); + mi_assert_internal(page->block_size == 0 /* empty heap */ || mi_page_block_size(page) >= size); + + // check the free list + mi_block_t* const block = page->free; + if mi_unlikely(block == NULL) { + return _mi_malloc_generic(heap, size, zero, 0); + } + mi_assert_internal(block != NULL && _mi_ptr_page(block) == page); + + // pop from the free list + page->free = mi_block_next(page, block); + page->used++; + mi_assert_internal(page->free == NULL || _mi_ptr_page(page->free) == page); + mi_assert_internal(page->block_size < MI_MAX_ALIGN_SIZE || _mi_is_aligned(block, MI_MAX_ALIGN_SIZE)); + + #if MI_DEBUG>3 + if (page->free_is_zero && size > sizeof(*block)) { + mi_assert_expensive(mi_mem_is_zero(block+1,size - sizeof(*block))); + } + #endif + + // allow use of the block internally + // note: when tracking we need to avoid ever touching the MI_PADDING since + // that is tracked by valgrind etc. as non-accessible (through the red-zone, see `mimalloc/track.h`) + mi_track_mem_undefined(block, mi_page_usable_block_size(page)); + + // zero the block? note: we need to zero the full block size (issue #63) + if mi_unlikely(zero) { + mi_assert_internal(page->block_size != 0); // do not call with zero'ing for huge blocks (see _mi_malloc_generic) + mi_assert_internal(!mi_page_is_huge(page)); + #if MI_PADDING + mi_assert_internal(page->block_size >= MI_PADDING_SIZE); + #endif + if (page->free_is_zero) { + block->next = 0; + mi_track_mem_defined(block, page->block_size - MI_PADDING_SIZE); + } + else { + _mi_memzero_aligned(block, page->block_size - MI_PADDING_SIZE); + } + } + + #if (MI_DEBUG>0) && !MI_TRACK_ENABLED && !MI_TSAN + if (!zero && !mi_page_is_huge(page)) { + memset(block, MI_DEBUG_UNINIT, mi_page_usable_block_size(page)); + } + #elif (MI_SECURE!=0) + if (!zero) { block->next = 0; } // don't leak internal data + #endif + + #if (MI_STAT>0) + const size_t bsize = mi_page_usable_block_size(page); + if (bsize <= MI_MEDIUM_OBJ_SIZE_MAX) { + mi_heap_stat_increase(heap, malloc_normal, bsize); + mi_heap_stat_counter_increase(heap, malloc_normal_count, 1); + #if (MI_STAT>1) + const size_t bin = _mi_bin(bsize); + mi_heap_stat_increase(heap, malloc_bins[bin], 1); + mi_heap_stat_increase(heap, malloc_requested, size - MI_PADDING_SIZE); + #endif + } + #endif + + #if MI_PADDING // && !MI_TRACK_ENABLED + mi_padding_t* const padding = (mi_padding_t*)((uint8_t*)block + mi_page_usable_block_size(page)); + ptrdiff_t delta = ((uint8_t*)padding - (uint8_t*)block - (size - MI_PADDING_SIZE)); + #if (MI_DEBUG>=2) + mi_assert_internal(delta >= 0 && mi_page_usable_block_size(page) >= (size - MI_PADDING_SIZE + delta)); + #endif + mi_track_mem_defined(padding,sizeof(mi_padding_t)); // note: re-enable since mi_page_usable_block_size may set noaccess + padding->canary = mi_ptr_encode_canary(page,block,page->keys); + padding->delta = (uint32_t)(delta); + #if MI_PADDING_CHECK + if (!mi_page_is_huge(page)) { + uint8_t* fill = (uint8_t*)padding - delta; + const size_t maxpad = (delta > MI_MAX_ALIGN_SIZE ? MI_MAX_ALIGN_SIZE : delta); // set at most N initial padding bytes + for (size_t i = 0; i < maxpad; i++) { fill[i] = MI_DEBUG_PADDING; } + } + #endif + #endif + + return block; +} + +// extra entries for improved efficiency in `alloc-aligned.c`. +extern void* _mi_page_malloc(mi_heap_t* heap, mi_page_t* page, size_t size) mi_attr_noexcept { + return _mi_page_malloc_zero(heap,page,size,false); +} +extern void* _mi_page_malloc_zeroed(mi_heap_t* heap, mi_page_t* page, size_t size) mi_attr_noexcept { + return _mi_page_malloc_zero(heap,page,size,true); +} + +#if MI_GUARDED +mi_decl_restrict void* _mi_heap_malloc_guarded(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept; +#endif + +static inline mi_decl_restrict void* mi_heap_malloc_small_zero(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept { + mi_assert(heap != NULL); + mi_assert(size <= MI_SMALL_SIZE_MAX); + #if MI_DEBUG + const uintptr_t tid = _mi_thread_id(); + mi_assert(heap->thread_id == 0 || heap->thread_id == tid); // heaps are thread local + #endif + #if (MI_PADDING || MI_GUARDED) + if (size == 0) { size = sizeof(void*); } + #endif + #if MI_GUARDED + if (mi_heap_malloc_use_guarded(heap,size)) { + return _mi_heap_malloc_guarded(heap, size, zero); + } + #endif + + // get page in constant time, and allocate from it + mi_page_t* page = _mi_heap_get_free_small_page(heap, size + MI_PADDING_SIZE); + void* const p = _mi_page_malloc_zero(heap, page, size + MI_PADDING_SIZE, zero); + mi_track_malloc(p,size,zero); + + #if MI_DEBUG>3 + if (p != NULL && zero) { + mi_assert_expensive(mi_mem_is_zero(p, size)); + } + #endif + return p; +} + +// allocate a small block +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_heap_malloc_small(mi_heap_t* heap, size_t size) mi_attr_noexcept { + return mi_heap_malloc_small_zero(heap, size, false); +} + +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_malloc_small(size_t size) mi_attr_noexcept { + return mi_heap_malloc_small(mi_prim_get_default_heap(), size); +} + +// The main allocation function +extern inline void* _mi_heap_malloc_zero_ex(mi_heap_t* heap, size_t size, bool zero, size_t huge_alignment) mi_attr_noexcept { + // fast path for small objects + if mi_likely(size <= MI_SMALL_SIZE_MAX) { + mi_assert_internal(huge_alignment == 0); + return mi_heap_malloc_small_zero(heap, size, zero); + } + #if MI_GUARDED + else if (huge_alignment==0 && mi_heap_malloc_use_guarded(heap,size)) { + return _mi_heap_malloc_guarded(heap, size, zero); + } + #endif + else { + // regular allocation + mi_assert(heap!=NULL); + mi_assert(heap->thread_id == 0 || heap->thread_id == _mi_thread_id()); // heaps are thread local + void* const p = _mi_malloc_generic(heap, size + MI_PADDING_SIZE, zero, huge_alignment); // note: size can overflow but it is detected in malloc_generic + mi_track_malloc(p,size,zero); + + #if MI_DEBUG>3 + if (p != NULL && zero) { + mi_assert_expensive(mi_mem_is_zero(p, size)); + } + #endif + return p; + } +} + +extern inline void* _mi_heap_malloc_zero(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept { + return _mi_heap_malloc_zero_ex(heap, size, zero, 0); +} + +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_heap_malloc(mi_heap_t* heap, size_t size) mi_attr_noexcept { + return _mi_heap_malloc_zero(heap, size, false); +} + +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_malloc(size_t size) mi_attr_noexcept { + return mi_heap_malloc(mi_prim_get_default_heap(), size); +} + +// zero initialized small block +mi_decl_nodiscard mi_decl_restrict void* mi_zalloc_small(size_t size) mi_attr_noexcept { + return mi_heap_malloc_small_zero(mi_prim_get_default_heap(), size, true); +} + +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_heap_zalloc(mi_heap_t* heap, size_t size) mi_attr_noexcept { + return _mi_heap_malloc_zero(heap, size, true); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_zalloc(size_t size) mi_attr_noexcept { + return mi_heap_zalloc(mi_prim_get_default_heap(),size); +} + + +mi_decl_nodiscard extern inline mi_decl_restrict void* mi_heap_calloc(mi_heap_t* heap, size_t count, size_t size) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(count,size,&total)) return NULL; + return mi_heap_zalloc(heap,total); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_calloc(size_t count, size_t size) mi_attr_noexcept { + return mi_heap_calloc(mi_prim_get_default_heap(),count,size); +} + +// Uninitialized `calloc` +mi_decl_nodiscard extern mi_decl_restrict void* mi_heap_mallocn(mi_heap_t* heap, size_t count, size_t size) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(count, size, &total)) return NULL; + return mi_heap_malloc(heap, total); +} + +mi_decl_nodiscard mi_decl_restrict void* mi_mallocn(size_t count, size_t size) mi_attr_noexcept { + return mi_heap_mallocn(mi_prim_get_default_heap(),count,size); +} + +// Expand (or shrink) in place (or fail) +void* mi_expand(void* p, size_t newsize) mi_attr_noexcept { + #if MI_PADDING + // we do not shrink/expand with padding enabled + MI_UNUSED(p); MI_UNUSED(newsize); + return NULL; + #else + if (p == NULL) return NULL; + const size_t size = _mi_usable_size(p,"mi_expand"); + if (newsize > size) return NULL; + return p; // it fits + #endif +} + +void* _mi_heap_realloc_zero(mi_heap_t* heap, void* p, size_t newsize, bool zero) mi_attr_noexcept { + // if p == NULL then behave as malloc. + // else if size == 0 then reallocate to a zero-sized block (and don't return NULL, just as mi_malloc(0)). + // (this means that returning NULL always indicates an error, and `p` will not have been freed in that case.) + const size_t size = _mi_usable_size(p,"mi_realloc"); // also works if p == NULL (with size 0) + if mi_unlikely(newsize <= size && newsize >= (size / 2) && newsize > 0) { // note: newsize must be > 0 or otherwise we return NULL for realloc(NULL,0) + mi_assert_internal(p!=NULL); + // todo: do not track as the usable size is still the same in the free; adjust potential padding? + // mi_track_resize(p,size,newsize) + // if (newsize < size) { mi_track_mem_noaccess((uint8_t*)p + newsize, size - newsize); } + return p; // reallocation still fits and not more than 50% waste + } + void* newp = mi_heap_malloc(heap,newsize); + if mi_likely(newp != NULL) { + if (zero && newsize > size) { + // also set last word in the previous allocation to zero to ensure any padding is zero-initialized + const size_t start = (size >= sizeof(intptr_t) ? size - sizeof(intptr_t) : 0); + _mi_memzero((uint8_t*)newp + start, newsize - start); + } + else if (newsize == 0) { + ((uint8_t*)newp)[0] = 0; // work around for applications that expect zero-reallocation to be zero initialized (issue #725) + } + if mi_likely(p != NULL) { + const size_t copysize = (newsize > size ? size : newsize); + mi_track_mem_defined(p,copysize); // _mi_useable_size may be too large for byte precise memory tracking.. + _mi_memcpy(newp, p, copysize); + mi_free(p); // only free the original pointer if successful + } + } + return newp; +} + +mi_decl_nodiscard void* mi_heap_realloc(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept { + return _mi_heap_realloc_zero(heap, p, newsize, false); +} + +mi_decl_nodiscard void* mi_heap_reallocn(mi_heap_t* heap, void* p, size_t count, size_t size) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(count, size, &total)) return NULL; + return mi_heap_realloc(heap, p, total); +} + + +// Reallocate but free `p` on errors +mi_decl_nodiscard void* mi_heap_reallocf(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept { + void* newp = mi_heap_realloc(heap, p, newsize); + if (newp==NULL && p!=NULL) mi_free(p); + return newp; +} + +mi_decl_nodiscard void* mi_heap_rezalloc(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept { + return _mi_heap_realloc_zero(heap, p, newsize, true); +} + +mi_decl_nodiscard void* mi_heap_recalloc(mi_heap_t* heap, void* p, size_t count, size_t size) mi_attr_noexcept { + size_t total; + if (mi_count_size_overflow(count, size, &total)) return NULL; + return mi_heap_rezalloc(heap, p, total); +} + + +mi_decl_nodiscard void* mi_realloc(void* p, size_t newsize) mi_attr_noexcept { + return mi_heap_realloc(mi_prim_get_default_heap(),p,newsize); +} + +mi_decl_nodiscard void* mi_reallocn(void* p, size_t count, size_t size) mi_attr_noexcept { + return mi_heap_reallocn(mi_prim_get_default_heap(),p,count,size); +} + +// Reallocate but free `p` on errors +mi_decl_nodiscard void* mi_reallocf(void* p, size_t newsize) mi_attr_noexcept { + return mi_heap_reallocf(mi_prim_get_default_heap(),p,newsize); +} + +mi_decl_nodiscard void* mi_rezalloc(void* p, size_t newsize) mi_attr_noexcept { + return mi_heap_rezalloc(mi_prim_get_default_heap(), p, newsize); +} + +mi_decl_nodiscard void* mi_recalloc(void* p, size_t count, size_t size) mi_attr_noexcept { + return mi_heap_recalloc(mi_prim_get_default_heap(), p, count, size); +} + + + +// ------------------------------------------------------ +// strdup, strndup, and realpath +// ------------------------------------------------------ + +// `strdup` using mi_malloc +mi_decl_nodiscard mi_decl_restrict char* mi_heap_strdup(mi_heap_t* heap, const char* s) mi_attr_noexcept { + if (s == NULL) return NULL; + size_t len = _mi_strlen(s); + char* t = (char*)mi_heap_malloc(heap,len+1); + if (t == NULL) return NULL; + _mi_memcpy(t, s, len); + t[len] = 0; + return t; +} + +mi_decl_nodiscard mi_decl_restrict char* mi_strdup(const char* s) mi_attr_noexcept { + return mi_heap_strdup(mi_prim_get_default_heap(), s); +} + +// `strndup` using mi_malloc +mi_decl_nodiscard mi_decl_restrict char* mi_heap_strndup(mi_heap_t* heap, const char* s, size_t n) mi_attr_noexcept { + if (s == NULL) return NULL; + const size_t len = _mi_strnlen(s,n); // len <= n + char* t = (char*)mi_heap_malloc(heap, len+1); + if (t == NULL) return NULL; + _mi_memcpy(t, s, len); + t[len] = 0; + return t; +} + +mi_decl_nodiscard mi_decl_restrict char* mi_strndup(const char* s, size_t n) mi_attr_noexcept { + return mi_heap_strndup(mi_prim_get_default_heap(),s,n); +} + +#ifndef __wasi__ +// `realpath` using mi_malloc +#ifdef _WIN32 +#ifndef PATH_MAX +#define PATH_MAX MAX_PATH +#endif + +mi_decl_nodiscard mi_decl_restrict char* mi_heap_realpath(mi_heap_t* heap, const char* fname, char* resolved_name) mi_attr_noexcept { + // todo: use GetFullPathNameW to allow longer file names + char buf[PATH_MAX]; + DWORD res = GetFullPathNameA(fname, PATH_MAX, (resolved_name == NULL ? buf : resolved_name), NULL); + if (res == 0) { + errno = GetLastError(); return NULL; + } + else if (res > PATH_MAX) { + errno = EINVAL; return NULL; + } + else if (resolved_name != NULL) { + return resolved_name; + } + else { + return mi_heap_strndup(heap, buf, PATH_MAX); + } +} +#else +/* +#include // pathconf +static size_t mi_path_max(void) { + static size_t path_max = 0; + if (path_max <= 0) { + long m = pathconf("/",_PC_PATH_MAX); + if (m <= 0) path_max = 4096; // guess + else if (m < 256) path_max = 256; // at least 256 + else path_max = m; + } + return path_max; +} +*/ +char* mi_heap_realpath(mi_heap_t* heap, const char* fname, char* resolved_name) mi_attr_noexcept { + if (resolved_name != NULL) { + return realpath(fname,resolved_name); + } + else { + char* rname = realpath(fname, NULL); + if (rname == NULL) return NULL; + char* result = mi_heap_strdup(heap, rname); + mi_cfree(rname); // use checked free (which may be redirected to our free but that's ok) + // note: with ASAN realpath is intercepted and mi_cfree may leak the returned pointer :-( + return result; + } + /* + const size_t n = mi_path_max(); + char* buf = (char*)mi_malloc(n+1); + if (buf == NULL) { + errno = ENOMEM; + return NULL; + } + char* rname = realpath(fname,buf); + char* result = mi_heap_strndup(heap,rname,n); // ok if `rname==NULL` + mi_free(buf); + return result; + } + */ +} +#endif + +mi_decl_nodiscard mi_decl_restrict char* mi_realpath(const char* fname, char* resolved_name) mi_attr_noexcept { + return mi_heap_realpath(mi_prim_get_default_heap(),fname,resolved_name); +} +#endif + +/*------------------------------------------------------- +C++ new and new_aligned +The standard requires calling into `get_new_handler` and +throwing the bad_alloc exception on failure. If we compile +with a C++ compiler we can implement this precisely. If we +use a C compiler we cannot throw a `bad_alloc` exception +but we call `exit` instead (i.e. not returning). +-------------------------------------------------------*/ + +#ifdef __cplusplus +#include +static bool mi_try_new_handler(bool nothrow) { + #if defined(_MSC_VER) || (__cplusplus >= 201103L) + std::new_handler h = std::get_new_handler(); + #else + std::new_handler h = std::set_new_handler(); + std::set_new_handler(h); + #endif + if (h==NULL) { + _mi_error_message(ENOMEM, "out of memory in 'new'"); + #if defined(_CPPUNWIND) || defined(__cpp_exceptions) // exceptions are not always enabled + if (!nothrow) { + throw std::bad_alloc(); + } + #else + MI_UNUSED(nothrow); + #endif + return false; + } + else { + h(); + return true; + } +} +#else +typedef void (*std_new_handler_t)(void); + +#if (defined(__GNUC__) || (defined(__clang__) && !defined(_MSC_VER))) // exclude clang-cl, see issue #631 +std_new_handler_t __attribute__((weak)) _ZSt15get_new_handlerv(void) { + return NULL; +} +static std_new_handler_t mi_get_new_handler(void) { + return _ZSt15get_new_handlerv(); +} +#else +// note: on windows we could dynamically link to `?get_new_handler@std@@YAP6AXXZXZ`. +static std_new_handler_t mi_get_new_handler() { + return NULL; +} +#endif + +static bool mi_try_new_handler(bool nothrow) { + std_new_handler_t h = mi_get_new_handler(); + if (h==NULL) { + _mi_error_message(ENOMEM, "out of memory in 'new'"); + if (!nothrow) { + abort(); // cannot throw in plain C, use abort + } + return false; + } + else { + h(); + return true; + } +} +#endif + +mi_decl_export mi_decl_noinline void* mi_heap_try_new(mi_heap_t* heap, size_t size, bool nothrow ) { + void* p = NULL; + while(p == NULL && mi_try_new_handler(nothrow)) { + p = mi_heap_malloc(heap,size); + } + return p; +} + +static mi_decl_noinline void* mi_try_new(size_t size, bool nothrow) { + return mi_heap_try_new(mi_prim_get_default_heap(), size, nothrow); +} + + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_alloc_new(mi_heap_t* heap, size_t size) { + void* p = mi_heap_malloc(heap,size); + if mi_unlikely(p == NULL) return mi_heap_try_new(heap, size, false); + return p; +} + +mi_decl_nodiscard mi_decl_restrict void* mi_new(size_t size) { + return mi_heap_alloc_new(mi_prim_get_default_heap(), size); +} + + +mi_decl_nodiscard mi_decl_restrict void* mi_heap_alloc_new_n(mi_heap_t* heap, size_t count, size_t size) { + size_t total; + if mi_unlikely(mi_count_size_overflow(count, size, &total)) { + mi_try_new_handler(false); // on overflow we invoke the try_new_handler once to potentially throw std::bad_alloc + return NULL; + } + else { + return mi_heap_alloc_new(heap,total); + } +} + +mi_decl_nodiscard mi_decl_restrict void* mi_new_n(size_t count, size_t size) { + return mi_heap_alloc_new_n(mi_prim_get_default_heap(), count, size); +} + + +mi_decl_nodiscard mi_decl_restrict void* mi_new_nothrow(size_t size) mi_attr_noexcept { + void* p = mi_malloc(size); + if mi_unlikely(p == NULL) return mi_try_new(size, true); + return p; +} + +mi_decl_nodiscard mi_decl_restrict void* mi_new_aligned(size_t size, size_t alignment) { + void* p; + do { + p = mi_malloc_aligned(size, alignment); + } + while(p == NULL && mi_try_new_handler(false)); + return p; +} + +mi_decl_nodiscard mi_decl_restrict void* mi_new_aligned_nothrow(size_t size, size_t alignment) mi_attr_noexcept { + void* p; + do { + p = mi_malloc_aligned(size, alignment); + } + while(p == NULL && mi_try_new_handler(true)); + return p; +} + +mi_decl_nodiscard void* mi_new_realloc(void* p, size_t newsize) { + void* q; + do { + q = mi_realloc(p, newsize); + } while (q == NULL && mi_try_new_handler(false)); + return q; +} + +mi_decl_nodiscard void* mi_new_reallocn(void* p, size_t newcount, size_t size) { + size_t total; + if mi_unlikely(mi_count_size_overflow(newcount, size, &total)) { + mi_try_new_handler(false); // on overflow we invoke the try_new_handler once to potentially throw std::bad_alloc + return NULL; + } + else { + return mi_new_realloc(p, total); + } +} + +#if MI_GUARDED +// We always allocate a guarded allocation at an offset (`mi_page_has_aligned` will be true). +// We then set the first word of the block to `0` for regular offset aligned allocations (in `alloc-aligned.c`) +// and the first word to `~0` for guarded allocations to have a correct `mi_usable_size` + +static void* mi_block_ptr_set_guarded(mi_block_t* block, size_t obj_size) { + // TODO: we can still make padding work by moving it out of the guard page area + mi_page_t* const page = _mi_ptr_page(block); + mi_page_set_has_aligned(page, true); + block->next = MI_BLOCK_TAG_GUARDED; + + // set guard page at the end of the block + mi_segment_t* const segment = _mi_page_segment(page); + const size_t block_size = mi_page_block_size(page); // must use `block_size` to match `mi_free_local` + const size_t os_page_size = _mi_os_page_size(); + mi_assert_internal(block_size >= obj_size + os_page_size + sizeof(mi_block_t)); + if (block_size < obj_size + os_page_size + sizeof(mi_block_t)) { + // should never happen + mi_free(block); + return NULL; + } + uint8_t* guard_page = (uint8_t*)block + block_size - os_page_size; + mi_assert_internal(_mi_is_aligned(guard_page, os_page_size)); + if (segment->allow_decommit && _mi_is_aligned(guard_page, os_page_size)) { + _mi_os_protect(guard_page, os_page_size); + } + else { + _mi_warning_message("unable to set a guard page behind an object due to pinned memory (large OS pages?) (object %p of size %zu)\n", block, block_size); + } + + // align pointer just in front of the guard page + size_t offset = block_size - os_page_size - obj_size; + mi_assert_internal(offset > sizeof(mi_block_t)); + if (offset > MI_BLOCK_ALIGNMENT_MAX) { + // give up to place it right in front of the guard page if the offset is too large for unalignment + offset = MI_BLOCK_ALIGNMENT_MAX; + } + void* p = (uint8_t*)block + offset; + mi_track_align(block, p, offset, obj_size); + mi_track_mem_defined(block, sizeof(mi_block_t)); + return p; +} + +mi_decl_restrict void* _mi_heap_malloc_guarded(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept +{ + #if defined(MI_PADDING_SIZE) + mi_assert(MI_PADDING_SIZE==0); + #endif + // allocate multiple of page size ending in a guard page + // ensure minimal alignment requirement? + const size_t os_page_size = _mi_os_page_size(); + const size_t obj_size = (mi_option_is_enabled(mi_option_guarded_precise) ? size : _mi_align_up(size, MI_MAX_ALIGN_SIZE)); + const size_t bsize = _mi_align_up(_mi_align_up(obj_size, MI_MAX_ALIGN_SIZE) + sizeof(mi_block_t), MI_MAX_ALIGN_SIZE); + const size_t req_size = _mi_align_up(bsize + os_page_size, os_page_size); + mi_block_t* const block = (mi_block_t*)_mi_malloc_generic(heap, req_size, zero, 0 /* huge_alignment */); + if (block==NULL) return NULL; + void* const p = mi_block_ptr_set_guarded(block, obj_size); + + // stats + mi_track_malloc(p, size, zero); + if (p != NULL) { + if (!mi_heap_is_initialized(heap)) { heap = mi_prim_get_default_heap(); } + #if MI_STAT>1 + mi_heap_stat_adjust_decrease(heap, malloc_requested, req_size); + mi_heap_stat_increase(heap, malloc_requested, size); + #endif + _mi_stat_counter_increase(&heap->tld->stats.malloc_guarded_count, 1); + } + #if MI_DEBUG>3 + if (p != NULL && zero) { + mi_assert_expensive(mi_mem_is_zero(p, size)); + } + #endif + return p; +} +#endif + +// ------------------------------------------------------ +// ensure explicit external inline definitions are emitted! +// ------------------------------------------------------ + +#ifdef __cplusplus +void* _mi_externs[] = { + (void*)&_mi_page_malloc, + (void*)&_mi_page_malloc_zero, + (void*)&_mi_heap_malloc_zero, + (void*)&_mi_heap_malloc_zero_ex, + (void*)&mi_malloc, + (void*)&mi_malloc_small, + (void*)&mi_zalloc_small, + (void*)&mi_heap_malloc, + (void*)&mi_heap_zalloc, + (void*)&mi_heap_malloc_small, + // (void*)&mi_heap_alloc_new, + // (void*)&mi_heap_alloc_new_n +}; +#endif diff --git a/compat/mimalloc/arena-abandon.c b/compat/mimalloc/arena-abandon.c new file mode 100644 index 00000000000000..460c80fc22782f --- /dev/null +++ b/compat/mimalloc/arena-abandon.c @@ -0,0 +1,346 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +#if !defined(MI_IN_ARENA_C) +#error "this file should be included from 'arena.c' (so mi_arena_t is visible)" +// add includes help an IDE +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "bitmap.h" +#endif + +// Minimal exports for arena-abandoned. +size_t mi_arena_id_index(mi_arena_id_t id); +mi_arena_t* mi_arena_from_index(size_t idx); +size_t mi_arena_get_count(void); +void* mi_arena_block_start(mi_arena_t* arena, mi_bitmap_index_t bindex); +bool mi_arena_memid_indices(mi_memid_t memid, size_t* arena_index, mi_bitmap_index_t* bitmap_index); + +/* ----------------------------------------------------------- + Abandoned blocks/segments: + + _mi_arena_segment_clear_abandoned + _mi_arena_segment_mark_abandoned + + This is used to atomically abandon/reclaim segments + (and crosses the arena API but it is convenient to have here). + + Abandoned segments still have live blocks; they get reclaimed + when a thread frees a block in it, or when a thread needs a fresh + segment. + + Abandoned segments are atomically marked in the `block_abandoned` + bitmap of arenas. Any segments allocated outside arenas are put + in the sub-process `abandoned_os_list`. This list is accessed + using locks but this should be uncommon and generally uncontended. + Reclaim and visiting either scan through the `block_abandoned` + bitmaps of the arena's, or visit the `abandoned_os_list` + + A potentially nicer design is to use arena's for everything + and perhaps have virtual arena's to map OS allocated memory + but this would lack the "density" of our current arena's. TBC. +----------------------------------------------------------- */ + + +// reclaim a specific OS abandoned segment; `true` on success. +// sets the thread_id. +static bool mi_arena_segment_os_clear_abandoned(mi_segment_t* segment, bool take_lock) { + mi_assert(segment->memid.memkind != MI_MEM_ARENA); + // not in an arena, remove from list of abandoned os segments + mi_subproc_t* const subproc = segment->subproc; + if (take_lock && !mi_lock_try_acquire(&subproc->abandoned_os_lock)) { + return false; // failed to acquire the lock, we just give up + } + // remove atomically from the abandoned os list (if possible!) + bool reclaimed = false; + mi_segment_t* const next = segment->abandoned_os_next; + mi_segment_t* const prev = segment->abandoned_os_prev; + if (next != NULL || prev != NULL || subproc->abandoned_os_list == segment) { + #if MI_DEBUG>3 + // find ourselves in the abandoned list (and check the count) + bool found = false; + size_t count = 0; + for (mi_segment_t* current = subproc->abandoned_os_list; current != NULL; current = current->abandoned_os_next) { + if (current == segment) { found = true; } + count++; + } + mi_assert_internal(found); + mi_assert_internal(count == mi_atomic_load_relaxed(&subproc->abandoned_os_list_count)); + #endif + // remove (atomically) from the list and reclaim + if (prev != NULL) { prev->abandoned_os_next = next; } + else { subproc->abandoned_os_list = next; } + if (next != NULL) { next->abandoned_os_prev = prev; } + else { subproc->abandoned_os_list_tail = prev; } + segment->abandoned_os_next = NULL; + segment->abandoned_os_prev = NULL; + mi_atomic_decrement_relaxed(&subproc->abandoned_count); + mi_atomic_decrement_relaxed(&subproc->abandoned_os_list_count); + if (take_lock) { // don't reset the thread_id when iterating + mi_atomic_store_release(&segment->thread_id, _mi_thread_id()); + } + reclaimed = true; + } + if (take_lock) { mi_lock_release(&segment->subproc->abandoned_os_lock); } + return reclaimed; +} + +// reclaim a specific abandoned segment; `true` on success. +// sets the thread_id. +bool _mi_arena_segment_clear_abandoned(mi_segment_t* segment) { + if mi_unlikely(segment->memid.memkind != MI_MEM_ARENA) { + return mi_arena_segment_os_clear_abandoned(segment, true /* take lock */); + } + // arena segment: use the blocks_abandoned bitmap. + size_t arena_idx; + size_t bitmap_idx; + mi_arena_memid_indices(segment->memid, &arena_idx, &bitmap_idx); + mi_arena_t* arena = mi_arena_from_index(arena_idx); + mi_assert_internal(arena != NULL); + // reclaim atomically + bool was_marked = _mi_bitmap_unclaim(arena->blocks_abandoned, arena->field_count, 1, bitmap_idx); + if (was_marked) { + mi_assert_internal(mi_atomic_load_acquire(&segment->thread_id) == 0); + mi_atomic_decrement_relaxed(&segment->subproc->abandoned_count); + mi_atomic_store_release(&segment->thread_id, _mi_thread_id()); + } + // mi_assert_internal(was_marked); + mi_assert_internal(!was_marked || _mi_bitmap_is_claimed(arena->blocks_inuse, arena->field_count, 1, bitmap_idx)); + //mi_assert_internal(arena->blocks_committed == NULL || _mi_bitmap_is_claimed(arena->blocks_committed, arena->field_count, 1, bitmap_idx)); + return was_marked; +} + + +// mark a specific OS segment as abandoned +static void mi_arena_segment_os_mark_abandoned(mi_segment_t* segment) { + mi_assert(segment->memid.memkind != MI_MEM_ARENA); + // not in an arena; we use a list of abandoned segments + mi_subproc_t* const subproc = segment->subproc; + mi_lock(&subproc->abandoned_os_lock) { + // push on the tail of the list (important for the visitor) + mi_segment_t* prev = subproc->abandoned_os_list_tail; + mi_assert_internal(prev == NULL || prev->abandoned_os_next == NULL); + mi_assert_internal(segment->abandoned_os_prev == NULL); + mi_assert_internal(segment->abandoned_os_next == NULL); + if (prev != NULL) { prev->abandoned_os_next = segment; } + else { subproc->abandoned_os_list = segment; } + subproc->abandoned_os_list_tail = segment; + segment->abandoned_os_prev = prev; + segment->abandoned_os_next = NULL; + mi_atomic_increment_relaxed(&subproc->abandoned_os_list_count); + mi_atomic_increment_relaxed(&subproc->abandoned_count); + // and release the lock + } + return; +} + +// mark a specific segment as abandoned +// clears the thread_id. +void _mi_arena_segment_mark_abandoned(mi_segment_t* segment) +{ + mi_assert_internal(segment->used == segment->abandoned); + mi_atomic_store_release(&segment->thread_id, (uintptr_t)0); // mark as abandoned for multi-thread free's + if mi_unlikely(segment->memid.memkind != MI_MEM_ARENA) { + mi_arena_segment_os_mark_abandoned(segment); + return; + } + // segment is in an arena, mark it in the arena `blocks_abandoned` bitmap + size_t arena_idx; + size_t bitmap_idx; + mi_arena_memid_indices(segment->memid, &arena_idx, &bitmap_idx); + mi_arena_t* arena = mi_arena_from_index(arena_idx); + mi_assert_internal(arena != NULL); + // set abandonment atomically + mi_subproc_t* const subproc = segment->subproc; // don't access the segment after setting it abandoned + const bool was_unmarked = _mi_bitmap_claim(arena->blocks_abandoned, arena->field_count, 1, bitmap_idx, NULL); + if (was_unmarked) { mi_atomic_increment_relaxed(&subproc->abandoned_count); } + mi_assert_internal(was_unmarked); + mi_assert_internal(_mi_bitmap_is_claimed(arena->blocks_inuse, arena->field_count, 1, bitmap_idx)); +} + + +/* ----------------------------------------------------------- + Iterate through the abandoned blocks/segments using a cursor. + This is used for reclaiming and abandoned block visiting. +----------------------------------------------------------- */ + +// start a cursor at a randomized arena +void _mi_arena_field_cursor_init(mi_heap_t* heap, mi_subproc_t* subproc, bool visit_all, mi_arena_field_cursor_t* current) { + mi_assert_internal(heap == NULL || heap->tld->segments.subproc == subproc); + current->bitmap_idx = 0; + current->subproc = subproc; + current->visit_all = visit_all; + current->hold_visit_lock = false; + const size_t abandoned_count = mi_atomic_load_relaxed(&subproc->abandoned_count); + const size_t abandoned_list_count = mi_atomic_load_relaxed(&subproc->abandoned_os_list_count); + const size_t max_arena = mi_arena_get_count(); + if (heap != NULL && heap->arena_id != _mi_arena_id_none()) { + // for a heap that is bound to one arena, only visit that arena + current->start = mi_arena_id_index(heap->arena_id); + current->end = current->start + 1; + current->os_list_count = 0; + } + else { + // otherwise visit all starting at a random location + if (abandoned_count > abandoned_list_count && max_arena > 0) { + current->start = (heap == NULL || max_arena == 0 ? 0 : (mi_arena_id_t)(_mi_heap_random_next(heap) % max_arena)); + current->end = current->start + max_arena; + } + else { + current->start = 0; + current->end = 0; + } + current->os_list_count = abandoned_list_count; // max entries to visit in the os abandoned list + } + mi_assert_internal(current->start <= max_arena); +} + +void _mi_arena_field_cursor_done(mi_arena_field_cursor_t* current) { + if (current->hold_visit_lock) { + mi_lock_release(¤t->subproc->abandoned_os_visit_lock); + current->hold_visit_lock = false; + } +} + +static mi_segment_t* mi_arena_segment_clear_abandoned_at(mi_arena_t* arena, mi_subproc_t* subproc, mi_bitmap_index_t bitmap_idx) { + // try to reclaim an abandoned segment in the arena atomically + if (!_mi_bitmap_unclaim(arena->blocks_abandoned, arena->field_count, 1, bitmap_idx)) return NULL; + mi_assert_internal(_mi_bitmap_is_claimed(arena->blocks_inuse, arena->field_count, 1, bitmap_idx)); + mi_segment_t* segment = (mi_segment_t*)mi_arena_block_start(arena, bitmap_idx); + mi_assert_internal(mi_atomic_load_relaxed(&segment->thread_id) == 0); + // check that the segment belongs to our sub-process + // note: this is the reason we need the `abandoned_visit` lock in the case abandoned visiting is enabled. + // without the lock an abandoned visit may otherwise fail to visit all abandoned segments in the sub-process. + // for regular reclaim it is fine to miss one sometimes so without abandoned visiting we don't need the `abandoned_visit` lock. + if (segment->subproc != subproc) { + // it is from another sub-process, re-mark it and continue searching + const bool was_zero = _mi_bitmap_claim(arena->blocks_abandoned, arena->field_count, 1, bitmap_idx, NULL); + mi_assert_internal(was_zero); MI_UNUSED(was_zero); + return NULL; + } + else { + // success, we unabandoned a segment in our sub-process + mi_atomic_decrement_relaxed(&subproc->abandoned_count); + return segment; + } +} + +static mi_segment_t* mi_arena_segment_clear_abandoned_next_field(mi_arena_field_cursor_t* previous) { + const size_t max_arena = mi_arena_get_count(); + size_t field_idx = mi_bitmap_index_field(previous->bitmap_idx); + size_t bit_idx = mi_bitmap_index_bit_in_field(previous->bitmap_idx); + // visit arena's (from the previous cursor) + for (; previous->start < previous->end; previous->start++, field_idx = 0, bit_idx = 0) { + // index wraps around + size_t arena_idx = (previous->start >= max_arena ? previous->start % max_arena : previous->start); + mi_arena_t* arena = mi_arena_from_index(arena_idx); + if (arena != NULL) { + bool has_lock = false; + // visit the abandoned fields (starting at previous_idx) + for (; field_idx < arena->field_count; field_idx++, bit_idx = 0) { + size_t field = mi_atomic_load_relaxed(&arena->blocks_abandoned[field_idx]); + if mi_unlikely(field != 0) { // skip zero fields quickly + // we only take the arena lock if there are actually abandoned segments present + if (!has_lock && mi_option_is_enabled(mi_option_visit_abandoned)) { + has_lock = (previous->visit_all ? (mi_lock_acquire(&arena->abandoned_visit_lock),true) : mi_lock_try_acquire(&arena->abandoned_visit_lock)); + if (!has_lock) { + if (previous->visit_all) { + _mi_error_message(EFAULT, "internal error: failed to visit all abandoned segments due to failure to acquire the visitor lock"); + } + // skip to next arena + break; + } + } + mi_assert_internal(has_lock || !mi_option_is_enabled(mi_option_visit_abandoned)); + // visit each set bit in the field (todo: maybe use `ctz` here?) + for (; bit_idx < MI_BITMAP_FIELD_BITS; bit_idx++) { + // pre-check if the bit is set + size_t mask = ((size_t)1 << bit_idx); + if mi_unlikely((field & mask) == mask) { + mi_bitmap_index_t bitmap_idx = mi_bitmap_index_create(field_idx, bit_idx); + mi_segment_t* const segment = mi_arena_segment_clear_abandoned_at(arena, previous->subproc, bitmap_idx); + if (segment != NULL) { + //mi_assert_internal(arena->blocks_committed == NULL || _mi_bitmap_is_claimed(arena->blocks_committed, arena->field_count, 1, bitmap_idx)); + if (has_lock) { mi_lock_release(&arena->abandoned_visit_lock); } + previous->bitmap_idx = mi_bitmap_index_create_ex(field_idx, bit_idx + 1); // start at next one for the next iteration + return segment; + } + } + } + } + } + if (has_lock) { mi_lock_release(&arena->abandoned_visit_lock); } + } + } + return NULL; +} + +static mi_segment_t* mi_arena_segment_clear_abandoned_next_list(mi_arena_field_cursor_t* previous) { + // go through the abandoned_os_list + // we only allow one thread per sub-process to do to visit guarded by the `abandoned_os_visit_lock`. + // The lock is released when the cursor is released. + if (!previous->hold_visit_lock) { + previous->hold_visit_lock = (previous->visit_all ? (mi_lock_acquire(&previous->subproc->abandoned_os_visit_lock),true) + : mi_lock_try_acquire(&previous->subproc->abandoned_os_visit_lock)); + if (!previous->hold_visit_lock) { + if (previous->visit_all) { + _mi_error_message(EFAULT, "internal error: failed to visit all abandoned segments due to failure to acquire the OS visitor lock"); + } + return NULL; // we cannot get the lock, give up + } + } + // One list entry at a time + while (previous->os_list_count > 0) { + previous->os_list_count--; + mi_lock_acquire(&previous->subproc->abandoned_os_lock); // this could contend with concurrent OS block abandonment and reclaim from `free` + mi_segment_t* segment = previous->subproc->abandoned_os_list; + // pop from head of the list, a subsequent mark will push at the end (and thus we iterate through os_list_count entries) + if (segment == NULL || mi_arena_segment_os_clear_abandoned(segment, false /* we already have the lock */)) { + mi_lock_release(&previous->subproc->abandoned_os_lock); + return segment; + } + // already abandoned, try again + mi_lock_release(&previous->subproc->abandoned_os_lock); + } + // done + mi_assert_internal(previous->os_list_count == 0); + return NULL; +} + + +// reclaim abandoned segments +// this does not set the thread id (so it appears as still abandoned) +mi_segment_t* _mi_arena_segment_clear_abandoned_next(mi_arena_field_cursor_t* previous) { + if (previous->start < previous->end) { + // walk the arena + mi_segment_t* segment = mi_arena_segment_clear_abandoned_next_field(previous); + if (segment != NULL) { return segment; } + } + // no entries in the arena's anymore, walk the abandoned OS list + mi_assert_internal(previous->start == previous->end); + return mi_arena_segment_clear_abandoned_next_list(previous); +} + + +bool mi_abandoned_visit_blocks(mi_subproc_id_t subproc_id, int heap_tag, bool visit_blocks, mi_block_visit_fun* visitor, void* arg) { + // (unfortunately) the visit_abandoned option must be enabled from the start. + // This is to avoid taking locks if abandoned list visiting is not required (as for most programs) + if (!mi_option_is_enabled(mi_option_visit_abandoned)) { + _mi_error_message(EFAULT, "internal error: can only visit abandoned blocks when MIMALLOC_VISIT_ABANDONED=ON"); + return false; + } + mi_arena_field_cursor_t current; + _mi_arena_field_cursor_init(NULL, _mi_subproc_from_id(subproc_id), true /* visit all (blocking) */, ¤t); + mi_segment_t* segment; + bool ok = true; + while (ok && (segment = _mi_arena_segment_clear_abandoned_next(¤t)) != NULL) { + ok = _mi_segment_visit_blocks(segment, heap_tag, visit_blocks, visitor, arg); + _mi_arena_segment_mark_abandoned(segment); + } + _mi_arena_field_cursor_done(¤t); + return ok; +} diff --git a/compat/mimalloc/arena.c b/compat/mimalloc/arena.c new file mode 100644 index 00000000000000..e97ca885fed86a --- /dev/null +++ b/compat/mimalloc/arena.c @@ -0,0 +1,1043 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ---------------------------------------------------------------------------- +"Arenas" are fixed area's of OS memory from which we can allocate +large blocks (>= MI_ARENA_MIN_BLOCK_SIZE, 4MiB). +In contrast to the rest of mimalloc, the arenas are shared between +threads and need to be accessed using atomic operations. + +Arenas are also used to for huge OS page (1GiB) reservations or for reserving +OS memory upfront which can be improve performance or is sometimes needed +on embedded devices. We can also employ this with WASI or `sbrk` systems +to reserve large arenas upfront and be able to reuse the memory more effectively. + +The arena allocation needs to be thread safe and we use an atomic bitmap to allocate. +-----------------------------------------------------------------------------*/ + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "bitmap.h" + + +/* ----------------------------------------------------------- + Arena allocation +----------------------------------------------------------- */ + +// A memory arena descriptor +typedef struct mi_arena_s { + mi_arena_id_t id; // arena id; 0 for non-specific + mi_memid_t memid; // memid of the memory area + _Atomic(uint8_t*) start; // the start of the memory area + size_t block_count; // size of the area in arena blocks (of `MI_ARENA_BLOCK_SIZE`) + size_t field_count; // number of bitmap fields (where `field_count * MI_BITMAP_FIELD_BITS >= block_count`) + size_t meta_size; // size of the arena structure itself (including its bitmaps) + mi_memid_t meta_memid; // memid of the arena structure itself (OS or static allocation) + int numa_node; // associated NUMA node + bool exclusive; // only allow allocations if specifically for this arena + bool is_large; // memory area consists of large- or huge OS pages (always committed) + mi_lock_t abandoned_visit_lock; // lock is only used when abandoned segments are being visited + _Atomic(size_t) search_idx; // optimization to start the search for free blocks + _Atomic(mi_msecs_t) purge_expire; // expiration time when blocks should be purged from `blocks_purge`. + + mi_bitmap_field_t* blocks_dirty; // are the blocks potentially non-zero? + mi_bitmap_field_t* blocks_committed; // are the blocks committed? (can be NULL for memory that cannot be decommitted) + mi_bitmap_field_t* blocks_purge; // blocks that can be (reset) decommitted. (can be NULL for memory that cannot be (reset) decommitted) + mi_bitmap_field_t* blocks_abandoned; // blocks that start with an abandoned segment. (This crosses API's but it is convenient to have here) + mi_bitmap_field_t blocks_inuse[1]; // in-place bitmap of in-use blocks (of size `field_count`) + // do not add further fields here as the dirty, committed, purged, and abandoned bitmaps follow the inuse bitmap fields. +} mi_arena_t; + + +#define MI_ARENA_BLOCK_SIZE (MI_SEGMENT_SIZE) // 64MiB (must be at least MI_SEGMENT_ALIGN) +#define MI_ARENA_MIN_OBJ_SIZE (MI_ARENA_BLOCK_SIZE/2) // 32MiB +#define MI_MAX_ARENAS (132) // Limited as the reservation exponentially increases (and takes up .bss) + +// The available arenas +static mi_decl_cache_align _Atomic(mi_arena_t*) mi_arenas[MI_MAX_ARENAS]; +static mi_decl_cache_align _Atomic(size_t) mi_arena_count; // = 0 +static mi_decl_cache_align _Atomic(int64_t) mi_arenas_purge_expire; // set if there exist purgeable arenas + +#define MI_IN_ARENA_C +#include "arena-abandon.c" +#undef MI_IN_ARENA_C + +/* ----------------------------------------------------------- + Arena id's + id = arena_index + 1 +----------------------------------------------------------- */ + +size_t mi_arena_id_index(mi_arena_id_t id) { + return (size_t)(id <= 0 ? MI_MAX_ARENAS : id - 1); +} + +static mi_arena_id_t mi_arena_id_create(size_t arena_index) { + mi_assert_internal(arena_index < MI_MAX_ARENAS); + return (int)arena_index + 1; +} + +mi_arena_id_t _mi_arena_id_none(void) { + return 0; +} + +static bool mi_arena_id_is_suitable(mi_arena_id_t arena_id, bool arena_is_exclusive, mi_arena_id_t req_arena_id) { + return ((!arena_is_exclusive && req_arena_id == _mi_arena_id_none()) || + (arena_id == req_arena_id)); +} + +bool _mi_arena_memid_is_suitable(mi_memid_t memid, mi_arena_id_t request_arena_id) { + if (memid.memkind == MI_MEM_ARENA) { + return mi_arena_id_is_suitable(memid.mem.arena.id, memid.mem.arena.is_exclusive, request_arena_id); + } + else { + return mi_arena_id_is_suitable(_mi_arena_id_none(), false, request_arena_id); + } +} + +bool _mi_arena_memid_is_os_allocated(mi_memid_t memid) { + return (memid.memkind == MI_MEM_OS); +} + +size_t mi_arena_get_count(void) { + return mi_atomic_load_relaxed(&mi_arena_count); +} + +mi_arena_t* mi_arena_from_index(size_t idx) { + mi_assert_internal(idx < mi_arena_get_count()); + return mi_atomic_load_ptr_acquire(mi_arena_t, &mi_arenas[idx]); +} + + +/* ----------------------------------------------------------- + Arena allocations get a (currently) 16-bit memory id where the + lower 8 bits are the arena id, and the upper bits the block index. +----------------------------------------------------------- */ + +static size_t mi_block_count_of_size(size_t size) { + return _mi_divide_up(size, MI_ARENA_BLOCK_SIZE); +} + +static size_t mi_arena_block_size(size_t bcount) { + return (bcount * MI_ARENA_BLOCK_SIZE); +} + +static size_t mi_arena_size(mi_arena_t* arena) { + return mi_arena_block_size(arena->block_count); +} + +static mi_memid_t mi_memid_create_arena(mi_arena_id_t id, bool is_exclusive, mi_bitmap_index_t bitmap_index) { + mi_memid_t memid = _mi_memid_create(MI_MEM_ARENA); + memid.mem.arena.id = id; + memid.mem.arena.block_index = bitmap_index; + memid.mem.arena.is_exclusive = is_exclusive; + return memid; +} + +bool mi_arena_memid_indices(mi_memid_t memid, size_t* arena_index, mi_bitmap_index_t* bitmap_index) { + mi_assert_internal(memid.memkind == MI_MEM_ARENA); + *arena_index = mi_arena_id_index(memid.mem.arena.id); + *bitmap_index = memid.mem.arena.block_index; + return memid.mem.arena.is_exclusive; +} + + + +/* ----------------------------------------------------------- + Special static area for mimalloc internal structures + to avoid OS calls (for example, for the arena metadata (~= 256b)) +----------------------------------------------------------- */ + +#define MI_ARENA_STATIC_MAX ((MI_INTPTR_SIZE/2)*MI_KiB) // 4 KiB on 64-bit + +static mi_decl_cache_align uint8_t mi_arena_static[MI_ARENA_STATIC_MAX]; // must be cache aligned, see issue #895 +static mi_decl_cache_align _Atomic(size_t) mi_arena_static_top; + +static void* mi_arena_static_zalloc(size_t size, size_t alignment, mi_memid_t* memid) { + *memid = _mi_memid_none(); + if (size == 0 || size > MI_ARENA_STATIC_MAX) return NULL; + const size_t toplow = mi_atomic_load_relaxed(&mi_arena_static_top); + if ((toplow + size) > MI_ARENA_STATIC_MAX) return NULL; + + // try to claim space + if (alignment < MI_MAX_ALIGN_SIZE) { alignment = MI_MAX_ALIGN_SIZE; } + const size_t oversize = size + alignment - 1; + if (toplow + oversize > MI_ARENA_STATIC_MAX) return NULL; + const size_t oldtop = mi_atomic_add_acq_rel(&mi_arena_static_top, oversize); + size_t top = oldtop + oversize; + if (top > MI_ARENA_STATIC_MAX) { + // try to roll back, ok if this fails + mi_atomic_cas_strong_acq_rel(&mi_arena_static_top, &top, oldtop); + return NULL; + } + + // success + *memid = _mi_memid_create(MI_MEM_STATIC); + memid->initially_zero = true; + const size_t start = _mi_align_up(oldtop, alignment); + uint8_t* const p = &mi_arena_static[start]; + _mi_memzero_aligned(p, size); + return p; +} + +void* _mi_arena_meta_zalloc(size_t size, mi_memid_t* memid) { + *memid = _mi_memid_none(); + + // try static + void* p = mi_arena_static_zalloc(size, MI_MAX_ALIGN_SIZE, memid); + if (p != NULL) return p; + + // or fall back to the OS + p = _mi_os_zalloc(size, memid); + if (p == NULL) return NULL; + + return p; +} + +void _mi_arena_meta_free(void* p, mi_memid_t memid, size_t size) { + if (mi_memkind_is_os(memid.memkind)) { + _mi_os_free(p, size, memid); + } + else { + mi_assert(memid.memkind == MI_MEM_STATIC); + } +} + +void* mi_arena_block_start(mi_arena_t* arena, mi_bitmap_index_t bindex) { + return (arena->start + mi_arena_block_size(mi_bitmap_index_bit(bindex))); +} + + +/* ----------------------------------------------------------- + Thread safe allocation in an arena +----------------------------------------------------------- */ + +// claim the `blocks_inuse` bits +static bool mi_arena_try_claim(mi_arena_t* arena, size_t blocks, mi_bitmap_index_t* bitmap_idx) +{ + size_t idx = 0; // mi_atomic_load_relaxed(&arena->search_idx); // start from last search; ok to be relaxed as the exact start does not matter + if (_mi_bitmap_try_find_from_claim_across(arena->blocks_inuse, arena->field_count, idx, blocks, bitmap_idx)) { + mi_atomic_store_relaxed(&arena->search_idx, mi_bitmap_index_field(*bitmap_idx)); // start search from found location next time around + return true; + }; + return false; +} + + +/* ----------------------------------------------------------- + Arena Allocation +----------------------------------------------------------- */ + +static mi_decl_noinline void* mi_arena_try_alloc_at(mi_arena_t* arena, size_t arena_index, size_t needed_bcount, + bool commit, mi_memid_t* memid) +{ + MI_UNUSED(arena_index); + mi_assert_internal(mi_arena_id_index(arena->id) == arena_index); + + mi_bitmap_index_t bitmap_index; + if (!mi_arena_try_claim(arena, needed_bcount, &bitmap_index)) return NULL; + + // claimed it! + void* p = mi_arena_block_start(arena, bitmap_index); + *memid = mi_memid_create_arena(arena->id, arena->exclusive, bitmap_index); + memid->is_pinned = arena->memid.is_pinned; + + // none of the claimed blocks should be scheduled for a decommit + if (arena->blocks_purge != NULL) { + // this is thread safe as a potential purge only decommits parts that are not yet claimed as used (in `blocks_inuse`). + _mi_bitmap_unclaim_across(arena->blocks_purge, arena->field_count, needed_bcount, bitmap_index); + } + + // set the dirty bits (todo: no need for an atomic op here?) + if (arena->memid.initially_zero && arena->blocks_dirty != NULL) { + memid->initially_zero = _mi_bitmap_claim_across(arena->blocks_dirty, arena->field_count, needed_bcount, bitmap_index, NULL, NULL); + } + + // set commit state + if (arena->blocks_committed == NULL) { + // always committed + memid->initially_committed = true; + } + else if (commit) { + // commit requested, but the range may not be committed as a whole: ensure it is committed now + memid->initially_committed = true; + const size_t commit_size = mi_arena_block_size(needed_bcount); + bool any_uncommitted; + size_t already_committed = 0; + _mi_bitmap_claim_across(arena->blocks_committed, arena->field_count, needed_bcount, bitmap_index, &any_uncommitted, &already_committed); + if (any_uncommitted) { + mi_assert_internal(already_committed < needed_bcount); + const size_t stat_commit_size = commit_size - mi_arena_block_size(already_committed); + bool commit_zero = false; + if (!_mi_os_commit_ex(p, commit_size, &commit_zero, stat_commit_size)) { + memid->initially_committed = false; + } + else { + if (commit_zero) { memid->initially_zero = true; } + } + } + else { + // all are already committed: signal that we are reusing memory in case it was purged before + _mi_os_reuse( p, commit_size ); + } + } + else { + // no need to commit, but check if already fully committed + size_t already_committed = 0; + memid->initially_committed = _mi_bitmap_is_claimed_across(arena->blocks_committed, arena->field_count, needed_bcount, bitmap_index, &already_committed); + if (!memid->initially_committed && already_committed > 0) { + // partially committed: as it will be committed at some time, adjust the stats and pretend the range is fully uncommitted. + mi_assert_internal(already_committed < needed_bcount); + _mi_stat_decrease(&_mi_stats_main.committed, mi_arena_block_size(already_committed)); + _mi_bitmap_unclaim_across(arena->blocks_committed, arena->field_count, needed_bcount, bitmap_index); + } + } + + return p; +} + +// allocate in a specific arena +static void* mi_arena_try_alloc_at_id(mi_arena_id_t arena_id, bool match_numa_node, int numa_node, size_t size, size_t alignment, + bool commit, bool allow_large, mi_arena_id_t req_arena_id, mi_memid_t* memid ) +{ + MI_UNUSED_RELEASE(alignment); + mi_assert(alignment <= MI_SEGMENT_ALIGN); + const size_t bcount = mi_block_count_of_size(size); + const size_t arena_index = mi_arena_id_index(arena_id); + mi_assert_internal(arena_index < mi_atomic_load_relaxed(&mi_arena_count)); + mi_assert_internal(size <= mi_arena_block_size(bcount)); + + // Check arena suitability + mi_arena_t* arena = mi_arena_from_index(arena_index); + if (arena == NULL) return NULL; + if (!allow_large && arena->is_large) return NULL; + if (!mi_arena_id_is_suitable(arena->id, arena->exclusive, req_arena_id)) return NULL; + if (req_arena_id == _mi_arena_id_none()) { // in not specific, check numa affinity + const bool numa_suitable = (numa_node < 0 || arena->numa_node < 0 || arena->numa_node == numa_node); + if (match_numa_node) { if (!numa_suitable) return NULL; } + else { if (numa_suitable) return NULL; } + } + + // try to allocate + void* p = mi_arena_try_alloc_at(arena, arena_index, bcount, commit, memid); + mi_assert_internal(p == NULL || _mi_is_aligned(p, alignment)); + return p; +} + + +// allocate from an arena with fallback to the OS +static mi_decl_noinline void* mi_arena_try_alloc(int numa_node, size_t size, size_t alignment, + bool commit, bool allow_large, + mi_arena_id_t req_arena_id, mi_memid_t* memid ) +{ + MI_UNUSED(alignment); + mi_assert_internal(alignment <= MI_SEGMENT_ALIGN); + const size_t max_arena = mi_atomic_load_relaxed(&mi_arena_count); + if mi_likely(max_arena == 0) return NULL; + + if (req_arena_id != _mi_arena_id_none()) { + // try a specific arena if requested + if (mi_arena_id_index(req_arena_id) < max_arena) { + void* p = mi_arena_try_alloc_at_id(req_arena_id, true, numa_node, size, alignment, commit, allow_large, req_arena_id, memid); + if (p != NULL) return p; + } + } + else { + // try numa affine allocation + for (size_t i = 0; i < max_arena; i++) { + void* p = mi_arena_try_alloc_at_id(mi_arena_id_create(i), true, numa_node, size, alignment, commit, allow_large, req_arena_id, memid); + if (p != NULL) return p; + } + + // try from another numa node instead.. + if (numa_node >= 0) { // if numa_node was < 0 (no specific affinity requested), all arena's have been tried already + for (size_t i = 0; i < max_arena; i++) { + void* p = mi_arena_try_alloc_at_id(mi_arena_id_create(i), false /* only proceed if not numa local */, numa_node, size, alignment, commit, allow_large, req_arena_id, memid); + if (p != NULL) return p; + } + } + } + return NULL; +} + +// try to reserve a fresh arena space +static bool mi_arena_reserve(size_t req_size, bool allow_large, mi_arena_id_t *arena_id) +{ + if (_mi_preloading()) return false; // use OS only while pre loading + + const size_t arena_count = mi_atomic_load_acquire(&mi_arena_count); + if (arena_count > (MI_MAX_ARENAS - 4)) return false; + + size_t arena_reserve = mi_option_get_size(mi_option_arena_reserve); + if (arena_reserve == 0) return false; + + if (!_mi_os_has_virtual_reserve()) { + arena_reserve = arena_reserve/4; // be conservative if virtual reserve is not supported (for WASM for example) + } + arena_reserve = _mi_align_up(arena_reserve, MI_ARENA_BLOCK_SIZE); + arena_reserve = _mi_align_up(arena_reserve, MI_SEGMENT_SIZE); + if (arena_count >= 8 && arena_count <= 128) { + // scale up the arena sizes exponentially every 8 entries (128 entries get to 589TiB) + const size_t multiplier = (size_t)1 << _mi_clamp(arena_count/8, 0, 16 ); + size_t reserve = 0; + if (!mi_mul_overflow(multiplier, arena_reserve, &reserve)) { + arena_reserve = reserve; + } + } + if (arena_reserve < req_size) return false; // should be able to at least handle the current allocation size + + // commit eagerly? + bool arena_commit = false; + if (mi_option_get(mi_option_arena_eager_commit) == 2) { arena_commit = _mi_os_has_overcommit(); } + else if (mi_option_get(mi_option_arena_eager_commit) == 1) { arena_commit = true; } + + return (mi_reserve_os_memory_ex(arena_reserve, arena_commit, allow_large, false /* exclusive? */, arena_id) == 0); +} + + +void* _mi_arena_alloc_aligned(size_t size, size_t alignment, size_t align_offset, bool commit, bool allow_large, + mi_arena_id_t req_arena_id, mi_memid_t* memid) +{ + mi_assert_internal(memid != NULL); + mi_assert_internal(size > 0); + *memid = _mi_memid_none(); + + const int numa_node = _mi_os_numa_node(); // current numa node + + // try to allocate in an arena if the alignment is small enough and the object is not too small (as for heap meta data) + if (!mi_option_is_enabled(mi_option_disallow_arena_alloc)) { // is arena allocation allowed? + if (size >= MI_ARENA_MIN_OBJ_SIZE && alignment <= MI_SEGMENT_ALIGN && align_offset == 0) + { + void* p = mi_arena_try_alloc(numa_node, size, alignment, commit, allow_large, req_arena_id, memid); + if (p != NULL) return p; + + // otherwise, try to first eagerly reserve a new arena + if (req_arena_id == _mi_arena_id_none()) { + mi_arena_id_t arena_id = 0; + if (mi_arena_reserve(size, allow_large, &arena_id)) { + // and try allocate in there + mi_assert_internal(req_arena_id == _mi_arena_id_none()); + p = mi_arena_try_alloc_at_id(arena_id, true, numa_node, size, alignment, commit, allow_large, req_arena_id, memid); + if (p != NULL) return p; + } + } + } + } + + // if we cannot use OS allocation, return NULL + if (mi_option_is_enabled(mi_option_disallow_os_alloc) || req_arena_id != _mi_arena_id_none()) { + errno = ENOMEM; + return NULL; + } + + // finally, fall back to the OS + if (align_offset > 0) { + return _mi_os_alloc_aligned_at_offset(size, alignment, align_offset, commit, allow_large, memid); + } + else { + return _mi_os_alloc_aligned(size, alignment, commit, allow_large, memid); + } +} + +void* _mi_arena_alloc(size_t size, bool commit, bool allow_large, mi_arena_id_t req_arena_id, mi_memid_t* memid) +{ + return _mi_arena_alloc_aligned(size, MI_ARENA_BLOCK_SIZE, 0, commit, allow_large, req_arena_id, memid); +} + + +void* mi_arena_area(mi_arena_id_t arena_id, size_t* size) { + if (size != NULL) *size = 0; + size_t arena_index = mi_arena_id_index(arena_id); + if (arena_index >= MI_MAX_ARENAS) return NULL; + mi_arena_t* arena = mi_atomic_load_ptr_acquire(mi_arena_t, &mi_arenas[arena_index]); + if (arena == NULL) return NULL; + if (size != NULL) { *size = mi_arena_block_size(arena->block_count); } + return arena->start; +} + + +/* ----------------------------------------------------------- + Arena purge +----------------------------------------------------------- */ + +static long mi_arena_purge_delay(void) { + // <0 = no purging allowed, 0=immediate purging, >0=milli-second delay + return (mi_option_get(mi_option_purge_delay) * mi_option_get(mi_option_arena_purge_mult)); +} + +// reset or decommit in an arena and update the committed/decommit bitmaps +// assumes we own the area (i.e. blocks_in_use is claimed by us) +static void mi_arena_purge(mi_arena_t* arena, size_t bitmap_idx, size_t blocks) { + mi_assert_internal(arena->blocks_committed != NULL); + mi_assert_internal(arena->blocks_purge != NULL); + mi_assert_internal(!arena->memid.is_pinned); + const size_t size = mi_arena_block_size(blocks); + void* const p = mi_arena_block_start(arena, bitmap_idx); + bool needs_recommit; + size_t already_committed = 0; + if (_mi_bitmap_is_claimed_across(arena->blocks_committed, arena->field_count, blocks, bitmap_idx, &already_committed)) { + // all blocks are committed, we can purge freely + mi_assert_internal(already_committed == blocks); + needs_recommit = _mi_os_purge(p, size); + } + else { + // some blocks are not committed -- this can happen when a partially committed block is freed + // in `_mi_arena_free` and it is conservatively marked as uncommitted but still scheduled for a purge + // we need to ensure we do not try to reset (as that may be invalid for uncommitted memory). + mi_assert_internal(already_committed < blocks); + mi_assert_internal(mi_option_is_enabled(mi_option_purge_decommits)); + needs_recommit = _mi_os_purge_ex(p, size, false /* allow reset? */, mi_arena_block_size(already_committed)); + } + + // clear the purged blocks + _mi_bitmap_unclaim_across(arena->blocks_purge, arena->field_count, blocks, bitmap_idx); + // update committed bitmap + if (needs_recommit) { + _mi_bitmap_unclaim_across(arena->blocks_committed, arena->field_count, blocks, bitmap_idx); + } +} + +// Schedule a purge. This is usually delayed to avoid repeated decommit/commit calls. +// Note: assumes we (still) own the area as we may purge immediately +static void mi_arena_schedule_purge(mi_arena_t* arena, size_t bitmap_idx, size_t blocks) { + mi_assert_internal(arena->blocks_purge != NULL); + const long delay = mi_arena_purge_delay(); + if (delay < 0) return; // is purging allowed at all? + + if (_mi_preloading() || delay == 0) { + // decommit directly + mi_arena_purge(arena, bitmap_idx, blocks); + } + else { + // schedule purge + const mi_msecs_t expire = _mi_clock_now() + delay; + mi_msecs_t expire0 = 0; + if (mi_atomic_casi64_strong_acq_rel(&arena->purge_expire, &expire0, expire)) { + // expiration was not yet set + // maybe set the global arenas expire as well (if it wasn't set already) + mi_atomic_casi64_strong_acq_rel(&mi_arenas_purge_expire, &expire0, expire); + } + else { + // already an expiration was set + } + _mi_bitmap_claim_across(arena->blocks_purge, arena->field_count, blocks, bitmap_idx, NULL, NULL); + } +} + +// purge a range of blocks +// return true if the full range was purged. +// assumes we own the area (i.e. blocks_in_use is claimed by us) +static bool mi_arena_purge_range(mi_arena_t* arena, size_t idx, size_t startidx, size_t bitlen, size_t purge) { + const size_t endidx = startidx + bitlen; + size_t bitidx = startidx; + bool all_purged = false; + while (bitidx < endidx) { + // count consecutive ones in the purge mask + size_t count = 0; + while (bitidx + count < endidx && (purge & ((size_t)1 << (bitidx + count))) != 0) { + count++; + } + if (count > 0) { + // found range to be purged + const mi_bitmap_index_t range_idx = mi_bitmap_index_create(idx, bitidx); + mi_arena_purge(arena, range_idx, count); + if (count == bitlen) { + all_purged = true; + } + } + bitidx += (count+1); // +1 to skip the zero bit (or end) + } + return all_purged; +} + +// returns true if anything was purged +static bool mi_arena_try_purge(mi_arena_t* arena, mi_msecs_t now, bool force) +{ + // check pre-conditions + if (arena->memid.is_pinned) return false; + + // expired yet? + mi_msecs_t expire = mi_atomic_loadi64_relaxed(&arena->purge_expire); + if (!force && (expire == 0 || expire > now)) return false; + + // reset expire (if not already set concurrently) + mi_atomic_casi64_strong_acq_rel(&arena->purge_expire, &expire, (mi_msecs_t)0); + _mi_stat_counter_increase(&_mi_stats_main.arena_purges, 1); + + // potential purges scheduled, walk through the bitmap + bool any_purged = false; + bool full_purge = true; + for (size_t i = 0; i < arena->field_count; i++) { + size_t purge = mi_atomic_load_relaxed(&arena->blocks_purge[i]); + if (purge != 0) { + size_t bitidx = 0; + while (bitidx < MI_BITMAP_FIELD_BITS) { + // find consecutive range of ones in the purge mask + size_t bitlen = 0; + while (bitidx + bitlen < MI_BITMAP_FIELD_BITS && (purge & ((size_t)1 << (bitidx + bitlen))) != 0) { + bitlen++; + } + // temporarily claim the purge range as "in-use" to be thread-safe with allocation + // try to claim the longest range of corresponding in_use bits + const mi_bitmap_index_t bitmap_index = mi_bitmap_index_create(i, bitidx); + while( bitlen > 0 ) { + if (_mi_bitmap_try_claim(arena->blocks_inuse, arena->field_count, bitlen, bitmap_index)) { + break; + } + bitlen--; + } + // actual claimed bits at `in_use` + if (bitlen > 0) { + // read purge again now that we have the in_use bits + purge = mi_atomic_load_acquire(&arena->blocks_purge[i]); + if (!mi_arena_purge_range(arena, i, bitidx, bitlen, purge)) { + full_purge = false; + } + any_purged = true; + // release the claimed `in_use` bits again + _mi_bitmap_unclaim(arena->blocks_inuse, arena->field_count, bitlen, bitmap_index); + } + bitidx += (bitlen+1); // +1 to skip the zero (or end) + } // while bitidx + } // purge != 0 + } + // if not fully purged, make sure to purge again in the future + if (!full_purge) { + const long delay = mi_arena_purge_delay(); + mi_msecs_t expected = 0; + mi_atomic_casi64_strong_acq_rel(&arena->purge_expire,&expected,_mi_clock_now() + delay); + } + return any_purged; +} + +static void mi_arenas_try_purge( bool force, bool visit_all ) +{ + if (_mi_preloading() || mi_arena_purge_delay() <= 0) return; // nothing will be scheduled + + // check if any arena needs purging? + const mi_msecs_t now = _mi_clock_now(); + mi_msecs_t arenas_expire = mi_atomic_loadi64_acquire(&mi_arenas_purge_expire); + if (!force && (arenas_expire == 0 || arenas_expire < now)) return; + + const size_t max_arena = mi_atomic_load_acquire(&mi_arena_count); + if (max_arena == 0) return; + + // allow only one thread to purge at a time + static mi_atomic_guard_t purge_guard; + mi_atomic_guard(&purge_guard) + { + // increase global expire: at most one purge per delay cycle + mi_atomic_storei64_release(&mi_arenas_purge_expire, now + mi_arena_purge_delay()); + size_t max_purge_count = (visit_all ? max_arena : 2); + bool all_visited = true; + for (size_t i = 0; i < max_arena; i++) { + mi_arena_t* arena = mi_atomic_load_ptr_acquire(mi_arena_t, &mi_arenas[i]); + if (arena != NULL) { + if (mi_arena_try_purge(arena, now, force)) { + if (max_purge_count <= 1) { + all_visited = false; + break; + } + max_purge_count--; + } + } + } + if (all_visited) { + // all arena's were visited and purged: reset global expire + mi_atomic_storei64_release(&mi_arenas_purge_expire, 0); + } + } +} + + +/* ----------------------------------------------------------- + Arena free +----------------------------------------------------------- */ + +void _mi_arena_free(void* p, size_t size, size_t committed_size, mi_memid_t memid) { + mi_assert_internal(size > 0); + mi_assert_internal(committed_size <= size); + if (p==NULL) return; + if (size==0) return; + const bool all_committed = (committed_size == size); + const size_t decommitted_size = (committed_size <= size ? size - committed_size : 0); + + // need to set all memory to undefined as some parts may still be marked as no_access (like padding etc.) + mi_track_mem_undefined(p,size); + + if (mi_memkind_is_os(memid.memkind)) { + // was a direct OS allocation, pass through + if (!all_committed && decommitted_size > 0) { + // if partially committed, adjust the committed stats (as `_mi_os_free` will decrease commit by the full size) + _mi_stat_increase(&_mi_stats_main.committed, decommitted_size); + } + _mi_os_free(p, size, memid); + } + else if (memid.memkind == MI_MEM_ARENA) { + // allocated in an arena + size_t arena_idx; + size_t bitmap_idx; + mi_arena_memid_indices(memid, &arena_idx, &bitmap_idx); + mi_assert_internal(arena_idx < MI_MAX_ARENAS); + mi_arena_t* arena = mi_atomic_load_ptr_acquire(mi_arena_t,&mi_arenas[arena_idx]); + mi_assert_internal(arena != NULL); + const size_t blocks = mi_block_count_of_size(size); + + // checks + if (arena == NULL) { + _mi_error_message(EINVAL, "trying to free from an invalid arena: %p, size %zu, memid: 0x%zx\n", p, size, memid); + return; + } + mi_assert_internal(arena->field_count > mi_bitmap_index_field(bitmap_idx)); + if (arena->field_count <= mi_bitmap_index_field(bitmap_idx)) { + _mi_error_message(EINVAL, "trying to free from an invalid arena block: %p, size %zu, memid: 0x%zx\n", p, size, memid); + return; + } + + // potentially decommit + if (arena->memid.is_pinned || arena->blocks_committed == NULL) { + mi_assert_internal(all_committed); + } + else { + mi_assert_internal(arena->blocks_committed != NULL); + mi_assert_internal(arena->blocks_purge != NULL); + + if (!all_committed) { + // mark the entire range as no longer committed (so we will recommit the full range when re-using) + _mi_bitmap_unclaim_across(arena->blocks_committed, arena->field_count, blocks, bitmap_idx); + mi_track_mem_noaccess(p,size); + //if (committed_size > 0) { + // if partially committed, adjust the committed stats (is it will be recommitted when re-using) + // in the delayed purge, we do no longer decrease the commit if the range is not marked entirely as committed. + _mi_stat_decrease(&_mi_stats_main.committed, committed_size); + //} + // note: if not all committed, it may be that the purge will reset/decommit the entire range + // that contains already decommitted parts. Since purge consistently uses reset or decommit that + // works (as we should never reset decommitted parts). + } + // (delay) purge the entire range + mi_arena_schedule_purge(arena, bitmap_idx, blocks); + } + + // and make it available to others again + bool all_inuse = _mi_bitmap_unclaim_across(arena->blocks_inuse, arena->field_count, blocks, bitmap_idx); + if (!all_inuse) { + _mi_error_message(EAGAIN, "trying to free an already freed arena block: %p, size %zu\n", p, size); + return; + }; + } + else { + // arena was none, external, or static; nothing to do + mi_assert_internal(memid.memkind < MI_MEM_OS); + } + + // purge expired decommits + mi_arenas_try_purge(false, false); +} + +// destroy owned arenas; this is unsafe and should only be done using `mi_option_destroy_on_exit` +// for dynamic libraries that are unloaded and need to release all their allocated memory. +static void mi_arenas_unsafe_destroy(void) { + const size_t max_arena = mi_atomic_load_relaxed(&mi_arena_count); + size_t new_max_arena = 0; + for (size_t i = 0; i < max_arena; i++) { + mi_arena_t* arena = mi_atomic_load_ptr_acquire(mi_arena_t, &mi_arenas[i]); + if (arena != NULL) { + mi_lock_done(&arena->abandoned_visit_lock); + if (arena->start != NULL && mi_memkind_is_os(arena->memid.memkind)) { + mi_atomic_store_ptr_release(mi_arena_t, &mi_arenas[i], NULL); + _mi_os_free(arena->start, mi_arena_size(arena), arena->memid); + } + else { + new_max_arena = i; + } + _mi_arena_meta_free(arena, arena->meta_memid, arena->meta_size); + } + } + + // try to lower the max arena. + size_t expected = max_arena; + mi_atomic_cas_strong_acq_rel(&mi_arena_count, &expected, new_max_arena); +} + +// Purge the arenas; if `force_purge` is true, amenable parts are purged even if not yet expired +void _mi_arenas_collect(bool force_purge) { + mi_arenas_try_purge(force_purge, force_purge /* visit all? */); +} + +// destroy owned arenas; this is unsafe and should only be done using `mi_option_destroy_on_exit` +// for dynamic libraries that are unloaded and need to release all their allocated memory. +void _mi_arena_unsafe_destroy_all(void) { + mi_arenas_unsafe_destroy(); + _mi_arenas_collect(true /* force purge */); // purge non-owned arenas +} + +// Is a pointer inside any of our arenas? +bool _mi_arena_contains(const void* p) { + const size_t max_arena = mi_atomic_load_relaxed(&mi_arena_count); + for (size_t i = 0; i < max_arena; i++) { + mi_arena_t* arena = mi_atomic_load_ptr_relaxed(mi_arena_t, &mi_arenas[i]); + if (arena != NULL && arena->start <= (const uint8_t*)p && arena->start + mi_arena_block_size(arena->block_count) > (const uint8_t*)p) { + return true; + } + } + return false; +} + +/* ----------------------------------------------------------- + Add an arena. +----------------------------------------------------------- */ + +static bool mi_arena_add(mi_arena_t* arena, mi_arena_id_t* arena_id, mi_stats_t* stats) { + mi_assert_internal(arena != NULL); + mi_assert_internal((uintptr_t)mi_atomic_load_ptr_relaxed(uint8_t,&arena->start) % MI_SEGMENT_ALIGN == 0); + mi_assert_internal(arena->block_count > 0); + if (arena_id != NULL) { *arena_id = -1; } + + size_t i = mi_atomic_increment_acq_rel(&mi_arena_count); + if (i >= MI_MAX_ARENAS) { + mi_atomic_decrement_acq_rel(&mi_arena_count); + return false; + } + _mi_stat_counter_increase(&stats->arena_count,1); + arena->id = mi_arena_id_create(i); + mi_atomic_store_ptr_release(mi_arena_t,&mi_arenas[i], arena); + if (arena_id != NULL) { *arena_id = arena->id; } + return true; +} + +static bool mi_manage_os_memory_ex2(void* start, size_t size, bool is_large, int numa_node, bool exclusive, mi_memid_t memid, mi_arena_id_t* arena_id) mi_attr_noexcept +{ + if (arena_id != NULL) *arena_id = _mi_arena_id_none(); + if (size < MI_ARENA_BLOCK_SIZE) { + _mi_warning_message("the arena size is too small (memory at %p with size %zu)\n", start, size); + return false; + } + if (is_large) { + mi_assert_internal(memid.initially_committed && memid.is_pinned); + } + if (!_mi_is_aligned(start, MI_SEGMENT_ALIGN)) { + void* const aligned_start = mi_align_up_ptr(start, MI_SEGMENT_ALIGN); + const size_t diff = (uint8_t*)aligned_start - (uint8_t*)start; + if (diff >= size || (size - diff) < MI_ARENA_BLOCK_SIZE) { + _mi_warning_message("after alignment, the size of the arena becomes too small (memory at %p with size %zu)\n", start, size); + return false; + } + start = aligned_start; + size = size - diff; + } + + const size_t bcount = size / MI_ARENA_BLOCK_SIZE; + const size_t fields = _mi_divide_up(bcount, MI_BITMAP_FIELD_BITS); + const size_t bitmaps = (memid.is_pinned ? 3 : 5); + const size_t asize = sizeof(mi_arena_t) + (bitmaps*fields*sizeof(mi_bitmap_field_t)); + mi_memid_t meta_memid; + mi_arena_t* arena = (mi_arena_t*)_mi_arena_meta_zalloc(asize, &meta_memid); + if (arena == NULL) return false; + + // already zero'd due to zalloc + // _mi_memzero(arena, asize); + arena->id = _mi_arena_id_none(); + arena->memid = memid; + arena->exclusive = exclusive; + arena->meta_size = asize; + arena->meta_memid = meta_memid; + arena->block_count = bcount; + arena->field_count = fields; + arena->start = (uint8_t*)start; + arena->numa_node = numa_node; // TODO: or get the current numa node if -1? (now it allows anyone to allocate on -1) + arena->is_large = is_large; + arena->purge_expire = 0; + arena->search_idx = 0; + mi_lock_init(&arena->abandoned_visit_lock); + // consecutive bitmaps + arena->blocks_dirty = &arena->blocks_inuse[fields]; // just after inuse bitmap + arena->blocks_abandoned = &arena->blocks_inuse[2 * fields]; // just after dirty bitmap + arena->blocks_committed = (arena->memid.is_pinned ? NULL : &arena->blocks_inuse[3*fields]); // just after abandoned bitmap + arena->blocks_purge = (arena->memid.is_pinned ? NULL : &arena->blocks_inuse[4*fields]); // just after committed bitmap + // initialize committed bitmap? + if (arena->blocks_committed != NULL && arena->memid.initially_committed) { + memset((void*)arena->blocks_committed, 0xFF, fields*sizeof(mi_bitmap_field_t)); // cast to void* to avoid atomic warning + } + + // and claim leftover blocks if needed (so we never allocate there) + ptrdiff_t post = (fields * MI_BITMAP_FIELD_BITS) - bcount; + mi_assert_internal(post >= 0); + if (post > 0) { + // don't use leftover bits at the end + mi_bitmap_index_t postidx = mi_bitmap_index_create(fields - 1, MI_BITMAP_FIELD_BITS - post); + _mi_bitmap_claim(arena->blocks_inuse, fields, post, postidx, NULL); + } + return mi_arena_add(arena, arena_id, &_mi_stats_main); + +} + +bool mi_manage_os_memory_ex(void* start, size_t size, bool is_committed, bool is_large, bool is_zero, int numa_node, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept { + mi_memid_t memid = _mi_memid_create(MI_MEM_EXTERNAL); + memid.initially_committed = is_committed; + memid.initially_zero = is_zero; + memid.is_pinned = is_large; + return mi_manage_os_memory_ex2(start,size,is_large,numa_node,exclusive,memid, arena_id); +} + +// Reserve a range of regular OS memory +int mi_reserve_os_memory_ex(size_t size, bool commit, bool allow_large, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept { + if (arena_id != NULL) *arena_id = _mi_arena_id_none(); + size = _mi_align_up(size, MI_ARENA_BLOCK_SIZE); // at least one block + mi_memid_t memid; + void* start = _mi_os_alloc_aligned(size, MI_SEGMENT_ALIGN, commit, allow_large, &memid); + if (start == NULL) return ENOMEM; + const bool is_large = memid.is_pinned; // todo: use separate is_large field? + if (!mi_manage_os_memory_ex2(start, size, is_large, -1 /* numa node */, exclusive, memid, arena_id)) { + _mi_os_free_ex(start, size, commit, memid); + _mi_verbose_message("failed to reserve %zu KiB memory\n", _mi_divide_up(size, 1024)); + return ENOMEM; + } + _mi_verbose_message("reserved %zu KiB memory%s\n", _mi_divide_up(size, 1024), is_large ? " (in large os pages)" : ""); + return 0; +} + + +// Manage a range of regular OS memory +bool mi_manage_os_memory(void* start, size_t size, bool is_committed, bool is_large, bool is_zero, int numa_node) mi_attr_noexcept { + return mi_manage_os_memory_ex(start, size, is_committed, is_large, is_zero, numa_node, false /* exclusive? */, NULL); +} + +// Reserve a range of regular OS memory +int mi_reserve_os_memory(size_t size, bool commit, bool allow_large) mi_attr_noexcept { + return mi_reserve_os_memory_ex(size, commit, allow_large, false, NULL); +} + + +/* ----------------------------------------------------------- + Debugging +----------------------------------------------------------- */ + +static size_t mi_debug_show_bitmap(const char* prefix, const char* header, size_t block_count, mi_bitmap_field_t* fields, size_t field_count ) { + _mi_message("%s%s:\n", prefix, header); + size_t bcount = 0; + size_t inuse_count = 0; + for (size_t i = 0; i < field_count; i++) { + char buf[MI_BITMAP_FIELD_BITS + 1]; + uintptr_t field = mi_atomic_load_relaxed(&fields[i]); + for (size_t bit = 0; bit < MI_BITMAP_FIELD_BITS; bit++, bcount++) { + if (bcount < block_count) { + bool inuse = ((((uintptr_t)1 << bit) & field) != 0); + if (inuse) inuse_count++; + buf[bit] = (inuse ? 'x' : '.'); + } + else { + buf[bit] = ' '; + } + } + buf[MI_BITMAP_FIELD_BITS] = 0; + _mi_message("%s %s\n", prefix, buf); + } + _mi_message("%s total ('x'): %zu\n", prefix, inuse_count); + return inuse_count; +} + +void mi_debug_show_arenas(void) mi_attr_noexcept { + const bool show_inuse = true; + size_t max_arenas = mi_atomic_load_relaxed(&mi_arena_count); + size_t inuse_total = 0; + //size_t abandoned_total = 0; + //size_t purge_total = 0; + for (size_t i = 0; i < max_arenas; i++) { + mi_arena_t* arena = mi_atomic_load_ptr_relaxed(mi_arena_t, &mi_arenas[i]); + if (arena == NULL) break; + _mi_message("arena %zu: %zu blocks of size %zuMiB (in %zu fields) %s\n", i, arena->block_count, (size_t)(MI_ARENA_BLOCK_SIZE / MI_MiB), arena->field_count, (arena->memid.is_pinned ? ", pinned" : "")); + if (show_inuse) { + inuse_total += mi_debug_show_bitmap(" ", "inuse blocks", arena->block_count, arena->blocks_inuse, arena->field_count); + } + if (arena->blocks_committed != NULL) { + mi_debug_show_bitmap(" ", "committed blocks", arena->block_count, arena->blocks_committed, arena->field_count); + } + //if (show_abandoned) { + // abandoned_total += mi_debug_show_bitmap(" ", "abandoned blocks", arena->block_count, arena->blocks_abandoned, arena->field_count); + //} + //if (show_purge && arena->blocks_purge != NULL) { + // purge_total += mi_debug_show_bitmap(" ", "purgeable blocks", arena->block_count, arena->blocks_purge, arena->field_count); + //} + } + if (show_inuse) _mi_message("total inuse blocks : %zu\n", inuse_total); + //if (show_abandoned) _mi_message("total abandoned blocks: %zu\n", abandoned_total); + //if (show_purge) _mi_message("total purgeable blocks: %zu\n", purge_total); +} + + +void mi_arenas_print(void) mi_attr_noexcept { + mi_debug_show_arenas(); +} + + +/* ----------------------------------------------------------- + Reserve a huge page arena. +----------------------------------------------------------- */ +// reserve at a specific numa node +int mi_reserve_huge_os_pages_at_ex(size_t pages, int numa_node, size_t timeout_msecs, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept { + if (arena_id != NULL) *arena_id = -1; + if (pages==0) return 0; + if (numa_node < -1) numa_node = -1; + if (numa_node >= 0) numa_node = numa_node % _mi_os_numa_node_count(); + size_t hsize = 0; + size_t pages_reserved = 0; + mi_memid_t memid; + void* p = _mi_os_alloc_huge_os_pages(pages, numa_node, timeout_msecs, &pages_reserved, &hsize, &memid); + if (p==NULL || pages_reserved==0) { + _mi_warning_message("failed to reserve %zu GiB huge pages\n", pages); + return ENOMEM; + } + _mi_verbose_message("numa node %i: reserved %zu GiB huge pages (of the %zu GiB requested)\n", numa_node, pages_reserved, pages); + + if (!mi_manage_os_memory_ex2(p, hsize, true, numa_node, exclusive, memid, arena_id)) { + _mi_os_free(p, hsize, memid); + return ENOMEM; + } + return 0; +} + +int mi_reserve_huge_os_pages_at(size_t pages, int numa_node, size_t timeout_msecs) mi_attr_noexcept { + return mi_reserve_huge_os_pages_at_ex(pages, numa_node, timeout_msecs, false, NULL); +} + +// reserve huge pages evenly among the given number of numa nodes (or use the available ones as detected) +int mi_reserve_huge_os_pages_interleave(size_t pages, size_t numa_nodes, size_t timeout_msecs) mi_attr_noexcept { + if (pages == 0) return 0; + + // pages per numa node + int numa_count = (numa_nodes > 0 && numa_nodes <= INT_MAX ? (int)numa_nodes : _mi_os_numa_node_count()); + if (numa_count == 0) numa_count = 1; + const size_t pages_per = pages / numa_count; + const size_t pages_mod = pages % numa_count; + const size_t timeout_per = (timeout_msecs==0 ? 0 : (timeout_msecs / numa_count) + 50); + + // reserve evenly among numa nodes + for (int numa_node = 0; numa_node < numa_count && pages > 0; numa_node++) { + size_t node_pages = pages_per; // can be 0 + if ((size_t)numa_node < pages_mod) node_pages++; + int err = mi_reserve_huge_os_pages_at(node_pages, numa_node, timeout_per); + if (err) return err; + if (pages < node_pages) { + pages = 0; + } + else { + pages -= node_pages; + } + } + + return 0; +} + +int mi_reserve_huge_os_pages(size_t pages, double max_secs, size_t* pages_reserved) mi_attr_noexcept { + MI_UNUSED(max_secs); + _mi_warning_message("mi_reserve_huge_os_pages is deprecated: use mi_reserve_huge_os_pages_interleave/at instead\n"); + if (pages_reserved != NULL) *pages_reserved = 0; + int err = mi_reserve_huge_os_pages_interleave(pages, 0, (size_t)(max_secs * 1000.0)); + if (err==0 && pages_reserved!=NULL) *pages_reserved = pages; + return err; +} diff --git a/compat/mimalloc/bitmap.c b/compat/mimalloc/bitmap.c new file mode 100644 index 00000000000000..32d1e9548d3e3b --- /dev/null +++ b/compat/mimalloc/bitmap.c @@ -0,0 +1,441 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2023 Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ---------------------------------------------------------------------------- +Concurrent bitmap that can set/reset sequences of bits atomically, +represented as an array of fields where each field is a machine word (`size_t`) + +There are two api's; the standard one cannot have sequences that cross +between the bitmap fields (and a sequence must be <= MI_BITMAP_FIELD_BITS). + +The `_across` postfixed functions do allow sequences that can cross over +between the fields. (This is used in arena allocation) +---------------------------------------------------------------------------- */ + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "bitmap.h" + +/* ----------------------------------------------------------- + Bitmap definition +----------------------------------------------------------- */ + +// The bit mask for a given number of blocks at a specified bit index. +static inline size_t mi_bitmap_mask_(size_t count, size_t bitidx) { + mi_assert_internal(count + bitidx <= MI_BITMAP_FIELD_BITS); + mi_assert_internal(count > 0); + if (count >= MI_BITMAP_FIELD_BITS) return MI_BITMAP_FIELD_FULL; + if (count == 0) return 0; + return ((((size_t)1 << count) - 1) << bitidx); +} + + +/* ----------------------------------------------------------- + Claim a bit sequence atomically +----------------------------------------------------------- */ + +// Try to atomically claim a sequence of `count` bits in a single +// field at `idx` in `bitmap`. Returns `true` on success. +inline bool _mi_bitmap_try_find_claim_field(mi_bitmap_t bitmap, size_t idx, const size_t count, mi_bitmap_index_t* bitmap_idx) +{ + mi_assert_internal(bitmap_idx != NULL); + mi_assert_internal(count <= MI_BITMAP_FIELD_BITS); + mi_assert_internal(count > 0); + mi_bitmap_field_t* field = &bitmap[idx]; + size_t map = mi_atomic_load_relaxed(field); + if (map==MI_BITMAP_FIELD_FULL) return false; // short cut + + // search for 0-bit sequence of length count + const size_t mask = mi_bitmap_mask_(count, 0); + const size_t bitidx_max = MI_BITMAP_FIELD_BITS - count; + +#ifdef MI_HAVE_FAST_BITSCAN + size_t bitidx = mi_ctz(~map); // quickly find the first zero bit if possible +#else + size_t bitidx = 0; // otherwise start at 0 +#endif + size_t m = (mask << bitidx); // invariant: m == mask shifted by bitidx + + // scan linearly for a free range of zero bits + while (bitidx <= bitidx_max) { + const size_t mapm = (map & m); + if (mapm == 0) { // are the mask bits free at bitidx? + mi_assert_internal((m >> bitidx) == mask); // no overflow? + const size_t newmap = (map | m); + mi_assert_internal((newmap^map) >> bitidx == mask); + if (!mi_atomic_cas_strong_acq_rel(field, &map, newmap)) { // TODO: use weak cas here? + // no success, another thread claimed concurrently.. keep going (with updated `map`) + continue; + } + else { + // success, we claimed the bits! + *bitmap_idx = mi_bitmap_index_create(idx, bitidx); + return true; + } + } + else { + // on to the next bit range +#ifdef MI_HAVE_FAST_BITSCAN + mi_assert_internal(mapm != 0); + const size_t shift = (count == 1 ? 1 : (MI_SIZE_BITS - mi_clz(mapm) - bitidx)); + mi_assert_internal(shift > 0 && shift <= count); +#else + const size_t shift = 1; +#endif + bitidx += shift; + m <<= shift; + } + } + // no bits found + return false; +} + +// Find `count` bits of 0 and set them to 1 atomically; returns `true` on success. +// Starts at idx, and wraps around to search in all `bitmap_fields` fields. +// `count` can be at most MI_BITMAP_FIELD_BITS and will never cross fields. +bool _mi_bitmap_try_find_from_claim(mi_bitmap_t bitmap, const size_t bitmap_fields, const size_t start_field_idx, const size_t count, mi_bitmap_index_t* bitmap_idx) { + size_t idx = start_field_idx; + for (size_t visited = 0; visited < bitmap_fields; visited++, idx++) { + if (idx >= bitmap_fields) { idx = 0; } // wrap + if (_mi_bitmap_try_find_claim_field(bitmap, idx, count, bitmap_idx)) { + return true; + } + } + return false; +} + +// Like _mi_bitmap_try_find_from_claim but with an extra predicate that must be fullfilled +bool _mi_bitmap_try_find_from_claim_pred(mi_bitmap_t bitmap, const size_t bitmap_fields, + const size_t start_field_idx, const size_t count, + mi_bitmap_pred_fun_t pred_fun, void* pred_arg, + mi_bitmap_index_t* bitmap_idx) { + size_t idx = start_field_idx; + for (size_t visited = 0; visited < bitmap_fields; visited++, idx++) { + if (idx >= bitmap_fields) idx = 0; // wrap + if (_mi_bitmap_try_find_claim_field(bitmap, idx, count, bitmap_idx)) { + if (pred_fun == NULL || pred_fun(*bitmap_idx, pred_arg)) { + return true; + } + // predicate returned false, unclaim and look further + _mi_bitmap_unclaim(bitmap, bitmap_fields, count, *bitmap_idx); + } + } + return false; +} + +// Set `count` bits at `bitmap_idx` to 0 atomically +// Returns `true` if all `count` bits were 1 previously. +bool _mi_bitmap_unclaim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + const size_t idx = mi_bitmap_index_field(bitmap_idx); + const size_t bitidx = mi_bitmap_index_bit_in_field(bitmap_idx); + const size_t mask = mi_bitmap_mask_(count, bitidx); + mi_assert_internal(bitmap_fields > idx); MI_UNUSED(bitmap_fields); + // mi_assert_internal((bitmap[idx] & mask) == mask); + const size_t prev = mi_atomic_and_acq_rel(&bitmap[idx], ~mask); + return ((prev & mask) == mask); +} + + +// Set `count` bits at `bitmap_idx` to 1 atomically +// Returns `true` if all `count` bits were 0 previously. `any_zero` is `true` if there was at least one zero bit. +bool _mi_bitmap_claim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* any_zero) { + const size_t idx = mi_bitmap_index_field(bitmap_idx); + const size_t bitidx = mi_bitmap_index_bit_in_field(bitmap_idx); + const size_t mask = mi_bitmap_mask_(count, bitidx); + mi_assert_internal(bitmap_fields > idx); MI_UNUSED(bitmap_fields); + //mi_assert_internal(any_zero != NULL || (bitmap[idx] & mask) == 0); + size_t prev = mi_atomic_or_acq_rel(&bitmap[idx], mask); + if (any_zero != NULL) { *any_zero = ((prev & mask) != mask); } + return ((prev & mask) == 0); +} + +// Returns `true` if all `count` bits were 1. `any_ones` is `true` if there was at least one bit set to one. +static bool mi_bitmap_is_claimedx(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* any_ones) { + const size_t idx = mi_bitmap_index_field(bitmap_idx); + const size_t bitidx = mi_bitmap_index_bit_in_field(bitmap_idx); + const size_t mask = mi_bitmap_mask_(count, bitidx); + mi_assert_internal(bitmap_fields > idx); MI_UNUSED(bitmap_fields); + const size_t field = mi_atomic_load_relaxed(&bitmap[idx]); + if (any_ones != NULL) { *any_ones = ((field & mask) != 0); } + return ((field & mask) == mask); +} + +// Try to set `count` bits at `bitmap_idx` from 0 to 1 atomically. +// Returns `true` if successful when all previous `count` bits were 0. +bool _mi_bitmap_try_claim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + const size_t idx = mi_bitmap_index_field(bitmap_idx); + const size_t bitidx = mi_bitmap_index_bit_in_field(bitmap_idx); + const size_t mask = mi_bitmap_mask_(count, bitidx); + mi_assert_internal(bitmap_fields > idx); MI_UNUSED(bitmap_fields); + size_t expected = mi_atomic_load_relaxed(&bitmap[idx]); + do { + if ((expected & mask) != 0) return false; + } + while (!mi_atomic_cas_strong_acq_rel(&bitmap[idx], &expected, expected | mask)); + mi_assert_internal((expected & mask) == 0); + return true; +} + + +bool _mi_bitmap_is_claimed(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + return mi_bitmap_is_claimedx(bitmap, bitmap_fields, count, bitmap_idx, NULL); +} + +bool _mi_bitmap_is_any_claimed(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + bool any_ones; + mi_bitmap_is_claimedx(bitmap, bitmap_fields, count, bitmap_idx, &any_ones); + return any_ones; +} + + +//-------------------------------------------------------------------------- +// the `_across` functions work on bitmaps where sequences can cross over +// between the fields. This is used in arena allocation +//-------------------------------------------------------------------------- + +// Try to atomically claim a sequence of `count` bits starting from the field +// at `idx` in `bitmap` and crossing into subsequent fields. Returns `true` on success. +// Only needs to consider crossing into the next fields (see `mi_bitmap_try_find_from_claim_across`) +static bool mi_bitmap_try_find_claim_field_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t idx, const size_t count, const size_t retries, mi_bitmap_index_t* bitmap_idx) +{ + mi_assert_internal(bitmap_idx != NULL); + + // check initial trailing zeros + mi_bitmap_field_t* field = &bitmap[idx]; + size_t map = mi_atomic_load_relaxed(field); + const size_t initial = mi_clz(map); // count of initial zeros starting at idx + mi_assert_internal(initial <= MI_BITMAP_FIELD_BITS); + if (initial == 0) return false; + if (initial >= count) return _mi_bitmap_try_find_claim_field(bitmap, idx, count, bitmap_idx); // no need to cross fields (this case won't happen for us) + if (_mi_divide_up(count - initial, MI_BITMAP_FIELD_BITS) >= (bitmap_fields - idx)) return false; // not enough entries + + // scan ahead + size_t found = initial; + size_t mask = 0; // mask bits for the final field + while(found < count) { + field++; + map = mi_atomic_load_relaxed(field); + const size_t mask_bits = (found + MI_BITMAP_FIELD_BITS <= count ? MI_BITMAP_FIELD_BITS : (count - found)); + mi_assert_internal(mask_bits > 0 && mask_bits <= MI_BITMAP_FIELD_BITS); + mask = mi_bitmap_mask_(mask_bits, 0); + if ((map & mask) != 0) return false; // some part is already claimed + found += mask_bits; + } + mi_assert_internal(field < &bitmap[bitmap_fields]); + + // we found a range of contiguous zeros up to the final field; mask contains mask in the final field + // now try to claim the range atomically + mi_bitmap_field_t* const final_field = field; + const size_t final_mask = mask; + mi_bitmap_field_t* const initial_field = &bitmap[idx]; + const size_t initial_idx = MI_BITMAP_FIELD_BITS - initial; + const size_t initial_mask = mi_bitmap_mask_(initial, initial_idx); + + // initial field + size_t newmap; + field = initial_field; + map = mi_atomic_load_relaxed(field); + do { + newmap = (map | initial_mask); + if ((map & initial_mask) != 0) { goto rollback; }; + } while (!mi_atomic_cas_strong_acq_rel(field, &map, newmap)); + + // intermediate fields + while (++field < final_field) { + newmap = MI_BITMAP_FIELD_FULL; + map = 0; + if (!mi_atomic_cas_strong_acq_rel(field, &map, newmap)) { goto rollback; } + } + + // final field + mi_assert_internal(field == final_field); + map = mi_atomic_load_relaxed(field); + do { + newmap = (map | final_mask); + if ((map & final_mask) != 0) { goto rollback; } + } while (!mi_atomic_cas_strong_acq_rel(field, &map, newmap)); + + // claimed! + *bitmap_idx = mi_bitmap_index_create(idx, initial_idx); + return true; + +rollback: + // roll back intermediate fields + // (we just failed to claim `field` so decrement first) + while (--field > initial_field) { + newmap = 0; + map = MI_BITMAP_FIELD_FULL; + mi_assert_internal(mi_atomic_load_relaxed(field) == map); + mi_atomic_store_release(field, newmap); + } + if (field == initial_field) { // (if we failed on the initial field, `field + 1 == initial_field`) + map = mi_atomic_load_relaxed(field); + do { + mi_assert_internal((map & initial_mask) == initial_mask); + newmap = (map & ~initial_mask); + } while (!mi_atomic_cas_strong_acq_rel(field, &map, newmap)); + } + mi_stat_counter_increase(_mi_stats_main.arena_rollback_count,1); + // retry? (we make a recursive call instead of goto to be able to use const declarations) + if (retries <= 2) { + return mi_bitmap_try_find_claim_field_across(bitmap, bitmap_fields, idx, count, retries+1, bitmap_idx); + } + else { + return false; + } +} + + +// Find `count` bits of zeros and set them to 1 atomically; returns `true` on success. +// Starts at idx, and wraps around to search in all `bitmap_fields` fields. +bool _mi_bitmap_try_find_from_claim_across(mi_bitmap_t bitmap, const size_t bitmap_fields, const size_t start_field_idx, const size_t count, mi_bitmap_index_t* bitmap_idx) { + mi_assert_internal(count > 0); + if (count <= 2) { + // we don't bother with crossover fields for small counts + return _mi_bitmap_try_find_from_claim(bitmap, bitmap_fields, start_field_idx, count, bitmap_idx); + } + + // visit the fields + size_t idx = start_field_idx; + for (size_t visited = 0; visited < bitmap_fields; visited++, idx++) { + if (idx >= bitmap_fields) { idx = 0; } // wrap + // first try to claim inside a field + /* + if (count <= MI_BITMAP_FIELD_BITS) { + if (_mi_bitmap_try_find_claim_field(bitmap, idx, count, bitmap_idx)) { + return true; + } + } + */ + // if that fails, then try to claim across fields + if (mi_bitmap_try_find_claim_field_across(bitmap, bitmap_fields, idx, count, 0, bitmap_idx)) { + return true; + } + } + return false; +} + +// Helper for masks across fields; returns the mid count, post_mask may be 0 +static size_t mi_bitmap_mask_across(mi_bitmap_index_t bitmap_idx, size_t bitmap_fields, size_t count, size_t* pre_mask, size_t* mid_mask, size_t* post_mask) { + MI_UNUSED(bitmap_fields); + const size_t bitidx = mi_bitmap_index_bit_in_field(bitmap_idx); + if mi_likely(bitidx + count <= MI_BITMAP_FIELD_BITS) { + *pre_mask = mi_bitmap_mask_(count, bitidx); + *mid_mask = 0; + *post_mask = 0; + mi_assert_internal(mi_bitmap_index_field(bitmap_idx) < bitmap_fields); + return 0; + } + else { + const size_t pre_bits = MI_BITMAP_FIELD_BITS - bitidx; + mi_assert_internal(pre_bits < count); + *pre_mask = mi_bitmap_mask_(pre_bits, bitidx); + count -= pre_bits; + const size_t mid_count = (count / MI_BITMAP_FIELD_BITS); + *mid_mask = MI_BITMAP_FIELD_FULL; + count %= MI_BITMAP_FIELD_BITS; + *post_mask = (count==0 ? 0 : mi_bitmap_mask_(count, 0)); + mi_assert_internal(mi_bitmap_index_field(bitmap_idx) + mid_count + (count==0 ? 0 : 1) < bitmap_fields); + return mid_count; + } +} + +// Set `count` bits at `bitmap_idx` to 0 atomically +// Returns `true` if all `count` bits were 1 previously. +bool _mi_bitmap_unclaim_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + size_t idx = mi_bitmap_index_field(bitmap_idx); + size_t pre_mask; + size_t mid_mask; + size_t post_mask; + size_t mid_count = mi_bitmap_mask_across(bitmap_idx, bitmap_fields, count, &pre_mask, &mid_mask, &post_mask); + bool all_one = true; + mi_bitmap_field_t* field = &bitmap[idx]; + size_t prev = mi_atomic_and_acq_rel(field++, ~pre_mask); // clear first part + if ((prev & pre_mask) != pre_mask) all_one = false; + while(mid_count-- > 0) { + prev = mi_atomic_and_acq_rel(field++, ~mid_mask); // clear mid part + if ((prev & mid_mask) != mid_mask) all_one = false; + } + if (post_mask!=0) { + prev = mi_atomic_and_acq_rel(field, ~post_mask); // clear end part + if ((prev & post_mask) != post_mask) all_one = false; + } + return all_one; +} + +// Set `count` bits at `bitmap_idx` to 1 atomically +// Returns `true` if all `count` bits were 0 previously. `any_zero` is `true` if there was at least one zero bit. +bool _mi_bitmap_claim_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* pany_zero, size_t* already_set) { + size_t idx = mi_bitmap_index_field(bitmap_idx); + size_t pre_mask; + size_t mid_mask; + size_t post_mask; + size_t mid_count = mi_bitmap_mask_across(bitmap_idx, bitmap_fields, count, &pre_mask, &mid_mask, &post_mask); + bool all_zero = true; + bool any_zero = false; + size_t one_count = 0; + _Atomic(size_t)*field = &bitmap[idx]; + size_t prev = mi_atomic_or_acq_rel(field++, pre_mask); + if ((prev & pre_mask) != 0) { all_zero = false; one_count += mi_popcount(prev & pre_mask); } + if ((prev & pre_mask) != pre_mask) any_zero = true; + while (mid_count-- > 0) { + prev = mi_atomic_or_acq_rel(field++, mid_mask); + if ((prev & mid_mask) != 0) { all_zero = false; one_count += mi_popcount(prev & mid_mask); } + if ((prev & mid_mask) != mid_mask) any_zero = true; + } + if (post_mask!=0) { + prev = mi_atomic_or_acq_rel(field, post_mask); + if ((prev & post_mask) != 0) { all_zero = false; one_count += mi_popcount(prev & post_mask); } + if ((prev & post_mask) != post_mask) any_zero = true; + } + if (pany_zero != NULL) { *pany_zero = any_zero; } + if (already_set != NULL) { *already_set = one_count; }; + mi_assert_internal(all_zero ? one_count == 0 : one_count <= count); + return all_zero; +} + + +// Returns `true` if all `count` bits were 1. +// `any_ones` is `true` if there was at least one bit set to one. +static bool mi_bitmap_is_claimedx_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* pany_ones, size_t* already_set) { + size_t idx = mi_bitmap_index_field(bitmap_idx); + size_t pre_mask; + size_t mid_mask; + size_t post_mask; + size_t mid_count = mi_bitmap_mask_across(bitmap_idx, bitmap_fields, count, &pre_mask, &mid_mask, &post_mask); + bool all_ones = true; + bool any_ones = false; + size_t one_count = 0; + mi_bitmap_field_t* field = &bitmap[idx]; + size_t prev = mi_atomic_load_relaxed(field++); + if ((prev & pre_mask) != pre_mask) all_ones = false; + if ((prev & pre_mask) != 0) { any_ones = true; one_count += mi_popcount(prev & pre_mask); } + while (mid_count-- > 0) { + prev = mi_atomic_load_relaxed(field++); + if ((prev & mid_mask) != mid_mask) all_ones = false; + if ((prev & mid_mask) != 0) { any_ones = true; one_count += mi_popcount(prev & mid_mask); } + } + if (post_mask!=0) { + prev = mi_atomic_load_relaxed(field); + if ((prev & post_mask) != post_mask) all_ones = false; + if ((prev & post_mask) != 0) { any_ones = true; one_count += mi_popcount(prev & post_mask); } + } + if (pany_ones != NULL) { *pany_ones = any_ones; } + if (already_set != NULL) { *already_set = one_count; } + mi_assert_internal(all_ones ? one_count == count : one_count < count); + return all_ones; +} + +bool _mi_bitmap_is_claimed_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, size_t* already_set) { + return mi_bitmap_is_claimedx_across(bitmap, bitmap_fields, count, bitmap_idx, NULL, already_set); +} + +bool _mi_bitmap_is_any_claimed_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx) { + bool any_ones; + mi_bitmap_is_claimedx_across(bitmap, bitmap_fields, count, bitmap_idx, &any_ones, NULL); + return any_ones; +} diff --git a/compat/mimalloc/bitmap.h b/compat/mimalloc/bitmap.h new file mode 100644 index 00000000000000..0f4744f4fc3ffd --- /dev/null +++ b/compat/mimalloc/bitmap.h @@ -0,0 +1,119 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2023 Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ---------------------------------------------------------------------------- +Concurrent bitmap that can set/reset sequences of bits atomically, +represented as an array of fields where each field is a machine word (`size_t`) + +There are two api's; the standard one cannot have sequences that cross +between the bitmap fields (and a sequence must be <= MI_BITMAP_FIELD_BITS). +(this is used in region allocation) + +The `_across` postfixed functions do allow sequences that can cross over +between the fields. (This is used in arena allocation) +---------------------------------------------------------------------------- */ +#pragma once +#ifndef MI_BITMAP_H +#define MI_BITMAP_H + +/* ----------------------------------------------------------- + Bitmap definition +----------------------------------------------------------- */ + +#define MI_BITMAP_FIELD_BITS (8*MI_SIZE_SIZE) +#define MI_BITMAP_FIELD_FULL (~((size_t)0)) // all bits set + +// An atomic bitmap of `size_t` fields +typedef _Atomic(size_t) mi_bitmap_field_t; +typedef mi_bitmap_field_t* mi_bitmap_t; + +// A bitmap index is the index of the bit in a bitmap. +typedef size_t mi_bitmap_index_t; + +// Create a bit index. +static inline mi_bitmap_index_t mi_bitmap_index_create_ex(size_t idx, size_t bitidx) { + mi_assert_internal(bitidx <= MI_BITMAP_FIELD_BITS); + return (idx*MI_BITMAP_FIELD_BITS) + bitidx; +} +static inline mi_bitmap_index_t mi_bitmap_index_create(size_t idx, size_t bitidx) { + mi_assert_internal(bitidx < MI_BITMAP_FIELD_BITS); + return mi_bitmap_index_create_ex(idx,bitidx); +} + +// Create a bit index. +static inline mi_bitmap_index_t mi_bitmap_index_create_from_bit(size_t full_bitidx) { + return mi_bitmap_index_create(full_bitidx / MI_BITMAP_FIELD_BITS, full_bitidx % MI_BITMAP_FIELD_BITS); +} + +// Get the field index from a bit index. +static inline size_t mi_bitmap_index_field(mi_bitmap_index_t bitmap_idx) { + return (bitmap_idx / MI_BITMAP_FIELD_BITS); +} + +// Get the bit index in a bitmap field +static inline size_t mi_bitmap_index_bit_in_field(mi_bitmap_index_t bitmap_idx) { + return (bitmap_idx % MI_BITMAP_FIELD_BITS); +} + +// Get the full bit index +static inline size_t mi_bitmap_index_bit(mi_bitmap_index_t bitmap_idx) { + return bitmap_idx; +} + +/* ----------------------------------------------------------- + Claim a bit sequence atomically +----------------------------------------------------------- */ + +// Try to atomically claim a sequence of `count` bits in a single +// field at `idx` in `bitmap`. Returns `true` on success. +bool _mi_bitmap_try_find_claim_field(mi_bitmap_t bitmap, size_t idx, const size_t count, mi_bitmap_index_t* bitmap_idx); + +// Starts at idx, and wraps around to search in all `bitmap_fields` fields. +// For now, `count` can be at most MI_BITMAP_FIELD_BITS and will never cross fields. +bool _mi_bitmap_try_find_from_claim(mi_bitmap_t bitmap, const size_t bitmap_fields, const size_t start_field_idx, const size_t count, mi_bitmap_index_t* bitmap_idx); + +// Like _mi_bitmap_try_find_from_claim but with an extra predicate that must be fullfilled +typedef bool (mi_cdecl *mi_bitmap_pred_fun_t)(mi_bitmap_index_t bitmap_idx, void* pred_arg); +bool _mi_bitmap_try_find_from_claim_pred(mi_bitmap_t bitmap, const size_t bitmap_fields, const size_t start_field_idx, const size_t count, mi_bitmap_pred_fun_t pred_fun, void* pred_arg, mi_bitmap_index_t* bitmap_idx); + +// Set `count` bits at `bitmap_idx` to 0 atomically +// Returns `true` if all `count` bits were 1 previously. +bool _mi_bitmap_unclaim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); + +// Try to set `count` bits at `bitmap_idx` from 0 to 1 atomically. +// Returns `true` if successful when all previous `count` bits were 0. +bool _mi_bitmap_try_claim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); + +// Set `count` bits at `bitmap_idx` to 1 atomically +// Returns `true` if all `count` bits were 0 previously. `any_zero` is `true` if there was at least one zero bit. +bool _mi_bitmap_claim(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* any_zero); + +bool _mi_bitmap_is_claimed(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); +bool _mi_bitmap_is_any_claimed(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); + + +//-------------------------------------------------------------------------- +// the `_across` functions work on bitmaps where sequences can cross over +// between the fields. This is used in arena allocation +//-------------------------------------------------------------------------- + +// Find `count` bits of zeros and set them to 1 atomically; returns `true` on success. +// Starts at idx, and wraps around to search in all `bitmap_fields` fields. +bool _mi_bitmap_try_find_from_claim_across(mi_bitmap_t bitmap, const size_t bitmap_fields, const size_t start_field_idx, const size_t count, mi_bitmap_index_t* bitmap_idx); + +// Set `count` bits at `bitmap_idx` to 0 atomically +// Returns `true` if all `count` bits were 1 previously. +bool _mi_bitmap_unclaim_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); + +// Set `count` bits at `bitmap_idx` to 1 atomically +// Returns `true` if all `count` bits were 0 previously. `any_zero` is `true` if there was at least one zero bit. +bool _mi_bitmap_claim_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, bool* pany_zero, size_t* already_set); + +bool _mi_bitmap_is_claimed_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx, size_t* already_set); +bool _mi_bitmap_is_any_claimed_across(mi_bitmap_t bitmap, size_t bitmap_fields, size_t count, mi_bitmap_index_t bitmap_idx); + +#endif diff --git a/compat/mimalloc/free.c b/compat/mimalloc/free.c new file mode 100644 index 00000000000000..5e5ae443f3a3a3 --- /dev/null +++ b/compat/mimalloc/free.c @@ -0,0 +1,572 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#if !defined(MI_IN_ALLOC_C) +#error "this file should be included from 'alloc.c' (so aliases can work from alloc-override)" +// add includes help an IDE +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" // _mi_prim_thread_id() +#endif + +// forward declarations +static void mi_check_padding(const mi_page_t* page, const mi_block_t* block); +static bool mi_check_is_double_free(const mi_page_t* page, const mi_block_t* block); +static size_t mi_page_usable_size_of(const mi_page_t* page, const mi_block_t* block); +static void mi_stat_free(const mi_page_t* page, const mi_block_t* block); + + +// ------------------------------------------------------ +// Free +// ------------------------------------------------------ + +// forward declaration of multi-threaded free (`_mt`) (or free in huge block if compiled with MI_HUGE_PAGE_ABANDON) +static mi_decl_noinline void mi_free_block_mt(mi_page_t* page, mi_segment_t* segment, mi_block_t* block); + +// regular free of a (thread local) block pointer +// fast path written carefully to prevent spilling on the stack +static inline void mi_free_block_local(mi_page_t* page, mi_block_t* block, bool track_stats, bool check_full) +{ + // checks + if mi_unlikely(mi_check_is_double_free(page, block)) return; + mi_check_padding(page, block); + if (track_stats) { mi_stat_free(page, block); } + #if (MI_DEBUG>0) && !MI_TRACK_ENABLED && !MI_TSAN && !MI_GUARDED + if (!mi_page_is_huge(page)) { // huge page content may be already decommitted + memset(block, MI_DEBUG_FREED, mi_page_block_size(page)); + } + #endif + if (track_stats) { mi_track_free_size(block, mi_page_usable_size_of(page, block)); } // faster then mi_usable_size as we already know the page and that p is unaligned + + // actual free: push on the local free list + mi_block_set_next(page, block, page->local_free); + page->local_free = block; + if mi_unlikely(--page->used == 0) { + _mi_page_retire(page); + } + else if mi_unlikely(check_full && mi_page_is_in_full(page)) { + _mi_page_unfull(page); + } +} + +// Adjust a block that was allocated aligned, to the actual start of the block in the page. +// note: this can be called from `mi_free_generic_mt` where a non-owning thread accesses the +// `page_start` and `block_size` fields; however these are constant and the page won't be +// deallocated (as the block we are freeing keeps it alive) and thus safe to read concurrently. +mi_block_t* _mi_page_ptr_unalign(const mi_page_t* page, const void* p) { + mi_assert_internal(page!=NULL && p!=NULL); + + size_t diff = (uint8_t*)p - page->page_start; + size_t adjust; + if mi_likely(page->block_size_shift != 0) { + adjust = diff & (((size_t)1 << page->block_size_shift) - 1); + } + else { + adjust = diff % mi_page_block_size(page); + } + + return (mi_block_t*)((uintptr_t)p - adjust); +} + +// forward declaration for a MI_GUARDED build +#if MI_GUARDED +static void mi_block_unguard(mi_page_t* page, mi_block_t* block, void* p); // forward declaration +static inline void mi_block_check_unguard(mi_page_t* page, mi_block_t* block, void* p) { + if (mi_block_ptr_is_guarded(block, p)) { mi_block_unguard(page, block, p); } +} +#else +static inline void mi_block_check_unguard(mi_page_t* page, mi_block_t* block, void* p) { + MI_UNUSED(page); MI_UNUSED(block); MI_UNUSED(p); +} +#endif + +// free a local pointer (page parameter comes first for better codegen) +static void mi_decl_noinline mi_free_generic_local(mi_page_t* page, mi_segment_t* segment, void* p) mi_attr_noexcept { + MI_UNUSED(segment); + mi_block_t* const block = (mi_page_has_aligned(page) ? _mi_page_ptr_unalign(page, p) : (mi_block_t*)p); + mi_block_check_unguard(page, block, p); + mi_free_block_local(page, block, true /* track stats */, true /* check for a full page */); +} + +// free a pointer owned by another thread (page parameter comes first for better codegen) +static void mi_decl_noinline mi_free_generic_mt(mi_page_t* page, mi_segment_t* segment, void* p) mi_attr_noexcept { + mi_block_t* const block = _mi_page_ptr_unalign(page, p); // don't check `has_aligned` flag to avoid a race (issue #865) + mi_block_check_unguard(page, block, p); + mi_free_block_mt(page, segment, block); +} + +// generic free (for runtime integration) +void mi_decl_noinline _mi_free_generic(mi_segment_t* segment, mi_page_t* page, bool is_local, void* p) mi_attr_noexcept { + if (is_local) mi_free_generic_local(page,segment,p); + else mi_free_generic_mt(page,segment,p); +} + +// Get the segment data belonging to a pointer +// This is just a single `and` in release mode but does further checks in debug mode +// (and secure mode) to see if this was a valid pointer. +static inline mi_segment_t* mi_checked_ptr_segment(const void* p, const char* msg) +{ + MI_UNUSED(msg); + + #if (MI_DEBUG>0) + if mi_unlikely(((uintptr_t)p & (MI_INTPTR_SIZE - 1)) != 0 && !mi_option_is_enabled(mi_option_guarded_precise)) { + _mi_error_message(EINVAL, "%s: invalid (unaligned) pointer: %p\n", msg, p); + return NULL; + } + #endif + + mi_segment_t* const segment = _mi_ptr_segment(p); + if mi_unlikely(segment==NULL) return segment; + + #if (MI_DEBUG>0) + if mi_unlikely(!mi_is_in_heap_region(p)) { + #if (MI_INTPTR_SIZE == 8 && defined(__linux__)) + if (((uintptr_t)p >> 40) != 0x7F) { // linux tends to align large blocks above 0x7F000000000 (issue #640) + #else + { + #endif + _mi_warning_message("%s: pointer might not point to a valid heap region: %p\n" + "(this may still be a valid very large allocation (over 64MiB))\n", msg, p); + if mi_likely(_mi_ptr_cookie(segment) == segment->cookie) { + _mi_warning_message("(yes, the previous pointer %p was valid after all)\n", p); + } + } + } + #endif + #if (MI_DEBUG>0 || MI_SECURE>=4) + if mi_unlikely(_mi_ptr_cookie(segment) != segment->cookie) { + _mi_error_message(EINVAL, "%s: pointer does not point to a valid heap space: %p\n", msg, p); + return NULL; + } + #endif + + return segment; +} + +// Free a block +// Fast path written carefully to prevent register spilling on the stack +void mi_free(void* p) mi_attr_noexcept +{ + mi_segment_t* const segment = mi_checked_ptr_segment(p,"mi_free"); + if mi_unlikely(segment==NULL) return; + + const bool is_local = (_mi_prim_thread_id() == mi_atomic_load_relaxed(&segment->thread_id)); + mi_page_t* const page = _mi_segment_page_of(segment, p); + + if mi_likely(is_local) { // thread-local free? + if mi_likely(page->flags.full_aligned == 0) { // and it is not a full page (full pages need to move from the full bin), nor has aligned blocks (aligned blocks need to be unaligned) + // thread-local, aligned, and not a full page + mi_block_t* const block = (mi_block_t*)p; + mi_free_block_local(page, block, true /* track stats */, false /* no need to check if the page is full */); + } + else { + // page is full or contains (inner) aligned blocks; use generic path + mi_free_generic_local(page, segment, p); + } + } + else { + // not thread-local; use generic path + mi_free_generic_mt(page, segment, p); + } +} + +// return true if successful +bool _mi_free_delayed_block(mi_block_t* block) { + // get segment and page + mi_assert_internal(block!=NULL); + const mi_segment_t* const segment = _mi_ptr_segment(block); + mi_assert_internal(_mi_ptr_cookie(segment) == segment->cookie); + mi_assert_internal(_mi_thread_id() == segment->thread_id); + mi_page_t* const page = _mi_segment_page_of(segment, block); + + // Clear the no-delayed flag so delayed freeing is used again for this page. + // This must be done before collecting the free lists on this page -- otherwise + // some blocks may end up in the page `thread_free` list with no blocks in the + // heap `thread_delayed_free` list which may cause the page to be never freed! + // (it would only be freed if we happen to scan it in `mi_page_queue_find_free_ex`) + if (!_mi_page_try_use_delayed_free(page, MI_USE_DELAYED_FREE, false /* dont overwrite never delayed */)) { + return false; + } + + // collect all other non-local frees (move from `thread_free` to `free`) to ensure up-to-date `used` count + _mi_page_free_collect(page, false); + + // and free the block (possibly freeing the page as well since `used` is updated) + mi_free_block_local(page, block, false /* stats have already been adjusted */, true /* check for a full page */); + return true; +} + +// ------------------------------------------------------ +// Multi-threaded Free (`_mt`) +// ------------------------------------------------------ + +// Push a block that is owned by another thread on its page-local thread free +// list or it's heap delayed free list. Such blocks are later collected by +// the owning thread in `_mi_free_delayed_block`. +static void mi_decl_noinline mi_free_block_delayed_mt( mi_page_t* page, mi_block_t* block ) +{ + // Try to put the block on either the page-local thread free list, + // or the heap delayed free list (if this is the first non-local free in that page) + mi_thread_free_t tfreex; + bool use_delayed; + mi_thread_free_t tfree = mi_atomic_load_relaxed(&page->xthread_free); + do { + use_delayed = (mi_tf_delayed(tfree) == MI_USE_DELAYED_FREE); + if mi_unlikely(use_delayed) { + // unlikely: this only happens on the first concurrent free in a page that is in the full list + tfreex = mi_tf_set_delayed(tfree,MI_DELAYED_FREEING); + } + else { + // usual: directly add to page thread_free list + mi_block_set_next(page, block, mi_tf_block(tfree)); + tfreex = mi_tf_set_block(tfree,block); + } + } while (!mi_atomic_cas_weak_release(&page->xthread_free, &tfree, tfreex)); + + // If this was the first non-local free, we need to push it on the heap delayed free list instead + if mi_unlikely(use_delayed) { + // racy read on `heap`, but ok because MI_DELAYED_FREEING is set (see `mi_heap_delete` and `mi_heap_collect_abandon`) + mi_heap_t* const heap = (mi_heap_t*)(mi_atomic_load_acquire(&page->xheap)); //mi_page_heap(page); + mi_assert_internal(heap != NULL); + if (heap != NULL) { + // add to the delayed free list of this heap. (do this atomically as the lock only protects heap memory validity) + mi_block_t* dfree = mi_atomic_load_ptr_relaxed(mi_block_t, &heap->thread_delayed_free); + do { + mi_block_set_nextx(heap,block,dfree, heap->keys); + } while (!mi_atomic_cas_ptr_weak_release(mi_block_t,&heap->thread_delayed_free, &dfree, block)); + } + + // and reset the MI_DELAYED_FREEING flag + tfree = mi_atomic_load_relaxed(&page->xthread_free); + do { + tfreex = tfree; + mi_assert_internal(mi_tf_delayed(tfree) == MI_DELAYED_FREEING); + tfreex = mi_tf_set_delayed(tfree,MI_NO_DELAYED_FREE); + } while (!mi_atomic_cas_weak_release(&page->xthread_free, &tfree, tfreex)); + } +} + +// Multi-threaded free (`_mt`) (or free in huge block if compiled with MI_HUGE_PAGE_ABANDON) +static void mi_decl_noinline mi_free_block_mt(mi_page_t* page, mi_segment_t* segment, mi_block_t* block) +{ + // first see if the segment was abandoned and if we can reclaim it into our thread + if (_mi_option_get_fast(mi_option_abandoned_reclaim_on_free) != 0 && + #if MI_HUGE_PAGE_ABANDON + segment->page_kind != MI_PAGE_HUGE && + #endif + mi_atomic_load_relaxed(&segment->thread_id) == 0 && // segment is abandoned? + mi_prim_get_default_heap() != (mi_heap_t*)&_mi_heap_empty) // and we did not already exit this thread (without this check, a fresh heap will be initalized (issue #944)) + { + // the segment is abandoned, try to reclaim it into our heap + if (_mi_segment_attempt_reclaim(mi_heap_get_default(), segment)) { + mi_assert_internal(_mi_thread_id() == mi_atomic_load_relaxed(&segment->thread_id)); + mi_assert_internal(mi_heap_get_default()->tld->segments.subproc == segment->subproc); + mi_free(block); // recursively free as now it will be a local free in our heap + return; + } + } + + // The padding check may access the non-thread-owned page for the key values. + // that is safe as these are constant and the page won't be freed (as the block is not freed yet). + mi_check_padding(page, block); + + // adjust stats (after padding check and potentially recursive `mi_free` above) + mi_stat_free(page, block); // stat_free may access the padding + mi_track_free_size(block, mi_page_usable_size_of(page,block)); + + // for small size, ensure we can fit the delayed thread pointers without triggering overflow detection + _mi_padding_shrink(page, block, sizeof(mi_block_t)); + + if (segment->kind == MI_SEGMENT_HUGE) { + #if MI_HUGE_PAGE_ABANDON + // huge page segments are always abandoned and can be freed immediately + _mi_segment_huge_page_free(segment, page, block); + return; + #else + // huge pages are special as they occupy the entire segment + // as these are large we reset the memory occupied by the page so it is available to other threads + // (as the owning thread needs to actually free the memory later). + _mi_segment_huge_page_reset(segment, page, block); + #endif + } + else { + #if (MI_DEBUG>0) && !MI_TRACK_ENABLED && !MI_TSAN // note: when tracking, cannot use mi_usable_size with multi-threading + memset(block, MI_DEBUG_FREED, mi_usable_size(block)); + #endif + } + + // and finally free the actual block by pushing it on the owning heap + // thread_delayed free list (or heap delayed free list) + mi_free_block_delayed_mt(page,block); +} + + +// ------------------------------------------------------ +// Usable size +// ------------------------------------------------------ + +// Bytes available in a block +static size_t mi_decl_noinline mi_page_usable_aligned_size_of(const mi_page_t* page, const void* p) mi_attr_noexcept { + const mi_block_t* block = _mi_page_ptr_unalign(page, p); + const size_t size = mi_page_usable_size_of(page, block); + const ptrdiff_t adjust = (uint8_t*)p - (uint8_t*)block; + mi_assert_internal(adjust >= 0 && (size_t)adjust <= size); + const size_t aligned_size = (size - adjust); + #if MI_GUARDED + if (mi_block_ptr_is_guarded(block, p)) { + return aligned_size - _mi_os_page_size(); + } + #endif + return aligned_size; +} + +static inline size_t _mi_usable_size(const void* p, const char* msg) mi_attr_noexcept { + const mi_segment_t* const segment = mi_checked_ptr_segment(p, msg); + if mi_unlikely(segment==NULL) return 0; + const mi_page_t* const page = _mi_segment_page_of(segment, p); + if mi_likely(!mi_page_has_aligned(page)) { + const mi_block_t* block = (const mi_block_t*)p; + return mi_page_usable_size_of(page, block); + } + else { + // split out to separate routine for improved code generation + return mi_page_usable_aligned_size_of(page, p); + } +} + +mi_decl_nodiscard size_t mi_usable_size(const void* p) mi_attr_noexcept { + return _mi_usable_size(p, "mi_usable_size"); +} + + +// ------------------------------------------------------ +// Free variants +// ------------------------------------------------------ + +void mi_free_size(void* p, size_t size) mi_attr_noexcept { + MI_UNUSED_RELEASE(size); + #if MI_DEBUG + const size_t available = _mi_usable_size(p,"mi_free_size"); + mi_assert(p == NULL || size <= available || available == 0 /* invalid pointer */ ); + #endif + mi_free(p); +} + +void mi_free_size_aligned(void* p, size_t size, size_t alignment) mi_attr_noexcept { + MI_UNUSED_RELEASE(alignment); + mi_assert(((uintptr_t)p % alignment) == 0); + mi_free_size(p,size); +} + +void mi_free_aligned(void* p, size_t alignment) mi_attr_noexcept { + MI_UNUSED_RELEASE(alignment); + mi_assert(((uintptr_t)p % alignment) == 0); + mi_free(p); +} + + +// ------------------------------------------------------ +// Check for double free in secure and debug mode +// This is somewhat expensive so only enabled for secure mode 4 +// ------------------------------------------------------ + +#if (MI_ENCODE_FREELIST && (MI_SECURE>=4 || MI_DEBUG!=0)) +// linear check if the free list contains a specific element +static bool mi_list_contains(const mi_page_t* page, const mi_block_t* list, const mi_block_t* elem) { + while (list != NULL) { + if (elem==list) return true; + list = mi_block_next(page, list); + } + return false; +} + +static mi_decl_noinline bool mi_check_is_double_freex(const mi_page_t* page, const mi_block_t* block) { + // The decoded value is in the same page (or NULL). + // Walk the free lists to verify positively if it is already freed + if (mi_list_contains(page, page->free, block) || + mi_list_contains(page, page->local_free, block) || + mi_list_contains(page, mi_page_thread_free(page), block)) + { + _mi_error_message(EAGAIN, "double free detected of block %p with size %zu\n", block, mi_page_block_size(page)); + return true; + } + return false; +} + +#define mi_track_page(page,access) { size_t psize; void* pstart = _mi_page_start(_mi_page_segment(page),page,&psize); mi_track_mem_##access( pstart, psize); } + +static inline bool mi_check_is_double_free(const mi_page_t* page, const mi_block_t* block) { + bool is_double_free = false; + mi_block_t* n = mi_block_nextx(page, block, page->keys); // pretend it is freed, and get the decoded first field + if (((uintptr_t)n & (MI_INTPTR_SIZE-1))==0 && // quick check: aligned pointer? + (n==NULL || mi_is_in_same_page(block, n))) // quick check: in same page or NULL? + { + // Suspicious: decoded value a in block is in the same page (or NULL) -- maybe a double free? + // (continue in separate function to improve code generation) + is_double_free = mi_check_is_double_freex(page, block); + } + return is_double_free; +} +#else +static inline bool mi_check_is_double_free(const mi_page_t* page, const mi_block_t* block) { + MI_UNUSED(page); + MI_UNUSED(block); + return false; +} +#endif + + +// --------------------------------------------------------------------------- +// Check for heap block overflow by setting up padding at the end of the block +// --------------------------------------------------------------------------- + +#if MI_PADDING // && !MI_TRACK_ENABLED +static bool mi_page_decode_padding(const mi_page_t* page, const mi_block_t* block, size_t* delta, size_t* bsize) { + *bsize = mi_page_usable_block_size(page); + const mi_padding_t* const padding = (mi_padding_t*)((uint8_t*)block + *bsize); + mi_track_mem_defined(padding,sizeof(mi_padding_t)); + *delta = padding->delta; + uint32_t canary = padding->canary; + uintptr_t keys[2]; + keys[0] = page->keys[0]; + keys[1] = page->keys[1]; + bool ok = (mi_ptr_encode_canary(page,block,keys) == canary && *delta <= *bsize); + mi_track_mem_noaccess(padding,sizeof(mi_padding_t)); + return ok; +} + +// Return the exact usable size of a block. +static size_t mi_page_usable_size_of(const mi_page_t* page, const mi_block_t* block) { + size_t bsize; + size_t delta; + bool ok = mi_page_decode_padding(page, block, &delta, &bsize); + mi_assert_internal(ok); mi_assert_internal(delta <= bsize); + return (ok ? bsize - delta : 0); +} + +// When a non-thread-local block is freed, it becomes part of the thread delayed free +// list that is freed later by the owning heap. If the exact usable size is too small to +// contain the pointer for the delayed list, then shrink the padding (by decreasing delta) +// so it will later not trigger an overflow error in `mi_free_block`. +void _mi_padding_shrink(const mi_page_t* page, const mi_block_t* block, const size_t min_size) { + size_t bsize; + size_t delta; + bool ok = mi_page_decode_padding(page, block, &delta, &bsize); + mi_assert_internal(ok); + if (!ok || (bsize - delta) >= min_size) return; // usually already enough space + mi_assert_internal(bsize >= min_size); + if (bsize < min_size) return; // should never happen + size_t new_delta = (bsize - min_size); + mi_assert_internal(new_delta < bsize); + mi_padding_t* padding = (mi_padding_t*)((uint8_t*)block + bsize); + mi_track_mem_defined(padding,sizeof(mi_padding_t)); + padding->delta = (uint32_t)new_delta; + mi_track_mem_noaccess(padding,sizeof(mi_padding_t)); +} +#else +static size_t mi_page_usable_size_of(const mi_page_t* page, const mi_block_t* block) { + MI_UNUSED(block); + return mi_page_usable_block_size(page); +} + +void _mi_padding_shrink(const mi_page_t* page, const mi_block_t* block, const size_t min_size) { + MI_UNUSED(page); + MI_UNUSED(block); + MI_UNUSED(min_size); +} +#endif + +#if MI_PADDING && MI_PADDING_CHECK + +static bool mi_verify_padding(const mi_page_t* page, const mi_block_t* block, size_t* size, size_t* wrong) { + size_t bsize; + size_t delta; + bool ok = mi_page_decode_padding(page, block, &delta, &bsize); + *size = *wrong = bsize; + if (!ok) return false; + mi_assert_internal(bsize >= delta); + *size = bsize - delta; + if (!mi_page_is_huge(page)) { + uint8_t* fill = (uint8_t*)block + bsize - delta; + const size_t maxpad = (delta > MI_MAX_ALIGN_SIZE ? MI_MAX_ALIGN_SIZE : delta); // check at most the first N padding bytes + mi_track_mem_defined(fill, maxpad); + for (size_t i = 0; i < maxpad; i++) { + if (fill[i] != MI_DEBUG_PADDING) { + *wrong = bsize - delta + i; + ok = false; + break; + } + } + mi_track_mem_noaccess(fill, maxpad); + } + return ok; +} + +static void mi_check_padding(const mi_page_t* page, const mi_block_t* block) { + size_t size; + size_t wrong; + if (!mi_verify_padding(page,block,&size,&wrong)) { + _mi_error_message(EFAULT, "buffer overflow in heap block %p of size %zu: write after %zu bytes\n", block, size, wrong ); + } +} + +#else + +static void mi_check_padding(const mi_page_t* page, const mi_block_t* block) { + MI_UNUSED(page); + MI_UNUSED(block); +} + +#endif + +// only maintain stats for smaller objects if requested +#if (MI_STAT>0) +static void mi_stat_free(const mi_page_t* page, const mi_block_t* block) { + MI_UNUSED(block); + mi_heap_t* const heap = mi_heap_get_default(); + const size_t bsize = mi_page_usable_block_size(page); + // #if (MI_STAT>1) + // const size_t usize = mi_page_usable_size_of(page, block); + // mi_heap_stat_decrease(heap, malloc_requested, usize); + // #endif + if (bsize <= MI_MEDIUM_OBJ_SIZE_MAX) { + mi_heap_stat_decrease(heap, malloc_normal, bsize); + #if (MI_STAT > 1) + mi_heap_stat_decrease(heap, malloc_bins[_mi_bin(bsize)], 1); + #endif + } + //else if (bsize <= MI_LARGE_OBJ_SIZE_MAX) { + // mi_heap_stat_decrease(heap, malloc_large, bsize); + //} + else { + mi_heap_stat_decrease(heap, malloc_huge, bsize); + } +} +#else +static void mi_stat_free(const mi_page_t* page, const mi_block_t* block) { + MI_UNUSED(page); MI_UNUSED(block); +} +#endif + + +// Remove guard page when building with MI_GUARDED +#if MI_GUARDED +static void mi_block_unguard(mi_page_t* page, mi_block_t* block, void* p) { + MI_UNUSED(p); + mi_assert_internal(mi_block_ptr_is_guarded(block, p)); + mi_assert_internal(mi_page_has_aligned(page)); + mi_assert_internal((uint8_t*)p - (uint8_t*)block >= (ptrdiff_t)sizeof(mi_block_t)); + mi_assert_internal(block->next == MI_BLOCK_TAG_GUARDED); + + const size_t bsize = mi_page_block_size(page); + const size_t psize = _mi_os_page_size(); + mi_assert_internal(bsize > psize); + mi_assert_internal(_mi_page_segment(page)->allow_decommit); + void* gpage = (uint8_t*)block + bsize - psize; + mi_assert_internal(_mi_is_aligned(gpage, psize)); + _mi_os_unprotect(gpage, psize); +} +#endif diff --git a/compat/mimalloc/heap.c b/compat/mimalloc/heap.c new file mode 100644 index 00000000000000..f96e60d0f8d94c --- /dev/null +++ b/compat/mimalloc/heap.c @@ -0,0 +1,733 @@ +/*---------------------------------------------------------------------------- +Copyright (c) 2018-2021, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "mimalloc/prim.h" // mi_prim_get_default_heap + +#include // memset, memcpy + +#if defined(_MSC_VER) && (_MSC_VER < 1920) +#pragma warning(disable:4204) // non-constant aggregate initializer +#endif + +/* ----------------------------------------------------------- + Helpers +----------------------------------------------------------- */ + +// return `true` if ok, `false` to break +typedef bool (heap_page_visitor_fun)(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* arg1, void* arg2); + +// Visit all pages in a heap; returns `false` if break was called. +static bool mi_heap_visit_pages(mi_heap_t* heap, heap_page_visitor_fun* fn, void* arg1, void* arg2) +{ + if (heap==NULL || heap->page_count==0) return 0; + + // visit all pages + #if MI_DEBUG>1 + size_t total = heap->page_count; + size_t count = 0; + #endif + + for (size_t i = 0; i <= MI_BIN_FULL; i++) { + mi_page_queue_t* pq = &heap->pages[i]; + mi_page_t* page = pq->first; + while(page != NULL) { + mi_page_t* next = page->next; // save next in case the page gets removed from the queue + mi_assert_internal(mi_page_heap(page) == heap); + #if MI_DEBUG>1 + count++; + #endif + if (!fn(heap, pq, page, arg1, arg2)) return false; + page = next; // and continue + } + } + mi_assert_internal(count == total); + return true; +} + + +#if MI_DEBUG>=2 +static bool mi_heap_page_is_valid(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* arg1, void* arg2) { + MI_UNUSED(arg1); + MI_UNUSED(arg2); + MI_UNUSED(pq); + mi_assert_internal(mi_page_heap(page) == heap); + mi_segment_t* segment = _mi_page_segment(page); + mi_assert_internal(mi_atomic_load_relaxed(&segment->thread_id) == heap->thread_id); + mi_assert_expensive(_mi_page_is_valid(page)); + return true; +} +#endif +#if MI_DEBUG>=3 +static bool mi_heap_is_valid(mi_heap_t* heap) { + mi_assert_internal(heap!=NULL); + mi_heap_visit_pages(heap, &mi_heap_page_is_valid, NULL, NULL); + return true; +} +#endif + + + + +/* ----------------------------------------------------------- + "Collect" pages by migrating `local_free` and `thread_free` + lists and freeing empty pages. This is done when a thread + stops (and in that case abandons pages if there are still + blocks alive) +----------------------------------------------------------- */ + +typedef enum mi_collect_e { + MI_NORMAL, + MI_FORCE, + MI_ABANDON +} mi_collect_t; + + +static bool mi_heap_page_collect(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* arg_collect, void* arg2 ) { + MI_UNUSED(arg2); + MI_UNUSED(heap); + mi_assert_internal(mi_heap_page_is_valid(heap, pq, page, NULL, NULL)); + mi_collect_t collect = *((mi_collect_t*)arg_collect); + _mi_page_free_collect(page, collect >= MI_FORCE); + if (collect == MI_FORCE) { + // note: call before a potential `_mi_page_free` as the segment may be freed if this was the last used page in that segment. + mi_segment_t* segment = _mi_page_segment(page); + _mi_segment_collect(segment, true /* force? */); + } + if (mi_page_all_free(page)) { + // no more used blocks, free the page. + // note: this will free retired pages as well. + _mi_page_free(page, pq, collect >= MI_FORCE); + } + else if (collect == MI_ABANDON) { + // still used blocks but the thread is done; abandon the page + _mi_page_abandon(page, pq); + } + return true; // don't break +} + +static bool mi_heap_page_never_delayed_free(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* arg1, void* arg2) { + MI_UNUSED(arg1); + MI_UNUSED(arg2); + MI_UNUSED(heap); + MI_UNUSED(pq); + _mi_page_use_delayed_free(page, MI_NEVER_DELAYED_FREE, false); + return true; // don't break +} + +static void mi_heap_collect_ex(mi_heap_t* heap, mi_collect_t collect) +{ + if (heap==NULL || !mi_heap_is_initialized(heap)) return; + + const bool force = (collect >= MI_FORCE); + _mi_deferred_free(heap, force); + + // python/cpython#112532: we may be called from a thread that is not the owner of the heap + const bool is_main_thread = (_mi_is_main_thread() && heap->thread_id == _mi_thread_id()); + + // note: never reclaim on collect but leave it to threads that need storage to reclaim + const bool force_main = + #ifdef NDEBUG + collect == MI_FORCE + #else + collect >= MI_FORCE + #endif + && is_main_thread && mi_heap_is_backing(heap) && !heap->no_reclaim; + + if (force_main) { + // the main thread is abandoned (end-of-program), try to reclaim all abandoned segments. + // if all memory is freed by now, all segments should be freed. + // note: this only collects in the current subprocess + _mi_abandoned_reclaim_all(heap, &heap->tld->segments); + } + + // if abandoning, mark all pages to no longer add to delayed_free + if (collect == MI_ABANDON) { + mi_heap_visit_pages(heap, &mi_heap_page_never_delayed_free, NULL, NULL); + } + + // free all current thread delayed blocks. + // (if abandoning, after this there are no more thread-delayed references into the pages.) + _mi_heap_delayed_free_all(heap); + + // collect retired pages + _mi_heap_collect_retired(heap, force); + + // collect all pages owned by this thread + mi_heap_visit_pages(heap, &mi_heap_page_collect, &collect, NULL); + mi_assert_internal( collect != MI_ABANDON || mi_atomic_load_ptr_acquire(mi_block_t,&heap->thread_delayed_free) == NULL ); + + // collect abandoned segments (in particular, purge expired parts of segments in the abandoned segment list) + // note: forced purge can be quite expensive if many threads are created/destroyed so we do not force on abandonment + _mi_abandoned_collect(heap, collect == MI_FORCE /* force? */, &heap->tld->segments); + + // if forced, collect thread data cache on program-exit (or shared library unload) + if (force && is_main_thread && mi_heap_is_backing(heap)) { + _mi_thread_data_collect(); // collect thread data cache + } + + // collect arenas (this is program wide so don't force purges on abandonment of threads) + _mi_arenas_collect(collect == MI_FORCE /* force purge? */); + + // merge statistics + if (collect <= MI_FORCE) { _mi_stats_merge_thread(heap->tld); } +} + +void _mi_heap_collect_abandon(mi_heap_t* heap) { + mi_heap_collect_ex(heap, MI_ABANDON); +} + +void mi_heap_collect(mi_heap_t* heap, bool force) mi_attr_noexcept { + mi_heap_collect_ex(heap, (force ? MI_FORCE : MI_NORMAL)); +} + +void mi_collect(bool force) mi_attr_noexcept { + mi_heap_collect(mi_prim_get_default_heap(), force); +} + + +/* ----------------------------------------------------------- + Heap new +----------------------------------------------------------- */ + +mi_heap_t* mi_heap_get_default(void) { + mi_thread_init(); + return mi_prim_get_default_heap(); +} + +static bool mi_heap_is_default(const mi_heap_t* heap) { + return (heap == mi_prim_get_default_heap()); +} + + +mi_heap_t* mi_heap_get_backing(void) { + mi_heap_t* heap = mi_heap_get_default(); + mi_assert_internal(heap!=NULL); + mi_heap_t* bheap = heap->tld->heap_backing; + mi_assert_internal(bheap!=NULL); + mi_assert_internal(bheap->thread_id == _mi_thread_id()); + return bheap; +} + +void _mi_heap_init(mi_heap_t* heap, mi_tld_t* tld, mi_arena_id_t arena_id, bool noreclaim, uint8_t tag) { + _mi_memcpy_aligned(heap, &_mi_heap_empty, sizeof(mi_heap_t)); + heap->tld = tld; + heap->thread_id = _mi_thread_id(); + heap->arena_id = arena_id; + heap->no_reclaim = noreclaim; + heap->tag = tag; + if (heap == tld->heap_backing) { + _mi_random_init(&heap->random); + } + else { + _mi_random_split(&tld->heap_backing->random, &heap->random); + } + heap->cookie = _mi_heap_random_next(heap) | 1; + heap->keys[0] = _mi_heap_random_next(heap); + heap->keys[1] = _mi_heap_random_next(heap); + _mi_heap_guarded_init(heap); + // push on the thread local heaps list + heap->next = heap->tld->heaps; + heap->tld->heaps = heap; +} + +mi_decl_nodiscard mi_heap_t* mi_heap_new_ex(int heap_tag, bool allow_destroy, mi_arena_id_t arena_id) { + mi_heap_t* bheap = mi_heap_get_backing(); + mi_heap_t* heap = mi_heap_malloc_tp(bheap, mi_heap_t); // todo: OS allocate in secure mode? + if (heap == NULL) return NULL; + mi_assert(heap_tag >= 0 && heap_tag < 256); + _mi_heap_init(heap, bheap->tld, arena_id, allow_destroy /* no reclaim? */, (uint8_t)heap_tag /* heap tag */); + return heap; +} + +mi_decl_nodiscard mi_heap_t* mi_heap_new_in_arena(mi_arena_id_t arena_id) { + return mi_heap_new_ex(0 /* default heap tag */, false /* don't allow `mi_heap_destroy` */, arena_id); +} + +mi_decl_nodiscard mi_heap_t* mi_heap_new(void) { + // don't reclaim abandoned memory or otherwise destroy is unsafe + return mi_heap_new_ex(0 /* default heap tag */, true /* no reclaim */, _mi_arena_id_none()); +} + +bool _mi_heap_memid_is_suitable(mi_heap_t* heap, mi_memid_t memid) { + return _mi_arena_memid_is_suitable(memid, heap->arena_id); +} + +uintptr_t _mi_heap_random_next(mi_heap_t* heap) { + return _mi_random_next(&heap->random); +} + +// zero out the page queues +static void mi_heap_reset_pages(mi_heap_t* heap) { + mi_assert_internal(heap != NULL); + mi_assert_internal(mi_heap_is_initialized(heap)); + // TODO: copy full empty heap instead? + memset(&heap->pages_free_direct, 0, sizeof(heap->pages_free_direct)); + _mi_memcpy_aligned(&heap->pages, &_mi_heap_empty.pages, sizeof(heap->pages)); + heap->thread_delayed_free = NULL; + heap->page_count = 0; +} + +// called from `mi_heap_destroy` and `mi_heap_delete` to free the internal heap resources. +static void mi_heap_free(mi_heap_t* heap) { + mi_assert(heap != NULL); + mi_assert_internal(mi_heap_is_initialized(heap)); + if (heap==NULL || !mi_heap_is_initialized(heap)) return; + if (mi_heap_is_backing(heap)) return; // dont free the backing heap + + // reset default + if (mi_heap_is_default(heap)) { + _mi_heap_set_default_direct(heap->tld->heap_backing); + } + + // remove ourselves from the thread local heaps list + // linear search but we expect the number of heaps to be relatively small + mi_heap_t* prev = NULL; + mi_heap_t* curr = heap->tld->heaps; + while (curr != heap && curr != NULL) { + prev = curr; + curr = curr->next; + } + mi_assert_internal(curr == heap); + if (curr == heap) { + if (prev != NULL) { prev->next = heap->next; } + else { heap->tld->heaps = heap->next; } + } + mi_assert_internal(heap->tld->heaps != NULL); + + // and free the used memory + mi_free(heap); +} + +// return a heap on the same thread as `heap` specialized for the specified tag (if it exists) +mi_heap_t* _mi_heap_by_tag(mi_heap_t* heap, uint8_t tag) { + if (heap->tag == tag) { + return heap; + } + for (mi_heap_t *curr = heap->tld->heaps; curr != NULL; curr = curr->next) { + if (curr->tag == tag) { + return curr; + } + } + return NULL; +} + +/* ----------------------------------------------------------- + Heap destroy +----------------------------------------------------------- */ + +static bool _mi_heap_page_destroy(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* arg1, void* arg2) { + MI_UNUSED(arg1); + MI_UNUSED(arg2); + MI_UNUSED(heap); + MI_UNUSED(pq); + + // ensure no more thread_delayed_free will be added + _mi_page_use_delayed_free(page, MI_NEVER_DELAYED_FREE, false); + + // stats + const size_t bsize = mi_page_block_size(page); + if (bsize > MI_MEDIUM_OBJ_SIZE_MAX) { + //if (bsize <= MI_LARGE_OBJ_SIZE_MAX) { + // mi_heap_stat_decrease(heap, malloc_large, bsize); + //} + //else + { + mi_heap_stat_decrease(heap, malloc_huge, bsize); + } + } + #if (MI_STAT>0) + _mi_page_free_collect(page, false); // update used count + const size_t inuse = page->used; + if (bsize <= MI_LARGE_OBJ_SIZE_MAX) { + mi_heap_stat_decrease(heap, malloc_normal, bsize * inuse); + #if (MI_STAT>1) + mi_heap_stat_decrease(heap, malloc_bins[_mi_bin(bsize)], inuse); + #endif + } + // mi_heap_stat_decrease(heap, malloc_requested, bsize * inuse); // todo: off for aligned blocks... + #endif + + /// pretend it is all free now + mi_assert_internal(mi_page_thread_free(page) == NULL); + page->used = 0; + + // and free the page + // mi_page_free(page,false); + page->next = NULL; + page->prev = NULL; + _mi_segment_page_free(page,false /* no force? */, &heap->tld->segments); + + return true; // keep going +} + +void _mi_heap_destroy_pages(mi_heap_t* heap) { + mi_heap_visit_pages(heap, &_mi_heap_page_destroy, NULL, NULL); + mi_heap_reset_pages(heap); +} + +#if MI_TRACK_HEAP_DESTROY +static bool mi_cdecl mi_heap_track_block_free(const mi_heap_t* heap, const mi_heap_area_t* area, void* block, size_t block_size, void* arg) { + MI_UNUSED(heap); MI_UNUSED(area); MI_UNUSED(arg); MI_UNUSED(block_size); + mi_track_free_size(block,mi_usable_size(block)); + return true; +} +#endif + +void mi_heap_destroy(mi_heap_t* heap) { + mi_assert(heap != NULL); + mi_assert(mi_heap_is_initialized(heap)); + mi_assert(heap->no_reclaim); + mi_assert_expensive(mi_heap_is_valid(heap)); + if (heap==NULL || !mi_heap_is_initialized(heap)) return; + #if MI_GUARDED + // _mi_warning_message("'mi_heap_destroy' called but MI_GUARDED is enabled -- using `mi_heap_delete` instead (heap at %p)\n", heap); + mi_heap_delete(heap); + return; + #else + if (!heap->no_reclaim) { + _mi_warning_message("'mi_heap_destroy' called but ignored as the heap was not created with 'allow_destroy' (heap at %p)\n", heap); + // don't free in case it may contain reclaimed pages + mi_heap_delete(heap); + } + else { + // track all blocks as freed + #if MI_TRACK_HEAP_DESTROY + mi_heap_visit_blocks(heap, true, mi_heap_track_block_free, NULL); + #endif + // free all pages + _mi_heap_destroy_pages(heap); + mi_heap_free(heap); + } + #endif +} + +// forcefully destroy all heaps in the current thread +void _mi_heap_unsafe_destroy_all(mi_heap_t* heap) { + mi_assert_internal(heap != NULL); + if (heap == NULL) return; + mi_heap_t* curr = heap->tld->heaps; + while (curr != NULL) { + mi_heap_t* next = curr->next; + if (curr->no_reclaim) { + mi_heap_destroy(curr); + } + else { + _mi_heap_destroy_pages(curr); + } + curr = next; + } +} + +/* ----------------------------------------------------------- + Safe Heap delete +----------------------------------------------------------- */ + +// Transfer the pages from one heap to the other +static void mi_heap_absorb(mi_heap_t* heap, mi_heap_t* from) { + mi_assert_internal(heap!=NULL); + if (from==NULL || from->page_count == 0) return; + + // reduce the size of the delayed frees + _mi_heap_delayed_free_partial(from); + + // transfer all pages by appending the queues; this will set a new heap field + // so threads may do delayed frees in either heap for a while. + // note: appending waits for each page to not be in the `MI_DELAYED_FREEING` state + // so after this only the new heap will get delayed frees + for (size_t i = 0; i <= MI_BIN_FULL; i++) { + mi_page_queue_t* pq = &heap->pages[i]; + mi_page_queue_t* append = &from->pages[i]; + size_t pcount = _mi_page_queue_append(heap, pq, append); + heap->page_count += pcount; + from->page_count -= pcount; + } + mi_assert_internal(from->page_count == 0); + + // and do outstanding delayed frees in the `from` heap + // note: be careful here as the `heap` field in all those pages no longer point to `from`, + // turns out to be ok as `_mi_heap_delayed_free` only visits the list and calls a + // the regular `_mi_free_delayed_block` which is safe. + _mi_heap_delayed_free_all(from); + #if !defined(_MSC_VER) || (_MSC_VER > 1900) // somehow the following line gives an error in VS2015, issue #353 + mi_assert_internal(mi_atomic_load_ptr_relaxed(mi_block_t,&from->thread_delayed_free) == NULL); + #endif + + // and reset the `from` heap + mi_heap_reset_pages(from); +} + +// are two heaps compatible with respect to heap-tag, exclusive arena etc. +static bool mi_heaps_are_compatible(mi_heap_t* heap1, mi_heap_t* heap2) { + return (heap1->tag == heap2->tag && // store same kind of objects + heap1->arena_id == heap2->arena_id); // same arena preference +} + +// Safe delete a heap without freeing any still allocated blocks in that heap. +void mi_heap_delete(mi_heap_t* heap) +{ + mi_assert(heap != NULL); + mi_assert(mi_heap_is_initialized(heap)); + mi_assert_expensive(mi_heap_is_valid(heap)); + if (heap==NULL || !mi_heap_is_initialized(heap)) return; + + mi_heap_t* bheap = heap->tld->heap_backing; + if (bheap != heap && mi_heaps_are_compatible(bheap,heap)) { + // transfer still used pages to the backing heap + mi_heap_absorb(bheap, heap); + } + else { + // the backing heap abandons its pages + _mi_heap_collect_abandon(heap); + } + mi_assert_internal(heap->page_count==0); + mi_heap_free(heap); +} + +mi_heap_t* mi_heap_set_default(mi_heap_t* heap) { + mi_assert(heap != NULL); + mi_assert(mi_heap_is_initialized(heap)); + if (heap==NULL || !mi_heap_is_initialized(heap)) return NULL; + mi_assert_expensive(mi_heap_is_valid(heap)); + mi_heap_t* old = mi_prim_get_default_heap(); + _mi_heap_set_default_direct(heap); + return old; +} + + + + +/* ----------------------------------------------------------- + Analysis +----------------------------------------------------------- */ + +// static since it is not thread safe to access heaps from other threads. +static mi_heap_t* mi_heap_of_block(const void* p) { + if (p == NULL) return NULL; + mi_segment_t* segment = _mi_ptr_segment(p); + bool valid = (_mi_ptr_cookie(segment) == segment->cookie); + mi_assert_internal(valid); + if mi_unlikely(!valid) return NULL; + return mi_page_heap(_mi_segment_page_of(segment,p)); +} + +bool mi_heap_contains_block(mi_heap_t* heap, const void* p) { + mi_assert(heap != NULL); + if (heap==NULL || !mi_heap_is_initialized(heap)) return false; + return (heap == mi_heap_of_block(p)); +} + + +static bool mi_heap_page_check_owned(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* p, void* vfound) { + MI_UNUSED(heap); + MI_UNUSED(pq); + bool* found = (bool*)vfound; + void* start = mi_page_start(page); + void* end = (uint8_t*)start + (page->capacity * mi_page_block_size(page)); + *found = (p >= start && p < end); + return (!*found); // continue if not found +} + +bool mi_heap_check_owned(mi_heap_t* heap, const void* p) { + mi_assert(heap != NULL); + if (heap==NULL || !mi_heap_is_initialized(heap)) return false; + if (((uintptr_t)p & (MI_INTPTR_SIZE - 1)) != 0) return false; // only aligned pointers + bool found = false; + mi_heap_visit_pages(heap, &mi_heap_page_check_owned, (void*)p, &found); + return found; +} + +bool mi_check_owned(const void* p) { + return mi_heap_check_owned(mi_prim_get_default_heap(), p); +} + +/* ----------------------------------------------------------- + Visit all heap blocks and areas + Todo: enable visiting abandoned pages, and + enable visiting all blocks of all heaps across threads +----------------------------------------------------------- */ + +void _mi_heap_area_init(mi_heap_area_t* area, mi_page_t* page) { + const size_t bsize = mi_page_block_size(page); + const size_t ubsize = mi_page_usable_block_size(page); + area->reserved = page->reserved * bsize; + area->committed = page->capacity * bsize; + area->blocks = mi_page_start(page); + area->used = page->used; // number of blocks in use (#553) + area->block_size = ubsize; + area->full_block_size = bsize; + area->heap_tag = page->heap_tag; +} + + +static void mi_get_fast_divisor(size_t divisor, uint64_t* magic, size_t* shift) { + mi_assert_internal(divisor > 0 && divisor <= UINT32_MAX); + *shift = MI_SIZE_BITS - mi_clz(divisor - 1); + *magic = ((((uint64_t)1 << 32) * (((uint64_t)1 << *shift) - divisor)) / divisor + 1); +} + +static size_t mi_fast_divide(size_t n, uint64_t magic, size_t shift) { + mi_assert_internal(n <= UINT32_MAX); + const uint64_t hi = ((uint64_t)n * magic) >> 32; + return (size_t)((hi + n) >> shift); +} + +bool _mi_heap_area_visit_blocks(const mi_heap_area_t* area, mi_page_t* page, mi_block_visit_fun* visitor, void* arg) { + mi_assert(area != NULL); + if (area==NULL) return true; + mi_assert(page != NULL); + if (page == NULL) return true; + + _mi_page_free_collect(page,true); // collect both thread_delayed and local_free + mi_assert_internal(page->local_free == NULL); + if (page->used == 0) return true; + + size_t psize; + uint8_t* const pstart = _mi_segment_page_start(_mi_page_segment(page), page, &psize); + mi_heap_t* const heap = mi_page_heap(page); + const size_t bsize = mi_page_block_size(page); + const size_t ubsize = mi_page_usable_block_size(page); // without padding + + // optimize page with one block + if (page->capacity == 1) { + mi_assert_internal(page->used == 1 && page->free == NULL); + return visitor(mi_page_heap(page), area, pstart, ubsize, arg); + } + mi_assert(bsize <= UINT32_MAX); + + // optimize full pages + if (page->used == page->capacity) { + uint8_t* block = pstart; + for (size_t i = 0; i < page->capacity; i++) { + if (!visitor(heap, area, block, ubsize, arg)) return false; + block += bsize; + } + return true; + } + + // create a bitmap of free blocks. + #define MI_MAX_BLOCKS (MI_SMALL_PAGE_SIZE / sizeof(void*)) + uintptr_t free_map[MI_MAX_BLOCKS / MI_INTPTR_BITS]; + const uintptr_t bmapsize = _mi_divide_up(page->capacity, MI_INTPTR_BITS); + memset(free_map, 0, bmapsize * sizeof(intptr_t)); + if (page->capacity % MI_INTPTR_BITS != 0) { + // mark left-over bits at the end as free + size_t shift = (page->capacity % MI_INTPTR_BITS); + uintptr_t mask = (UINTPTR_MAX << shift); + free_map[bmapsize - 1] = mask; + } + + // fast repeated division by the block size + uint64_t magic; + size_t shift; + mi_get_fast_divisor(bsize, &magic, &shift); + + #if MI_DEBUG>1 + size_t free_count = 0; + #endif + for (mi_block_t* block = page->free; block != NULL; block = mi_block_next(page, block)) { + #if MI_DEBUG>1 + free_count++; + #endif + mi_assert_internal((uint8_t*)block >= pstart && (uint8_t*)block < (pstart + psize)); + size_t offset = (uint8_t*)block - pstart; + mi_assert_internal(offset % bsize == 0); + mi_assert_internal(offset <= UINT32_MAX); + size_t blockidx = mi_fast_divide(offset, magic, shift); + mi_assert_internal(blockidx == offset / bsize); + mi_assert_internal(blockidx < MI_MAX_BLOCKS); + size_t bitidx = (blockidx / MI_INTPTR_BITS); + size_t bit = blockidx - (bitidx * MI_INTPTR_BITS); + free_map[bitidx] |= ((uintptr_t)1 << bit); + } + mi_assert_internal(page->capacity == (free_count + page->used)); + + // walk through all blocks skipping the free ones + #if MI_DEBUG>1 + size_t used_count = 0; + #endif + uint8_t* block = pstart; + for (size_t i = 0; i < bmapsize; i++) { + if (free_map[i] == 0) { + // every block is in use + for (size_t j = 0; j < MI_INTPTR_BITS; j++) { + #if MI_DEBUG>1 + used_count++; + #endif + if (!visitor(heap, area, block, ubsize, arg)) return false; + block += bsize; + } + } + else { + // visit the used blocks in the mask + uintptr_t m = ~free_map[i]; + while (m != 0) { + #if MI_DEBUG>1 + used_count++; + #endif + size_t bitidx = mi_ctz(m); + if (!visitor(heap, area, block + (bitidx * bsize), ubsize, arg)) return false; + m &= m - 1; // clear least significant bit + } + block += bsize * MI_INTPTR_BITS; + } + } + mi_assert_internal(page->used == used_count); + return true; +} + + + +// Separate struct to keep `mi_page_t` out of the public interface +typedef struct mi_heap_area_ex_s { + mi_heap_area_t area; + mi_page_t* page; +} mi_heap_area_ex_t; + +typedef bool (mi_heap_area_visit_fun)(const mi_heap_t* heap, const mi_heap_area_ex_t* area, void* arg); + +static bool mi_heap_visit_areas_page(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_t* page, void* vfun, void* arg) { + MI_UNUSED(heap); + MI_UNUSED(pq); + mi_heap_area_visit_fun* fun = (mi_heap_area_visit_fun*)vfun; + mi_heap_area_ex_t xarea; + xarea.page = page; + _mi_heap_area_init(&xarea.area, page); + return fun(heap, &xarea, arg); +} + +// Visit all heap pages as areas +static bool mi_heap_visit_areas(const mi_heap_t* heap, mi_heap_area_visit_fun* visitor, void* arg) { + if (visitor == NULL) return false; + return mi_heap_visit_pages((mi_heap_t*)heap, &mi_heap_visit_areas_page, (void*)(visitor), arg); // note: function pointer to void* :-{ +} + +// Just to pass arguments +typedef struct mi_visit_blocks_args_s { + bool visit_blocks; + mi_block_visit_fun* visitor; + void* arg; +} mi_visit_blocks_args_t; + +static bool mi_heap_area_visitor(const mi_heap_t* heap, const mi_heap_area_ex_t* xarea, void* arg) { + mi_visit_blocks_args_t* args = (mi_visit_blocks_args_t*)arg; + if (!args->visitor(heap, &xarea->area, NULL, xarea->area.block_size, args->arg)) return false; + if (args->visit_blocks) { + return _mi_heap_area_visit_blocks(&xarea->area, xarea->page, args->visitor, args->arg); + } + else { + return true; + } +} + +// Visit all blocks in a heap +bool mi_heap_visit_blocks(const mi_heap_t* heap, bool visit_blocks, mi_block_visit_fun* visitor, void* arg) { + mi_visit_blocks_args_t args = { visit_blocks, visitor, arg }; + return mi_heap_visit_areas(heap, &mi_heap_area_visitor, &args); +} diff --git a/compat/mimalloc/init.c b/compat/mimalloc/init.c new file mode 100644 index 00000000000000..3fc8b033695a38 --- /dev/null +++ b/compat/mimalloc/init.c @@ -0,0 +1,714 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2022, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" + +#include // memcpy, memset +#include // atexit + + +// Empty page used to initialize the small free pages array +const mi_page_t _mi_page_empty = { + 0, + false, false, false, false, + 0, // capacity + 0, // reserved capacity + { 0 }, // flags + false, // is_zero + 0, // retire_expire + NULL, // free + NULL, // local_free + 0, // used + 0, // block size shift + 0, // heap tag + 0, // block_size + NULL, // page_start + #if (MI_PADDING || MI_ENCODE_FREELIST) + { 0, 0 }, + #endif + MI_ATOMIC_VAR_INIT(0), // xthread_free + MI_ATOMIC_VAR_INIT(0), // xheap + NULL, NULL + , { 0 } // padding +}; + +#define MI_PAGE_EMPTY() ((mi_page_t*)&_mi_page_empty) + +#if (MI_SMALL_WSIZE_MAX==128) +#if (MI_PADDING>0) && (MI_INTPTR_SIZE >= 8) +#define MI_SMALL_PAGES_EMPTY { MI_INIT128(MI_PAGE_EMPTY), MI_PAGE_EMPTY(), MI_PAGE_EMPTY() } +#elif (MI_PADDING>0) +#define MI_SMALL_PAGES_EMPTY { MI_INIT128(MI_PAGE_EMPTY), MI_PAGE_EMPTY(), MI_PAGE_EMPTY(), MI_PAGE_EMPTY() } +#else +#define MI_SMALL_PAGES_EMPTY { MI_INIT128(MI_PAGE_EMPTY), MI_PAGE_EMPTY() } +#endif +#else +#error "define right initialization sizes corresponding to MI_SMALL_WSIZE_MAX" +#endif + +// Empty page queues for every bin +#define QNULL(sz) { NULL, NULL, (sz)*sizeof(uintptr_t) } +#define MI_PAGE_QUEUES_EMPTY \ + { QNULL(1), \ + QNULL( 1), QNULL( 2), QNULL( 3), QNULL( 4), QNULL( 5), QNULL( 6), QNULL( 7), QNULL( 8), /* 8 */ \ + QNULL( 10), QNULL( 12), QNULL( 14), QNULL( 16), QNULL( 20), QNULL( 24), QNULL( 28), QNULL( 32), /* 16 */ \ + QNULL( 40), QNULL( 48), QNULL( 56), QNULL( 64), QNULL( 80), QNULL( 96), QNULL( 112), QNULL( 128), /* 24 */ \ + QNULL( 160), QNULL( 192), QNULL( 224), QNULL( 256), QNULL( 320), QNULL( 384), QNULL( 448), QNULL( 512), /* 32 */ \ + QNULL( 640), QNULL( 768), QNULL( 896), QNULL( 1024), QNULL( 1280), QNULL( 1536), QNULL( 1792), QNULL( 2048), /* 40 */ \ + QNULL( 2560), QNULL( 3072), QNULL( 3584), QNULL( 4096), QNULL( 5120), QNULL( 6144), QNULL( 7168), QNULL( 8192), /* 48 */ \ + QNULL( 10240), QNULL( 12288), QNULL( 14336), QNULL( 16384), QNULL( 20480), QNULL( 24576), QNULL( 28672), QNULL( 32768), /* 56 */ \ + QNULL( 40960), QNULL( 49152), QNULL( 57344), QNULL( 65536), QNULL( 81920), QNULL( 98304), QNULL(114688), QNULL(131072), /* 64 */ \ + QNULL(163840), QNULL(196608), QNULL(229376), QNULL(262144), QNULL(327680), QNULL(393216), QNULL(458752), QNULL(524288), /* 72 */ \ + QNULL(MI_MEDIUM_OBJ_WSIZE_MAX + 1 /* 655360, Huge queue */), \ + QNULL(MI_MEDIUM_OBJ_WSIZE_MAX + 2) /* Full queue */ } + +#define MI_STAT_COUNT_NULL() {0,0,0} + +// Empty statistics +#define MI_STATS_NULL \ + MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), \ + MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), \ + MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), MI_STAT_COUNT_NULL(), \ + { 0 }, { 0 }, { 0 }, { 0 }, \ + { 0 }, { 0 }, { 0 }, { 0 }, \ + \ + { 0 }, { 0 }, { 0 }, { 0 }, { 0 }, \ + MI_INIT4(MI_STAT_COUNT_NULL), \ + { 0 }, { 0 }, { 0 }, { 0 }, \ + \ + { MI_INIT4(MI_STAT_COUNT_NULL) }, \ + { { 0 }, { 0 }, { 0 }, { 0 } }, \ + \ + { MI_INIT74(MI_STAT_COUNT_NULL) }, \ + { MI_INIT74(MI_STAT_COUNT_NULL) } + + +// Empty slice span queues for every bin +#define SQNULL(sz) { NULL, NULL, sz } +#define MI_SEGMENT_SPAN_QUEUES_EMPTY \ + { SQNULL(1), \ + SQNULL( 1), SQNULL( 2), SQNULL( 3), SQNULL( 4), SQNULL( 5), SQNULL( 6), SQNULL( 7), SQNULL( 10), /* 8 */ \ + SQNULL( 12), SQNULL( 14), SQNULL( 16), SQNULL( 20), SQNULL( 24), SQNULL( 28), SQNULL( 32), SQNULL( 40), /* 16 */ \ + SQNULL( 48), SQNULL( 56), SQNULL( 64), SQNULL( 80), SQNULL( 96), SQNULL( 112), SQNULL( 128), SQNULL( 160), /* 24 */ \ + SQNULL( 192), SQNULL( 224), SQNULL( 256), SQNULL( 320), SQNULL( 384), SQNULL( 448), SQNULL( 512), SQNULL( 640), /* 32 */ \ + SQNULL( 768), SQNULL( 896), SQNULL( 1024) /* 35 */ } + + +// -------------------------------------------------------- +// Statically allocate an empty heap as the initial +// thread local value for the default heap, +// and statically allocate the backing heap for the main +// thread so it can function without doing any allocation +// itself (as accessing a thread local for the first time +// may lead to allocation itself on some platforms) +// -------------------------------------------------------- + +mi_decl_cache_align const mi_heap_t _mi_heap_empty = { + NULL, + MI_ATOMIC_VAR_INIT(NULL), + 0, // tid + 0, // cookie + 0, // arena id + { 0, 0 }, // keys + { {0}, {0}, 0, true }, // random + 0, // page count + MI_BIN_FULL, 0, // page retired min/max + 0, 0, // generic count + NULL, // next + false, // can reclaim + 0, // tag + #if MI_GUARDED + 0, 0, 0, 1, // count is 1 so we never write to it (see `internal.h:mi_heap_malloc_use_guarded`) + #endif + MI_SMALL_PAGES_EMPTY, + MI_PAGE_QUEUES_EMPTY +}; + +static mi_decl_cache_align mi_subproc_t mi_subproc_default; + +#define tld_empty_stats ((mi_stats_t*)((uint8_t*)&tld_empty + offsetof(mi_tld_t,stats))) + +mi_decl_cache_align static const mi_tld_t tld_empty = { + 0, + false, + NULL, NULL, + { MI_SEGMENT_SPAN_QUEUES_EMPTY, 0, 0, 0, 0, 0, &mi_subproc_default, tld_empty_stats }, // segments + { MI_STAT_VERSION, MI_STATS_NULL } // stats +}; + +mi_threadid_t _mi_thread_id(void) mi_attr_noexcept { + return _mi_prim_thread_id(); +} + +// the thread-local default heap for allocation +mi_decl_thread mi_heap_t* _mi_heap_default = (mi_heap_t*)&_mi_heap_empty; + +extern mi_decl_hidden mi_heap_t _mi_heap_main; + +static mi_decl_cache_align mi_tld_t tld_main = { + 0, false, + &_mi_heap_main, & _mi_heap_main, + { MI_SEGMENT_SPAN_QUEUES_EMPTY, 0, 0, 0, 0, 0, &mi_subproc_default, &tld_main.stats }, // segments + { MI_STAT_VERSION, MI_STATS_NULL } // stats +}; + +mi_decl_cache_align mi_heap_t _mi_heap_main = { + &tld_main, + MI_ATOMIC_VAR_INIT(NULL), + 0, // thread id + 0, // initial cookie + 0, // arena id + { 0, 0 }, // the key of the main heap can be fixed (unlike page keys that need to be secure!) + { {0x846ca68b}, {0}, 0, true }, // random + 0, // page count + MI_BIN_FULL, 0, // page retired min/max + 0, 0, // generic count + NULL, // next heap + false, // can reclaim + 0, // tag + #if MI_GUARDED + 0, 0, 0, 0, + #endif + MI_SMALL_PAGES_EMPTY, + MI_PAGE_QUEUES_EMPTY +}; + +bool _mi_process_is_initialized = false; // set to `true` in `mi_process_init`. + +mi_stats_t _mi_stats_main = { MI_STAT_VERSION, MI_STATS_NULL }; + +#if MI_GUARDED +mi_decl_export void mi_heap_guarded_set_sample_rate(mi_heap_t* heap, size_t sample_rate, size_t seed) { + heap->guarded_sample_rate = sample_rate; + heap->guarded_sample_count = sample_rate; // count down samples + if (heap->guarded_sample_rate > 1) { + if (seed == 0) { + seed = _mi_heap_random_next(heap); + } + heap->guarded_sample_count = (seed % heap->guarded_sample_rate) + 1; // start at random count between 1 and `sample_rate` + } +} + +mi_decl_export void mi_heap_guarded_set_size_bound(mi_heap_t* heap, size_t min, size_t max) { + heap->guarded_size_min = min; + heap->guarded_size_max = (min > max ? min : max); +} + +void _mi_heap_guarded_init(mi_heap_t* heap) { + mi_heap_guarded_set_sample_rate(heap, + (size_t)mi_option_get_clamp(mi_option_guarded_sample_rate, 0, LONG_MAX), + (size_t)mi_option_get(mi_option_guarded_sample_seed)); + mi_heap_guarded_set_size_bound(heap, + (size_t)mi_option_get_clamp(mi_option_guarded_min, 0, LONG_MAX), + (size_t)mi_option_get_clamp(mi_option_guarded_max, 0, LONG_MAX) ); +} +#else +mi_decl_export void mi_heap_guarded_set_sample_rate(mi_heap_t* heap, size_t sample_rate, size_t seed) { + MI_UNUSED(heap); MI_UNUSED(sample_rate); MI_UNUSED(seed); +} + +mi_decl_export void mi_heap_guarded_set_size_bound(mi_heap_t* heap, size_t min, size_t max) { + MI_UNUSED(heap); MI_UNUSED(min); MI_UNUSED(max); +} +void _mi_heap_guarded_init(mi_heap_t* heap) { + MI_UNUSED(heap); +} +#endif + + +static void mi_heap_main_init(void) { + if (_mi_heap_main.cookie == 0) { + _mi_heap_main.thread_id = _mi_thread_id(); + _mi_heap_main.cookie = 1; + #if defined(_WIN32) && !defined(MI_SHARED_LIB) + _mi_random_init_weak(&_mi_heap_main.random); // prevent allocation failure during bcrypt dll initialization with static linking + #else + _mi_random_init(&_mi_heap_main.random); + #endif + _mi_heap_main.cookie = _mi_heap_random_next(&_mi_heap_main); + _mi_heap_main.keys[0] = _mi_heap_random_next(&_mi_heap_main); + _mi_heap_main.keys[1] = _mi_heap_random_next(&_mi_heap_main); + mi_lock_init(&mi_subproc_default.abandoned_os_lock); + mi_lock_init(&mi_subproc_default.abandoned_os_visit_lock); + _mi_heap_guarded_init(&_mi_heap_main); + } +} + +mi_heap_t* _mi_heap_main_get(void) { + mi_heap_main_init(); + return &_mi_heap_main; +} + +/* ----------------------------------------------------------- + Sub process +----------------------------------------------------------- */ + +mi_subproc_id_t mi_subproc_main(void) { + return NULL; +} + +mi_subproc_id_t mi_subproc_new(void) { + mi_memid_t memid = _mi_memid_none(); + mi_subproc_t* subproc = (mi_subproc_t*)_mi_arena_meta_zalloc(sizeof(mi_subproc_t), &memid); + if (subproc == NULL) return NULL; + subproc->memid = memid; + subproc->abandoned_os_list = NULL; + mi_lock_init(&subproc->abandoned_os_lock); + mi_lock_init(&subproc->abandoned_os_visit_lock); + return subproc; +} + +mi_subproc_t* _mi_subproc_from_id(mi_subproc_id_t subproc_id) { + return (subproc_id == NULL ? &mi_subproc_default : (mi_subproc_t*)subproc_id); +} + +void mi_subproc_delete(mi_subproc_id_t subproc_id) { + if (subproc_id == NULL) return; + mi_subproc_t* subproc = _mi_subproc_from_id(subproc_id); + // check if there are no abandoned segments still.. + bool safe_to_delete = false; + mi_lock(&subproc->abandoned_os_lock) { + if (subproc->abandoned_os_list == NULL) { + safe_to_delete = true; + } + } + if (!safe_to_delete) return; + // safe to release + // todo: should we refcount subprocesses? + mi_lock_done(&subproc->abandoned_os_lock); + mi_lock_done(&subproc->abandoned_os_visit_lock); + _mi_arena_meta_free(subproc, subproc->memid, sizeof(mi_subproc_t)); +} + +void mi_subproc_add_current_thread(mi_subproc_id_t subproc_id) { + mi_heap_t* heap = mi_heap_get_default(); + if (heap == NULL) return; + mi_assert(heap->tld->segments.subproc == &mi_subproc_default); + if (heap->tld->segments.subproc != &mi_subproc_default) return; + heap->tld->segments.subproc = _mi_subproc_from_id(subproc_id); +} + + + +/* ----------------------------------------------------------- + Initialization and freeing of the thread local heaps +----------------------------------------------------------- */ + +// note: in x64 in release build `sizeof(mi_thread_data_t)` is under 4KiB (= OS page size). +typedef struct mi_thread_data_s { + mi_heap_t heap; // must come first due to cast in `_mi_heap_done` + mi_tld_t tld; + mi_memid_t memid; // must come last due to zero'ing +} mi_thread_data_t; + + +// Thread meta-data is allocated directly from the OS. For +// some programs that do not use thread pools and allocate and +// destroy many OS threads, this may causes too much overhead +// per thread so we maintain a small cache of recently freed metadata. + +#define TD_CACHE_SIZE (32) +static _Atomic(mi_thread_data_t*) td_cache[TD_CACHE_SIZE]; + +static mi_thread_data_t* mi_thread_data_zalloc(void) { + // try to find thread metadata in the cache + mi_thread_data_t* td = NULL; + for (int i = 0; i < TD_CACHE_SIZE; i++) { + td = mi_atomic_load_ptr_relaxed(mi_thread_data_t, &td_cache[i]); + if (td != NULL) { + // found cached allocation, try use it + td = mi_atomic_exchange_ptr_acq_rel(mi_thread_data_t, &td_cache[i], NULL); + if (td != NULL) { + _mi_memzero(td, offsetof(mi_thread_data_t,memid)); + return td; + } + } + } + + // if that fails, allocate as meta data + mi_memid_t memid; + td = (mi_thread_data_t*)_mi_os_zalloc(sizeof(mi_thread_data_t), &memid); + if (td == NULL) { + // if this fails, try once more. (issue #257) + td = (mi_thread_data_t*)_mi_os_zalloc(sizeof(mi_thread_data_t), &memid); + if (td == NULL) { + // really out of memory + _mi_error_message(ENOMEM, "unable to allocate thread local heap metadata (%zu bytes)\n", sizeof(mi_thread_data_t)); + return NULL; + } + } + td->memid = memid; + return td; +} + +static void mi_thread_data_free( mi_thread_data_t* tdfree ) { + // try to add the thread metadata to the cache + for (int i = 0; i < TD_CACHE_SIZE; i++) { + mi_thread_data_t* td = mi_atomic_load_ptr_relaxed(mi_thread_data_t, &td_cache[i]); + if (td == NULL) { + mi_thread_data_t* expected = NULL; + if (mi_atomic_cas_ptr_weak_acq_rel(mi_thread_data_t, &td_cache[i], &expected, tdfree)) { + return; + } + } + } + // if that fails, just free it directly + _mi_os_free(tdfree, sizeof(mi_thread_data_t), tdfree->memid); +} + +void _mi_thread_data_collect(void) { + // free all thread metadata from the cache + for (int i = 0; i < TD_CACHE_SIZE; i++) { + mi_thread_data_t* td = mi_atomic_load_ptr_relaxed(mi_thread_data_t, &td_cache[i]); + if (td != NULL) { + td = mi_atomic_exchange_ptr_acq_rel(mi_thread_data_t, &td_cache[i], NULL); + if (td != NULL) { + _mi_os_free(td, sizeof(mi_thread_data_t), td->memid); + } + } + } +} + +// Initialize the thread local default heap, called from `mi_thread_init` +static bool _mi_thread_heap_init(void) { + if (mi_heap_is_initialized(mi_prim_get_default_heap())) return true; + if (_mi_is_main_thread()) { + // mi_assert_internal(_mi_heap_main.thread_id != 0); // can happen on freeBSD where alloc is called before any initialization + // the main heap is statically allocated + mi_heap_main_init(); + _mi_heap_set_default_direct(&_mi_heap_main); + //mi_assert_internal(_mi_heap_default->tld->heap_backing == mi_prim_get_default_heap()); + } + else { + // use `_mi_os_alloc` to allocate directly from the OS + mi_thread_data_t* td = mi_thread_data_zalloc(); + if (td == NULL) return false; + + mi_tld_t* tld = &td->tld; + mi_heap_t* heap = &td->heap; + _mi_tld_init(tld, heap); // must be before `_mi_heap_init` + _mi_heap_init(heap, tld, _mi_arena_id_none(), false /* can reclaim */, 0 /* default tag */); + _mi_heap_set_default_direct(heap); + } + return false; +} + +// initialize thread local data +void _mi_tld_init(mi_tld_t* tld, mi_heap_t* bheap) { + _mi_memcpy_aligned(tld, &tld_empty, sizeof(mi_tld_t)); + tld->heap_backing = bheap; + tld->heaps = NULL; + tld->segments.subproc = &mi_subproc_default; + tld->segments.stats = &tld->stats; +} + +// Free the thread local default heap (called from `mi_thread_done`) +static bool _mi_thread_heap_done(mi_heap_t* heap) { + if (!mi_heap_is_initialized(heap)) return true; + + // reset default heap + _mi_heap_set_default_direct(_mi_is_main_thread() ? &_mi_heap_main : (mi_heap_t*)&_mi_heap_empty); + + // switch to backing heap + heap = heap->tld->heap_backing; + if (!mi_heap_is_initialized(heap)) return false; + + // delete all non-backing heaps in this thread + mi_heap_t* curr = heap->tld->heaps; + while (curr != NULL) { + mi_heap_t* next = curr->next; // save `next` as `curr` will be freed + if (curr != heap) { + mi_assert_internal(!mi_heap_is_backing(curr)); + mi_heap_delete(curr); + } + curr = next; + } + mi_assert_internal(heap->tld->heaps == heap && heap->next == NULL); + mi_assert_internal(mi_heap_is_backing(heap)); + + // collect if not the main thread + if (heap != &_mi_heap_main) { + _mi_heap_collect_abandon(heap); + } + + // merge stats + _mi_stats_done(&heap->tld->stats); + + // free if not the main thread + if (heap != &_mi_heap_main) { + // the following assertion does not always hold for huge segments as those are always treated + // as abondened: one may allocate it in one thread, but deallocate in another in which case + // the count can be too large or negative. todo: perhaps not count huge segments? see issue #363 + // mi_assert_internal(heap->tld->segments.count == 0 || heap->thread_id != _mi_thread_id()); + mi_thread_data_free((mi_thread_data_t*)heap); + } + else { + #if 0 + // never free the main thread even in debug mode; if a dll is linked statically with mimalloc, + // there may still be delete/free calls after the mi_fls_done is called. Issue #207 + _mi_heap_destroy_pages(heap); + mi_assert_internal(heap->tld->heap_backing == &_mi_heap_main); + #endif + } + return false; +} + + + +// -------------------------------------------------------- +// Try to run `mi_thread_done()` automatically so any memory +// owned by the thread but not yet released can be abandoned +// and re-owned by another thread. +// +// 1. windows dynamic library: +// call from DllMain on DLL_THREAD_DETACH +// 2. windows static library: +// use `FlsAlloc` to call a destructor when the thread is done +// 3. unix, pthreads: +// use a pthread key to call a destructor when a pthread is done +// +// In the last two cases we also need to call `mi_process_init` +// to set up the thread local keys. +// -------------------------------------------------------- + +// Set up handlers so `mi_thread_done` is called automatically +static void mi_process_setup_auto_thread_done(void) { + static bool tls_initialized = false; // fine if it races + if (tls_initialized) return; + tls_initialized = true; + _mi_prim_thread_init_auto_done(); + _mi_heap_set_default_direct(&_mi_heap_main); +} + + +bool _mi_is_main_thread(void) { + return (_mi_heap_main.thread_id==0 || _mi_heap_main.thread_id == _mi_thread_id()); +} + +static _Atomic(size_t) thread_count = MI_ATOMIC_VAR_INIT(1); + +size_t _mi_current_thread_count(void) { + return mi_atomic_load_relaxed(&thread_count); +} + +// This is called from the `mi_malloc_generic` +void mi_thread_init(void) mi_attr_noexcept +{ + // ensure our process has started already + mi_process_init(); + + // initialize the thread local default heap + // (this will call `_mi_heap_set_default_direct` and thus set the + // fiber/pthread key to a non-zero value, ensuring `_mi_thread_done` is called) + if (_mi_thread_heap_init()) return; // returns true if already initialized + + _mi_stat_increase(&_mi_stats_main.threads, 1); + mi_atomic_increment_relaxed(&thread_count); + //_mi_verbose_message("thread init: 0x%zx\n", _mi_thread_id()); +} + +void mi_thread_done(void) mi_attr_noexcept { + _mi_thread_done(NULL); +} + +void _mi_thread_done(mi_heap_t* heap) +{ + // calling with NULL implies using the default heap + if (heap == NULL) { + heap = mi_prim_get_default_heap(); + if (heap == NULL) return; + } + + // prevent re-entrancy through heap_done/heap_set_default_direct (issue #699) + if (!mi_heap_is_initialized(heap)) { + return; + } + + // adjust stats + mi_atomic_decrement_relaxed(&thread_count); + _mi_stat_decrease(&_mi_stats_main.threads, 1); + + // check thread-id as on Windows shutdown with FLS the main (exit) thread may call this on thread-local heaps... + if (heap->thread_id != _mi_thread_id()) return; + + // abandon the thread local heap + if (_mi_thread_heap_done(heap)) return; // returns true if already ran +} + +void _mi_heap_set_default_direct(mi_heap_t* heap) { + mi_assert_internal(heap != NULL); + #if defined(MI_TLS_SLOT) + mi_prim_tls_slot_set(MI_TLS_SLOT,heap); + #elif defined(MI_TLS_PTHREAD_SLOT_OFS) + *mi_prim_tls_pthread_heap_slot() = heap; + #elif defined(MI_TLS_PTHREAD) + // we use _mi_heap_default_key + #else + _mi_heap_default = heap; + #endif + + // ensure the default heap is passed to `_mi_thread_done` + // setting to a non-NULL value also ensures `mi_thread_done` is called. + _mi_prim_thread_associate_default_heap(heap); +} + +void mi_thread_set_in_threadpool(void) mi_attr_noexcept { + // nothing +} + +// -------------------------------------------------------- +// Run functions on process init/done, and thread init/done +// -------------------------------------------------------- +static bool os_preloading = true; // true until this module is initialized + +// Returns true if this module has not been initialized; Don't use C runtime routines until it returns false. +bool mi_decl_noinline _mi_preloading(void) { + return os_preloading; +} + +// Returns true if mimalloc was redirected +mi_decl_nodiscard bool mi_is_redirected(void) mi_attr_noexcept { + return _mi_is_redirected(); +} + +// Called once by the process loader from `src/prim/prim.c` +void _mi_auto_process_init(void) { + mi_heap_main_init(); + #if defined(__APPLE__) || defined(MI_TLS_RECURSE_GUARD) + volatile mi_heap_t* dummy = _mi_heap_default; // access TLS to allocate it before setting tls_initialized to true; + if (dummy == NULL) return; // use dummy or otherwise the access may get optimized away (issue #697) + #endif + os_preloading = false; + mi_assert_internal(_mi_is_main_thread()); + _mi_options_init(); + mi_process_setup_auto_thread_done(); + mi_process_init(); + if (_mi_is_redirected()) _mi_verbose_message("malloc is redirected.\n"); + + // show message from the redirector (if present) + const char* msg = NULL; + _mi_allocator_init(&msg); + if (msg != NULL && (mi_option_is_enabled(mi_option_verbose) || mi_option_is_enabled(mi_option_show_errors))) { + _mi_fputs(NULL,NULL,NULL,msg); + } + + // reseed random + _mi_random_reinit_if_weak(&_mi_heap_main.random); +} + +#if defined(_WIN32) && (defined(_M_IX86) || defined(_M_X64)) +#include +mi_decl_cache_align bool _mi_cpu_has_fsrm = false; +mi_decl_cache_align bool _mi_cpu_has_erms = false; + +static void mi_detect_cpu_features(void) { + // FSRM for fast short rep movsb/stosb support (AMD Zen3+ (~2020) or Intel Ice Lake+ (~2017)) + // EMRS for fast enhanced rep movsb/stosb support + int32_t cpu_info[4]; + __cpuid(cpu_info, 7); + _mi_cpu_has_fsrm = ((cpu_info[3] & (1 << 4)) != 0); // bit 4 of EDX : see + _mi_cpu_has_erms = ((cpu_info[1] & (1 << 9)) != 0); // bit 9 of EBX : see +} +#else +static void mi_detect_cpu_features(void) { + // nothing +} +#endif + +// Initialize the process; called by thread_init or the process loader +void mi_process_init(void) mi_attr_noexcept { + // ensure we are called once + static mi_atomic_once_t process_init; + #if _MSC_VER < 1920 + mi_heap_main_init(); // vs2017 can dynamically re-initialize _mi_heap_main + #endif + if (!mi_atomic_once(&process_init)) return; + _mi_process_is_initialized = true; + _mi_verbose_message("process init: 0x%zx\n", _mi_thread_id()); + mi_process_setup_auto_thread_done(); + + mi_detect_cpu_features(); + _mi_os_init(); + mi_heap_main_init(); + mi_thread_init(); + + #if defined(_WIN32) + // On windows, when building as a static lib the FLS cleanup happens to early for the main thread. + // To avoid this, set the FLS value for the main thread to NULL so the fls cleanup + // will not call _mi_thread_done on the (still executing) main thread. See issue #508. + _mi_prim_thread_associate_default_heap(NULL); + #endif + + mi_stats_reset(); // only call stat reset *after* thread init (or the heap tld == NULL) + mi_track_init(); + + if (mi_option_is_enabled(mi_option_reserve_huge_os_pages)) { + size_t pages = mi_option_get_clamp(mi_option_reserve_huge_os_pages, 0, 128*1024); + long reserve_at = mi_option_get(mi_option_reserve_huge_os_pages_at); + if (reserve_at != -1) { + mi_reserve_huge_os_pages_at(pages, reserve_at, pages*500); + } else { + mi_reserve_huge_os_pages_interleave(pages, 0, pages*500); + } + } + if (mi_option_is_enabled(mi_option_reserve_os_memory)) { + long ksize = mi_option_get(mi_option_reserve_os_memory); + if (ksize > 0) { + mi_reserve_os_memory((size_t)ksize*MI_KiB, true /* commit? */, true /* allow large pages? */); + } + } +} + +// Called when the process is done (cdecl as it is used with `at_exit` on some platforms) +void mi_cdecl mi_process_done(void) mi_attr_noexcept { + // only shutdown if we were initialized + if (!_mi_process_is_initialized) return; + // ensure we are called once + static bool process_done = false; + if (process_done) return; + process_done = true; + + // get the default heap so we don't need to acces thread locals anymore + mi_heap_t* heap = mi_prim_get_default_heap(); // use prim to not initialize any heap + mi_assert_internal(heap != NULL); + + // release any thread specific resources and ensure _mi_thread_done is called on all but the main thread + _mi_prim_thread_done_auto_done(); + + + #ifndef MI_SKIP_COLLECT_ON_EXIT + #if (MI_DEBUG || !defined(MI_SHARED_LIB)) + // free all memory if possible on process exit. This is not needed for a stand-alone process + // but should be done if mimalloc is statically linked into another shared library which + // is repeatedly loaded/unloaded, see issue #281. + mi_heap_collect(heap, true /* force */ ); + #endif + #endif + + // Forcefully release all retained memory; this can be dangerous in general if overriding regular malloc/free + // since after process_done there might still be other code running that calls `free` (like at_exit routines, + // or C-runtime termination code. + if (mi_option_is_enabled(mi_option_destroy_on_exit)) { + mi_heap_collect(heap, true /* force */); + _mi_heap_unsafe_destroy_all(heap); // forcefully release all memory held by all heaps (of this thread only!) + _mi_arena_unsafe_destroy_all(); + _mi_segment_map_unsafe_destroy(); + } + + if (mi_option_is_enabled(mi_option_show_stats) || mi_option_is_enabled(mi_option_verbose)) { + mi_stats_print(NULL); + } + _mi_allocator_done(); + _mi_verbose_message("process done: 0x%zx\n", _mi_heap_main.thread_id); + os_preloading = true; // don't call the C runtime anymore +} + +void mi_cdecl _mi_auto_process_done(void) mi_attr_noexcept { + if (_mi_option_get_fast(mi_option_destroy_on_exit)>1) return; + mi_process_done(); +} diff --git a/compat/mimalloc/libc.c b/compat/mimalloc/libc.c new file mode 100644 index 00000000000000..52d095eb240dc1 --- /dev/null +++ b/compat/mimalloc/libc.c @@ -0,0 +1,334 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +// -------------------------------------------------------- +// This module defines various std libc functions to reduce +// the dependency on libc, and also prevent errors caused +// by some libc implementations when called before `main` +// executes (due to malloc redirection) +// -------------------------------------------------------- + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" // mi_prim_getenv + +char _mi_toupper(char c) { + if (c >= 'a' && c <= 'z') return (c - 'a' + 'A'); + else return c; +} + +int _mi_strnicmp(const char* s, const char* t, size_t n) { + if (n == 0) return 0; + for (; *s != 0 && *t != 0 && n > 0; s++, t++, n--) { + if (_mi_toupper(*s) != _mi_toupper(*t)) break; + } + return (n == 0 ? 0 : *s - *t); +} + +void _mi_strlcpy(char* dest, const char* src, size_t dest_size) { + if (dest==NULL || src==NULL || dest_size == 0) return; + // copy until end of src, or when dest is (almost) full + while (*src != 0 && dest_size > 1) { + *dest++ = *src++; + dest_size--; + } + // always zero terminate + *dest = 0; +} + +void _mi_strlcat(char* dest, const char* src, size_t dest_size) { + if (dest==NULL || src==NULL || dest_size == 0) return; + // find end of string in the dest buffer + while (*dest != 0 && dest_size > 1) { + dest++; + dest_size--; + } + // and catenate + _mi_strlcpy(dest, src, dest_size); +} + +size_t _mi_strlen(const char* s) { + if (s==NULL) return 0; + size_t len = 0; + while(s[len] != 0) { len++; } + return len; +} + +size_t _mi_strnlen(const char* s, size_t max_len) { + if (s==NULL) return 0; + size_t len = 0; + while(s[len] != 0 && len < max_len) { len++; } + return len; +} + +#ifdef MI_NO_GETENV +bool _mi_getenv(const char* name, char* result, size_t result_size) { + MI_UNUSED(name); + MI_UNUSED(result); + MI_UNUSED(result_size); + return false; +} +#else +bool _mi_getenv(const char* name, char* result, size_t result_size) { + if (name==NULL || result == NULL || result_size < 64) return false; + return _mi_prim_getenv(name,result,result_size); +} +#endif + +// -------------------------------------------------------- +// Define our own limited `_mi_vsnprintf` and `_mi_snprintf` +// This is mostly to avoid calling these when libc is not yet +// initialized (and to reduce dependencies) +// +// format: d i, p x u, s +// prec: z l ll L +// width: 10 +// align-left: - +// fill: 0 +// plus: + +// -------------------------------------------------------- + +static void mi_outc(char c, char** out, char* end) { + char* p = *out; + if (p >= end) return; + *p = c; + *out = p + 1; +} + +static void mi_outs(const char* s, char** out, char* end) { + if (s == NULL) return; + char* p = *out; + while (*s != 0 && p < end) { + *p++ = *s++; + } + *out = p; +} + +static void mi_out_fill(char fill, size_t len, char** out, char* end) { + char* p = *out; + for (size_t i = 0; i < len && p < end; i++) { + *p++ = fill; + } + *out = p; +} + +static void mi_out_alignright(char fill, char* start, size_t len, size_t extra, char* end) { + if (len == 0 || extra == 0) return; + if (start + len + extra >= end) return; + // move `len` characters to the right (in reverse since it can overlap) + for (size_t i = 1; i <= len; i++) { + start[len + extra - i] = start[len - i]; + } + // and fill the start + for (size_t i = 0; i < extra; i++) { + start[i] = fill; + } +} + + +static void mi_out_num(uintmax_t x, size_t base, char prefix, char** out, char* end) +{ + if (x == 0 || base == 0 || base > 16) { + if (prefix != 0) { mi_outc(prefix, out, end); } + mi_outc('0',out,end); + } + else { + // output digits in reverse + char* start = *out; + while (x > 0) { + char digit = (char)(x % base); + mi_outc((digit <= 9 ? '0' + digit : 'A' + digit - 10),out,end); + x = x / base; + } + if (prefix != 0) { + mi_outc(prefix, out, end); + } + size_t len = *out - start; + // and reverse in-place + for (size_t i = 0; i < (len / 2); i++) { + char c = start[len - i - 1]; + start[len - i - 1] = start[i]; + start[i] = c; + } + } +} + + +#define MI_NEXTC() c = *in; if (c==0) break; in++; + +int _mi_vsnprintf(char* buf, size_t bufsize, const char* fmt, va_list args) { + if (buf == NULL || bufsize == 0 || fmt == NULL) return 0; + buf[bufsize - 1] = 0; + char* const end = buf + (bufsize - 1); + const char* in = fmt; + char* out = buf; + while (true) { + if (out >= end) break; + char c; + MI_NEXTC(); + if (c != '%') { + if ((c >= ' ' && c <= '~') || c=='\n' || c=='\r' || c=='\t') { // output visible ascii or standard control only + mi_outc(c, &out, end); + } + } + else { + MI_NEXTC(); + char fill = ' '; + size_t width = 0; + char numtype = 'd'; + char numplus = 0; + bool alignright = true; + if (c == '+' || c == ' ') { numplus = c; MI_NEXTC(); } + if (c == '-') { alignright = false; MI_NEXTC(); } + if (c == '0') { fill = '0'; MI_NEXTC(); } + if (c >= '1' && c <= '9') { + width = (c - '0'); MI_NEXTC(); + while (c >= '0' && c <= '9') { + width = (10 * width) + (c - '0'); MI_NEXTC(); + } + if (c == 0) break; // extra check due to while + } + if (c == 'z' || c == 't' || c == 'L') { numtype = c; MI_NEXTC(); } + else if (c == 'l') { + numtype = c; MI_NEXTC(); + if (c == 'l') { numtype = 'L'; MI_NEXTC(); } + } + + char* start = out; + if (c == 's') { + // string + const char* s = va_arg(args, const char*); + mi_outs(s, &out, end); + } + else if (c == 'p' || c == 'x' || c == 'u') { + // unsigned + uintmax_t x = 0; + if (c == 'x' || c == 'u') { + if (numtype == 'z') x = va_arg(args, size_t); + else if (numtype == 't') x = va_arg(args, uintptr_t); // unsigned ptrdiff_t + else if (numtype == 'L') x = va_arg(args, unsigned long long); + else if (numtype == 'l') x = va_arg(args, unsigned long); + else x = va_arg(args, unsigned int); + } + else if (c == 'p') { + x = va_arg(args, uintptr_t); + mi_outs("0x", &out, end); + start = out; + width = (width >= 2 ? width - 2 : 0); + } + if (width == 0 && (c == 'x' || c == 'p')) { + if (c == 'p') { width = 2 * (x <= UINT32_MAX ? 4 : ((x >> 16) <= UINT32_MAX ? 6 : sizeof(void*))); } + if (width == 0) { width = 2; } + fill = '0'; + } + mi_out_num(x, (c == 'x' || c == 'p' ? 16 : 10), numplus, &out, end); + } + else if (c == 'i' || c == 'd') { + // signed + intmax_t x = 0; + if (numtype == 'z') x = va_arg(args, intptr_t ); + else if (numtype == 't') x = va_arg(args, ptrdiff_t); + else if (numtype == 'L') x = va_arg(args, long long); + else if (numtype == 'l') x = va_arg(args, long); + else x = va_arg(args, int); + char pre = 0; + if (x < 0) { + pre = '-'; + if (x > INTMAX_MIN) { x = -x; } + } + else if (numplus != 0) { + pre = numplus; + } + mi_out_num((uintmax_t)x, 10, pre, &out, end); + } + else if (c >= ' ' && c <= '~') { + // unknown format + mi_outc('%', &out, end); + mi_outc(c, &out, end); + } + + // fill & align + mi_assert_internal(out <= end); + mi_assert_internal(out >= start); + const size_t len = out - start; + if (len < width) { + mi_out_fill(fill, width - len, &out, end); + if (alignright && out <= end) { + mi_out_alignright(fill, start, len, width - len, end); + } + } + } + } + mi_assert_internal(out <= end); + *out = 0; + return (int)(out - buf); +} + +int _mi_snprintf(char* buf, size_t buflen, const char* fmt, ...) { + va_list args; + va_start(args, fmt); + const int written = _mi_vsnprintf(buf, buflen, fmt, args); + va_end(args); + return written; +} + + +#if MI_SIZE_SIZE == 4 +#define mi_mask_even_bits32 (0x55555555) +#define mi_mask_even_pairs32 (0x33333333) +#define mi_mask_even_nibbles32 (0x0F0F0F0F) + +// sum of all the bytes in `x` if it is guaranteed that the sum < 256! +static size_t mi_byte_sum32(uint32_t x) { + // perform `x * 0x01010101`: the highest byte contains the sum of all bytes. + x += (x << 8); + x += (x << 16); + return (size_t)(x >> 24); +} + +static size_t mi_popcount_generic32(uint32_t x) { + // first count each 2-bit group `a`, where: a==0b00 -> 00, a==0b01 -> 01, a==0b10 -> 01, a==0b11 -> 10 + // in other words, `a - (a>>1)`; to do this in parallel, we need to mask to prevent spilling a bit pair + // into the lower bit-pair: + x = x - ((x >> 1) & mi_mask_even_bits32); + // add the 2-bit pair results + x = (x & mi_mask_even_pairs32) + ((x >> 2) & mi_mask_even_pairs32); + // add the 4-bit nibble results + x = (x + (x >> 4)) & mi_mask_even_nibbles32; + // each byte now has a count of its bits, we can sum them now: + return mi_byte_sum32(x); +} + +mi_decl_noinline size_t _mi_popcount_generic(size_t x) { + return mi_popcount_generic32(x); +} + +#else +#define mi_mask_even_bits64 (0x5555555555555555) +#define mi_mask_even_pairs64 (0x3333333333333333) +#define mi_mask_even_nibbles64 (0x0F0F0F0F0F0F0F0F) + +// sum of all the bytes in `x` if it is guaranteed that the sum < 256! +static size_t mi_byte_sum64(uint64_t x) { + x += (x << 8); + x += (x << 16); + x += (x << 32); + return (size_t)(x >> 56); +} + +static size_t mi_popcount_generic64(uint64_t x) { + x = x - ((x >> 1) & mi_mask_even_bits64); + x = (x & mi_mask_even_pairs64) + ((x >> 2) & mi_mask_even_pairs64); + x = (x + (x >> 4)) & mi_mask_even_nibbles64; + return mi_byte_sum64(x); +} + +mi_decl_noinline size_t _mi_popcount_generic(size_t x) { + return mi_popcount_generic64(x); +} +#endif + diff --git a/compat/mimalloc/mimalloc-stats.h b/compat/mimalloc/mimalloc-stats.h new file mode 100644 index 00000000000000..44c4886f88a0c7 --- /dev/null +++ b/compat/mimalloc/mimalloc-stats.h @@ -0,0 +1,103 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2025, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_STATS_H +#define MIMALLOC_STATS_H + +#include +#include + +#define MI_STAT_VERSION 1 // increased on every backward incompatible change + +// count allocation over time +typedef struct mi_stat_count_s { + int64_t total; // total allocated + int64_t peak; // peak allocation + int64_t current; // current allocation +} mi_stat_count_t; + +// counters only increase +typedef struct mi_stat_counter_s { + int64_t total; // total count +} mi_stat_counter_t; + +#define MI_STAT_FIELDS() \ + MI_STAT_COUNT(pages) /* count of mimalloc pages */ \ + MI_STAT_COUNT(reserved) /* reserved memory bytes */ \ + MI_STAT_COUNT(committed) /* committed bytes */ \ + MI_STAT_COUNT(reset) /* reset bytes */ \ + MI_STAT_COUNT(purged) /* purged bytes */ \ + MI_STAT_COUNT(page_committed) /* committed memory inside pages */ \ + MI_STAT_COUNT(pages_abandoned) /* abandonded pages count */ \ + MI_STAT_COUNT(threads) /* number of threads */ \ + MI_STAT_COUNT(malloc_normal) /* allocated bytes <= MI_LARGE_OBJ_SIZE_MAX */ \ + MI_STAT_COUNT(malloc_huge) /* allocated bytes in huge pages */ \ + MI_STAT_COUNT(malloc_requested) /* malloc requested bytes */ \ + \ + MI_STAT_COUNTER(mmap_calls) \ + MI_STAT_COUNTER(commit_calls) \ + MI_STAT_COUNTER(reset_calls) \ + MI_STAT_COUNTER(purge_calls) \ + MI_STAT_COUNTER(arena_count) /* number of memory arena's */ \ + MI_STAT_COUNTER(malloc_normal_count) /* number of blocks <= MI_LARGE_OBJ_SIZE_MAX */ \ + MI_STAT_COUNTER(malloc_huge_count) /* number of huge bloks */ \ + MI_STAT_COUNTER(malloc_guarded_count) /* number of allocations with guard pages */ \ + \ + /* internal statistics */ \ + MI_STAT_COUNTER(arena_rollback_count) \ + MI_STAT_COUNTER(arena_purges) \ + MI_STAT_COUNTER(pages_extended) /* number of page extensions */ \ + MI_STAT_COUNTER(pages_retire) /* number of pages that are retired */ \ + MI_STAT_COUNTER(page_searches) /* searches for a fresh page */ \ + /* only on v1 and v2 */ \ + MI_STAT_COUNT(segments) \ + MI_STAT_COUNT(segments_abandoned) \ + MI_STAT_COUNT(segments_cache) \ + MI_STAT_COUNT(_segments_reserved) \ + /* only on v3 */ \ + MI_STAT_COUNTER(pages_reclaim_on_alloc) \ + MI_STAT_COUNTER(pages_reclaim_on_free) \ + MI_STAT_COUNTER(pages_reabandon_full) \ + MI_STAT_COUNTER(pages_unabandon_busy_wait) \ + + +// Define the statistics structure +#define MI_BIN_HUGE (73U) // see types.h +#define MI_STAT_COUNT(stat) mi_stat_count_t stat; +#define MI_STAT_COUNTER(stat) mi_stat_counter_t stat; + +typedef struct mi_stats_s +{ + int version; + + MI_STAT_FIELDS() + + // future extension + mi_stat_count_t _stat_reserved[4]; + mi_stat_counter_t _stat_counter_reserved[4]; + + // size segregated statistics + mi_stat_count_t malloc_bins[MI_BIN_HUGE+1]; // allocation per size bin + mi_stat_count_t page_bins[MI_BIN_HUGE+1]; // pages allocated per size bin +} mi_stats_t; + +#undef MI_STAT_COUNT +#undef MI_STAT_COUNTER + +// Exported definitions +#ifdef __cplusplus +extern "C" { +#endif + +mi_decl_export void mi_stats_get( size_t stats_size, mi_stats_t* stats ) mi_attr_noexcept; +mi_decl_export char* mi_stats_get_json( size_t buf_size, char* buf ) mi_attr_noexcept; // use mi_free to free the result if the input buf == NULL + +#ifdef __cplusplus +} +#endif + +#endif // MIMALLOC_STATS_H diff --git a/compat/mimalloc/mimalloc.h b/compat/mimalloc/mimalloc.h new file mode 100644 index 00000000000000..f887278a98879d --- /dev/null +++ b/compat/mimalloc/mimalloc.h @@ -0,0 +1,612 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2025, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_H +#define MIMALLOC_H + +#define MI_MALLOC_VERSION 224 // major + 2 digits minor + +// ------------------------------------------------------ +// Compiler specific attributes +// ------------------------------------------------------ + +#ifdef __cplusplus + #if (__cplusplus >= 201103L) || (_MSC_VER > 1900) // C++11 + #define mi_attr_noexcept noexcept + #else + #define mi_attr_noexcept throw() + #endif +#else + #define mi_attr_noexcept +#endif + +#if defined(__cplusplus) && (__cplusplus >= 201703) + #define mi_decl_nodiscard [[nodiscard]] +#elif (defined(__GNUC__) && (__GNUC__ >= 4)) || defined(__clang__) // includes clang, icc, and clang-cl + #define mi_decl_nodiscard __attribute__((warn_unused_result)) +#elif defined(_HAS_NODISCARD) + #define mi_decl_nodiscard _NODISCARD +#elif (_MSC_VER >= 1700) + #define mi_decl_nodiscard _Check_return_ +#else + #define mi_decl_nodiscard +#endif + +#if defined(_MSC_VER) || defined(__MINGW32__) + #if !defined(MI_SHARED_LIB) + #define mi_decl_export + #elif defined(MI_SHARED_LIB_EXPORT) + #define mi_decl_export __declspec(dllexport) + #else + #define mi_decl_export __declspec(dllimport) + #endif + #if defined(__MINGW32__) + #define mi_decl_restrict + #define mi_attr_malloc __attribute__((malloc)) + #else + #if (_MSC_VER >= 1900) && !defined(__EDG__) + #define mi_decl_restrict __declspec(allocator) __declspec(restrict) + #else + #define mi_decl_restrict __declspec(restrict) + #endif + #define mi_attr_malloc + #endif + #define mi_cdecl __cdecl + #define mi_attr_alloc_size(s) + #define mi_attr_alloc_size2(s1,s2) + #define mi_attr_alloc_align(p) +#elif defined(__GNUC__) // includes clang and icc + #if defined(MI_SHARED_LIB) && defined(MI_SHARED_LIB_EXPORT) + #define mi_decl_export __attribute__((visibility("default"))) + #else + #define mi_decl_export + #endif + #define mi_cdecl // leads to warnings... __attribute__((cdecl)) + #define mi_decl_restrict + #define mi_attr_malloc __attribute__((malloc)) + #if (defined(__clang_major__) && (__clang_major__ < 4)) || (__GNUC__ < 5) + #define mi_attr_alloc_size(s) + #define mi_attr_alloc_size2(s1,s2) + #define mi_attr_alloc_align(p) + #elif defined(__INTEL_COMPILER) + #define mi_attr_alloc_size(s) __attribute__((alloc_size(s))) + #define mi_attr_alloc_size2(s1,s2) __attribute__((alloc_size(s1,s2))) + #define mi_attr_alloc_align(p) + #else + #define mi_attr_alloc_size(s) __attribute__((alloc_size(s))) + #define mi_attr_alloc_size2(s1,s2) __attribute__((alloc_size(s1,s2))) + #define mi_attr_alloc_align(p) __attribute__((alloc_align(p))) + #endif +#else + #define mi_cdecl + #define mi_decl_export + #define mi_decl_restrict + #define mi_attr_malloc + #define mi_attr_alloc_size(s) + #define mi_attr_alloc_size2(s1,s2) + #define mi_attr_alloc_align(p) +#endif + +// ------------------------------------------------------ +// Includes +// ------------------------------------------------------ + +#include // size_t +#include // bool +#include // INTPTR_MAX + +#ifdef __cplusplus +extern "C" { +#endif + +// ------------------------------------------------------ +// Standard malloc interface +// ------------------------------------------------------ + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_malloc(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_calloc(size_t count, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(1,2); +mi_decl_nodiscard mi_decl_export void* mi_realloc(void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(2); +mi_decl_export void* mi_expand(void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(2); + +mi_decl_export void mi_free(void* p) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_strdup(const char* s) mi_attr_noexcept mi_attr_malloc; +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_strndup(const char* s, size_t n) mi_attr_noexcept mi_attr_malloc; +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_realpath(const char* fname, char* resolved_name) mi_attr_noexcept mi_attr_malloc; + +// ------------------------------------------------------ +// Extended functionality +// ------------------------------------------------------ +#define MI_SMALL_WSIZE_MAX (128) +#define MI_SMALL_SIZE_MAX (MI_SMALL_WSIZE_MAX*sizeof(void*)) + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_malloc_small(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_zalloc_small(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_zalloc(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_mallocn(size_t count, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(1,2); +mi_decl_nodiscard mi_decl_export void* mi_reallocn(void* p, size_t count, size_t size) mi_attr_noexcept mi_attr_alloc_size2(2,3); +mi_decl_nodiscard mi_decl_export void* mi_reallocf(void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(2); + +mi_decl_nodiscard mi_decl_export size_t mi_usable_size(const void* p) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export size_t mi_good_size(size_t size) mi_attr_noexcept; + + +// ------------------------------------------------------ +// Internals +// ------------------------------------------------------ + +typedef void (mi_cdecl mi_deferred_free_fun)(bool force, unsigned long long heartbeat, void* arg); +mi_decl_export void mi_register_deferred_free(mi_deferred_free_fun* deferred_free, void* arg) mi_attr_noexcept; + +typedef void (mi_cdecl mi_output_fun)(const char* msg, void* arg); +mi_decl_export void mi_register_output(mi_output_fun* out, void* arg) mi_attr_noexcept; + +typedef void (mi_cdecl mi_error_fun)(int err, void* arg); +mi_decl_export void mi_register_error(mi_error_fun* fun, void* arg); + +mi_decl_export void mi_collect(bool force) mi_attr_noexcept; +mi_decl_export int mi_version(void) mi_attr_noexcept; +mi_decl_export void mi_stats_reset(void) mi_attr_noexcept; +mi_decl_export void mi_stats_merge(void) mi_attr_noexcept; +mi_decl_export void mi_stats_print(void* out) mi_attr_noexcept; // backward compatibility: `out` is ignored and should be NULL +mi_decl_export void mi_stats_print_out(mi_output_fun* out, void* arg) mi_attr_noexcept; +mi_decl_export void mi_thread_stats_print_out(mi_output_fun* out, void* arg) mi_attr_noexcept; +mi_decl_export void mi_options_print(void) mi_attr_noexcept; + +mi_decl_export void mi_process_info(size_t* elapsed_msecs, size_t* user_msecs, size_t* system_msecs, + size_t* current_rss, size_t* peak_rss, + size_t* current_commit, size_t* peak_commit, size_t* page_faults) mi_attr_noexcept; + + +// Generally do not use the following as these are usually called automatically +mi_decl_export void mi_process_init(void) mi_attr_noexcept; +mi_decl_export void mi_cdecl mi_process_done(void) mi_attr_noexcept; +mi_decl_export void mi_thread_init(void) mi_attr_noexcept; +mi_decl_export void mi_thread_done(void) mi_attr_noexcept; + + +// ------------------------------------------------------------------------------------- +// Aligned allocation +// Note that `alignment` always follows `size` for consistency with unaligned +// allocation, but unfortunately this differs from `posix_memalign` and `aligned_alloc`. +// ------------------------------------------------------------------------------------- + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_malloc_aligned(size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1) mi_attr_alloc_align(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_malloc_aligned_at(size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_zalloc_aligned(size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1) mi_attr_alloc_align(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_zalloc_aligned_at(size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_calloc_aligned(size_t count, size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(1,2) mi_attr_alloc_align(3); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_calloc_aligned_at(size_t count, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(1,2); +mi_decl_nodiscard mi_decl_export void* mi_realloc_aligned(void* p, size_t newsize, size_t alignment) mi_attr_noexcept mi_attr_alloc_size(2) mi_attr_alloc_align(3); +mi_decl_nodiscard mi_decl_export void* mi_realloc_aligned_at(void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size(2); + + +// ------------------------------------------------------------------------------------- +// Heaps: first-class, but can only allocate from the same thread that created it. +// ------------------------------------------------------------------------------------- + +struct mi_heap_s; +typedef struct mi_heap_s mi_heap_t; + +mi_decl_nodiscard mi_decl_export mi_heap_t* mi_heap_new(void); +mi_decl_export void mi_heap_delete(mi_heap_t* heap); +mi_decl_export void mi_heap_destroy(mi_heap_t* heap); +mi_decl_export mi_heap_t* mi_heap_set_default(mi_heap_t* heap); +mi_decl_export mi_heap_t* mi_heap_get_default(void); +mi_decl_export mi_heap_t* mi_heap_get_backing(void); +mi_decl_export void mi_heap_collect(mi_heap_t* heap, bool force) mi_attr_noexcept; + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_malloc(mi_heap_t* heap, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_zalloc(mi_heap_t* heap, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_calloc(mi_heap_t* heap, size_t count, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(2, 3); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_mallocn(mi_heap_t* heap, size_t count, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(2, 3); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_malloc_small(mi_heap_t* heap, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2); + +mi_decl_nodiscard mi_decl_export void* mi_heap_realloc(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(3); +mi_decl_nodiscard mi_decl_export void* mi_heap_reallocn(mi_heap_t* heap, void* p, size_t count, size_t size) mi_attr_noexcept mi_attr_alloc_size2(3,4); +mi_decl_nodiscard mi_decl_export void* mi_heap_reallocf(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(3); + +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_heap_strdup(mi_heap_t* heap, const char* s) mi_attr_noexcept mi_attr_malloc; +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_heap_strndup(mi_heap_t* heap, const char* s, size_t n) mi_attr_noexcept mi_attr_malloc; +mi_decl_nodiscard mi_decl_export mi_decl_restrict char* mi_heap_realpath(mi_heap_t* heap, const char* fname, char* resolved_name) mi_attr_noexcept mi_attr_malloc; + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_malloc_aligned(mi_heap_t* heap, size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2) mi_attr_alloc_align(3); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_malloc_aligned_at(mi_heap_t* heap, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_zalloc_aligned(mi_heap_t* heap, size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2) mi_attr_alloc_align(3); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_zalloc_aligned_at(mi_heap_t* heap, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_calloc_aligned(mi_heap_t* heap, size_t count, size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(2, 3) mi_attr_alloc_align(4); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_calloc_aligned_at(mi_heap_t* heap, size_t count, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size2(2, 3); +mi_decl_nodiscard mi_decl_export void* mi_heap_realloc_aligned(mi_heap_t* heap, void* p, size_t newsize, size_t alignment) mi_attr_noexcept mi_attr_alloc_size(3) mi_attr_alloc_align(4); +mi_decl_nodiscard mi_decl_export void* mi_heap_realloc_aligned_at(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size(3); + + +// -------------------------------------------------------------------------------- +// Zero initialized re-allocation. +// Only valid on memory that was originally allocated with zero initialization too. +// e.g. `mi_calloc`, `mi_zalloc`, `mi_zalloc_aligned` etc. +// see +// -------------------------------------------------------------------------------- + +mi_decl_nodiscard mi_decl_export void* mi_rezalloc(void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export void* mi_recalloc(void* p, size_t newcount, size_t size) mi_attr_noexcept mi_attr_alloc_size2(2,3); + +mi_decl_nodiscard mi_decl_export void* mi_rezalloc_aligned(void* p, size_t newsize, size_t alignment) mi_attr_noexcept mi_attr_alloc_size(2) mi_attr_alloc_align(3); +mi_decl_nodiscard mi_decl_export void* mi_rezalloc_aligned_at(void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export void* mi_recalloc_aligned(void* p, size_t newcount, size_t size, size_t alignment) mi_attr_noexcept mi_attr_alloc_size2(2,3) mi_attr_alloc_align(4); +mi_decl_nodiscard mi_decl_export void* mi_recalloc_aligned_at(void* p, size_t newcount, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size2(2,3); + +mi_decl_nodiscard mi_decl_export void* mi_heap_rezalloc(mi_heap_t* heap, void* p, size_t newsize) mi_attr_noexcept mi_attr_alloc_size(3); +mi_decl_nodiscard mi_decl_export void* mi_heap_recalloc(mi_heap_t* heap, void* p, size_t newcount, size_t size) mi_attr_noexcept mi_attr_alloc_size2(3,4); + +mi_decl_nodiscard mi_decl_export void* mi_heap_rezalloc_aligned(mi_heap_t* heap, void* p, size_t newsize, size_t alignment) mi_attr_noexcept mi_attr_alloc_size(3) mi_attr_alloc_align(4); +mi_decl_nodiscard mi_decl_export void* mi_heap_rezalloc_aligned_at(mi_heap_t* heap, void* p, size_t newsize, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size(3); +mi_decl_nodiscard mi_decl_export void* mi_heap_recalloc_aligned(mi_heap_t* heap, void* p, size_t newcount, size_t size, size_t alignment) mi_attr_noexcept mi_attr_alloc_size2(3,4) mi_attr_alloc_align(5); +mi_decl_nodiscard mi_decl_export void* mi_heap_recalloc_aligned_at(mi_heap_t* heap, void* p, size_t newcount, size_t size, size_t alignment, size_t offset) mi_attr_noexcept mi_attr_alloc_size2(3,4); + + +// ------------------------------------------------------ +// Analysis +// ------------------------------------------------------ + +mi_decl_export bool mi_heap_contains_block(mi_heap_t* heap, const void* p); +mi_decl_export bool mi_heap_check_owned(mi_heap_t* heap, const void* p); +mi_decl_export bool mi_check_owned(const void* p); + +// An area of heap space contains blocks of a single size. +typedef struct mi_heap_area_s { + void* blocks; // start of the area containing heap blocks + size_t reserved; // bytes reserved for this area (virtual) + size_t committed; // current available bytes for this area + size_t used; // number of allocated blocks + size_t block_size; // size in bytes of each block + size_t full_block_size; // size in bytes of a full block including padding and metadata. + int heap_tag; // heap tag associated with this area +} mi_heap_area_t; + +typedef bool (mi_cdecl mi_block_visit_fun)(const mi_heap_t* heap, const mi_heap_area_t* area, void* block, size_t block_size, void* arg); + +mi_decl_export bool mi_heap_visit_blocks(const mi_heap_t* heap, bool visit_blocks, mi_block_visit_fun* visitor, void* arg); + +// Experimental +mi_decl_nodiscard mi_decl_export bool mi_is_in_heap_region(const void* p) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export bool mi_is_redirected(void) mi_attr_noexcept; + +mi_decl_export int mi_reserve_huge_os_pages_interleave(size_t pages, size_t numa_nodes, size_t timeout_msecs) mi_attr_noexcept; +mi_decl_export int mi_reserve_huge_os_pages_at(size_t pages, int numa_node, size_t timeout_msecs) mi_attr_noexcept; + +mi_decl_export int mi_reserve_os_memory(size_t size, bool commit, bool allow_large) mi_attr_noexcept; +mi_decl_export bool mi_manage_os_memory(void* start, size_t size, bool is_committed, bool is_large, bool is_zero, int numa_node) mi_attr_noexcept; + +mi_decl_export void mi_debug_show_arenas(void) mi_attr_noexcept; +mi_decl_export void mi_arenas_print(void) mi_attr_noexcept; + +// Experimental: heaps associated with specific memory arena's +typedef int mi_arena_id_t; +mi_decl_export void* mi_arena_area(mi_arena_id_t arena_id, size_t* size); +mi_decl_export int mi_reserve_huge_os_pages_at_ex(size_t pages, int numa_node, size_t timeout_msecs, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept; +mi_decl_export int mi_reserve_os_memory_ex(size_t size, bool commit, bool allow_large, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept; +mi_decl_export bool mi_manage_os_memory_ex(void* start, size_t size, bool is_committed, bool is_large, bool is_zero, int numa_node, bool exclusive, mi_arena_id_t* arena_id) mi_attr_noexcept; + +#if MI_MALLOC_VERSION >= 182 +// Create a heap that only allocates in the specified arena +mi_decl_nodiscard mi_decl_export mi_heap_t* mi_heap_new_in_arena(mi_arena_id_t arena_id); +#endif + + +// Experimental: allow sub-processes whose memory areas stay separated (and no reclamation between them) +// Used for example for separate interpreters in one process. +typedef void* mi_subproc_id_t; +mi_decl_export mi_subproc_id_t mi_subproc_main(void); +mi_decl_export mi_subproc_id_t mi_subproc_new(void); +mi_decl_export void mi_subproc_delete(mi_subproc_id_t subproc); +mi_decl_export void mi_subproc_add_current_thread(mi_subproc_id_t subproc); // this should be called right after a thread is created (and no allocation has taken place yet) + +// Experimental: visit abandoned heap areas (that are not owned by a specific heap) +mi_decl_export bool mi_abandoned_visit_blocks(mi_subproc_id_t subproc_id, int heap_tag, bool visit_blocks, mi_block_visit_fun* visitor, void* arg); + +// Experimental: objects followed by a guard page. +// A sample rate of 0 disables guarded objects, while 1 uses a guard page for every object. +// A seed of 0 uses a random start point. Only objects within the size bound are eligable for guard pages. +mi_decl_export void mi_heap_guarded_set_sample_rate(mi_heap_t* heap, size_t sample_rate, size_t seed); +mi_decl_export void mi_heap_guarded_set_size_bound(mi_heap_t* heap, size_t min, size_t max); + +// Experimental: communicate that the thread is part of a threadpool +mi_decl_export void mi_thread_set_in_threadpool(void) mi_attr_noexcept; + +// Experimental: create a new heap with a specified heap tag. Set `allow_destroy` to false to allow the thread +// to reclaim abandoned memory (with a compatible heap_tag and arena_id) but in that case `mi_heap_destroy` will +// fall back to `mi_heap_delete`. +mi_decl_nodiscard mi_decl_export mi_heap_t* mi_heap_new_ex(int heap_tag, bool allow_destroy, mi_arena_id_t arena_id); + +// deprecated +mi_decl_export int mi_reserve_huge_os_pages(size_t pages, double max_secs, size_t* pages_reserved) mi_attr_noexcept; +mi_decl_export void mi_collect_reduce(size_t target_thread_owned) mi_attr_noexcept; + + + +// ------------------------------------------------------ +// Convenience +// ------------------------------------------------------ + +#define mi_malloc_tp(tp) ((tp*)mi_malloc(sizeof(tp))) +#define mi_zalloc_tp(tp) ((tp*)mi_zalloc(sizeof(tp))) +#define mi_calloc_tp(tp,n) ((tp*)mi_calloc(n,sizeof(tp))) +#define mi_mallocn_tp(tp,n) ((tp*)mi_mallocn(n,sizeof(tp))) +#define mi_reallocn_tp(p,tp,n) ((tp*)mi_reallocn(p,n,sizeof(tp))) +#define mi_recalloc_tp(p,tp,n) ((tp*)mi_recalloc(p,n,sizeof(tp))) + +#define mi_heap_malloc_tp(hp,tp) ((tp*)mi_heap_malloc(hp,sizeof(tp))) +#define mi_heap_zalloc_tp(hp,tp) ((tp*)mi_heap_zalloc(hp,sizeof(tp))) +#define mi_heap_calloc_tp(hp,tp,n) ((tp*)mi_heap_calloc(hp,n,sizeof(tp))) +#define mi_heap_mallocn_tp(hp,tp,n) ((tp*)mi_heap_mallocn(hp,n,sizeof(tp))) +#define mi_heap_reallocn_tp(hp,p,tp,n) ((tp*)mi_heap_reallocn(hp,p,n,sizeof(tp))) +#define mi_heap_recalloc_tp(hp,p,tp,n) ((tp*)mi_heap_recalloc(hp,p,n,sizeof(tp))) + + +// ------------------------------------------------------ +// Options +// ------------------------------------------------------ + +typedef enum mi_option_e { + // stable options + mi_option_show_errors, // print error messages + mi_option_show_stats, // print statistics on termination + mi_option_verbose, // print verbose messages + // advanced options + mi_option_eager_commit, // eager commit segments? (after `eager_commit_delay` segments) (=1) + mi_option_arena_eager_commit, // eager commit arenas? Use 2 to enable just on overcommit systems (=2) + mi_option_purge_decommits, // should a memory purge decommit? (=1). Set to 0 to use memory reset on a purge (instead of decommit) + mi_option_allow_large_os_pages, // allow large (2 or 4 MiB) OS pages, implies eager commit. If false, also disables THP for the process. + mi_option_reserve_huge_os_pages, // reserve N huge OS pages (1GiB pages) at startup + mi_option_reserve_huge_os_pages_at, // reserve huge OS pages at a specific NUMA node + mi_option_reserve_os_memory, // reserve specified amount of OS memory in an arena at startup (internally, this value is in KiB; use `mi_option_get_size`) + mi_option_deprecated_segment_cache, + mi_option_deprecated_page_reset, + mi_option_abandoned_page_purge, // immediately purge delayed purges on thread termination + mi_option_deprecated_segment_reset, + mi_option_eager_commit_delay, // the first N segments per thread are not eagerly committed (but per page in the segment on demand) + mi_option_purge_delay, // memory purging is delayed by N milli seconds; use 0 for immediate purging or -1 for no purging at all. (=10) + mi_option_use_numa_nodes, // 0 = use all available numa nodes, otherwise use at most N nodes. + mi_option_disallow_os_alloc, // 1 = do not use OS memory for allocation (but only programmatically reserved arenas) + mi_option_os_tag, // tag used for OS logging (macOS only for now) (=100) + mi_option_max_errors, // issue at most N error messages + mi_option_max_warnings, // issue at most N warning messages + mi_option_max_segment_reclaim, // max. percentage of the abandoned segments can be reclaimed per try (=10%) + mi_option_destroy_on_exit, // if set, release all memory on exit; sometimes used for dynamic unloading but can be unsafe + mi_option_arena_reserve, // initial memory size for arena reservation (= 1 GiB on 64-bit) (internally, this value is in KiB; use `mi_option_get_size`) + mi_option_arena_purge_mult, // multiplier for `purge_delay` for the purging delay for arenas (=10) + mi_option_purge_extend_delay, + mi_option_abandoned_reclaim_on_free, // allow to reclaim an abandoned segment on a free (=1) + mi_option_disallow_arena_alloc, // 1 = do not use arena's for allocation (except if using specific arena id's) + mi_option_retry_on_oom, // retry on out-of-memory for N milli seconds (=400), set to 0 to disable retries. (only on windows) + mi_option_visit_abandoned, // allow visiting heap blocks from abandoned threads (=0) + mi_option_guarded_min, // only used when building with MI_GUARDED: minimal rounded object size for guarded objects (=0) + mi_option_guarded_max, // only used when building with MI_GUARDED: maximal rounded object size for guarded objects (=0) + mi_option_guarded_precise, // disregard minimal alignment requirement to always place guarded blocks exactly in front of a guard page (=0) + mi_option_guarded_sample_rate, // 1 out of N allocations in the min/max range will be guarded (=1000) + mi_option_guarded_sample_seed, // can be set to allow for a (more) deterministic re-execution when a guard page is triggered (=0) + mi_option_target_segments_per_thread, // experimental (=0) + mi_option_generic_collect, // collect heaps every N (=10000) generic allocation calls + _mi_option_last, + // legacy option names + mi_option_large_os_pages = mi_option_allow_large_os_pages, + mi_option_eager_region_commit = mi_option_arena_eager_commit, + mi_option_reset_decommits = mi_option_purge_decommits, + mi_option_reset_delay = mi_option_purge_delay, + mi_option_abandoned_page_reset = mi_option_abandoned_page_purge, + mi_option_limit_os_alloc = mi_option_disallow_os_alloc +} mi_option_t; + + +mi_decl_nodiscard mi_decl_export bool mi_option_is_enabled(mi_option_t option); +mi_decl_export void mi_option_enable(mi_option_t option); +mi_decl_export void mi_option_disable(mi_option_t option); +mi_decl_export void mi_option_set_enabled(mi_option_t option, bool enable); +mi_decl_export void mi_option_set_enabled_default(mi_option_t option, bool enable); + +mi_decl_nodiscard mi_decl_export long mi_option_get(mi_option_t option); +mi_decl_nodiscard mi_decl_export long mi_option_get_clamp(mi_option_t option, long min, long max); +mi_decl_nodiscard mi_decl_export size_t mi_option_get_size(mi_option_t option); +mi_decl_export void mi_option_set(mi_option_t option, long value); +mi_decl_export void mi_option_set_default(mi_option_t option, long value); + + +// ------------------------------------------------------------------------------------------------------- +// "mi" prefixed implementations of various posix, Unix, Windows, and C++ allocation functions. +// (This can be convenient when providing overrides of these functions as done in `mimalloc-override.h`.) +// note: we use `mi_cfree` as "checked free" and it checks if the pointer is in our heap before free-ing. +// ------------------------------------------------------------------------------------------------------- + +mi_decl_export void mi_cfree(void* p) mi_attr_noexcept; +mi_decl_export void* mi__expand(void* p, size_t newsize) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export size_t mi_malloc_size(const void* p) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export size_t mi_malloc_good_size(size_t size) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export size_t mi_malloc_usable_size(const void *p) mi_attr_noexcept; + +mi_decl_export int mi_posix_memalign(void** p, size_t alignment, size_t size) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_memalign(size_t alignment, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2) mi_attr_alloc_align(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_valloc(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_pvalloc(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_aligned_alloc(size_t alignment, size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(2) mi_attr_alloc_align(1); + +mi_decl_nodiscard mi_decl_export void* mi_reallocarray(void* p, size_t count, size_t size) mi_attr_noexcept mi_attr_alloc_size2(2,3); +mi_decl_nodiscard mi_decl_export int mi_reallocarr(void* p, size_t count, size_t size) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export void* mi_aligned_recalloc(void* p, size_t newcount, size_t size, size_t alignment) mi_attr_noexcept; +mi_decl_nodiscard mi_decl_export void* mi_aligned_offset_recalloc(void* p, size_t newcount, size_t size, size_t alignment, size_t offset) mi_attr_noexcept; + +mi_decl_nodiscard mi_decl_export mi_decl_restrict unsigned short* mi_wcsdup(const unsigned short* s) mi_attr_noexcept mi_attr_malloc; +mi_decl_nodiscard mi_decl_export mi_decl_restrict unsigned char* mi_mbsdup(const unsigned char* s) mi_attr_noexcept mi_attr_malloc; +mi_decl_export int mi_dupenv_s(char** buf, size_t* size, const char* name) mi_attr_noexcept; +mi_decl_export int mi_wdupenv_s(unsigned short** buf, size_t* size, const unsigned short* name) mi_attr_noexcept; + +mi_decl_export void mi_free_size(void* p, size_t size) mi_attr_noexcept; +mi_decl_export void mi_free_size_aligned(void* p, size_t size, size_t alignment) mi_attr_noexcept; +mi_decl_export void mi_free_aligned(void* p, size_t alignment) mi_attr_noexcept; + +// The `mi_new` wrappers implement C++ semantics on out-of-memory instead of directly returning `NULL`. +// (and call `std::get_new_handler` and potentially raise a `std::bad_alloc` exception). +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_new(size_t size) mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_new_aligned(size_t size, size_t alignment) mi_attr_malloc mi_attr_alloc_size(1) mi_attr_alloc_align(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_new_nothrow(size_t size) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_new_aligned_nothrow(size_t size, size_t alignment) mi_attr_noexcept mi_attr_malloc mi_attr_alloc_size(1) mi_attr_alloc_align(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_new_n(size_t count, size_t size) mi_attr_malloc mi_attr_alloc_size2(1, 2); +mi_decl_nodiscard mi_decl_export void* mi_new_realloc(void* p, size_t newsize) mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export void* mi_new_reallocn(void* p, size_t newcount, size_t size) mi_attr_alloc_size2(2, 3); + +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_alloc_new(mi_heap_t* heap, size_t size) mi_attr_malloc mi_attr_alloc_size(2); +mi_decl_nodiscard mi_decl_export mi_decl_restrict void* mi_heap_alloc_new_n(mi_heap_t* heap, size_t count, size_t size) mi_attr_malloc mi_attr_alloc_size2(2, 3); + +#ifdef __cplusplus +} +#endif + +// --------------------------------------------------------------------------------------------- +// Implement the C++ std::allocator interface for use in STL containers. +// (note: see `mimalloc-new-delete.h` for overriding the new/delete operators globally) +// --------------------------------------------------------------------------------------------- +#ifdef __cplusplus + +#include // std::size_t +#include // PTRDIFF_MAX +#if (__cplusplus >= 201103L) || (_MSC_VER > 1900) // C++11 +#include // std::true_type +#include // std::forward +#endif + +template struct _mi_stl_allocator_common { + typedef T value_type; + typedef std::size_t size_type; + typedef std::ptrdiff_t difference_type; + typedef value_type& reference; + typedef value_type const& const_reference; + typedef value_type* pointer; + typedef value_type const* const_pointer; + + #if ((__cplusplus >= 201103L) || (_MSC_VER > 1900)) // C++11 + using propagate_on_container_copy_assignment = std::true_type; + using propagate_on_container_move_assignment = std::true_type; + using propagate_on_container_swap = std::true_type; + template void construct(U* p, Args&& ...args) { ::new(p) U(std::forward(args)...); } + template void destroy(U* p) mi_attr_noexcept { p->~U(); } + #else + void construct(pointer p, value_type const& val) { ::new(p) value_type(val); } + void destroy(pointer p) { p->~value_type(); } + #endif + + size_type max_size() const mi_attr_noexcept { return (PTRDIFF_MAX/sizeof(value_type)); } + pointer address(reference x) const { return &x; } + const_pointer address(const_reference x) const { return &x; } +}; + +template struct mi_stl_allocator : public _mi_stl_allocator_common { + using typename _mi_stl_allocator_common::size_type; + using typename _mi_stl_allocator_common::value_type; + using typename _mi_stl_allocator_common::pointer; + template struct rebind { typedef mi_stl_allocator other; }; + + mi_stl_allocator() mi_attr_noexcept = default; + mi_stl_allocator(const mi_stl_allocator&) mi_attr_noexcept = default; + template mi_stl_allocator(const mi_stl_allocator&) mi_attr_noexcept { } + mi_stl_allocator select_on_container_copy_construction() const { return *this; } + void deallocate(T* p, size_type) { mi_free(p); } + + #if (__cplusplus >= 201703L) // C++17 + mi_decl_nodiscard T* allocate(size_type count) { return static_cast(mi_new_n(count, sizeof(T))); } + mi_decl_nodiscard T* allocate(size_type count, const void*) { return allocate(count); } + #else + mi_decl_nodiscard pointer allocate(size_type count, const void* = 0) { return static_cast(mi_new_n(count, sizeof(value_type))); } + #endif + + #if ((__cplusplus >= 201103L) || (_MSC_VER > 1900)) // C++11 + using is_always_equal = std::true_type; + #endif +}; + +template bool operator==(const mi_stl_allocator& , const mi_stl_allocator& ) mi_attr_noexcept { return true; } +template bool operator!=(const mi_stl_allocator& , const mi_stl_allocator& ) mi_attr_noexcept { return false; } + + +#if (__cplusplus >= 201103L) || (_MSC_VER >= 1900) // C++11 +#define MI_HAS_HEAP_STL_ALLOCATOR 1 + +#include // std::shared_ptr + +// Common base class for STL allocators in a specific heap +template struct _mi_heap_stl_allocator_common : public _mi_stl_allocator_common { + using typename _mi_stl_allocator_common::size_type; + using typename _mi_stl_allocator_common::value_type; + using typename _mi_stl_allocator_common::pointer; + + _mi_heap_stl_allocator_common(mi_heap_t* hp) : heap(hp, [](mi_heap_t*) {}) {} /* will not delete nor destroy the passed in heap */ + + #if (__cplusplus >= 201703L) // C++17 + mi_decl_nodiscard T* allocate(size_type count) { return static_cast(mi_heap_alloc_new_n(this->heap.get(), count, sizeof(T))); } + mi_decl_nodiscard T* allocate(size_type count, const void*) { return allocate(count); } + #else + mi_decl_nodiscard pointer allocate(size_type count, const void* = 0) { return static_cast(mi_heap_alloc_new_n(this->heap.get(), count, sizeof(value_type))); } + #endif + + #if ((__cplusplus >= 201103L) || (_MSC_VER > 1900)) // C++11 + using is_always_equal = std::false_type; + #endif + + void collect(bool force) { mi_heap_collect(this->heap.get(), force); } + template bool is_equal(const _mi_heap_stl_allocator_common& x) const { return (this->heap == x.heap); } + +protected: + std::shared_ptr heap; + template friend struct _mi_heap_stl_allocator_common; + + _mi_heap_stl_allocator_common() { + mi_heap_t* hp = mi_heap_new(); + this->heap.reset(hp, (_mi_destroy ? &heap_destroy : &heap_delete)); /* calls heap_delete/destroy when the refcount drops to zero */ + } + _mi_heap_stl_allocator_common(const _mi_heap_stl_allocator_common& x) mi_attr_noexcept : heap(x.heap) { } + template _mi_heap_stl_allocator_common(const _mi_heap_stl_allocator_common& x) mi_attr_noexcept : heap(x.heap) { } + +private: + static void heap_delete(mi_heap_t* hp) { if (hp != NULL) { mi_heap_delete(hp); } } + static void heap_destroy(mi_heap_t* hp) { if (hp != NULL) { mi_heap_destroy(hp); } } +}; + +// STL allocator allocation in a specific heap +template struct mi_heap_stl_allocator : public _mi_heap_stl_allocator_common { + using typename _mi_heap_stl_allocator_common::size_type; + mi_heap_stl_allocator() : _mi_heap_stl_allocator_common() { } // creates fresh heap that is deleted when the destructor is called + mi_heap_stl_allocator(mi_heap_t* hp) : _mi_heap_stl_allocator_common(hp) { } // no delete nor destroy on the passed in heap + template mi_heap_stl_allocator(const mi_heap_stl_allocator& x) mi_attr_noexcept : _mi_heap_stl_allocator_common(x) { } + + mi_heap_stl_allocator select_on_container_copy_construction() const { return *this; } + void deallocate(T* p, size_type) { mi_free(p); } + template struct rebind { typedef mi_heap_stl_allocator other; }; +}; + +template bool operator==(const mi_heap_stl_allocator& x, const mi_heap_stl_allocator& y) mi_attr_noexcept { return (x.is_equal(y)); } +template bool operator!=(const mi_heap_stl_allocator& x, const mi_heap_stl_allocator& y) mi_attr_noexcept { return (!x.is_equal(y)); } + + +// STL allocator allocation in a specific heap, where `free` does nothing and +// the heap is destroyed in one go on destruction -- use with care! +template struct mi_heap_destroy_stl_allocator : public _mi_heap_stl_allocator_common { + using typename _mi_heap_stl_allocator_common::size_type; + mi_heap_destroy_stl_allocator() : _mi_heap_stl_allocator_common() { } // creates fresh heap that is destroyed when the destructor is called + mi_heap_destroy_stl_allocator(mi_heap_t* hp) : _mi_heap_stl_allocator_common(hp) { } // no delete nor destroy on the passed in heap + template mi_heap_destroy_stl_allocator(const mi_heap_destroy_stl_allocator& x) mi_attr_noexcept : _mi_heap_stl_allocator_common(x) { } + + mi_heap_destroy_stl_allocator select_on_container_copy_construction() const { return *this; } + void deallocate(T*, size_type) { /* do nothing as we destroy the heap on destruct. */ } + template struct rebind { typedef mi_heap_destroy_stl_allocator other; }; +}; + +template bool operator==(const mi_heap_destroy_stl_allocator& x, const mi_heap_destroy_stl_allocator& y) mi_attr_noexcept { return (x.is_equal(y)); } +template bool operator!=(const mi_heap_destroy_stl_allocator& x, const mi_heap_destroy_stl_allocator& y) mi_attr_noexcept { return (!x.is_equal(y)); } + +#endif // C++11 + +#endif // __cplusplus + +#endif diff --git a/compat/mimalloc/mimalloc/atomic.h b/compat/mimalloc/mimalloc/atomic.h new file mode 100644 index 00000000000000..e8bac316b3a6f3 --- /dev/null +++ b/compat/mimalloc/mimalloc/atomic.h @@ -0,0 +1,557 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024 Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_ATOMIC_H +#define MIMALLOC_ATOMIC_H + +// include windows.h or pthreads.h +#if defined(_WIN32) +#ifndef WIN32_LEAN_AND_MEAN +#define WIN32_LEAN_AND_MEAN +#endif +#include +#elif !defined(__wasi__) && (!defined(__EMSCRIPTEN__) || defined(__EMSCRIPTEN_PTHREADS__)) +#define MI_USE_PTHREADS +#include +#endif + +// -------------------------------------------------------------------------------------------- +// Atomics +// We need to be portable between C, C++, and MSVC. +// We base the primitives on the C/C++ atomics and create a minimal wrapper for MSVC in C compilation mode. +// This is why we try to use only `uintptr_t` and `*` as atomic types. +// To gain better insight in the range of used atomics, we use explicitly named memory order operations +// instead of passing the memory order as a parameter. +// ----------------------------------------------------------------------------------------------- + +#if defined(__cplusplus) +// Use C++ atomics +#include +#define _Atomic(tp) std::atomic +#define mi_atomic(name) std::atomic_##name +#define mi_memory_order(name) std::memory_order_##name +#if (__cplusplus >= 202002L) // c++20, see issue #571 + #define MI_ATOMIC_VAR_INIT(x) x +#elif !defined(ATOMIC_VAR_INIT) + #define MI_ATOMIC_VAR_INIT(x) x +#else + #define MI_ATOMIC_VAR_INIT(x) ATOMIC_VAR_INIT(x) +#endif +#elif defined(_MSC_VER) +// Use MSVC C wrapper for C11 atomics +#define _Atomic(tp) tp +#define MI_ATOMIC_VAR_INIT(x) x +#define mi_atomic(name) mi_atomic_##name +#define mi_memory_order(name) mi_memory_order_##name +#else +// Use C11 atomics +#include +#define mi_atomic(name) atomic_##name +#define mi_memory_order(name) memory_order_##name +#if (__STDC_VERSION__ >= 201710L) // c17, see issue #735 + #define MI_ATOMIC_VAR_INIT(x) x +#elif !defined(ATOMIC_VAR_INIT) + #define MI_ATOMIC_VAR_INIT(x) x +#else + #define MI_ATOMIC_VAR_INIT(x) ATOMIC_VAR_INIT(x) +#endif +#endif + +// Various defines for all used memory orders in mimalloc +#define mi_atomic_cas_weak(p,expected,desired,mem_success,mem_fail) \ + mi_atomic(compare_exchange_weak_explicit)(p,expected,desired,mem_success,mem_fail) + +#define mi_atomic_cas_strong(p,expected,desired,mem_success,mem_fail) \ + mi_atomic(compare_exchange_strong_explicit)(p,expected,desired,mem_success,mem_fail) + +#define mi_atomic_load_acquire(p) mi_atomic(load_explicit)(p,mi_memory_order(acquire)) +#define mi_atomic_load_relaxed(p) mi_atomic(load_explicit)(p,mi_memory_order(relaxed)) +#define mi_atomic_store_release(p,x) mi_atomic(store_explicit)(p,x,mi_memory_order(release)) +#define mi_atomic_store_relaxed(p,x) mi_atomic(store_explicit)(p,x,mi_memory_order(relaxed)) +#define mi_atomic_exchange_relaxed(p,x) mi_atomic(exchange_explicit)(p,x,mi_memory_order(relaxed)) +#define mi_atomic_exchange_release(p,x) mi_atomic(exchange_explicit)(p,x,mi_memory_order(release)) +#define mi_atomic_exchange_acq_rel(p,x) mi_atomic(exchange_explicit)(p,x,mi_memory_order(acq_rel)) +#define mi_atomic_cas_weak_release(p,exp,des) mi_atomic_cas_weak(p,exp,des,mi_memory_order(release),mi_memory_order(relaxed)) +#define mi_atomic_cas_weak_acq_rel(p,exp,des) mi_atomic_cas_weak(p,exp,des,mi_memory_order(acq_rel),mi_memory_order(acquire)) +#define mi_atomic_cas_strong_release(p,exp,des) mi_atomic_cas_strong(p,exp,des,mi_memory_order(release),mi_memory_order(relaxed)) +#define mi_atomic_cas_strong_acq_rel(p,exp,des) mi_atomic_cas_strong(p,exp,des,mi_memory_order(acq_rel),mi_memory_order(acquire)) + +#define mi_atomic_add_relaxed(p,x) mi_atomic(fetch_add_explicit)(p,x,mi_memory_order(relaxed)) +#define mi_atomic_sub_relaxed(p,x) mi_atomic(fetch_sub_explicit)(p,x,mi_memory_order(relaxed)) +#define mi_atomic_add_acq_rel(p,x) mi_atomic(fetch_add_explicit)(p,x,mi_memory_order(acq_rel)) +#define mi_atomic_sub_acq_rel(p,x) mi_atomic(fetch_sub_explicit)(p,x,mi_memory_order(acq_rel)) +#define mi_atomic_and_acq_rel(p,x) mi_atomic(fetch_and_explicit)(p,x,mi_memory_order(acq_rel)) +#define mi_atomic_or_acq_rel(p,x) mi_atomic(fetch_or_explicit)(p,x,mi_memory_order(acq_rel)) + +#define mi_atomic_increment_relaxed(p) mi_atomic_add_relaxed(p,(uintptr_t)1) +#define mi_atomic_decrement_relaxed(p) mi_atomic_sub_relaxed(p,(uintptr_t)1) +#define mi_atomic_increment_acq_rel(p) mi_atomic_add_acq_rel(p,(uintptr_t)1) +#define mi_atomic_decrement_acq_rel(p) mi_atomic_sub_acq_rel(p,(uintptr_t)1) + +static inline void mi_atomic_yield(void); +static inline intptr_t mi_atomic_addi(_Atomic(intptr_t)*p, intptr_t add); +static inline intptr_t mi_atomic_subi(_Atomic(intptr_t)*p, intptr_t sub); + + +#if defined(__cplusplus) || !defined(_MSC_VER) + +// In C++/C11 atomics we have polymorphic atomics so can use the typed `ptr` variants (where `tp` is the type of atomic value) +// We use these macros so we can provide a typed wrapper in MSVC in C compilation mode as well +#define mi_atomic_load_ptr_acquire(tp,p) mi_atomic_load_acquire(p) +#define mi_atomic_load_ptr_relaxed(tp,p) mi_atomic_load_relaxed(p) + +// In C++ we need to add casts to help resolve templates if NULL is passed +#if defined(__cplusplus) +#define mi_atomic_store_ptr_release(tp,p,x) mi_atomic_store_release(p,(tp*)x) +#define mi_atomic_store_ptr_relaxed(tp,p,x) mi_atomic_store_relaxed(p,(tp*)x) +#define mi_atomic_cas_ptr_weak_release(tp,p,exp,des) mi_atomic_cas_weak_release(p,exp,(tp*)des) +#define mi_atomic_cas_ptr_weak_acq_rel(tp,p,exp,des) mi_atomic_cas_weak_acq_rel(p,exp,(tp*)des) +#define mi_atomic_cas_ptr_strong_release(tp,p,exp,des) mi_atomic_cas_strong_release(p,exp,(tp*)des) +#define mi_atomic_cas_ptr_strong_acq_rel(tp,p,exp,des) mi_atomic_cas_strong_acq_rel(p,exp,(tp*)des) +#define mi_atomic_exchange_ptr_relaxed(tp,p,x) mi_atomic_exchange_relaxed(p,(tp*)x) +#define mi_atomic_exchange_ptr_release(tp,p,x) mi_atomic_exchange_release(p,(tp*)x) +#define mi_atomic_exchange_ptr_acq_rel(tp,p,x) mi_atomic_exchange_acq_rel(p,(tp*)x) +#else +#define mi_atomic_store_ptr_release(tp,p,x) mi_atomic_store_release(p,x) +#define mi_atomic_store_ptr_relaxed(tp,p,x) mi_atomic_store_relaxed(p,x) +#define mi_atomic_cas_ptr_weak_release(tp,p,exp,des) mi_atomic_cas_weak_release(p,exp,des) +#define mi_atomic_cas_ptr_weak_acq_rel(tp,p,exp,des) mi_atomic_cas_weak_acq_rel(p,exp,des) +#define mi_atomic_cas_ptr_strong_release(tp,p,exp,des) mi_atomic_cas_strong_release(p,exp,des) +#define mi_atomic_cas_ptr_strong_acq_rel(tp,p,exp,des) mi_atomic_cas_strong_acq_rel(p,exp,des) +#define mi_atomic_exchange_ptr_relaxed(tp,p,x) mi_atomic_exchange_relaxed(p,x) +#define mi_atomic_exchange_ptr_release(tp,p,x) mi_atomic_exchange_release(p,x) +#define mi_atomic_exchange_ptr_acq_rel(tp,p,x) mi_atomic_exchange_acq_rel(p,x) +#endif + +// These are used by the statistics +static inline int64_t mi_atomic_addi64_relaxed(volatile int64_t* p, int64_t add) { + return mi_atomic(fetch_add_explicit)((_Atomic(int64_t)*)p, add, mi_memory_order(relaxed)); +} +static inline void mi_atomic_void_addi64_relaxed(volatile int64_t* p, const volatile int64_t* padd) { + const int64_t add = mi_atomic_load_relaxed((_Atomic(int64_t)*)padd); + if (add != 0) { + mi_atomic(fetch_add_explicit)((_Atomic(int64_t)*)p, add, mi_memory_order(relaxed)); + } +} +static inline void mi_atomic_maxi64_relaxed(volatile int64_t* p, int64_t x) { + int64_t current = mi_atomic_load_relaxed((_Atomic(int64_t)*)p); + while (current < x && !mi_atomic_cas_weak_release((_Atomic(int64_t)*)p, ¤t, x)) { /* nothing */ }; +} + +// Used by timers +#define mi_atomic_loadi64_acquire(p) mi_atomic(load_explicit)(p,mi_memory_order(acquire)) +#define mi_atomic_loadi64_relaxed(p) mi_atomic(load_explicit)(p,mi_memory_order(relaxed)) +#define mi_atomic_storei64_release(p,x) mi_atomic(store_explicit)(p,x,mi_memory_order(release)) +#define mi_atomic_storei64_relaxed(p,x) mi_atomic(store_explicit)(p,x,mi_memory_order(relaxed)) + +#define mi_atomic_casi64_strong_acq_rel(p,e,d) mi_atomic_cas_strong_acq_rel(p,e,d) +#define mi_atomic_addi64_acq_rel(p,i) mi_atomic_add_acq_rel(p,i) + + +#elif defined(_MSC_VER) + +// Legacy MSVC plain C compilation wrapper that uses Interlocked operations to model C11 atomics. +#include +#ifdef _WIN64 +typedef LONG64 msc_intptr_t; +#define MI_64(f) f##64 +#else +typedef LONG msc_intptr_t; +#define MI_64(f) f +#endif + +typedef enum mi_memory_order_e { + mi_memory_order_relaxed, + mi_memory_order_consume, + mi_memory_order_acquire, + mi_memory_order_release, + mi_memory_order_acq_rel, + mi_memory_order_seq_cst +} mi_memory_order; + +static inline uintptr_t mi_atomic_fetch_add_explicit(_Atomic(uintptr_t)*p, uintptr_t add, mi_memory_order mo) { + (void)(mo); + return (uintptr_t)MI_64(_InterlockedExchangeAdd)((volatile msc_intptr_t*)p, (msc_intptr_t)add); +} +static inline uintptr_t mi_atomic_fetch_sub_explicit(_Atomic(uintptr_t)*p, uintptr_t sub, mi_memory_order mo) { + (void)(mo); + return (uintptr_t)MI_64(_InterlockedExchangeAdd)((volatile msc_intptr_t*)p, -((msc_intptr_t)sub)); +} +static inline uintptr_t mi_atomic_fetch_and_explicit(_Atomic(uintptr_t)*p, uintptr_t x, mi_memory_order mo) { + (void)(mo); + return (uintptr_t)MI_64(_InterlockedAnd)((volatile msc_intptr_t*)p, (msc_intptr_t)x); +} +static inline uintptr_t mi_atomic_fetch_or_explicit(_Atomic(uintptr_t)*p, uintptr_t x, mi_memory_order mo) { + (void)(mo); + return (uintptr_t)MI_64(_InterlockedOr)((volatile msc_intptr_t*)p, (msc_intptr_t)x); +} +static inline bool mi_atomic_compare_exchange_strong_explicit(_Atomic(uintptr_t)*p, uintptr_t* expected, uintptr_t desired, mi_memory_order mo1, mi_memory_order mo2) { + (void)(mo1); (void)(mo2); + uintptr_t read = (uintptr_t)MI_64(_InterlockedCompareExchange)((volatile msc_intptr_t*)p, (msc_intptr_t)desired, (msc_intptr_t)(*expected)); + if (read == *expected) { + return true; + } + else { + *expected = read; + return false; + } +} +static inline bool mi_atomic_compare_exchange_weak_explicit(_Atomic(uintptr_t)*p, uintptr_t* expected, uintptr_t desired, mi_memory_order mo1, mi_memory_order mo2) { + return mi_atomic_compare_exchange_strong_explicit(p, expected, desired, mo1, mo2); +} +static inline uintptr_t mi_atomic_exchange_explicit(_Atomic(uintptr_t)*p, uintptr_t exchange, mi_memory_order mo) { + (void)(mo); + return (uintptr_t)MI_64(_InterlockedExchange)((volatile msc_intptr_t*)p, (msc_intptr_t)exchange); +} +static inline void mi_atomic_thread_fence(mi_memory_order mo) { + (void)(mo); + _Atomic(uintptr_t) x = 0; + mi_atomic_exchange_explicit(&x, 1, mo); +} +static inline uintptr_t mi_atomic_load_explicit(_Atomic(uintptr_t) const* p, mi_memory_order mo) { + (void)(mo); +#if defined(_M_IX86) || defined(_M_X64) + return *p; +#else + uintptr_t x = *p; + if (mo > mi_memory_order_relaxed) { + while (!mi_atomic_compare_exchange_weak_explicit((_Atomic(uintptr_t)*)p, &x, x, mo, mi_memory_order_relaxed)) { /* nothing */ }; + } + return x; +#endif +} +static inline void mi_atomic_store_explicit(_Atomic(uintptr_t)*p, uintptr_t x, mi_memory_order mo) { + (void)(mo); +#if defined(_M_IX86) || defined(_M_X64) + *p = x; +#else + mi_atomic_exchange_explicit(p, x, mo); +#endif +} +static inline int64_t mi_atomic_loadi64_explicit(_Atomic(int64_t)*p, mi_memory_order mo) { + (void)(mo); +#if defined(_M_X64) + return *p; +#else + int64_t old = *p; + int64_t x = old; + while ((old = InterlockedCompareExchange64(p, x, old)) != x) { + x = old; + } + return x; +#endif +} +static inline void mi_atomic_storei64_explicit(_Atomic(int64_t)*p, int64_t x, mi_memory_order mo) { + (void)(mo); +#if defined(x_M_IX86) || defined(_M_X64) + *p = x; +#else + InterlockedExchange64(p, x); +#endif +} + +// These are used by the statistics +static inline int64_t mi_atomic_addi64_relaxed(volatile _Atomic(int64_t)*p, int64_t add) { +#ifdef _WIN64 + return (int64_t)mi_atomic_addi((int64_t*)p, add); +#else + int64_t current; + int64_t sum; + do { + current = *p; + sum = current + add; + } while (_InterlockedCompareExchange64(p, sum, current) != current); + return current; +#endif +} +static inline void mi_atomic_void_addi64_relaxed(volatile int64_t* p, const volatile int64_t* padd) { + const int64_t add = *padd; + if (add != 0) { + mi_atomic_addi64_relaxed((volatile _Atomic(int64_t)*)p, add); + } +} + +static inline void mi_atomic_maxi64_relaxed(volatile _Atomic(int64_t)*p, int64_t x) { + int64_t current; + do { + current = *p; + } while (current < x && _InterlockedCompareExchange64(p, x, current) != current); +} + +static inline void mi_atomic_addi64_acq_rel(volatile _Atomic(int64_t*)p, int64_t i) { + mi_atomic_addi64_relaxed(p, i); +} + +static inline bool mi_atomic_casi64_strong_acq_rel(volatile _Atomic(int64_t*)p, int64_t* exp, int64_t des) { + int64_t read = _InterlockedCompareExchange64(p, des, *exp); + if (read == *exp) { + return true; + } + else { + *exp = read; + return false; + } +} + +// The pointer macros cast to `uintptr_t`. +#define mi_atomic_load_ptr_acquire(tp,p) (tp*)mi_atomic_load_acquire((_Atomic(uintptr_t)*)(p)) +#define mi_atomic_load_ptr_relaxed(tp,p) (tp*)mi_atomic_load_relaxed((_Atomic(uintptr_t)*)(p)) +#define mi_atomic_store_ptr_release(tp,p,x) mi_atomic_store_release((_Atomic(uintptr_t)*)(p),(uintptr_t)(x)) +#define mi_atomic_store_ptr_relaxed(tp,p,x) mi_atomic_store_relaxed((_Atomic(uintptr_t)*)(p),(uintptr_t)(x)) +#define mi_atomic_cas_ptr_weak_release(tp,p,exp,des) mi_atomic_cas_weak_release((_Atomic(uintptr_t)*)(p),(uintptr_t*)exp,(uintptr_t)des) +#define mi_atomic_cas_ptr_weak_acq_rel(tp,p,exp,des) mi_atomic_cas_weak_acq_rel((_Atomic(uintptr_t)*)(p),(uintptr_t*)exp,(uintptr_t)des) +#define mi_atomic_cas_ptr_strong_release(tp,p,exp,des) mi_atomic_cas_strong_release((_Atomic(uintptr_t)*)(p),(uintptr_t*)exp,(uintptr_t)des) +#define mi_atomic_cas_ptr_strong_acq_rel(tp,p,exp,des) mi_atomic_cas_strong_acq_rel((_Atomic(uintptr_t)*)(p),(uintptr_t*)exp,(uintptr_t)des) +#define mi_atomic_exchange_ptr_relaxed(tp,p,x) (tp*)mi_atomic_exchange_relaxed((_Atomic(uintptr_t)*)(p),(uintptr_t)x) +#define mi_atomic_exchange_ptr_release(tp,p,x) (tp*)mi_atomic_exchange_release((_Atomic(uintptr_t)*)(p),(uintptr_t)x) +#define mi_atomic_exchange_ptr_acq_rel(tp,p,x) (tp*)mi_atomic_exchange_acq_rel((_Atomic(uintptr_t)*)(p),(uintptr_t)x) + +#define mi_atomic_loadi64_acquire(p) mi_atomic(loadi64_explicit)(p,mi_memory_order(acquire)) +#define mi_atomic_loadi64_relaxed(p) mi_atomic(loadi64_explicit)(p,mi_memory_order(relaxed)) +#define mi_atomic_storei64_release(p,x) mi_atomic(storei64_explicit)(p,x,mi_memory_order(release)) +#define mi_atomic_storei64_relaxed(p,x) mi_atomic(storei64_explicit)(p,x,mi_memory_order(relaxed)) + + +#endif + + +// Atomically add a signed value; returns the previous value. +static inline intptr_t mi_atomic_addi(_Atomic(intptr_t)*p, intptr_t add) { + return (intptr_t)mi_atomic_add_acq_rel((_Atomic(uintptr_t)*)p, (uintptr_t)add); +} + +// Atomically subtract a signed value; returns the previous value. +static inline intptr_t mi_atomic_subi(_Atomic(intptr_t)*p, intptr_t sub) { + return (intptr_t)mi_atomic_addi(p, -sub); +} + + +// ---------------------------------------------------------------------- +// Once and Guard +// ---------------------------------------------------------------------- + +typedef _Atomic(uintptr_t) mi_atomic_once_t; + +// Returns true only on the first invocation +static inline bool mi_atomic_once( mi_atomic_once_t* once ) { + if (mi_atomic_load_relaxed(once) != 0) return false; // quick test + uintptr_t expected = 0; + return mi_atomic_cas_strong_acq_rel(once, &expected, (uintptr_t)1); // try to set to 1 +} + +typedef _Atomic(uintptr_t) mi_atomic_guard_t; + +// Allows only one thread to execute at a time +#define mi_atomic_guard(guard) \ + uintptr_t _mi_guard_expected = 0; \ + for(bool _mi_guard_once = true; \ + _mi_guard_once && mi_atomic_cas_strong_acq_rel(guard,&_mi_guard_expected,(uintptr_t)1); \ + (mi_atomic_store_release(guard,(uintptr_t)0), _mi_guard_once = false) ) + + + +// ---------------------------------------------------------------------- +// Yield +// ---------------------------------------------------------------------- + +#if defined(__cplusplus) +#include +static inline void mi_atomic_yield(void) { + std::this_thread::yield(); +} +#elif defined(_WIN32) +static inline void mi_atomic_yield(void) { + YieldProcessor(); +} +#elif defined(__SSE2__) +#include +static inline void mi_atomic_yield(void) { + _mm_pause(); +} +#elif (defined(__GNUC__) || defined(__clang__)) && \ + (defined(__x86_64__) || defined(__i386__) || \ + defined(__aarch64__) || defined(__arm__) || \ + defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) || defined(__POWERPC__)) +#if defined(__x86_64__) || defined(__i386__) +static inline void mi_atomic_yield(void) { + __asm__ volatile ("pause" ::: "memory"); +} +#elif defined(__aarch64__) +static inline void mi_atomic_yield(void) { + __asm__ volatile("wfe"); +} +#elif defined(__arm__) +#if __ARM_ARCH >= 7 +static inline void mi_atomic_yield(void) { + __asm__ volatile("yield" ::: "memory"); +} +#else +static inline void mi_atomic_yield(void) { + __asm__ volatile ("nop" ::: "memory"); +} +#endif +#elif defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) || defined(__POWERPC__) +#ifdef __APPLE__ +static inline void mi_atomic_yield(void) { + __asm__ volatile ("or r27,r27,r27" ::: "memory"); +} +#else +static inline void mi_atomic_yield(void) { + __asm__ __volatile__ ("or 27,27,27" ::: "memory"); +} +#endif +#endif +#elif defined(__sun) +// Fallback for other archs +#include +static inline void mi_atomic_yield(void) { + smt_pause(); +} +#elif defined(__wasi__) +#include +static inline void mi_atomic_yield(void) { + sched_yield(); +} +#else +#include +static inline void mi_atomic_yield(void) { + sleep(0); +} +#endif + + +// ---------------------------------------------------------------------- +// Locks +// These do not have to be recursive and should be light-weight +// in-process only locks. Only used for reserving arena's and to +// maintain the abandoned list. +// ---------------------------------------------------------------------- +#if _MSC_VER +#pragma warning(disable:26110) // unlock with holding lock +#endif + +#define mi_lock(lock) for(bool _go = (mi_lock_acquire(lock),true); _go; (mi_lock_release(lock), _go=false) ) + +#if defined(_WIN32) + +#if 1 +#define mi_lock_t SRWLOCK // slim reader-writer lock + +static inline bool mi_lock_try_acquire(mi_lock_t* lock) { + return TryAcquireSRWLockExclusive(lock); +} +static inline void mi_lock_acquire(mi_lock_t* lock) { + AcquireSRWLockExclusive(lock); +} +static inline void mi_lock_release(mi_lock_t* lock) { + ReleaseSRWLockExclusive(lock); +} +static inline void mi_lock_init(mi_lock_t* lock) { + InitializeSRWLock(lock); +} +static inline void mi_lock_done(mi_lock_t* lock) { + (void)(lock); +} + +#else +#define mi_lock_t CRITICAL_SECTION + +static inline bool mi_lock_try_acquire(mi_lock_t* lock) { + return TryEnterCriticalSection(lock); +} +static inline void mi_lock_acquire(mi_lock_t* lock) { + EnterCriticalSection(lock); +} +static inline void mi_lock_release(mi_lock_t* lock) { + LeaveCriticalSection(lock); +} +static inline void mi_lock_init(mi_lock_t* lock) { + InitializeCriticalSection(lock); +} +static inline void mi_lock_done(mi_lock_t* lock) { + DeleteCriticalSection(lock); +} + +#endif + +#elif defined(MI_USE_PTHREADS) + +void _mi_error_message(int err, const char* fmt, ...); + +#define mi_lock_t pthread_mutex_t + +static inline bool mi_lock_try_acquire(mi_lock_t* lock) { + return (pthread_mutex_trylock(lock) == 0); +} +static inline void mi_lock_acquire(mi_lock_t* lock) { + const int err = pthread_mutex_lock(lock); + if (err != 0) { + _mi_error_message(err, "internal error: lock cannot be acquired\n"); + } +} +static inline void mi_lock_release(mi_lock_t* lock) { + pthread_mutex_unlock(lock); +} +static inline void mi_lock_init(mi_lock_t* lock) { + pthread_mutex_init(lock, NULL); +} +static inline void mi_lock_done(mi_lock_t* lock) { + pthread_mutex_destroy(lock); +} + +#elif defined(__cplusplus) + +#include +#define mi_lock_t std::mutex + +static inline bool mi_lock_try_acquire(mi_lock_t* lock) { + return lock->try_lock(); +} +static inline void mi_lock_acquire(mi_lock_t* lock) { + lock->lock(); +} +static inline void mi_lock_release(mi_lock_t* lock) { + lock->unlock(); +} +static inline void mi_lock_init(mi_lock_t* lock) { + (void)(lock); +} +static inline void mi_lock_done(mi_lock_t* lock) { + (void)(lock); +} + +#else + +// fall back to poor man's locks. +// this should only be the case in a single-threaded environment (like __wasi__) + +#define mi_lock_t _Atomic(uintptr_t) + +static inline bool mi_lock_try_acquire(mi_lock_t* lock) { + uintptr_t expected = 0; + return mi_atomic_cas_strong_acq_rel(lock, &expected, (uintptr_t)1); +} +static inline void mi_lock_acquire(mi_lock_t* lock) { + for (int i = 0; i < 1000; i++) { // for at most 1000 tries? + if (mi_lock_try_acquire(lock)) return; + mi_atomic_yield(); + } +} +static inline void mi_lock_release(mi_lock_t* lock) { + mi_atomic_store_release(lock, (uintptr_t)0); +} +static inline void mi_lock_init(mi_lock_t* lock) { + mi_lock_release(lock); +} +static inline void mi_lock_done(mi_lock_t* lock) { + (void)(lock); +} + +#endif + + +#endif // __MIMALLOC_ATOMIC_H diff --git a/compat/mimalloc/mimalloc/internal.h b/compat/mimalloc/mimalloc/internal.h new file mode 100644 index 00000000000000..ca5be9304a8bae --- /dev/null +++ b/compat/mimalloc/mimalloc/internal.h @@ -0,0 +1,1153 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_INTERNAL_H +#define MIMALLOC_INTERNAL_H + +// -------------------------------------------------------------------------- +// This file contains the internal API's of mimalloc and various utility +// functions and macros. +// -------------------------------------------------------------------------- + +#include "types.h" +#include "track.h" + + +// -------------------------------------------------------------------------- +// Compiler defines +// -------------------------------------------------------------------------- + +#if (MI_DEBUG>0) +#define mi_trace_message(...) _mi_trace_message(__VA_ARGS__) +#else +#define mi_trace_message(...) +#endif + +#define mi_decl_cache_align mi_decl_align(64) + +#if defined(_MSC_VER) +#pragma warning(disable:4127) // suppress constant conditional warning (due to MI_SECURE paths) +#pragma warning(disable:26812) // unscoped enum warning +#define mi_decl_noinline __declspec(noinline) +#define mi_decl_thread __declspec(thread) +#define mi_decl_align(a) __declspec(align(a)) +#define mi_decl_noreturn __declspec(noreturn) +#define mi_decl_weak +#define mi_decl_hidden +#define mi_decl_cold +#elif (defined(__GNUC__) && (__GNUC__ >= 3)) || defined(__clang__) // includes clang and icc +#define mi_decl_noinline __attribute__((noinline)) +#define mi_decl_thread __thread +#define mi_decl_align(a) __attribute__((aligned(a))) +#define mi_decl_noreturn __attribute__((noreturn)) +#define mi_decl_weak __attribute__((weak)) +#define mi_decl_hidden __attribute__((visibility("hidden"))) +#if (__GNUC__ >= 4) || defined(__clang__) +#define mi_decl_cold __attribute__((cold)) +#else +#define mi_decl_cold +#endif +#elif __cplusplus >= 201103L // c++11 +#define mi_decl_noinline +#define mi_decl_thread thread_local +#define mi_decl_align(a) alignas(a) +#define mi_decl_noreturn [[noreturn]] +#define mi_decl_weak +#define mi_decl_hidden +#define mi_decl_cold +#else +#define mi_decl_noinline +#define mi_decl_thread __thread // hope for the best :-) +#define mi_decl_align(a) +#define mi_decl_noreturn +#define mi_decl_weak +#define mi_decl_hidden +#define mi_decl_cold +#endif + +#if defined(__GNUC__) || defined(__clang__) +#define mi_unlikely(x) (__builtin_expect(!!(x),false)) +#define mi_likely(x) (__builtin_expect(!!(x),true)) +#elif (defined(__cplusplus) && (__cplusplus >= 202002L)) || (defined(_MSVC_LANG) && _MSVC_LANG >= 202002L) +#define mi_unlikely(x) (x) [[unlikely]] +#define mi_likely(x) (x) [[likely]] +#else +#define mi_unlikely(x) (x) +#define mi_likely(x) (x) +#endif + +#ifndef __has_builtin +#define __has_builtin(x) 0 +#endif + +#if defined(__cplusplus) +#define mi_decl_externc extern "C" +#else +#define mi_decl_externc +#endif + +#if defined(__EMSCRIPTEN__) && !defined(__wasi__) +#define __wasi__ +#endif + + +// -------------------------------------------------------------------------- +// Internal functions +// -------------------------------------------------------------------------- + +// "libc.c" +#include +int _mi_vsnprintf(char* buf, size_t bufsize, const char* fmt, va_list args); +int _mi_snprintf(char* buf, size_t buflen, const char* fmt, ...); +char _mi_toupper(char c); +int _mi_strnicmp(const char* s, const char* t, size_t n); +void _mi_strlcpy(char* dest, const char* src, size_t dest_size); +void _mi_strlcat(char* dest, const char* src, size_t dest_size); +size_t _mi_strlen(const char* s); +size_t _mi_strnlen(const char* s, size_t max_len); +bool _mi_getenv(const char* name, char* result, size_t result_size); + +// "options.c" +void _mi_fputs(mi_output_fun* out, void* arg, const char* prefix, const char* message); +void _mi_fprintf(mi_output_fun* out, void* arg, const char* fmt, ...); +void _mi_message(const char* fmt, ...); +void _mi_warning_message(const char* fmt, ...); +void _mi_verbose_message(const char* fmt, ...); +void _mi_trace_message(const char* fmt, ...); +void _mi_options_init(void); +long _mi_option_get_fast(mi_option_t option); +void _mi_error_message(int err, const char* fmt, ...); + +// random.c +void _mi_random_init(mi_random_ctx_t* ctx); +void _mi_random_init_weak(mi_random_ctx_t* ctx); +void _mi_random_reinit_if_weak(mi_random_ctx_t * ctx); +void _mi_random_split(mi_random_ctx_t* ctx, mi_random_ctx_t* new_ctx); +uintptr_t _mi_random_next(mi_random_ctx_t* ctx); +uintptr_t _mi_heap_random_next(mi_heap_t* heap); +uintptr_t _mi_os_random_weak(uintptr_t extra_seed); +static inline uintptr_t _mi_random_shuffle(uintptr_t x); + +// init.c +extern mi_decl_hidden mi_decl_cache_align mi_stats_t _mi_stats_main; +extern mi_decl_hidden mi_decl_cache_align const mi_page_t _mi_page_empty; +void _mi_auto_process_init(void); +void mi_cdecl _mi_auto_process_done(void) mi_attr_noexcept; +bool _mi_is_redirected(void); +bool _mi_allocator_init(const char** message); +void _mi_allocator_done(void); +bool _mi_is_main_thread(void); +size_t _mi_current_thread_count(void); +bool _mi_preloading(void); // true while the C runtime is not initialized yet +void _mi_thread_done(mi_heap_t* heap); +void _mi_thread_data_collect(void); +void _mi_tld_init(mi_tld_t* tld, mi_heap_t* bheap); +mi_threadid_t _mi_thread_id(void) mi_attr_noexcept; +mi_heap_t* _mi_heap_main_get(void); // statically allocated main backing heap +mi_subproc_t* _mi_subproc_from_id(mi_subproc_id_t subproc_id); +void _mi_heap_guarded_init(mi_heap_t* heap); + +// os.c +void _mi_os_init(void); // called from process init +void* _mi_os_alloc(size_t size, mi_memid_t* memid); +void* _mi_os_zalloc(size_t size, mi_memid_t* memid); +void _mi_os_free(void* p, size_t size, mi_memid_t memid); +void _mi_os_free_ex(void* p, size_t size, bool still_committed, mi_memid_t memid); + +size_t _mi_os_page_size(void); +size_t _mi_os_good_alloc_size(size_t size); +bool _mi_os_has_overcommit(void); +bool _mi_os_has_virtual_reserve(void); + +bool _mi_os_reset(void* addr, size_t size); +bool _mi_os_decommit(void* addr, size_t size); +bool _mi_os_unprotect(void* addr, size_t size); +bool _mi_os_purge(void* p, size_t size); +bool _mi_os_purge_ex(void* p, size_t size, bool allow_reset, size_t stat_size); +void _mi_os_reuse(void* p, size_t size); +mi_decl_nodiscard bool _mi_os_commit(void* p, size_t size, bool* is_zero); +mi_decl_nodiscard bool _mi_os_commit_ex(void* addr, size_t size, bool* is_zero, size_t stat_size); +bool _mi_os_protect(void* addr, size_t size); + +void* _mi_os_alloc_aligned(size_t size, size_t alignment, bool commit, bool allow_large, mi_memid_t* memid); +void* _mi_os_alloc_aligned_at_offset(size_t size, size_t alignment, size_t align_offset, bool commit, bool allow_large, mi_memid_t* memid); + +void* _mi_os_get_aligned_hint(size_t try_alignment, size_t size); +bool _mi_os_use_large_page(size_t size, size_t alignment); +size_t _mi_os_large_page_size(void); +void* _mi_os_alloc_huge_os_pages(size_t pages, int numa_node, mi_msecs_t max_secs, size_t* pages_reserved, size_t* psize, mi_memid_t* memid); + +int _mi_os_numa_node_count(void); +int _mi_os_numa_node(void); + +// arena.c +mi_arena_id_t _mi_arena_id_none(void); +void _mi_arena_free(void* p, size_t size, size_t still_committed_size, mi_memid_t memid); +void* _mi_arena_alloc(size_t size, bool commit, bool allow_large, mi_arena_id_t req_arena_id, mi_memid_t* memid); +void* _mi_arena_alloc_aligned(size_t size, size_t alignment, size_t align_offset, bool commit, bool allow_large, mi_arena_id_t req_arena_id, mi_memid_t* memid); +bool _mi_arena_memid_is_suitable(mi_memid_t memid, mi_arena_id_t request_arena_id); +bool _mi_arena_contains(const void* p); +void _mi_arenas_collect(bool force_purge); +void _mi_arena_unsafe_destroy_all(void); + +bool _mi_arena_segment_clear_abandoned(mi_segment_t* segment); +void _mi_arena_segment_mark_abandoned(mi_segment_t* segment); + +void* _mi_arena_meta_zalloc(size_t size, mi_memid_t* memid); +void _mi_arena_meta_free(void* p, mi_memid_t memid, size_t size); + +typedef struct mi_arena_field_cursor_s { // abstract struct + size_t os_list_count; // max entries to visit in the OS abandoned list + size_t start; // start arena idx (may need to be wrapped) + size_t end; // end arena idx (exclusive, may need to be wrapped) + size_t bitmap_idx; // current bit idx for an arena + mi_subproc_t* subproc; // only visit blocks in this sub-process + bool visit_all; // ensure all abandoned blocks are seen (blocking) + bool hold_visit_lock; // if the subproc->abandoned_os_visit_lock is held +} mi_arena_field_cursor_t; +void _mi_arena_field_cursor_init(mi_heap_t* heap, mi_subproc_t* subproc, bool visit_all, mi_arena_field_cursor_t* current); +mi_segment_t* _mi_arena_segment_clear_abandoned_next(mi_arena_field_cursor_t* previous); +void _mi_arena_field_cursor_done(mi_arena_field_cursor_t* current); + +// "segment-map.c" +void _mi_segment_map_allocated_at(const mi_segment_t* segment); +void _mi_segment_map_freed_at(const mi_segment_t* segment); +void _mi_segment_map_unsafe_destroy(void); + +// "segment.c" +mi_page_t* _mi_segment_page_alloc(mi_heap_t* heap, size_t block_size, size_t page_alignment, mi_segments_tld_t* tld); +void _mi_segment_page_free(mi_page_t* page, bool force, mi_segments_tld_t* tld); +void _mi_segment_page_abandon(mi_page_t* page, mi_segments_tld_t* tld); +bool _mi_segment_try_reclaim_abandoned( mi_heap_t* heap, bool try_all, mi_segments_tld_t* tld); +void _mi_segment_collect(mi_segment_t* segment, bool force); + +#if MI_HUGE_PAGE_ABANDON +void _mi_segment_huge_page_free(mi_segment_t* segment, mi_page_t* page, mi_block_t* block); +#else +void _mi_segment_huge_page_reset(mi_segment_t* segment, mi_page_t* page, mi_block_t* block); +#endif + +uint8_t* _mi_segment_page_start(const mi_segment_t* segment, const mi_page_t* page, size_t* page_size); // page start for any page +void _mi_abandoned_reclaim_all(mi_heap_t* heap, mi_segments_tld_t* tld); +void _mi_abandoned_collect(mi_heap_t* heap, bool force, mi_segments_tld_t* tld); +bool _mi_segment_attempt_reclaim(mi_heap_t* heap, mi_segment_t* segment); +bool _mi_segment_visit_blocks(mi_segment_t* segment, int heap_tag, bool visit_blocks, mi_block_visit_fun* visitor, void* arg); + +// "page.c" +void* _mi_malloc_generic(mi_heap_t* heap, size_t size, bool zero, size_t huge_alignment) mi_attr_noexcept mi_attr_malloc; + +void _mi_page_retire(mi_page_t* page) mi_attr_noexcept; // free the page if there are no other pages with many free blocks +void _mi_page_unfull(mi_page_t* page); +void _mi_page_free(mi_page_t* page, mi_page_queue_t* pq, bool force); // free the page +void _mi_page_abandon(mi_page_t* page, mi_page_queue_t* pq); // abandon the page, to be picked up by another thread... +void _mi_page_force_abandon(mi_page_t* page); + +void _mi_heap_delayed_free_all(mi_heap_t* heap); +bool _mi_heap_delayed_free_partial(mi_heap_t* heap); +void _mi_heap_collect_retired(mi_heap_t* heap, bool force); + +void _mi_page_use_delayed_free(mi_page_t* page, mi_delayed_t delay, bool override_never); +bool _mi_page_try_use_delayed_free(mi_page_t* page, mi_delayed_t delay, bool override_never); +size_t _mi_page_queue_append(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_queue_t* append); +void _mi_deferred_free(mi_heap_t* heap, bool force); + +void _mi_page_free_collect(mi_page_t* page,bool force); +void _mi_page_reclaim(mi_heap_t* heap, mi_page_t* page); // callback from segments + +size_t _mi_page_bin(const mi_page_t* page); // for stats +size_t _mi_bin_size(size_t bin); // for stats +size_t _mi_bin(size_t size); // for stats + +// "heap.c" +void _mi_heap_init(mi_heap_t* heap, mi_tld_t* tld, mi_arena_id_t arena_id, bool noreclaim, uint8_t tag); +void _mi_heap_destroy_pages(mi_heap_t* heap); +void _mi_heap_collect_abandon(mi_heap_t* heap); +void _mi_heap_set_default_direct(mi_heap_t* heap); +bool _mi_heap_memid_is_suitable(mi_heap_t* heap, mi_memid_t memid); +void _mi_heap_unsafe_destroy_all(mi_heap_t* heap); +mi_heap_t* _mi_heap_by_tag(mi_heap_t* heap, uint8_t tag); +void _mi_heap_area_init(mi_heap_area_t* area, mi_page_t* page); +bool _mi_heap_area_visit_blocks(const mi_heap_area_t* area, mi_page_t* page, mi_block_visit_fun* visitor, void* arg); + +// "stats.c" +void _mi_stats_done(mi_stats_t* stats); +void _mi_stats_merge_thread(mi_tld_t* tld); +mi_msecs_t _mi_clock_now(void); +mi_msecs_t _mi_clock_end(mi_msecs_t start); +mi_msecs_t _mi_clock_start(void); + +// "alloc.c" +void* _mi_page_malloc_zero(mi_heap_t* heap, mi_page_t* page, size_t size, bool zero) mi_attr_noexcept; // called from `_mi_malloc_generic` +void* _mi_page_malloc(mi_heap_t* heap, mi_page_t* page, size_t size) mi_attr_noexcept; // called from `_mi_heap_malloc_aligned` +void* _mi_page_malloc_zeroed(mi_heap_t* heap, mi_page_t* page, size_t size) mi_attr_noexcept; // called from `_mi_heap_malloc_aligned` +void* _mi_heap_malloc_zero(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept; +void* _mi_heap_malloc_zero_ex(mi_heap_t* heap, size_t size, bool zero, size_t huge_alignment) mi_attr_noexcept; // called from `_mi_heap_malloc_aligned` +void* _mi_heap_realloc_zero(mi_heap_t* heap, void* p, size_t newsize, bool zero) mi_attr_noexcept; +mi_block_t* _mi_page_ptr_unalign(const mi_page_t* page, const void* p); +bool _mi_free_delayed_block(mi_block_t* block); +void _mi_free_generic(mi_segment_t* segment, mi_page_t* page, bool is_local, void* p) mi_attr_noexcept; // for runtime integration +void _mi_padding_shrink(const mi_page_t* page, const mi_block_t* block, const size_t min_size); + +#if MI_DEBUG>1 +bool _mi_page_is_valid(mi_page_t* page); +#endif + + +/* ----------------------------------------------------------- + Error codes passed to `_mi_fatal_error` + All are recoverable but EFAULT is a serious error and aborts by default in secure mode. + For portability define undefined error codes using common Unix codes: + +----------------------------------------------------------- */ +#include +#ifndef EAGAIN // double free +#define EAGAIN (11) +#endif +#ifndef ENOMEM // out of memory +#define ENOMEM (12) +#endif +#ifndef EFAULT // corrupted free-list or meta-data +#define EFAULT (14) +#endif +#ifndef EINVAL // trying to free an invalid pointer +#define EINVAL (22) +#endif +#ifndef EOVERFLOW // count*size overflow +#define EOVERFLOW (75) +#endif + + +// ------------------------------------------------------ +// Assertions +// ------------------------------------------------------ + +#if (MI_DEBUG) +// use our own assertion to print without memory allocation +mi_decl_noreturn mi_decl_cold void _mi_assert_fail(const char* assertion, const char* fname, unsigned int line, const char* func) mi_attr_noexcept; +#define mi_assert(expr) ((expr) ? (void)0 : _mi_assert_fail(#expr,__FILE__,__LINE__,__func__)) +#else +#define mi_assert(x) +#endif + +#if (MI_DEBUG>1) +#define mi_assert_internal mi_assert +#else +#define mi_assert_internal(x) +#endif + +#if (MI_DEBUG>2) +#define mi_assert_expensive mi_assert +#else +#define mi_assert_expensive(x) +#endif + + + +/* ----------------------------------------------------------- + Inlined definitions +----------------------------------------------------------- */ +#define MI_UNUSED(x) (void)(x) +#if (MI_DEBUG>0) +#define MI_UNUSED_RELEASE(x) +#else +#define MI_UNUSED_RELEASE(x) MI_UNUSED(x) +#endif + +#define MI_INIT4(x) x(),x(),x(),x() +#define MI_INIT8(x) MI_INIT4(x),MI_INIT4(x) +#define MI_INIT16(x) MI_INIT8(x),MI_INIT8(x) +#define MI_INIT32(x) MI_INIT16(x),MI_INIT16(x) +#define MI_INIT64(x) MI_INIT32(x),MI_INIT32(x) +#define MI_INIT128(x) MI_INIT64(x),MI_INIT64(x) +#define MI_INIT256(x) MI_INIT128(x),MI_INIT128(x) +#define MI_INIT74(x) MI_INIT64(x),MI_INIT8(x),x(),x() + +#include +// initialize a local variable to zero; use memset as compilers optimize constant sized memset's +#define _mi_memzero_var(x) memset(&x,0,sizeof(x)) + +// Is `x` a power of two? (0 is considered a power of two) +static inline bool _mi_is_power_of_two(uintptr_t x) { + return ((x & (x - 1)) == 0); +} + +// Is a pointer aligned? +static inline bool _mi_is_aligned(void* p, size_t alignment) { + mi_assert_internal(alignment != 0); + return (((uintptr_t)p % alignment) == 0); +} + +// Align upwards +static inline uintptr_t _mi_align_up(uintptr_t sz, size_t alignment) { + mi_assert_internal(alignment != 0); + uintptr_t mask = alignment - 1; + if ((alignment & mask) == 0) { // power of two? + return ((sz + mask) & ~mask); + } + else { + return (((sz + mask)/alignment)*alignment); + } +} + +// Align downwards +static inline uintptr_t _mi_align_down(uintptr_t sz, size_t alignment) { + mi_assert_internal(alignment != 0); + uintptr_t mask = alignment - 1; + if ((alignment & mask) == 0) { // power of two? + return (sz & ~mask); + } + else { + return ((sz / alignment) * alignment); + } +} + +// Align a pointer upwards +static inline void* mi_align_up_ptr(void* p, size_t alignment) { + return (void*)_mi_align_up((uintptr_t)p, alignment); +} + +// Align a pointer downwards +static inline void* mi_align_down_ptr(void* p, size_t alignment) { + return (void*)_mi_align_down((uintptr_t)p, alignment); +} + + +// Divide upwards: `s <= _mi_divide_up(s,d)*d < s+d`. +static inline uintptr_t _mi_divide_up(uintptr_t size, size_t divider) { + mi_assert_internal(divider != 0); + return (divider == 0 ? size : ((size + divider - 1) / divider)); +} + + +// clamp an integer +static inline size_t _mi_clamp(size_t sz, size_t min, size_t max) { + if (sz < min) return min; + else if (sz > max) return max; + else return sz; +} + +// Is memory zero initialized? +static inline bool mi_mem_is_zero(const void* p, size_t size) { + for (size_t i = 0; i < size; i++) { + if (((uint8_t*)p)[i] != 0) return false; + } + return true; +} + + +// Align a byte size to a size in _machine words_, +// i.e. byte size == `wsize*sizeof(void*)`. +static inline size_t _mi_wsize_from_size(size_t size) { + mi_assert_internal(size <= SIZE_MAX - sizeof(uintptr_t)); + return (size + sizeof(uintptr_t) - 1) / sizeof(uintptr_t); +} + +// Overflow detecting multiply +#if __has_builtin(__builtin_umul_overflow) || (defined(__GNUC__) && (__GNUC__ >= 5)) +#include // UINT_MAX, ULONG_MAX +#if defined(_CLOCK_T) // for Illumos +#undef _CLOCK_T +#endif +static inline bool mi_mul_overflow(size_t count, size_t size, size_t* total) { + #if (SIZE_MAX == ULONG_MAX) + return __builtin_umull_overflow(count, size, (unsigned long *)total); + #elif (SIZE_MAX == UINT_MAX) + return __builtin_umul_overflow(count, size, (unsigned int *)total); + #else + return __builtin_umulll_overflow(count, size, (unsigned long long *)total); + #endif +} +#else /* __builtin_umul_overflow is unavailable */ +static inline bool mi_mul_overflow(size_t count, size_t size, size_t* total) { + #define MI_MUL_COULD_OVERFLOW ((size_t)1 << (4*sizeof(size_t))) // sqrt(SIZE_MAX) + *total = count * size; + // note: gcc/clang optimize this to directly check the overflow flag + return ((size >= MI_MUL_COULD_OVERFLOW || count >= MI_MUL_COULD_OVERFLOW) && size > 0 && (SIZE_MAX / size) < count); +} +#endif + +// Safe multiply `count*size` into `total`; return `true` on overflow. +static inline bool mi_count_size_overflow(size_t count, size_t size, size_t* total) { + if (count==1) { // quick check for the case where count is one (common for C++ allocators) + *total = size; + return false; + } + else if mi_unlikely(mi_mul_overflow(count, size, total)) { + #if MI_DEBUG > 0 + _mi_error_message(EOVERFLOW, "allocation request is too large (%zu * %zu bytes)\n", count, size); + #endif + *total = SIZE_MAX; + return true; + } + else return false; +} + + +/*---------------------------------------------------------------------------------------- + Heap functions +------------------------------------------------------------------------------------------- */ + +extern mi_decl_hidden const mi_heap_t _mi_heap_empty; // read-only empty heap, initial value of the thread local default heap + +static inline bool mi_heap_is_backing(const mi_heap_t* heap) { + return (heap->tld->heap_backing == heap); +} + +static inline bool mi_heap_is_initialized(mi_heap_t* heap) { + mi_assert_internal(heap != NULL); + return (heap != NULL && heap != &_mi_heap_empty); +} + +static inline uintptr_t _mi_ptr_cookie(const void* p) { + extern mi_decl_hidden mi_heap_t _mi_heap_main; + mi_assert_internal(_mi_heap_main.cookie != 0); + return ((uintptr_t)p ^ _mi_heap_main.cookie); +} + +/* ----------------------------------------------------------- + Pages +----------------------------------------------------------- */ + +static inline mi_page_t* _mi_heap_get_free_small_page(mi_heap_t* heap, size_t size) { + mi_assert_internal(size <= (MI_SMALL_SIZE_MAX + MI_PADDING_SIZE)); + const size_t idx = _mi_wsize_from_size(size); + mi_assert_internal(idx < MI_PAGES_DIRECT); + return heap->pages_free_direct[idx]; +} + +// Segment that contains the pointer +// Large aligned blocks may be aligned at N*MI_SEGMENT_SIZE (inside a huge segment > MI_SEGMENT_SIZE), +// and we need align "down" to the segment info which is `MI_SEGMENT_SIZE` bytes before it; +// therefore we align one byte before `p`. +// We check for NULL afterwards on 64-bit systems to improve codegen for `mi_free`. +static inline mi_segment_t* _mi_ptr_segment(const void* p) { + mi_segment_t* const segment = (mi_segment_t*)(((uintptr_t)p - 1) & ~MI_SEGMENT_MASK); + #if MI_INTPTR_SIZE <= 4 + return (p==NULL ? NULL : segment); + #else + return ((intptr_t)segment <= 0 ? NULL : segment); + #endif +} + +static inline mi_page_t* mi_slice_to_page(mi_slice_t* s) { + mi_assert_internal(s->slice_offset== 0 && s->slice_count > 0); + return (mi_page_t*)(s); +} + +static inline mi_slice_t* mi_page_to_slice(mi_page_t* p) { + mi_assert_internal(p->slice_offset== 0 && p->slice_count > 0); + return (mi_slice_t*)(p); +} + +// Segment belonging to a page +static inline mi_segment_t* _mi_page_segment(const mi_page_t* page) { + mi_assert_internal(page!=NULL); + mi_segment_t* segment = _mi_ptr_segment(page); + mi_assert_internal(segment == NULL || ((mi_slice_t*)page >= segment->slices && (mi_slice_t*)page < segment->slices + segment->slice_entries)); + return segment; +} + +static inline mi_slice_t* mi_slice_first(const mi_slice_t* slice) { + mi_slice_t* start = (mi_slice_t*)((uint8_t*)slice - slice->slice_offset); + mi_assert_internal(start >= _mi_ptr_segment(slice)->slices); + mi_assert_internal(start->slice_offset == 0); + mi_assert_internal(start + start->slice_count > slice); + return start; +} + +// Get the page containing the pointer (performance critical as it is called in mi_free) +static inline mi_page_t* _mi_segment_page_of(const mi_segment_t* segment, const void* p) { + mi_assert_internal(p > (void*)segment); + ptrdiff_t diff = (uint8_t*)p - (uint8_t*)segment; + mi_assert_internal(diff > 0 && diff <= (ptrdiff_t)MI_SEGMENT_SIZE); + size_t idx = (size_t)diff >> MI_SEGMENT_SLICE_SHIFT; + mi_assert_internal(idx <= segment->slice_entries); + mi_slice_t* slice0 = (mi_slice_t*)&segment->slices[idx]; + mi_slice_t* slice = mi_slice_first(slice0); // adjust to the block that holds the page data + mi_assert_internal(slice->slice_offset == 0); + mi_assert_internal(slice >= segment->slices && slice < segment->slices + segment->slice_entries); + return mi_slice_to_page(slice); +} + +// Quick page start for initialized pages +static inline uint8_t* mi_page_start(const mi_page_t* page) { + mi_assert_internal(page->page_start != NULL); + mi_assert_expensive(_mi_segment_page_start(_mi_page_segment(page),page,NULL) == page->page_start); + return page->page_start; +} + +// Get the page containing the pointer +static inline mi_page_t* _mi_ptr_page(void* p) { + mi_assert_internal(p!=NULL); + return _mi_segment_page_of(_mi_ptr_segment(p), p); +} + +// Get the block size of a page (special case for huge objects) +static inline size_t mi_page_block_size(const mi_page_t* page) { + mi_assert_internal(page->block_size > 0); + return page->block_size; +} + +static inline bool mi_page_is_huge(const mi_page_t* page) { + mi_assert_internal((page->is_huge && _mi_page_segment(page)->kind == MI_SEGMENT_HUGE) || + (!page->is_huge && _mi_page_segment(page)->kind != MI_SEGMENT_HUGE)); + return page->is_huge; +} + +// Get the usable block size of a page without fixed padding. +// This may still include internal padding due to alignment and rounding up size classes. +static inline size_t mi_page_usable_block_size(const mi_page_t* page) { + return mi_page_block_size(page) - MI_PADDING_SIZE; +} + +// size of a segment +static inline size_t mi_segment_size(mi_segment_t* segment) { + return segment->segment_slices * MI_SEGMENT_SLICE_SIZE; +} + +static inline uint8_t* mi_segment_end(mi_segment_t* segment) { + return (uint8_t*)segment + mi_segment_size(segment); +} + +// Thread free access +static inline mi_block_t* mi_page_thread_free(const mi_page_t* page) { + return (mi_block_t*)(mi_atomic_load_relaxed(&((mi_page_t*)page)->xthread_free) & ~3); +} + +static inline mi_delayed_t mi_page_thread_free_flag(const mi_page_t* page) { + return (mi_delayed_t)(mi_atomic_load_relaxed(&((mi_page_t*)page)->xthread_free) & 3); +} + +// Heap access +static inline mi_heap_t* mi_page_heap(const mi_page_t* page) { + return (mi_heap_t*)(mi_atomic_load_relaxed(&((mi_page_t*)page)->xheap)); +} + +static inline void mi_page_set_heap(mi_page_t* page, mi_heap_t* heap) { + mi_assert_internal(mi_page_thread_free_flag(page) != MI_DELAYED_FREEING); + mi_atomic_store_release(&page->xheap,(uintptr_t)heap); + if (heap != NULL) { page->heap_tag = heap->tag; } +} + +// Thread free flag helpers +static inline mi_block_t* mi_tf_block(mi_thread_free_t tf) { + return (mi_block_t*)(tf & ~0x03); +} +static inline mi_delayed_t mi_tf_delayed(mi_thread_free_t tf) { + return (mi_delayed_t)(tf & 0x03); +} +static inline mi_thread_free_t mi_tf_make(mi_block_t* block, mi_delayed_t delayed) { + return (mi_thread_free_t)((uintptr_t)block | (uintptr_t)delayed); +} +static inline mi_thread_free_t mi_tf_set_delayed(mi_thread_free_t tf, mi_delayed_t delayed) { + return mi_tf_make(mi_tf_block(tf),delayed); +} +static inline mi_thread_free_t mi_tf_set_block(mi_thread_free_t tf, mi_block_t* block) { + return mi_tf_make(block, mi_tf_delayed(tf)); +} + +// are all blocks in a page freed? +// note: needs up-to-date used count, (as the `xthread_free` list may not be empty). see `_mi_page_collect_free`. +static inline bool mi_page_all_free(const mi_page_t* page) { + mi_assert_internal(page != NULL); + return (page->used == 0); +} + +// are there any available blocks? +static inline bool mi_page_has_any_available(const mi_page_t* page) { + mi_assert_internal(page != NULL && page->reserved > 0); + return (page->used < page->reserved || (mi_page_thread_free(page) != NULL)); +} + +// are there immediately available blocks, i.e. blocks available on the free list. +static inline bool mi_page_immediate_available(const mi_page_t* page) { + mi_assert_internal(page != NULL); + return (page->free != NULL); +} + +// is more than 7/8th of a page in use? +static inline bool mi_page_is_mostly_used(const mi_page_t* page) { + if (page==NULL) return true; + uint16_t frac = page->reserved / 8U; + return (page->reserved - page->used <= frac); +} + +static inline mi_page_queue_t* mi_page_queue(const mi_heap_t* heap, size_t size) { + return &((mi_heap_t*)heap)->pages[_mi_bin(size)]; +} + + + +//----------------------------------------------------------- +// Page flags +//----------------------------------------------------------- +static inline bool mi_page_is_in_full(const mi_page_t* page) { + return page->flags.x.in_full; +} + +static inline void mi_page_set_in_full(mi_page_t* page, bool in_full) { + page->flags.x.in_full = in_full; +} + +static inline bool mi_page_has_aligned(const mi_page_t* page) { + return page->flags.x.has_aligned; +} + +static inline void mi_page_set_has_aligned(mi_page_t* page, bool has_aligned) { + page->flags.x.has_aligned = has_aligned; +} + +/* ------------------------------------------------------------------- + Guarded objects +------------------------------------------------------------------- */ +#if MI_GUARDED +static inline bool mi_block_ptr_is_guarded(const mi_block_t* block, const void* p) { + const ptrdiff_t offset = (uint8_t*)p - (uint8_t*)block; + return (offset >= (ptrdiff_t)(sizeof(mi_block_t)) && block->next == MI_BLOCK_TAG_GUARDED); +} + +static inline bool mi_heap_malloc_use_guarded(mi_heap_t* heap, size_t size) { + // this code is written to result in fast assembly as it is on the hot path for allocation + const size_t count = heap->guarded_sample_count - 1; // if the rate was 0, this will underflow and count for a long time.. + if mi_likely(count != 0) { + // no sample + heap->guarded_sample_count = count; + return false; + } + else if (size >= heap->guarded_size_min && size <= heap->guarded_size_max) { + // use guarded allocation + heap->guarded_sample_count = heap->guarded_sample_rate; // reset + return (heap->guarded_sample_rate != 0); + } + else { + // failed size criteria, rewind count (but don't write to an empty heap) + if (heap->guarded_sample_rate != 0) { heap->guarded_sample_count = 1; } + return false; + } +} + +mi_decl_restrict void* _mi_heap_malloc_guarded(mi_heap_t* heap, size_t size, bool zero) mi_attr_noexcept; + +#endif + + +/* ------------------------------------------------------------------- +Encoding/Decoding the free list next pointers + +This is to protect against buffer overflow exploits where the +free list is mutated. Many hardened allocators xor the next pointer `p` +with a secret key `k1`, as `p^k1`. This prevents overwriting with known +values but might be still too weak: if the attacker can guess +the pointer `p` this can reveal `k1` (since `p^k1^p == k1`). +Moreover, if multiple blocks can be read as well, the attacker can +xor both as `(p1^k1) ^ (p2^k1) == p1^p2` which may reveal a lot +about the pointers (and subsequently `k1`). + +Instead mimalloc uses an extra key `k2` and encodes as `((p^k2)<<> (MI_INTPTR_BITS - shift)))); +} +static inline uintptr_t mi_rotr(uintptr_t x, uintptr_t shift) { + shift %= MI_INTPTR_BITS; + return (shift==0 ? x : ((x >> shift) | (x << (MI_INTPTR_BITS - shift)))); +} + +static inline void* mi_ptr_decode(const void* null, const mi_encoded_t x, const uintptr_t* keys) { + void* p = (void*)(mi_rotr(x - keys[0], keys[0]) ^ keys[1]); + return (p==null ? NULL : p); +} + +static inline mi_encoded_t mi_ptr_encode(const void* null, const void* p, const uintptr_t* keys) { + uintptr_t x = (uintptr_t)(p==NULL ? null : p); + return mi_rotl(x ^ keys[1], keys[0]) + keys[0]; +} + +static inline uint32_t mi_ptr_encode_canary(const void* null, const void* p, const uintptr_t* keys) { + const uint32_t x = (uint32_t)(mi_ptr_encode(null,p,keys)); + // make the lowest byte 0 to prevent spurious read overflows which could be a security issue (issue #951) + #ifdef MI_BIG_ENDIAN + return (x & 0x00FFFFFF); + #else + return (x & 0xFFFFFF00); + #endif +} + +static inline mi_block_t* mi_block_nextx( const void* null, const mi_block_t* block, const uintptr_t* keys ) { + mi_track_mem_defined(block,sizeof(mi_block_t)); + mi_block_t* next; + #ifdef MI_ENCODE_FREELIST + next = (mi_block_t*)mi_ptr_decode(null, block->next, keys); + #else + MI_UNUSED(keys); MI_UNUSED(null); + next = (mi_block_t*)block->next; + #endif + mi_track_mem_noaccess(block,sizeof(mi_block_t)); + return next; +} + +static inline void mi_block_set_nextx(const void* null, mi_block_t* block, const mi_block_t* next, const uintptr_t* keys) { + mi_track_mem_undefined(block,sizeof(mi_block_t)); + #ifdef MI_ENCODE_FREELIST + block->next = mi_ptr_encode(null, next, keys); + #else + MI_UNUSED(keys); MI_UNUSED(null); + block->next = (mi_encoded_t)next; + #endif + mi_track_mem_noaccess(block,sizeof(mi_block_t)); +} + +static inline mi_block_t* mi_block_next(const mi_page_t* page, const mi_block_t* block) { + #ifdef MI_ENCODE_FREELIST + mi_block_t* next = mi_block_nextx(page,block,page->keys); + // check for free list corruption: is `next` at least in the same page? + // TODO: check if `next` is `page->block_size` aligned? + if mi_unlikely(next!=NULL && !mi_is_in_same_page(block, next)) { + _mi_error_message(EFAULT, "corrupted free list entry of size %zub at %p: value 0x%zx\n", mi_page_block_size(page), block, (uintptr_t)next); + next = NULL; + } + return next; + #else + MI_UNUSED(page); + return mi_block_nextx(page,block,NULL); + #endif +} + +static inline void mi_block_set_next(const mi_page_t* page, mi_block_t* block, const mi_block_t* next) { + #ifdef MI_ENCODE_FREELIST + mi_block_set_nextx(page,block,next, page->keys); + #else + MI_UNUSED(page); + mi_block_set_nextx(page,block,next,NULL); + #endif +} + + +// ------------------------------------------------------------------- +// commit mask +// ------------------------------------------------------------------- + +static inline void mi_commit_mask_create_empty(mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + cm->mask[i] = 0; + } +} + +static inline void mi_commit_mask_create_full(mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + cm->mask[i] = ~((size_t)0); + } +} + +static inline bool mi_commit_mask_is_empty(const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + if (cm->mask[i] != 0) return false; + } + return true; +} + +static inline bool mi_commit_mask_is_full(const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + if (cm->mask[i] != ~((size_t)0)) return false; + } + return true; +} + +// defined in `segment.c`: +size_t _mi_commit_mask_committed_size(const mi_commit_mask_t* cm, size_t total); +size_t _mi_commit_mask_next_run(const mi_commit_mask_t* cm, size_t* idx); + +#define mi_commit_mask_foreach(cm,idx,count) \ + idx = 0; \ + while ((count = _mi_commit_mask_next_run(cm,&idx)) > 0) { + +#define mi_commit_mask_foreach_end() \ + idx += count; \ + } + + + +/* ----------------------------------------------------------- + memory id's +----------------------------------------------------------- */ + +static inline mi_memid_t _mi_memid_create(mi_memkind_t memkind) { + mi_memid_t memid; + _mi_memzero_var(memid); + memid.memkind = memkind; + return memid; +} + +static inline mi_memid_t _mi_memid_none(void) { + return _mi_memid_create(MI_MEM_NONE); +} + +static inline mi_memid_t _mi_memid_create_os(void* base, size_t size, bool committed, bool is_zero, bool is_large) { + mi_memid_t memid = _mi_memid_create(MI_MEM_OS); + memid.mem.os.base = base; + memid.mem.os.size = size; + memid.initially_committed = committed; + memid.initially_zero = is_zero; + memid.is_pinned = is_large; + return memid; +} + + +// ------------------------------------------------------------------- +// Fast "random" shuffle +// ------------------------------------------------------------------- + +static inline uintptr_t _mi_random_shuffle(uintptr_t x) { + if (x==0) { x = 17; } // ensure we don't get stuck in generating zeros +#if (MI_INTPTR_SIZE>=8) + // by Sebastiano Vigna, see: + x ^= x >> 30; + x *= 0xbf58476d1ce4e5b9UL; + x ^= x >> 27; + x *= 0x94d049bb133111ebUL; + x ^= x >> 31; +#elif (MI_INTPTR_SIZE==4) + // by Chris Wellons, see: + x ^= x >> 16; + x *= 0x7feb352dUL; + x ^= x >> 15; + x *= 0x846ca68bUL; + x ^= x >> 16; +#endif + return x; +} + + + +// ----------------------------------------------------------------------- +// Count bits: trailing or leading zeros (with MI_INTPTR_BITS on all zero) +// ----------------------------------------------------------------------- + +#if defined(__GNUC__) + +#include // LONG_MAX +#define MI_HAVE_FAST_BITSCAN +static inline size_t mi_clz(size_t x) { + if (x==0) return MI_SIZE_BITS; + #if (SIZE_MAX == ULONG_MAX) + return __builtin_clzl(x); + #else + return __builtin_clzll(x); + #endif +} +static inline size_t mi_ctz(size_t x) { + if (x==0) return MI_SIZE_BITS; + #if (SIZE_MAX == ULONG_MAX) + return __builtin_ctzl(x); + #else + return __builtin_ctzll(x); + #endif +} + +#elif defined(_MSC_VER) + +#include // LONG_MAX +#include // BitScanReverse64 +#define MI_HAVE_FAST_BITSCAN +static inline size_t mi_clz(size_t x) { + if (x==0) return MI_SIZE_BITS; + unsigned long idx; + #if (SIZE_MAX == ULONG_MAX) + _BitScanReverse(&idx, x); + #else + _BitScanReverse64(&idx, x); + #endif + return ((MI_SIZE_BITS - 1) - (size_t)idx); +} +static inline size_t mi_ctz(size_t x) { + if (x==0) return MI_SIZE_BITS; + unsigned long idx; + #if (SIZE_MAX == ULONG_MAX) + _BitScanForward(&idx, x); + #else + _BitScanForward64(&idx, x); + #endif + return (size_t)idx; +} + +#else + +static inline size_t mi_ctz_generic32(uint32_t x) { + // de Bruijn multiplication, see + static const uint8_t debruijn[32] = { + 0, 1, 28, 2, 29, 14, 24, 3, 30, 22, 20, 15, 25, 17, 4, 8, + 31, 27, 13, 23, 21, 19, 16, 7, 26, 12, 18, 6, 11, 5, 10, 9 + }; + if (x==0) return 32; + return debruijn[(uint32_t)((x & -(int32_t)x) * (uint32_t)(0x077CB531U)) >> 27]; +} + +static inline size_t mi_clz_generic32(uint32_t x) { + // de Bruijn multiplication, see + static const uint8_t debruijn[32] = { + 31, 22, 30, 21, 18, 10, 29, 2, 20, 17, 15, 13, 9, 6, 28, 1, + 23, 19, 11, 3, 16, 14, 7, 24, 12, 4, 8, 25, 5, 26, 27, 0 + }; + if (x==0) return 32; + x |= x >> 1; + x |= x >> 2; + x |= x >> 4; + x |= x >> 8; + x |= x >> 16; + return debruijn[(uint32_t)(x * (uint32_t)(0x07C4ACDDU)) >> 27]; +} + +static inline size_t mi_ctz(size_t x) { + if (x==0) return MI_SIZE_BITS; + #if (MI_SIZE_BITS <= 32) + return mi_ctz_generic32((uint32_t)x); + #else + const uint32_t lo = (uint32_t)x; + if (lo != 0) { + return mi_ctz_generic32(lo); + } + else { + return (32 + mi_ctz_generic32((uint32_t)(x>>32))); + } + #endif +} + +static inline size_t mi_clz(size_t x) { + if (x==0) return MI_SIZE_BITS; + #if (MI_SIZE_BITS <= 32) + return mi_clz_generic32((uint32_t)x); + #else + const uint32_t hi = (uint32_t)(x>>32); + if (hi != 0) { + return mi_clz_generic32(hi); + } + else { + return 32 + mi_clz_generic32((uint32_t)x); + } + #endif +} + +#endif + +// "bit scan reverse": Return index of the highest bit (or MI_SIZE_BITS if `x` is zero) +static inline size_t mi_bsr(size_t x) { + return (x==0 ? MI_SIZE_BITS : MI_SIZE_BITS - 1 - mi_clz(x)); +} + +size_t _mi_popcount_generic(size_t x); + +static inline size_t mi_popcount(size_t x) { + if (x<=1) return x; + if (x==SIZE_MAX) return MI_SIZE_BITS; + #if defined(__GNUC__) + #if (SIZE_MAX == ULONG_MAX) + return __builtin_popcountl(x); + #else + return __builtin_popcountll(x); + #endif + #else + return _mi_popcount_generic(x); + #endif +} + +// --------------------------------------------------------------------------------- +// Provide our own `_mi_memcpy` for potential performance optimizations. +// +// For now, only on Windows with msvc/clang-cl we optimize to `rep movsb` if +// we happen to run on x86/x64 cpu's that have "fast short rep movsb" (FSRM) support +// (AMD Zen3+ (~2020) or Intel Ice Lake+ (~2017). See also issue #201 and pr #253. +// --------------------------------------------------------------------------------- + +#if !MI_TRACK_ENABLED && defined(_WIN32) && (defined(_M_IX86) || defined(_M_X64)) +#include +extern mi_decl_hidden bool _mi_cpu_has_fsrm; +extern mi_decl_hidden bool _mi_cpu_has_erms; +static inline void _mi_memcpy(void* dst, const void* src, size_t n) { + if ((_mi_cpu_has_fsrm && n <= 128) || (_mi_cpu_has_erms && n > 128)) { + __movsb((unsigned char*)dst, (const unsigned char*)src, n); + } + else { + memcpy(dst, src, n); + } +} +static inline void _mi_memzero(void* dst, size_t n) { + if ((_mi_cpu_has_fsrm && n <= 128) || (_mi_cpu_has_erms && n > 128)) { + __stosb((unsigned char*)dst, 0, n); + } + else { + memset(dst, 0, n); + } +} +#else +static inline void _mi_memcpy(void* dst, const void* src, size_t n) { + memcpy(dst, src, n); +} +static inline void _mi_memzero(void* dst, size_t n) { + memset(dst, 0, n); +} +#endif + +// ------------------------------------------------------------------------------- +// The `_mi_memcpy_aligned` can be used if the pointers are machine-word aligned +// This is used for example in `mi_realloc`. +// ------------------------------------------------------------------------------- + +#if (defined(__GNUC__) && (__GNUC__ >= 4)) || defined(__clang__) +// On GCC/CLang we provide a hint that the pointers are word aligned. +static inline void _mi_memcpy_aligned(void* dst, const void* src, size_t n) { + mi_assert_internal(((uintptr_t)dst % MI_INTPTR_SIZE == 0) && ((uintptr_t)src % MI_INTPTR_SIZE == 0)); + void* adst = __builtin_assume_aligned(dst, MI_INTPTR_SIZE); + const void* asrc = __builtin_assume_aligned(src, MI_INTPTR_SIZE); + _mi_memcpy(adst, asrc, n); +} + +static inline void _mi_memzero_aligned(void* dst, size_t n) { + mi_assert_internal((uintptr_t)dst % MI_INTPTR_SIZE == 0); + void* adst = __builtin_assume_aligned(dst, MI_INTPTR_SIZE); + _mi_memzero(adst, n); +} +#else +// Default fallback on `_mi_memcpy` +static inline void _mi_memcpy_aligned(void* dst, const void* src, size_t n) { + mi_assert_internal(((uintptr_t)dst % MI_INTPTR_SIZE == 0) && ((uintptr_t)src % MI_INTPTR_SIZE == 0)); + _mi_memcpy(dst, src, n); +} + +static inline void _mi_memzero_aligned(void* dst, size_t n) { + mi_assert_internal((uintptr_t)dst % MI_INTPTR_SIZE == 0); + _mi_memzero(dst, n); +} +#endif + + +#endif diff --git a/compat/mimalloc/mimalloc/prim.h b/compat/mimalloc/mimalloc/prim.h new file mode 100644 index 00000000000000..1087d9b8dad304 --- /dev/null +++ b/compat/mimalloc/mimalloc/prim.h @@ -0,0 +1,421 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_PRIM_H +#define MIMALLOC_PRIM_H + + +// -------------------------------------------------------------------------- +// This file specifies the primitive portability API. +// Each OS/host needs to implement these primitives, see `src/prim` +// for implementations on Window, macOS, WASI, and Linux/Unix. +// +// note: on all primitive functions, we always have result parameters != NULL, and: +// addr != NULL and page aligned +// size > 0 and page aligned +// the return value is an error code as an `int` where 0 is success +// -------------------------------------------------------------------------- + +// OS memory configuration +typedef struct mi_os_mem_config_s { + size_t page_size; // default to 4KiB + size_t large_page_size; // 0 if not supported, usually 2MiB (4MiB on Windows) + size_t alloc_granularity; // smallest allocation size (usually 4KiB, on Windows 64KiB) + size_t physical_memory_in_kib; // physical memory size in KiB + size_t virtual_address_bits; // usually 48 or 56 bits on 64-bit systems. (used to determine secure randomization) + bool has_overcommit; // can we reserve more memory than can be actually committed? + bool has_partial_free; // can allocated blocks be freed partially? (true for mmap, false for VirtualAlloc) + bool has_virtual_reserve; // supports virtual address space reservation? (if true we can reserve virtual address space without using commit or physical memory) +} mi_os_mem_config_t; + +// Initialize +void _mi_prim_mem_init( mi_os_mem_config_t* config ); + +// Free OS memory +int _mi_prim_free(void* addr, size_t size ); + +// Allocate OS memory. Return NULL on error. +// The `try_alignment` is just a hint and the returned pointer does not have to be aligned. +// If `commit` is false, the virtual memory range only needs to be reserved (with no access) +// which will later be committed explicitly using `_mi_prim_commit`. +// `is_zero` is set to true if the memory was zero initialized (as on most OS's) +// The `hint_addr` address is either `NULL` or a preferred allocation address but can be ignored. +// pre: !commit => !allow_large +// try_alignment >= _mi_os_page_size() and a power of 2 +int _mi_prim_alloc(void* hint_addr, size_t size, size_t try_alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero, void** addr); + +// Commit memory. Returns error code or 0 on success. +// For example, on Linux this would make the memory PROT_READ|PROT_WRITE. +// `is_zero` is set to true if the memory was zero initialized (e.g. on Windows) +int _mi_prim_commit(void* addr, size_t size, bool* is_zero); + +// Decommit memory. Returns error code or 0 on success. The `needs_recommit` result is true +// if the memory would need to be re-committed. For example, on Windows this is always true, +// but on Linux we could use MADV_DONTNEED to decommit which does not need a recommit. +// pre: needs_recommit != NULL +int _mi_prim_decommit(void* addr, size_t size, bool* needs_recommit); + +// Reset memory. The range keeps being accessible but the content might be reset to zero at any moment. +// Returns error code or 0 on success. +int _mi_prim_reset(void* addr, size_t size); + +// Reuse memory. This is called for memory that is already committed but +// may have been reset (`_mi_prim_reset`) or decommitted (`_mi_prim_decommit`) where `needs_recommit` was false. +// Returns error code or 0 on success. On most platforms this is a no-op. +int _mi_prim_reuse(void* addr, size_t size); + +// Protect memory. Returns error code or 0 on success. +int _mi_prim_protect(void* addr, size_t size, bool protect); + +// Allocate huge (1GiB) pages possibly associated with a NUMA node. +// `is_zero` is set to true if the memory was zero initialized (as on most OS's) +// pre: size > 0 and a multiple of 1GiB. +// numa_node is either negative (don't care), or a numa node number. +int _mi_prim_alloc_huge_os_pages(void* hint_addr, size_t size, int numa_node, bool* is_zero, void** addr); + +// Return the current NUMA node +size_t _mi_prim_numa_node(void); + +// Return the number of logical NUMA nodes +size_t _mi_prim_numa_node_count(void); + +// Clock ticks +mi_msecs_t _mi_prim_clock_now(void); + +// Return process information (only for statistics) +typedef struct mi_process_info_s { + mi_msecs_t elapsed; + mi_msecs_t utime; + mi_msecs_t stime; + size_t current_rss; + size_t peak_rss; + size_t current_commit; + size_t peak_commit; + size_t page_faults; +} mi_process_info_t; + +void _mi_prim_process_info(mi_process_info_t* pinfo); + +// Default stderr output. (only for warnings etc. with verbose enabled) +// msg != NULL && _mi_strlen(msg) > 0 +void _mi_prim_out_stderr( const char* msg ); + +// Get an environment variable. (only for options) +// name != NULL, result != NULL, result_size >= 64 +bool _mi_prim_getenv(const char* name, char* result, size_t result_size); + + +// Fill a buffer with strong randomness; return `false` on error or if +// there is no strong randomization available. +bool _mi_prim_random_buf(void* buf, size_t buf_len); + +// Called on the first thread start, and should ensure `_mi_thread_done` is called on thread termination. +void _mi_prim_thread_init_auto_done(void); + +// Called on process exit and may take action to clean up resources associated with the thread auto done. +void _mi_prim_thread_done_auto_done(void); + +// Called when the default heap for a thread changes +void _mi_prim_thread_associate_default_heap(mi_heap_t* heap); + + +//------------------------------------------------------------------- +// Access to TLS (thread local storage) slots. +// We need fast access to both a unique thread id (in `free.c:mi_free`) and +// to a thread-local heap pointer (in `alloc.c:mi_malloc`). +// To achieve this we use specialized code for various platforms. +//------------------------------------------------------------------- + +// On some libc + platform combinations we can directly access a thread-local storage (TLS) slot. +// The TLS layout depends on both the OS and libc implementation so we use specific tests for each main platform. +// If you test on another platform and it works please send a PR :-) +// see also https://akkadia.org/drepper/tls.pdf for more info on the TLS register. +// +// Note: we would like to prefer `__builtin_thread_pointer()` nowadays instead of using assembly, +// but unfortunately we can not detect support reliably (see issue #883) +// We also use it on Apple OS as we use a TLS slot for the default heap there. +#if defined(__GNUC__) && ( \ + (defined(__GLIBC__) && (defined(__x86_64__) || defined(__i386__) || (defined(__arm__) && __ARM_ARCH >= 7) || defined(__aarch64__))) \ + || (defined(__APPLE__) && (defined(__x86_64__) || defined(__aarch64__) || defined(__POWERPC__))) \ + || (defined(__BIONIC__) && (defined(__x86_64__) || defined(__i386__) || (defined(__arm__) && __ARM_ARCH >= 7) || defined(__aarch64__))) \ + || (defined(__FreeBSD__) && (defined(__x86_64__) || defined(__i386__) || defined(__aarch64__))) \ + || (defined(__OpenBSD__) && (defined(__x86_64__) || defined(__i386__) || defined(__aarch64__))) \ + ) + +#define MI_HAS_TLS_SLOT 1 + +static inline void* mi_prim_tls_slot(size_t slot) mi_attr_noexcept { + void* res; + const size_t ofs = (slot*sizeof(void*)); + #if defined(__i386__) + __asm__("movl %%gs:%1, %0" : "=r" (res) : "m" (*((void**)ofs)) : ); // x86 32-bit always uses GS + #elif defined(__APPLE__) && defined(__x86_64__) + __asm__("movq %%gs:%1, %0" : "=r" (res) : "m" (*((void**)ofs)) : ); // x86_64 macOSX uses GS + #elif defined(__x86_64__) && (MI_INTPTR_SIZE==4) + __asm__("movl %%fs:%1, %0" : "=r" (res) : "m" (*((void**)ofs)) : ); // x32 ABI + #elif defined(__x86_64__) + __asm__("movq %%fs:%1, %0" : "=r" (res) : "m" (*((void**)ofs)) : ); // x86_64 Linux, BSD uses FS + #elif defined(__arm__) + void** tcb; MI_UNUSED(ofs); + __asm__ volatile ("mrc p15, 0, %0, c13, c0, 3\nbic %0, %0, #3" : "=r" (tcb)); + res = tcb[slot]; + #elif defined(__aarch64__) + void** tcb; MI_UNUSED(ofs); + #if defined(__APPLE__) // M1, issue #343 + __asm__ volatile ("mrs %0, tpidrro_el0\nbic %0, %0, #7" : "=r" (tcb)); + #else + __asm__ volatile ("mrs %0, tpidr_el0" : "=r" (tcb)); + #endif + res = tcb[slot]; + #elif defined(__APPLE__) && defined(__POWERPC__) // ppc, issue #781 + MI_UNUSED(ofs); + res = pthread_getspecific(slot); + #endif + return res; +} + +// setting a tls slot is only used on macOS for now +static inline void mi_prim_tls_slot_set(size_t slot, void* value) mi_attr_noexcept { + const size_t ofs = (slot*sizeof(void*)); + #if defined(__i386__) + __asm__("movl %1,%%gs:%0" : "=m" (*((void**)ofs)) : "rn" (value) : ); // 32-bit always uses GS + #elif defined(__APPLE__) && defined(__x86_64__) + __asm__("movq %1,%%gs:%0" : "=m" (*((void**)ofs)) : "rn" (value) : ); // x86_64 macOS uses GS + #elif defined(__x86_64__) && (MI_INTPTR_SIZE==4) + __asm__("movl %1,%%fs:%0" : "=m" (*((void**)ofs)) : "rn" (value) : ); // x32 ABI + #elif defined(__x86_64__) + __asm__("movq %1,%%fs:%0" : "=m" (*((void**)ofs)) : "rn" (value) : ); // x86_64 Linux, BSD uses FS + #elif defined(__arm__) + void** tcb; MI_UNUSED(ofs); + __asm__ volatile ("mrc p15, 0, %0, c13, c0, 3\nbic %0, %0, #3" : "=r" (tcb)); + tcb[slot] = value; + #elif defined(__aarch64__) + void** tcb; MI_UNUSED(ofs); + #if defined(__APPLE__) // M1, issue #343 + __asm__ volatile ("mrs %0, tpidrro_el0\nbic %0, %0, #7" : "=r" (tcb)); + #else + __asm__ volatile ("mrs %0, tpidr_el0" : "=r" (tcb)); + #endif + tcb[slot] = value; + #elif defined(__APPLE__) && defined(__POWERPC__) // ppc, issue #781 + MI_UNUSED(ofs); + pthread_setspecific(slot, value); + #endif +} + +#elif _WIN32 && MI_WIN_USE_FIXED_TLS && !defined(MI_WIN_USE_FLS) + +// On windows we can store the thread-local heap at a fixed TLS slot to avoid +// thread-local initialization checks in the fast path. +// We allocate a user TLS slot at process initialization (see `windows/prim.c`) +// and store the offset `_mi_win_tls_offset`. +#define MI_HAS_TLS_SLOT 1 // 2 = we can reliably initialize the slot (saving a test on each malloc) + +extern mi_decl_hidden size_t _mi_win_tls_offset; + +#if MI_WIN_USE_FIXED_TLS > 1 +#define MI_TLS_SLOT (MI_WIN_USE_FIXED_TLS) +#elif MI_SIZE_SIZE == 4 +#define MI_TLS_SLOT (0x0E10 + _mi_win_tls_offset) // User TLS slots +#else +#define MI_TLS_SLOT (0x1480 + _mi_win_tls_offset) // User TLS slots +#endif + +static inline void* mi_prim_tls_slot(size_t slot) mi_attr_noexcept { + #if (_M_X64 || _M_AMD64) && !defined(_M_ARM64EC) + return (void*)__readgsqword((unsigned long)slot); // direct load at offset from gs + #elif _M_IX86 && !defined(_M_ARM64EC) + return (void*)__readfsdword((unsigned long)slot); // direct load at offset from fs + #else + return ((void**)NtCurrentTeb())[slot / sizeof(void*)]; + #endif +} +static inline void mi_prim_tls_slot_set(size_t slot, void* value) mi_attr_noexcept { + ((void**)NtCurrentTeb())[slot / sizeof(void*)] = value; +} + +#endif + + + +//------------------------------------------------------------------- +// Get a fast unique thread id. +// +// Getting the thread id should be performant as it is called in the +// fast path of `_mi_free` and we specialize for various platforms as +// inlined definitions. Regular code should call `init.c:_mi_thread_id()`. +// We only require _mi_prim_thread_id() to return a unique id +// for each thread (unequal to zero). +//------------------------------------------------------------------- + + +// Do we have __builtin_thread_pointer? This would be the preferred way to get a unique thread id +// but unfortunately, it seems we cannot test for this reliably at this time (see issue #883) +// Nevertheless, it seems needed on older graviton platforms (see issue #851). +// For now, we only enable this for specific platforms. +#if !defined(__APPLE__) /* on apple (M1) the wrong register is read (tpidr_el0 instead of tpidrro_el0) so fall back to TLS slot assembly ()*/ \ + && !defined(__CYGWIN__) \ + && !defined(MI_LIBC_MUSL) \ + && (!defined(__clang_major__) || __clang_major__ >= 14) /* older clang versions emit bad code; fall back to using the TLS slot () */ + #if (defined(__GNUC__) && (__GNUC__ >= 7) && defined(__aarch64__)) /* aarch64 for older gcc versions (issue #851) */ \ + || (defined(__GNUC__) && (__GNUC__ >= 11) && defined(__x86_64__)) \ + || (defined(__clang_major__) && (__clang_major__ >= 14) && (defined(__aarch64__) || defined(__x86_64__))) + #define MI_USE_BUILTIN_THREAD_POINTER 1 + #endif +#endif + + + +// defined in `init.c`; do not use these directly +extern mi_decl_hidden mi_decl_thread mi_heap_t* _mi_heap_default; // default heap to allocate from +extern mi_decl_hidden bool _mi_process_is_initialized; // has mi_process_init been called? + +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept; + +// Get a unique id for the current thread. +#if defined(MI_PRIM_THREAD_ID) + +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept { + return MI_PRIM_THREAD_ID(); // used for example by CPython for a free threaded build (see python/cpython#115488) +} + +#elif defined(_WIN32) + +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept { + // Windows: works on Intel and ARM in both 32- and 64-bit + return (uintptr_t)NtCurrentTeb(); +} + +#elif MI_USE_BUILTIN_THREAD_POINTER + +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept { + // Works on most Unix based platforms with recent compilers + return (uintptr_t)__builtin_thread_pointer(); +} + +#elif MI_HAS_TLS_SLOT + +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept { + #if defined(__BIONIC__) + // issue #384, #495: on the Bionic libc (Android), slot 1 is the thread id + // see: https://github.com/aosp-mirror/platform_bionic/blob/c44b1d0676ded732df4b3b21c5f798eacae93228/libc/platform/bionic/tls_defines.h#L86 + return (uintptr_t)mi_prim_tls_slot(1); + #else + // in all our other targets, slot 0 is the thread id + // glibc: https://sourceware.org/git/?p=glibc.git;a=blob_plain;f=sysdeps/x86_64/nptl/tls.h + // apple: https://github.com/apple/darwin-xnu/blob/main/libsyscall/os/tsd.h#L36 + return (uintptr_t)mi_prim_tls_slot(0); + #endif +} + +#else + +// otherwise use portable C, taking the address of a thread local variable (this is still very fast on most platforms). +static inline mi_threadid_t _mi_prim_thread_id(void) mi_attr_noexcept { + return (uintptr_t)&_mi_heap_default; +} + +#endif + + + +/* ---------------------------------------------------------------------------------------- +Get the thread local default heap: `_mi_prim_get_default_heap()` + +This is inlined here as it is on the fast path for allocation functions. + +On most platforms (Windows, Linux, FreeBSD, NetBSD, etc), this just returns a +__thread local variable (`_mi_heap_default`). With the initial-exec TLS model this ensures +that the storage will always be available (allocated on the thread stacks). + +On some platforms though we cannot use that when overriding `malloc` since the underlying +TLS implementation (or the loader) will call itself `malloc` on a first access and recurse. +We try to circumvent this in an efficient way: +- macOSX : we use an unused TLS slot from the OS allocated slots (MI_TLS_SLOT). On OSX, the + loader itself calls `malloc` even before the modules are initialized. +- OpenBSD: we use an unused slot from the pthread block (MI_TLS_PTHREAD_SLOT_OFS). +- DragonFly: defaults are working but seem slow compared to freeBSD (see PR #323) +------------------------------------------------------------------------------------------- */ + +static inline mi_heap_t* mi_prim_get_default_heap(void); + +#if defined(MI_MALLOC_OVERRIDE) +#if defined(__APPLE__) // macOS + #define MI_TLS_SLOT 89 // seems unused? + // other possible unused ones are 9, 29, __PTK_FRAMEWORK_JAVASCRIPTCORE_KEY4 (94), __PTK_FRAMEWORK_GC_KEY9 (112) and __PTK_FRAMEWORK_OLDGC_KEY9 (89) + // see +#elif defined(__OpenBSD__) + // use end bytes of a name; goes wrong if anyone uses names > 23 characters (ptrhread specifies 16) + // see + #define MI_TLS_PTHREAD_SLOT_OFS (6*sizeof(int) + 4*sizeof(void*) + 24) + // #elif defined(__DragonFly__) + // #warning "mimalloc is not working correctly on DragonFly yet." + // #define MI_TLS_PTHREAD_SLOT_OFS (4 + 1*sizeof(void*)) // offset `uniqueid` (also used by gdb?) +#elif defined(__ANDROID__) + // See issue #381 + #define MI_TLS_PTHREAD +#endif +#endif + + +#if MI_TLS_SLOT +# if !defined(MI_HAS_TLS_SLOT) +# error "trying to use a TLS slot for the default heap, but the mi_prim_tls_slot primitives are not defined" +# endif + +static inline mi_heap_t* mi_prim_get_default_heap(void) { + mi_heap_t* heap = (mi_heap_t*)mi_prim_tls_slot(MI_TLS_SLOT); + #if MI_HAS_TLS_SLOT == 1 // check if the TLS slot is initialized + if mi_unlikely(heap == NULL) { + #ifdef __GNUC__ + __asm(""); // prevent conditional load of the address of _mi_heap_empty + #endif + heap = (mi_heap_t*)&_mi_heap_empty; + } + #endif + return heap; +} + +#elif defined(MI_TLS_PTHREAD_SLOT_OFS) + +static inline mi_heap_t** mi_prim_tls_pthread_heap_slot(void) { + pthread_t self = pthread_self(); + #if defined(__DragonFly__) + if (self==NULL) return NULL; + #endif + return (mi_heap_t**)((uint8_t*)self + MI_TLS_PTHREAD_SLOT_OFS); +} + +static inline mi_heap_t* mi_prim_get_default_heap(void) { + mi_heap_t** pheap = mi_prim_tls_pthread_heap_slot(); + if mi_unlikely(pheap == NULL) return _mi_heap_main_get(); + mi_heap_t* heap = *pheap; + if mi_unlikely(heap == NULL) return (mi_heap_t*)&_mi_heap_empty; + return heap; +} + +#elif defined(MI_TLS_PTHREAD) + +extern mi_decl_hidden pthread_key_t _mi_heap_default_key; +static inline mi_heap_t* mi_prim_get_default_heap(void) { + mi_heap_t* heap = (mi_unlikely(_mi_heap_default_key == (pthread_key_t)(-1)) ? _mi_heap_main_get() : (mi_heap_t*)pthread_getspecific(_mi_heap_default_key)); + return (mi_unlikely(heap == NULL) ? (mi_heap_t*)&_mi_heap_empty : heap); +} + +#else // default using a thread local variable; used on most platforms. + +static inline mi_heap_t* mi_prim_get_default_heap(void) { + #if defined(MI_TLS_RECURSE_GUARD) + if (mi_unlikely(!_mi_process_is_initialized)) return _mi_heap_main_get(); + #endif + return _mi_heap_default; +} + +#endif // mi_prim_get_default_heap() + + +#endif // MIMALLOC_PRIM_H diff --git a/compat/mimalloc/mimalloc/track.h b/compat/mimalloc/mimalloc/track.h new file mode 100644 index 00000000000000..4b5709e2b54110 --- /dev/null +++ b/compat/mimalloc/mimalloc/track.h @@ -0,0 +1,145 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_TRACK_H +#define MIMALLOC_TRACK_H + +/* ------------------------------------------------------------------------------------------------------ +Track memory ranges with macros for tools like Valgrind address sanitizer, or other memory checkers. +These can be defined for tracking allocation: + + #define mi_track_malloc_size(p,reqsize,size,zero) + #define mi_track_free_size(p,_size) + +The macros are set up such that the size passed to `mi_track_free_size` +always matches the size of `mi_track_malloc_size`. (currently, `size == mi_usable_size(p)`). +The `reqsize` is what the user requested, and `size >= reqsize`. +The `size` is either byte precise (and `size==reqsize`) if `MI_PADDING` is enabled, +or otherwise it is the usable block size which may be larger than the original request. +Use `_mi_block_size_of(void* p)` to get the full block size that was allocated (including padding etc). +The `zero` parameter is `true` if the allocated block is zero initialized. + +Optional: + + #define mi_track_align(p,alignedp,offset,size) + #define mi_track_resize(p,oldsize,newsize) + #define mi_track_init() + +The `mi_track_align` is called right after a `mi_track_malloc` for aligned pointers in a block. +The corresponding `mi_track_free` still uses the block start pointer and original size (corresponding to the `mi_track_malloc`). +The `mi_track_resize` is currently unused but could be called on reallocations within a block. +`mi_track_init` is called at program start. + +The following macros are for tools like asan and valgrind to track whether memory is +defined, undefined, or not accessible at all: + + #define mi_track_mem_defined(p,size) + #define mi_track_mem_undefined(p,size) + #define mi_track_mem_noaccess(p,size) + +-------------------------------------------------------------------------------------------------------*/ + +#if MI_TRACK_VALGRIND +// valgrind tool + +#define MI_TRACK_ENABLED 1 +#define MI_TRACK_HEAP_DESTROY 1 // track free of individual blocks on heap_destroy +#define MI_TRACK_TOOL "valgrind" + +#include +#include + +#define mi_track_malloc_size(p,reqsize,size,zero) VALGRIND_MALLOCLIKE_BLOCK(p,size,MI_PADDING_SIZE /*red zone*/,zero) +#define mi_track_free_size(p,_size) VALGRIND_FREELIKE_BLOCK(p,MI_PADDING_SIZE /*red zone*/) +#define mi_track_resize(p,oldsize,newsize) VALGRIND_RESIZEINPLACE_BLOCK(p,oldsize,newsize,MI_PADDING_SIZE /*red zone*/) +#define mi_track_mem_defined(p,size) VALGRIND_MAKE_MEM_DEFINED(p,size) +#define mi_track_mem_undefined(p,size) VALGRIND_MAKE_MEM_UNDEFINED(p,size) +#define mi_track_mem_noaccess(p,size) VALGRIND_MAKE_MEM_NOACCESS(p,size) + +#elif MI_TRACK_ASAN +// address sanitizer + +#define MI_TRACK_ENABLED 1 +#define MI_TRACK_HEAP_DESTROY 0 +#define MI_TRACK_TOOL "asan" + +#include + +#define mi_track_malloc_size(p,reqsize,size,zero) ASAN_UNPOISON_MEMORY_REGION(p,size) +#define mi_track_free_size(p,size) ASAN_POISON_MEMORY_REGION(p,size) +#define mi_track_mem_defined(p,size) ASAN_UNPOISON_MEMORY_REGION(p,size) +#define mi_track_mem_undefined(p,size) ASAN_UNPOISON_MEMORY_REGION(p,size) +#define mi_track_mem_noaccess(p,size) ASAN_POISON_MEMORY_REGION(p,size) + +#elif MI_TRACK_ETW +// windows event tracing + +#define MI_TRACK_ENABLED 1 +#define MI_TRACK_HEAP_DESTROY 1 +#define MI_TRACK_TOOL "ETW" + +#include "../src/prim/windows/etw.h" + +#define mi_track_init() EventRegistermicrosoft_windows_mimalloc(); +#define mi_track_malloc_size(p,reqsize,size,zero) EventWriteETW_MI_ALLOC((UINT64)(p), size) +#define mi_track_free_size(p,size) EventWriteETW_MI_FREE((UINT64)(p), size) + +#else +// no tracking + +#define MI_TRACK_ENABLED 0 +#define MI_TRACK_HEAP_DESTROY 0 +#define MI_TRACK_TOOL "none" + +#define mi_track_malloc_size(p,reqsize,size,zero) +#define mi_track_free_size(p,_size) + +#endif + +// ------------------- +// Utility definitions + +#ifndef mi_track_resize +#define mi_track_resize(p,oldsize,newsize) mi_track_free_size(p,oldsize); mi_track_malloc(p,newsize,false) +#endif + +#ifndef mi_track_align +#define mi_track_align(p,alignedp,offset,size) mi_track_mem_noaccess(p,offset) +#endif + +#ifndef mi_track_init +#define mi_track_init() +#endif + +#ifndef mi_track_mem_defined +#define mi_track_mem_defined(p,size) +#endif + +#ifndef mi_track_mem_undefined +#define mi_track_mem_undefined(p,size) +#endif + +#ifndef mi_track_mem_noaccess +#define mi_track_mem_noaccess(p,size) +#endif + + +#if MI_PADDING +#define mi_track_malloc(p,reqsize,zero) \ + if ((p)!=NULL) { \ + mi_assert_internal(mi_usable_size(p)==(reqsize)); \ + mi_track_malloc_size(p,reqsize,reqsize,zero); \ + } +#else +#define mi_track_malloc(p,reqsize,zero) \ + if ((p)!=NULL) { \ + mi_assert_internal(mi_usable_size(p)>=(reqsize)); \ + mi_track_malloc_size(p,reqsize,mi_usable_size(p),zero); \ + } +#endif + +#endif diff --git a/compat/mimalloc/mimalloc/types.h b/compat/mimalloc/mimalloc/types.h new file mode 100644 index 00000000000000..a15d9cba4658cb --- /dev/null +++ b/compat/mimalloc/mimalloc/types.h @@ -0,0 +1,685 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#pragma once +#ifndef MIMALLOC_TYPES_H +#define MIMALLOC_TYPES_H + +// -------------------------------------------------------------------------- +// This file contains the main type definitions for mimalloc: +// mi_heap_t : all data for a thread-local heap, contains +// lists of all managed heap pages. +// mi_segment_t : a larger chunk of memory (32GiB) from where pages +// are allocated. A segment is divided in slices (64KiB) from +// which pages are allocated. +// mi_page_t : a "mimalloc" page (usually 64KiB or 512KiB) from +// where objects are allocated. +// Note: we write "OS page" for OS memory pages while +// using plain "page" for mimalloc pages (`mi_page_t`). +// -------------------------------------------------------------------------- + + +#include +#include // ptrdiff_t +#include // uintptr_t, uint16_t, etc +#include "atomic.h" // _Atomic + +#ifdef _MSC_VER +#pragma warning(disable:4214) // bitfield is not int +#endif + +// Minimal alignment necessary. On most platforms 16 bytes are needed +// due to SSE registers for example. This must be at least `sizeof(void*)` +#ifndef MI_MAX_ALIGN_SIZE +#define MI_MAX_ALIGN_SIZE 16 // sizeof(max_align_t) +#endif + +// ------------------------------------------------------ +// Variants +// ------------------------------------------------------ + +// Define NDEBUG in the release version to disable assertions. +// #define NDEBUG + +// Define MI_TRACK_ to enable tracking support +// #define MI_TRACK_VALGRIND 1 +// #define MI_TRACK_ASAN 1 +// #define MI_TRACK_ETW 1 + +// Define MI_STAT as 1 to maintain statistics; set it to 2 to have detailed statistics (but costs some performance). +// #define MI_STAT 1 + +// Define MI_SECURE to enable security mitigations +// #define MI_SECURE 1 // guard page around metadata +// #define MI_SECURE 2 // guard page around each mimalloc page +// #define MI_SECURE 3 // encode free lists (detect corrupted free list (buffer overflow), and invalid pointer free) +// #define MI_SECURE 4 // checks for double free. (may be more expensive) + +#if !defined(MI_SECURE) +#define MI_SECURE 0 +#endif + +// Define MI_DEBUG for debug mode +// #define MI_DEBUG 1 // basic assertion checks and statistics, check double free, corrupted free list, and invalid pointer free. +// #define MI_DEBUG 2 // + internal assertion checks +// #define MI_DEBUG 3 // + extensive internal invariant checking (cmake -DMI_DEBUG_FULL=ON) +#if !defined(MI_DEBUG) +#if defined(MI_BUILD_RELEASE) || defined(NDEBUG) +#define MI_DEBUG 0 +#else +#define MI_DEBUG 2 +#endif +#endif + +// Use guard pages behind objects of a certain size (set by the MIMALLOC_DEBUG_GUARDED_MIN/MAX options) +// Padding should be disabled when using guard pages +// #define MI_GUARDED 1 +#if defined(MI_GUARDED) +#define MI_PADDING 0 +#endif + +// Reserve extra padding at the end of each block to be more resilient against heap block overflows. +// The padding can detect buffer overflow on free. +#if !defined(MI_PADDING) && (MI_SECURE>=3 || MI_DEBUG>=1 || (MI_TRACK_VALGRIND || MI_TRACK_ASAN || MI_TRACK_ETW)) +#define MI_PADDING 1 +#endif + +// Check padding bytes; allows byte-precise buffer overflow detection +#if !defined(MI_PADDING_CHECK) && MI_PADDING && (MI_SECURE>=3 || MI_DEBUG>=1) +#define MI_PADDING_CHECK 1 +#endif + + +// Encoded free lists allow detection of corrupted free lists +// and can detect buffer overflows, modify after free, and double `free`s. +#if (MI_SECURE>=3 || MI_DEBUG>=1) +#define MI_ENCODE_FREELIST 1 +#endif + + +// We used to abandon huge pages in order to eagerly deallocate it if freed from another thread. +// Unfortunately, that makes it not possible to visit them during a heap walk or include them in a +// `mi_heap_destroy`. We therefore instead reset/decommit the huge blocks nowadays if freed from +// another thread so the memory becomes "virtually" available (and eventually gets properly freed by +// the owning thread). +// #define MI_HUGE_PAGE_ABANDON 1 + + +// ------------------------------------------------------ +// Platform specific values +// ------------------------------------------------------ + +// ------------------------------------------------------ +// Size of a pointer. +// We assume that `sizeof(void*)==sizeof(intptr_t)` +// and it holds for all platforms we know of. +// +// However, the C standard only requires that: +// p == (void*)((intptr_t)p)) +// but we also need: +// i == (intptr_t)((void*)i) +// or otherwise one might define an intptr_t type that is larger than a pointer... +// ------------------------------------------------------ + +#if INTPTR_MAX > INT64_MAX +# define MI_INTPTR_SHIFT (4) // assume 128-bit (as on arm CHERI for example) +#elif INTPTR_MAX == INT64_MAX +# define MI_INTPTR_SHIFT (3) +#elif INTPTR_MAX == INT32_MAX +# define MI_INTPTR_SHIFT (2) +#else +#error platform pointers must be 32, 64, or 128 bits +#endif + +#if SIZE_MAX == UINT64_MAX +# define MI_SIZE_SHIFT (3) +typedef int64_t mi_ssize_t; +#elif SIZE_MAX == UINT32_MAX +# define MI_SIZE_SHIFT (2) +typedef int32_t mi_ssize_t; +#else +#error platform objects must be 32 or 64 bits +#endif + +#if (SIZE_MAX/2) > LONG_MAX +# define MI_ZU(x) x##ULL +# define MI_ZI(x) x##LL +#else +# define MI_ZU(x) x##UL +# define MI_ZI(x) x##L +#endif + +#define MI_INTPTR_SIZE (1< 4 +#define MI_SEGMENT_SHIFT ( 9 + MI_SEGMENT_SLICE_SHIFT) // 32MiB +#else +#define MI_SEGMENT_SHIFT ( 7 + MI_SEGMENT_SLICE_SHIFT) // 4MiB on 32-bit +#endif +#endif + +#ifndef MI_SMALL_PAGE_SHIFT +#define MI_SMALL_PAGE_SHIFT (MI_SEGMENT_SLICE_SHIFT) // 64KiB +#endif +#ifndef MI_MEDIUM_PAGE_SHIFT +#define MI_MEDIUM_PAGE_SHIFT ( 3 + MI_SMALL_PAGE_SHIFT) // 512KiB +#endif + +// Derived constants +#define MI_SEGMENT_SIZE (MI_ZU(1)<= 655360) +#error "mimalloc internal: define more bins" +#endif + +// Maximum block size for which blocks are guaranteed to be block size aligned. (see `segment.c:_mi_segment_page_start`) +#define MI_MAX_ALIGN_GUARANTEE (MI_MEDIUM_OBJ_SIZE_MAX) + +// Alignments over MI_BLOCK_ALIGNMENT_MAX are allocated in dedicated huge page segments +#define MI_BLOCK_ALIGNMENT_MAX (MI_SEGMENT_SIZE >> 1) + +// Maximum slice count (255) for which we can find the page for interior pointers +#define MI_MAX_SLICE_OFFSET_COUNT ((MI_BLOCK_ALIGNMENT_MAX / MI_SEGMENT_SLICE_SIZE) - 1) + +// we never allocate more than PTRDIFF_MAX (see also ) +// on 64-bit+ systems we also limit the maximum allocation size such that the slice count fits in 32-bits. (issue #877) +#if (PTRDIFF_MAX > INT32_MAX) && (PTRDIFF_MAX >= (MI_SEGMENT_SLIZE_SIZE * UINT32_MAX)) +#define MI_MAX_ALLOC_SIZE (MI_SEGMENT_SLICE_SIZE * (UINT32_MAX-1)) +#else +#define MI_MAX_ALLOC_SIZE PTRDIFF_MAX +#endif + + +// ------------------------------------------------------ +// Mimalloc pages contain allocated blocks +// ------------------------------------------------------ + +// The free lists use encoded next fields +// (Only actually encodes when MI_ENCODED_FREELIST is defined.) +typedef uintptr_t mi_encoded_t; + +// thread id's +typedef size_t mi_threadid_t; + +// free lists contain blocks +typedef struct mi_block_s { + mi_encoded_t next; +} mi_block_t; + +#if MI_GUARDED +// we always align guarded pointers in a block at an offset +// the block `next` field is then used as a tag to distinguish regular offset aligned blocks from guarded ones +#define MI_BLOCK_TAG_ALIGNED ((mi_encoded_t)(0)) +#define MI_BLOCK_TAG_GUARDED (~MI_BLOCK_TAG_ALIGNED) +#endif + + +// The delayed flags are used for efficient multi-threaded free-ing +typedef enum mi_delayed_e { + MI_USE_DELAYED_FREE = 0, // push on the owning heap thread delayed list + MI_DELAYED_FREEING = 1, // temporary: another thread is accessing the owning heap + MI_NO_DELAYED_FREE = 2, // optimize: push on page local thread free queue if another block is already in the heap thread delayed free list + MI_NEVER_DELAYED_FREE = 3 // sticky: used for abandoned pages without a owning heap; this only resets on page reclaim +} mi_delayed_t; + + +// The `in_full` and `has_aligned` page flags are put in a union to efficiently +// test if both are false (`full_aligned == 0`) in the `mi_free` routine. +#if !MI_TSAN +typedef union mi_page_flags_s { + uint8_t full_aligned; + struct { + uint8_t in_full : 1; + uint8_t has_aligned : 1; + } x; +} mi_page_flags_t; +#else +// under thread sanitizer, use a byte for each flag to suppress warning, issue #130 +typedef union mi_page_flags_s { + uint32_t full_aligned; + struct { + uint8_t in_full; + uint8_t has_aligned; + } x; +} mi_page_flags_t; +#endif + +// Thread free list. +// We use the bottom 2 bits of the pointer for mi_delayed_t flags +typedef uintptr_t mi_thread_free_t; + +// A page contains blocks of one specific size (`block_size`). +// Each page has three list of free blocks: +// `free` for blocks that can be allocated, +// `local_free` for freed blocks that are not yet available to `mi_malloc` +// `thread_free` for freed blocks by other threads +// The `local_free` and `thread_free` lists are migrated to the `free` list +// when it is exhausted. The separate `local_free` list is necessary to +// implement a monotonic heartbeat. The `thread_free` list is needed for +// avoiding atomic operations in the common case. +// +// `used - |thread_free|` == actual blocks that are in use (alive) +// `used - |thread_free| + |free| + |local_free| == capacity` +// +// We don't count `freed` (as |free|) but use `used` to reduce +// the number of memory accesses in the `mi_page_all_free` function(s). +// +// Notes: +// - Access is optimized for `free.c:mi_free` and `alloc.c:mi_page_alloc` +// - Using `uint16_t` does not seem to slow things down +// - The size is 12 words on 64-bit which helps the page index calculations +// (and 14 words on 32-bit, and encoded free lists add 2 words) +// - `xthread_free` uses the bottom bits as a delayed-free flags to optimize +// concurrent frees where only the first concurrent free adds to the owning +// heap `thread_delayed_free` list (see `free.c:mi_free_block_mt`). +// The invariant is that no-delayed-free is only set if there is +// at least one block that will be added, or as already been added, to +// the owning heap `thread_delayed_free` list. This guarantees that pages +// will be freed correctly even if only other threads free blocks. +typedef struct mi_page_s { + // "owned" by the segment + uint32_t slice_count; // slices in this page (0 if not a page) + uint32_t slice_offset; // distance from the actual page data slice (0 if a page) + uint8_t is_committed:1; // `true` if the page virtual memory is committed + uint8_t is_zero_init:1; // `true` if the page was initially zero initialized + uint8_t is_huge:1; // `true` if the page is in a huge segment (`segment->kind == MI_SEGMENT_HUGE`) + // padding + // layout like this to optimize access in `mi_malloc` and `mi_free` + uint16_t capacity; // number of blocks committed, must be the first field, see `segment.c:page_clear` + uint16_t reserved; // number of blocks reserved in memory + mi_page_flags_t flags; // `in_full` and `has_aligned` flags (8 bits) + uint8_t free_is_zero:1; // `true` if the blocks in the free list are zero initialized + uint8_t retire_expire:7; // expiration count for retired blocks + + mi_block_t* free; // list of available free blocks (`malloc` allocates from this list) + mi_block_t* local_free; // list of deferred free blocks by this thread (migrates to `free`) + uint16_t used; // number of blocks in use (including blocks in `thread_free`) + uint8_t block_size_shift; // if not zero, then `(1 << block_size_shift) == block_size` (only used for fast path in `free.c:_mi_page_ptr_unalign`) + uint8_t heap_tag; // tag of the owning heap, used to separate heaps by object type + // padding + size_t block_size; // size available in each block (always `>0`) + uint8_t* page_start; // start of the page area containing the blocks + + #if (MI_ENCODE_FREELIST || MI_PADDING) + uintptr_t keys[2]; // two random keys to encode the free lists (see `_mi_block_next`) or padding canary + #endif + + _Atomic(mi_thread_free_t) xthread_free; // list of deferred free blocks freed by other threads + _Atomic(uintptr_t) xheap; + + struct mi_page_s* next; // next page owned by this thread with the same `block_size` + struct mi_page_s* prev; // previous page owned by this thread with the same `block_size` + + // 64-bit 11 words, 32-bit 13 words, (+2 for secure) + void* padding[1]; +} mi_page_t; + + + +// ------------------------------------------------------ +// Mimalloc segments contain mimalloc pages +// ------------------------------------------------------ + +typedef enum mi_page_kind_e { + MI_PAGE_SMALL, // small blocks go into 64KiB pages inside a segment + MI_PAGE_MEDIUM, // medium blocks go into 512KiB pages inside a segment + MI_PAGE_LARGE, // larger blocks go into a single page spanning a whole segment + MI_PAGE_HUGE // a huge page is a single page in a segment of variable size + // used for blocks `> MI_LARGE_OBJ_SIZE_MAX` or an aligment `> MI_BLOCK_ALIGNMENT_MAX`. +} mi_page_kind_t; + +typedef enum mi_segment_kind_e { + MI_SEGMENT_NORMAL, // MI_SEGMENT_SIZE size with pages inside. + MI_SEGMENT_HUGE, // segment with just one huge page inside. +} mi_segment_kind_t; + +// ------------------------------------------------------ +// A segment holds a commit mask where a bit is set if +// the corresponding MI_COMMIT_SIZE area is committed. +// The MI_COMMIT_SIZE must be a multiple of the slice +// size. If it is equal we have the most fine grained +// decommit (but setting it higher can be more efficient). +// The MI_MINIMAL_COMMIT_SIZE is the minimal amount that will +// be committed in one go which can be set higher than +// MI_COMMIT_SIZE for efficiency (while the decommit mask +// is still tracked in fine-grained MI_COMMIT_SIZE chunks) +// ------------------------------------------------------ + +#define MI_MINIMAL_COMMIT_SIZE (1*MI_SEGMENT_SLICE_SIZE) +#define MI_COMMIT_SIZE (MI_SEGMENT_SLICE_SIZE) // 64KiB +#define MI_COMMIT_MASK_BITS (MI_SEGMENT_SIZE / MI_COMMIT_SIZE) +#define MI_COMMIT_MASK_FIELD_BITS MI_SIZE_BITS +#define MI_COMMIT_MASK_FIELD_COUNT (MI_COMMIT_MASK_BITS / MI_COMMIT_MASK_FIELD_BITS) + +#if (MI_COMMIT_MASK_BITS != (MI_COMMIT_MASK_FIELD_COUNT * MI_COMMIT_MASK_FIELD_BITS)) +#error "the segment size must be exactly divisible by the (commit size * size_t bits)" +#endif + +typedef struct mi_commit_mask_s { + size_t mask[MI_COMMIT_MASK_FIELD_COUNT]; +} mi_commit_mask_t; + +typedef mi_page_t mi_slice_t; +typedef int64_t mi_msecs_t; + + +// --------------------------------------------------------------- +// a memory id tracks the provenance of arena/OS allocated memory +// --------------------------------------------------------------- + +// Memory can reside in arena's, direct OS allocated, or statically allocated. The memid keeps track of this. +typedef enum mi_memkind_e { + MI_MEM_NONE, // not allocated + MI_MEM_EXTERNAL, // not owned by mimalloc but provided externally (via `mi_manage_os_memory` for example) + MI_MEM_STATIC, // allocated in a static area and should not be freed (for arena meta data for example) + MI_MEM_OS, // allocated from the OS + MI_MEM_OS_HUGE, // allocated as huge OS pages (usually 1GiB, pinned to physical memory) + MI_MEM_OS_REMAP, // allocated in a remapable area (i.e. using `mremap`) + MI_MEM_ARENA // allocated from an arena (the usual case) +} mi_memkind_t; + +static inline bool mi_memkind_is_os(mi_memkind_t memkind) { + return (memkind >= MI_MEM_OS && memkind <= MI_MEM_OS_REMAP); +} + +typedef struct mi_memid_os_info { + void* base; // actual base address of the block (used for offset aligned allocations) + size_t size; // full allocation size +} mi_memid_os_info_t; + +typedef struct mi_memid_arena_info { + size_t block_index; // index in the arena + mi_arena_id_t id; // arena id (>= 1) + bool is_exclusive; // this arena can only be used for specific arena allocations +} mi_memid_arena_info_t; + +typedef struct mi_memid_s { + union { + mi_memid_os_info_t os; // only used for MI_MEM_OS + mi_memid_arena_info_t arena; // only used for MI_MEM_ARENA + } mem; + bool is_pinned; // `true` if we cannot decommit/reset/protect in this memory (e.g. when allocated using large (2Mib) or huge (1GiB) OS pages) + bool initially_committed;// `true` if the memory was originally allocated as committed + bool initially_zero; // `true` if the memory was originally zero initialized + mi_memkind_t memkind; +} mi_memid_t; + + +// ----------------------------------------------------------------------------------------- +// Segments are large allocated memory blocks (32mb on 64 bit) from arenas or the OS. +// +// Inside segments we allocated fixed size mimalloc pages (`mi_page_t`) that contain blocks. +// The start of a segment is this structure with a fixed number of slice entries (`slices`) +// usually followed by a guard OS page and the actual allocation area with pages. +// While a page is not allocated, we view it's data as a `mi_slice_t` (instead of a `mi_page_t`). +// Of any free area, the first slice has the info and `slice_offset == 0`; for any subsequent +// slices part of the area, the `slice_offset` is the byte offset back to the first slice +// (so we can quickly find the page info on a free, `internal.h:_mi_segment_page_of`). +// For slices, the `block_size` field is repurposed to signify if a slice is used (`1`) or not (`0`). +// Small and medium pages use a fixed amount of slices to reduce slice fragmentation, while +// large and huge pages span a variable amount of slices. + +typedef struct mi_subproc_s mi_subproc_t; + +typedef struct mi_segment_s { + // constant fields + mi_memid_t memid; // memory id for arena/OS allocation + bool allow_decommit; // can we decommmit the memory + bool allow_purge; // can we purge the memory (reset or decommit) + size_t segment_size; + mi_subproc_t* subproc; // segment belongs to sub process + + // segment fields + mi_msecs_t purge_expire; // purge slices in the `purge_mask` after this time + mi_commit_mask_t purge_mask; // slices that can be purged + mi_commit_mask_t commit_mask; // slices that are currently committed + + // from here is zero initialized + struct mi_segment_s* next; // the list of freed segments in the cache (must be first field, see `segment.c:mi_segment_init`) + bool was_reclaimed; // true if it was reclaimed (used to limit on-free reclamation) + bool dont_free; // can be temporarily true to ensure the segment is not freed + + size_t abandoned; // abandoned pages (i.e. the original owning thread stopped) (`abandoned <= used`) + size_t abandoned_visits; // count how often this segment is visited during abondoned reclamation (to force reclaim if it takes too long) + size_t used; // count of pages in use + uintptr_t cookie; // verify addresses in debug mode: `mi_ptr_cookie(segment) == segment->cookie` + + struct mi_segment_s* abandoned_os_next; // only used for abandoned segments outside arena's, and only if `mi_option_visit_abandoned` is enabled + struct mi_segment_s* abandoned_os_prev; + + size_t segment_slices; // for huge segments this may be different from `MI_SLICES_PER_SEGMENT` + size_t segment_info_slices; // initial count of slices that we are using for segment info and possible guard pages. + + // layout like this to optimize access in `mi_free` + mi_segment_kind_t kind; + size_t slice_entries; // entries in the `slices` array, at most `MI_SLICES_PER_SEGMENT` + _Atomic(mi_threadid_t) thread_id; // unique id of the thread owning this segment + + mi_slice_t slices[MI_SLICES_PER_SEGMENT+1]; // one extra final entry for huge blocks with large alignment +} mi_segment_t; + + +// ------------------------------------------------------ +// Heaps +// Provide first-class heaps to allocate from. +// A heap just owns a set of pages for allocation and +// can only be allocate/reallocate from the thread that created it. +// Freeing blocks can be done from any thread though. +// Per thread, the segments are shared among its heaps. +// Per thread, there is always a default heap that is +// used for allocation; it is initialized to statically +// point to an empty heap to avoid initialization checks +// in the fast path. +// ------------------------------------------------------ + +// Thread local data +typedef struct mi_tld_s mi_tld_t; + +// Pages of a certain block size are held in a queue. +typedef struct mi_page_queue_s { + mi_page_t* first; + mi_page_t* last; + size_t block_size; +} mi_page_queue_t; + +#define MI_BIN_FULL (MI_BIN_HUGE+1) + +// Random context +typedef struct mi_random_cxt_s { + uint32_t input[16]; + uint32_t output[16]; + int output_available; + bool weak; +} mi_random_ctx_t; + + +// In debug mode there is a padding structure at the end of the blocks to check for buffer overflows +#if (MI_PADDING) +typedef struct mi_padding_s { + uint32_t canary; // encoded block value to check validity of the padding (in case of overflow) + uint32_t delta; // padding bytes before the block. (mi_usable_size(p) - delta == exact allocated bytes) +} mi_padding_t; +#define MI_PADDING_SIZE (sizeof(mi_padding_t)) +#define MI_PADDING_WSIZE ((MI_PADDING_SIZE + MI_INTPTR_SIZE - 1) / MI_INTPTR_SIZE) +#else +#define MI_PADDING_SIZE 0 +#define MI_PADDING_WSIZE 0 +#endif + +#define MI_PAGES_DIRECT (MI_SMALL_WSIZE_MAX + MI_PADDING_WSIZE + 1) + + +// A heap owns a set of pages. +struct mi_heap_s { + mi_tld_t* tld; + _Atomic(mi_block_t*) thread_delayed_free; + mi_threadid_t thread_id; // thread this heap belongs too + mi_arena_id_t arena_id; // arena id if the heap belongs to a specific arena (or 0) + uintptr_t cookie; // random cookie to verify pointers (see `_mi_ptr_cookie`) + uintptr_t keys[2]; // two random keys used to encode the `thread_delayed_free` list + mi_random_ctx_t random; // random number context used for secure allocation + size_t page_count; // total number of pages in the `pages` queues. + size_t page_retired_min; // smallest retired index (retired pages are fully free, but still in the page queues) + size_t page_retired_max; // largest retired index into the `pages` array. + long generic_count; // how often is `_mi_malloc_generic` called? + long generic_collect_count; // how often is `_mi_malloc_generic` called without collecting? + mi_heap_t* next; // list of heaps per thread + bool no_reclaim; // `true` if this heap should not reclaim abandoned pages + uint8_t tag; // custom tag, can be used for separating heaps based on the object types + #if MI_GUARDED + size_t guarded_size_min; // minimal size for guarded objects + size_t guarded_size_max; // maximal size for guarded objects + size_t guarded_sample_rate; // sample rate (set to 0 to disable guarded pages) + size_t guarded_sample_count; // current sample count (counting down to 0) + #endif + mi_page_t* pages_free_direct[MI_PAGES_DIRECT]; // optimize: array where every entry points a page with possibly free blocks in the corresponding queue for that size. + mi_page_queue_t pages[MI_BIN_FULL + 1]; // queue of pages for each size class (or "bin") +}; + + +// ------------------------------------------------------ +// Sub processes do not reclaim or visit segments +// from other sub processes. These are essentially the +// static variables of a process. +// ------------------------------------------------------ + +struct mi_subproc_s { + _Atomic(size_t) abandoned_count; // count of abandoned segments for this sub-process + _Atomic(size_t) abandoned_os_list_count; // count of abandoned segments in the os-list + mi_lock_t abandoned_os_lock; // lock for the abandoned os segment list (outside of arena's) (this lock protect list operations) + mi_lock_t abandoned_os_visit_lock; // ensure only one thread per subproc visits the abandoned os list + mi_segment_t* abandoned_os_list; // doubly-linked list of abandoned segments outside of arena's (in OS allocated memory) + mi_segment_t* abandoned_os_list_tail; // the tail-end of the list + mi_memid_t memid; // provenance of this memory block +}; + + +// ------------------------------------------------------ +// Thread Local data +// ------------------------------------------------------ + +// A "span" is is an available range of slices. The span queues keep +// track of slice spans of at most the given `slice_count` (but more than the previous size class). +typedef struct mi_span_queue_s { + mi_slice_t* first; + mi_slice_t* last; + size_t slice_count; +} mi_span_queue_t; + +#define MI_SEGMENT_BIN_MAX (35) // 35 == mi_segment_bin(MI_SLICES_PER_SEGMENT) + +// Segments thread local data +typedef struct mi_segments_tld_s { + mi_span_queue_t spans[MI_SEGMENT_BIN_MAX+1]; // free slice spans inside segments + size_t count; // current number of segments; + size_t peak_count; // peak number of segments + size_t current_size; // current size of all segments + size_t peak_size; // peak size of all segments + size_t reclaim_count;// number of reclaimed (abandoned) segments + mi_subproc_t* subproc; // sub-process this thread belongs to. + mi_stats_t* stats; // points to tld stats +} mi_segments_tld_t; + +// Thread local data +struct mi_tld_s { + unsigned long long heartbeat; // monotonic heartbeat count + bool recurse; // true if deferred was called; used to prevent infinite recursion. + mi_heap_t* heap_backing; // backing heap of this thread (cannot be deleted) + mi_heap_t* heaps; // list of heaps in this thread (so we can abandon all when the thread terminates) + mi_segments_tld_t segments; // segment tld + mi_stats_t stats; // statistics +}; + + +// ------------------------------------------------------ +// Debug +// ------------------------------------------------------ + +#if !defined(MI_DEBUG_UNINIT) +#define MI_DEBUG_UNINIT (0xD0) +#endif +#if !defined(MI_DEBUG_FREED) +#define MI_DEBUG_FREED (0xDF) +#endif +#if !defined(MI_DEBUG_PADDING) +#define MI_DEBUG_PADDING (0xDE) +#endif + + +// ------------------------------------------------------ +// Statistics +// ------------------------------------------------------ +#ifndef MI_STAT +#if (MI_DEBUG>0) +#define MI_STAT 2 +#else +#define MI_STAT 0 +#endif +#endif + +// add to stat keeping track of the peak +void _mi_stat_increase(mi_stat_count_t* stat, size_t amount); +void _mi_stat_decrease(mi_stat_count_t* stat, size_t amount); +void _mi_stat_adjust_decrease(mi_stat_count_t* stat, size_t amount); +// counters can just be increased +void _mi_stat_counter_increase(mi_stat_counter_t* stat, size_t amount); + +#if (MI_STAT) +#define mi_stat_increase(stat,amount) _mi_stat_increase( &(stat), amount) +#define mi_stat_decrease(stat,amount) _mi_stat_decrease( &(stat), amount) +#define mi_stat_adjust_decrease(stat,amount) _mi_stat_adjust_decrease( &(stat), amount) +#define mi_stat_counter_increase(stat,amount) _mi_stat_counter_increase( &(stat), amount) +#else +#define mi_stat_increase(stat,amount) ((void)0) +#define mi_stat_decrease(stat,amount) ((void)0) +#define mi_stat_adjust_decrease(stat,amount) ((void)0) +#define mi_stat_counter_increase(stat,amount) ((void)0) +#endif + +#define mi_heap_stat_counter_increase(heap,stat,amount) mi_stat_counter_increase( (heap)->tld->stats.stat, amount) +#define mi_heap_stat_increase(heap,stat,amount) mi_stat_increase( (heap)->tld->stats.stat, amount) +#define mi_heap_stat_decrease(heap,stat,amount) mi_stat_decrease( (heap)->tld->stats.stat, amount) +#define mi_heap_stat_adjust_decrease(heap,stat,amount) mi_stat_adjust_decrease( (heap)->tld->stats.stat, amount) + +#endif diff --git a/compat/mimalloc/options.c b/compat/mimalloc/options.c new file mode 100644 index 00000000000000..af2a0e70c4c7ac --- /dev/null +++ b/compat/mimalloc/options.c @@ -0,0 +1,663 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2021, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "mimalloc/prim.h" // mi_prim_out_stderr + +#include // stdin/stdout +#include // abort + + + +static long mi_max_error_count = 16; // stop outputting errors after this (use < 0 for no limit) +static long mi_max_warning_count = 16; // stop outputting warnings after this (use < 0 for no limit) + +static void mi_add_stderr_output(void); + +int mi_version(void) mi_attr_noexcept { + return MI_MALLOC_VERSION; +} + + +// -------------------------------------------------------- +// Options +// These can be accessed by multiple threads and may be +// concurrently initialized, but an initializing data race +// is ok since they resolve to the same value. +// -------------------------------------------------------- +typedef enum mi_init_e { + UNINIT, // not yet initialized + DEFAULTED, // not found in the environment, use default value + INITIALIZED // found in environment or set explicitly +} mi_init_t; + +typedef struct mi_option_desc_s { + long value; // the value + mi_init_t init; // is it initialized yet? (from the environment) + mi_option_t option; // for debugging: the option index should match the option + const char* name; // option name without `mimalloc_` prefix + const char* legacy_name; // potential legacy option name +} mi_option_desc_t; + +#define MI_OPTION(opt) mi_option_##opt, #opt, NULL +#define MI_OPTION_LEGACY(opt,legacy) mi_option_##opt, #opt, #legacy + +// Some options can be set at build time for statically linked libraries +// (use `-DMI_EXTRA_CPPDEFS="opt1=val1;opt2=val2"`) +// +// This is useful if we cannot pass them as environment variables +// (and setting them programmatically would be too late) + +#ifndef MI_DEFAULT_VERBOSE +#define MI_DEFAULT_VERBOSE 0 +#endif + +#ifndef MI_DEFAULT_EAGER_COMMIT +#define MI_DEFAULT_EAGER_COMMIT 1 +#endif + +#ifndef MI_DEFAULT_ARENA_EAGER_COMMIT +#define MI_DEFAULT_ARENA_EAGER_COMMIT 2 +#endif + +// in KiB +#ifndef MI_DEFAULT_ARENA_RESERVE + #if (MI_INTPTR_SIZE>4) + #define MI_DEFAULT_ARENA_RESERVE 1024L*1024L + #else + #define MI_DEFAULT_ARENA_RESERVE 128L*1024L + #endif +#endif + +#ifndef MI_DEFAULT_DISALLOW_ARENA_ALLOC +#define MI_DEFAULT_DISALLOW_ARENA_ALLOC 0 +#endif + +#ifndef MI_DEFAULT_ALLOW_LARGE_OS_PAGES +#if defined(__linux__) && !defined(__ANDROID__) +#define MI_DEFAULT_ALLOW_LARGE_OS_PAGES 2 // enabled, but only use transparent huge pages through madvise +#else +#define MI_DEFAULT_ALLOW_LARGE_OS_PAGES 0 +#endif +#endif + +#ifndef MI_DEFAULT_RESERVE_HUGE_OS_PAGES +#define MI_DEFAULT_RESERVE_HUGE_OS_PAGES 0 +#endif + +#ifndef MI_DEFAULT_RESERVE_OS_MEMORY +#define MI_DEFAULT_RESERVE_OS_MEMORY 0 +#endif + +#ifndef MI_DEFAULT_GUARDED_SAMPLE_RATE +#if MI_GUARDED +#define MI_DEFAULT_GUARDED_SAMPLE_RATE 4000 +#else +#define MI_DEFAULT_GUARDED_SAMPLE_RATE 0 +#endif +#endif + + +static mi_option_desc_t options[_mi_option_last] = +{ + // stable options + #if MI_DEBUG || defined(MI_SHOW_ERRORS) + { 1, UNINIT, MI_OPTION(show_errors) }, + #else + { 0, UNINIT, MI_OPTION(show_errors) }, + #endif + { 0, UNINIT, MI_OPTION(show_stats) }, + { MI_DEFAULT_VERBOSE, UNINIT, MI_OPTION(verbose) }, + + // some of the following options are experimental and not all combinations are allowed. + { MI_DEFAULT_EAGER_COMMIT, + UNINIT, MI_OPTION(eager_commit) }, // commit per segment directly (4MiB) (but see also `eager_commit_delay`) + { MI_DEFAULT_ARENA_EAGER_COMMIT, + UNINIT, MI_OPTION_LEGACY(arena_eager_commit,eager_region_commit) }, // eager commit arena's? 2 is used to enable this only on an OS that has overcommit (i.e. linux) + { 1, UNINIT, MI_OPTION_LEGACY(purge_decommits,reset_decommits) }, // purge decommits memory (instead of reset) (note: on linux this uses MADV_DONTNEED for decommit) + { MI_DEFAULT_ALLOW_LARGE_OS_PAGES, + UNINIT, MI_OPTION_LEGACY(allow_large_os_pages,large_os_pages) }, // use large OS pages, use only with eager commit to prevent fragmentation of VMA's + { MI_DEFAULT_RESERVE_HUGE_OS_PAGES, + UNINIT, MI_OPTION(reserve_huge_os_pages) }, // per 1GiB huge pages + {-1, UNINIT, MI_OPTION(reserve_huge_os_pages_at) }, // reserve huge pages at node N + { MI_DEFAULT_RESERVE_OS_MEMORY, + UNINIT, MI_OPTION(reserve_os_memory) }, // reserve N KiB OS memory in advance (use `option_get_size`) + { 0, UNINIT, MI_OPTION(deprecated_segment_cache) }, // cache N segments per thread + { 0, UNINIT, MI_OPTION(deprecated_page_reset) }, // reset page memory on free + { 0, UNINIT, MI_OPTION_LEGACY(abandoned_page_purge,abandoned_page_reset) }, // reset free page memory when a thread terminates + { 0, UNINIT, MI_OPTION(deprecated_segment_reset) }, // reset segment memory on free (needs eager commit) +#if defined(__NetBSD__) + { 0, UNINIT, MI_OPTION(eager_commit_delay) }, // the first N segments per thread are not eagerly committed +#else + { 1, UNINIT, MI_OPTION(eager_commit_delay) }, // the first N segments per thread are not eagerly committed (but per page in the segment on demand) +#endif + { 10, UNINIT, MI_OPTION_LEGACY(purge_delay,reset_delay) }, // purge delay in milli-seconds + { 0, UNINIT, MI_OPTION(use_numa_nodes) }, // 0 = use available numa nodes, otherwise use at most N nodes. + { 0, UNINIT, MI_OPTION_LEGACY(disallow_os_alloc,limit_os_alloc) }, // 1 = do not use OS memory for allocation (but only reserved arenas) + { 100, UNINIT, MI_OPTION(os_tag) }, // only apple specific for now but might serve more or less related purpose + { 32, UNINIT, MI_OPTION(max_errors) }, // maximum errors that are output + { 32, UNINIT, MI_OPTION(max_warnings) }, // maximum warnings that are output + { 10, UNINIT, MI_OPTION(max_segment_reclaim)}, // max. percentage of the abandoned segments to be reclaimed per try. + { 0, UNINIT, MI_OPTION(destroy_on_exit)}, // release all OS memory on process exit; careful with dangling pointer or after-exit frees! + { MI_DEFAULT_ARENA_RESERVE, UNINIT, MI_OPTION(arena_reserve) }, // reserve memory N KiB at a time (=1GiB) (use `option_get_size`) + { 10, UNINIT, MI_OPTION(arena_purge_mult) }, // purge delay multiplier for arena's + { 1, UNINIT, MI_OPTION_LEGACY(purge_extend_delay, decommit_extend_delay) }, + { 0, UNINIT, MI_OPTION(abandoned_reclaim_on_free) },// reclaim an abandoned segment on a free + { MI_DEFAULT_DISALLOW_ARENA_ALLOC, UNINIT, MI_OPTION(disallow_arena_alloc) }, // 1 = do not use arena's for allocation (except if using specific arena id's) + { 400, UNINIT, MI_OPTION(retry_on_oom) }, // windows only: retry on out-of-memory for N milli seconds (=400), set to 0 to disable retries. +#if defined(MI_VISIT_ABANDONED) + { 1, INITIALIZED, MI_OPTION(visit_abandoned) }, // allow visiting heap blocks in abandoned segments; requires taking locks during reclaim. +#else + { 0, UNINIT, MI_OPTION(visit_abandoned) }, +#endif + { 0, UNINIT, MI_OPTION(guarded_min) }, // only used when building with MI_GUARDED: minimal rounded object size for guarded objects + { MI_GiB, UNINIT, MI_OPTION(guarded_max) }, // only used when building with MI_GUARDED: maximal rounded object size for guarded objects + { 0, UNINIT, MI_OPTION(guarded_precise) }, // disregard minimal alignment requirement to always place guarded blocks exactly in front of a guard page (=0) + { MI_DEFAULT_GUARDED_SAMPLE_RATE, + UNINIT, MI_OPTION(guarded_sample_rate)}, // 1 out of N allocations in the min/max range will be guarded (=4000) + { 0, UNINIT, MI_OPTION(guarded_sample_seed)}, + { 0, UNINIT, MI_OPTION(target_segments_per_thread) }, // abandon segments beyond this point, or 0 to disable. + { 10000, UNINIT, MI_OPTION(generic_collect) }, // collect heaps every N (=10000) generic allocation calls +}; + +static void mi_option_init(mi_option_desc_t* desc); + +static bool mi_option_has_size_in_kib(mi_option_t option) { + return (option == mi_option_reserve_os_memory || option == mi_option_arena_reserve); +} + +void _mi_options_init(void) { + // called on process load + mi_add_stderr_output(); // now it safe to use stderr for output + for(int i = 0; i < _mi_option_last; i++ ) { + mi_option_t option = (mi_option_t)i; + long l = mi_option_get(option); MI_UNUSED(l); // initialize + } + mi_max_error_count = mi_option_get(mi_option_max_errors); + mi_max_warning_count = mi_option_get(mi_option_max_warnings); + #if MI_GUARDED + if (mi_option_get(mi_option_guarded_sample_rate) > 0) { + if (mi_option_is_enabled(mi_option_allow_large_os_pages)) { + mi_option_disable(mi_option_allow_large_os_pages); + _mi_warning_message("option 'allow_large_os_pages' is disabled to allow for guarded objects\n"); + } + } + #endif + if (mi_option_is_enabled(mi_option_verbose)) { mi_options_print(); } +} + +#define mi_stringifyx(str) #str // and stringify +#define mi_stringify(str) mi_stringifyx(str) // expand + +void mi_options_print(void) mi_attr_noexcept +{ + // show version + const int vermajor = MI_MALLOC_VERSION/100; + const int verminor = (MI_MALLOC_VERSION%100)/10; + const int verpatch = (MI_MALLOC_VERSION%10); + _mi_message("v%i.%i.%i%s%s (built on %s, %s)\n", vermajor, verminor, verpatch, + #if defined(MI_CMAKE_BUILD_TYPE) + ", " mi_stringify(MI_CMAKE_BUILD_TYPE) + #else + "" + #endif + , + #if defined(MI_GIT_DESCRIBE) + ", git " mi_stringify(MI_GIT_DESCRIBE) + #else + "" + #endif + , __DATE__, __TIME__); + + // show options + for (int i = 0; i < _mi_option_last; i++) { + mi_option_t option = (mi_option_t)i; + long l = mi_option_get(option); MI_UNUSED(l); // possibly initialize + mi_option_desc_t* desc = &options[option]; + _mi_message("option '%s': %ld %s\n", desc->name, desc->value, (mi_option_has_size_in_kib(option) ? "KiB" : "")); + } + + // show build configuration + _mi_message("debug level : %d\n", MI_DEBUG ); + _mi_message("secure level: %d\n", MI_SECURE ); + _mi_message("mem tracking: %s\n", MI_TRACK_TOOL); + #if MI_GUARDED + _mi_message("guarded build: %s\n", mi_option_get(mi_option_guarded_sample_rate) != 0 ? "enabled" : "disabled"); + #endif + #if MI_TSAN + _mi_message("thread santizer enabled\n"); + #endif +} + +long _mi_option_get_fast(mi_option_t option) { + mi_assert(option >= 0 && option < _mi_option_last); + mi_option_desc_t* desc = &options[option]; + mi_assert(desc->option == option); // index should match the option + //mi_assert(desc->init != UNINIT); + return desc->value; +} + + +mi_decl_nodiscard long mi_option_get(mi_option_t option) { + mi_assert(option >= 0 && option < _mi_option_last); + if (option < 0 || option >= _mi_option_last) return 0; + mi_option_desc_t* desc = &options[option]; + mi_assert(desc->option == option); // index should match the option + if mi_unlikely(desc->init == UNINIT) { + mi_option_init(desc); + } + return desc->value; +} + +mi_decl_nodiscard long mi_option_get_clamp(mi_option_t option, long min, long max) { + long x = mi_option_get(option); + return (x < min ? min : (x > max ? max : x)); +} + +mi_decl_nodiscard size_t mi_option_get_size(mi_option_t option) { + const long x = mi_option_get(option); + size_t size = (x < 0 ? 0 : (size_t)x); + if (mi_option_has_size_in_kib(option)) { + size *= MI_KiB; + } + return size; +} + +void mi_option_set(mi_option_t option, long value) { + mi_assert(option >= 0 && option < _mi_option_last); + if (option < 0 || option >= _mi_option_last) return; + mi_option_desc_t* desc = &options[option]; + mi_assert(desc->option == option); // index should match the option + desc->value = value; + desc->init = INITIALIZED; + // ensure min/max range; be careful to not recurse. + if (desc->option == mi_option_guarded_min && _mi_option_get_fast(mi_option_guarded_max) < value) { + mi_option_set(mi_option_guarded_max, value); + } + else if (desc->option == mi_option_guarded_max && _mi_option_get_fast(mi_option_guarded_min) > value) { + mi_option_set(mi_option_guarded_min, value); + } +} + +void mi_option_set_default(mi_option_t option, long value) { + mi_assert(option >= 0 && option < _mi_option_last); + if (option < 0 || option >= _mi_option_last) return; + mi_option_desc_t* desc = &options[option]; + if (desc->init != INITIALIZED) { + desc->value = value; + } +} + +mi_decl_nodiscard bool mi_option_is_enabled(mi_option_t option) { + return (mi_option_get(option) != 0); +} + +void mi_option_set_enabled(mi_option_t option, bool enable) { + mi_option_set(option, (enable ? 1 : 0)); +} + +void mi_option_set_enabled_default(mi_option_t option, bool enable) { + mi_option_set_default(option, (enable ? 1 : 0)); +} + +void mi_option_enable(mi_option_t option) { + mi_option_set_enabled(option,true); +} + +void mi_option_disable(mi_option_t option) { + mi_option_set_enabled(option,false); +} + +static void mi_cdecl mi_out_stderr(const char* msg, void* arg) { + MI_UNUSED(arg); + if (msg != NULL && msg[0] != 0) { + _mi_prim_out_stderr(msg); + } +} + +// Since an output function can be registered earliest in the `main` +// function we also buffer output that happens earlier. When +// an output function is registered it is called immediately with +// the output up to that point. +#ifndef MI_MAX_DELAY_OUTPUT +#define MI_MAX_DELAY_OUTPUT ((size_t)(16*1024)) +#endif +static char out_buf[MI_MAX_DELAY_OUTPUT+1]; +static _Atomic(size_t) out_len; + +static void mi_cdecl mi_out_buf(const char* msg, void* arg) { + MI_UNUSED(arg); + if (msg==NULL) return; + if (mi_atomic_load_relaxed(&out_len)>=MI_MAX_DELAY_OUTPUT) return; + size_t n = _mi_strlen(msg); + if (n==0) return; + // claim space + size_t start = mi_atomic_add_acq_rel(&out_len, n); + if (start >= MI_MAX_DELAY_OUTPUT) return; + // check bound + if (start+n >= MI_MAX_DELAY_OUTPUT) { + n = MI_MAX_DELAY_OUTPUT-start-1; + } + _mi_memcpy(&out_buf[start], msg, n); +} + +static void mi_out_buf_flush(mi_output_fun* out, bool no_more_buf, void* arg) { + if (out==NULL) return; + // claim (if `no_more_buf == true`, no more output will be added after this point) + size_t count = mi_atomic_add_acq_rel(&out_len, (no_more_buf ? MI_MAX_DELAY_OUTPUT : 1)); + // and output the current contents + if (count>MI_MAX_DELAY_OUTPUT) count = MI_MAX_DELAY_OUTPUT; + out_buf[count] = 0; + out(out_buf,arg); + if (!no_more_buf) { + out_buf[count] = '\n'; // if continue with the buffer, insert a newline + } +} + + +// Once this module is loaded, switch to this routine +// which outputs to stderr and the delayed output buffer. +static void mi_cdecl mi_out_buf_stderr(const char* msg, void* arg) { + mi_out_stderr(msg,arg); + mi_out_buf(msg,arg); +} + + + +// -------------------------------------------------------- +// Default output handler +// -------------------------------------------------------- + +// Should be atomic but gives errors on many platforms as generally we cannot cast a function pointer to a uintptr_t. +// For now, don't register output from multiple threads. +static mi_output_fun* volatile mi_out_default; // = NULL +static _Atomic(void*) mi_out_arg; // = NULL + +static mi_output_fun* mi_out_get_default(void** parg) { + if (parg != NULL) { *parg = mi_atomic_load_ptr_acquire(void,&mi_out_arg); } + mi_output_fun* out = mi_out_default; + return (out == NULL ? &mi_out_buf : out); +} + +void mi_register_output(mi_output_fun* out, void* arg) mi_attr_noexcept { + mi_out_default = (out == NULL ? &mi_out_stderr : out); // stop using the delayed output buffer + mi_atomic_store_ptr_release(void,&mi_out_arg, arg); + if (out!=NULL) mi_out_buf_flush(out,true,arg); // output all the delayed output now +} + +// add stderr to the delayed output after the module is loaded +static void mi_add_stderr_output(void) { + mi_assert_internal(mi_out_default == NULL); + mi_out_buf_flush(&mi_out_stderr, false, NULL); // flush current contents to stderr + mi_out_default = &mi_out_buf_stderr; // and add stderr to the delayed output +} + +// -------------------------------------------------------- +// Messages, all end up calling `_mi_fputs`. +// -------------------------------------------------------- +static _Atomic(size_t) error_count; // = 0; // when >= max_error_count stop emitting errors +static _Atomic(size_t) warning_count; // = 0; // when >= max_warning_count stop emitting warnings + +// When overriding malloc, we may recurse into mi_vfprintf if an allocation +// inside the C runtime causes another message. +// In some cases (like on macOS) the loader already allocates which +// calls into mimalloc; if we then access thread locals (like `recurse`) +// this may crash as the access may call _tlv_bootstrap that tries to +// (recursively) invoke malloc again to allocate space for the thread local +// variables on demand. This is why we use a _mi_preloading test on such +// platforms. However, C code generator may move the initial thread local address +// load before the `if` and we therefore split it out in a separate function. +static mi_decl_thread bool recurse = false; + +static mi_decl_noinline bool mi_recurse_enter_prim(void) { + if (recurse) return false; + recurse = true; + return true; +} + +static mi_decl_noinline void mi_recurse_exit_prim(void) { + recurse = false; +} + +static bool mi_recurse_enter(void) { + #if defined(__APPLE__) || defined(__ANDROID__) || defined(MI_TLS_RECURSE_GUARD) + if (_mi_preloading()) return false; + #endif + return mi_recurse_enter_prim(); +} + +static void mi_recurse_exit(void) { + #if defined(__APPLE__) || defined(__ANDROID__) || defined(MI_TLS_RECURSE_GUARD) + if (_mi_preloading()) return; + #endif + mi_recurse_exit_prim(); +} + +void _mi_fputs(mi_output_fun* out, void* arg, const char* prefix, const char* message) { + if (out==NULL || (void*)out==(void*)stdout || (void*)out==(void*)stderr) { // TODO: use mi_out_stderr for stderr? + if (!mi_recurse_enter()) return; + out = mi_out_get_default(&arg); + if (prefix != NULL) out(prefix, arg); + out(message, arg); + mi_recurse_exit(); + } + else { + if (prefix != NULL) out(prefix, arg); + out(message, arg); + } +} + +// Define our own limited `fprintf` that avoids memory allocation. +// We do this using `_mi_vsnprintf` with a limited buffer. +static void mi_vfprintf( mi_output_fun* out, void* arg, const char* prefix, const char* fmt, va_list args ) { + char buf[512]; + if (fmt==NULL) return; + if (!mi_recurse_enter()) return; + _mi_vsnprintf(buf, sizeof(buf)-1, fmt, args); + mi_recurse_exit(); + _mi_fputs(out,arg,prefix,buf); +} + +void _mi_fprintf( mi_output_fun* out, void* arg, const char* fmt, ... ) { + va_list args; + va_start(args,fmt); + mi_vfprintf(out,arg,NULL,fmt,args); + va_end(args); +} + +static void mi_vfprintf_thread(mi_output_fun* out, void* arg, const char* prefix, const char* fmt, va_list args) { + if (prefix != NULL && _mi_strnlen(prefix,33) <= 32 && !_mi_is_main_thread()) { + char tprefix[64]; + _mi_snprintf(tprefix, sizeof(tprefix), "%sthread 0x%tx: ", prefix, (uintptr_t)_mi_thread_id()); + mi_vfprintf(out, arg, tprefix, fmt, args); + } + else { + mi_vfprintf(out, arg, prefix, fmt, args); + } +} + +void _mi_message(const char* fmt, ...) { + va_list args; + va_start(args, fmt); + mi_vfprintf_thread(NULL, NULL, "mimalloc: ", fmt, args); + va_end(args); +} + +void _mi_trace_message(const char* fmt, ...) { + if (mi_option_get(mi_option_verbose) <= 1) return; // only with verbose level 2 or higher + va_list args; + va_start(args, fmt); + mi_vfprintf_thread(NULL, NULL, "mimalloc: ", fmt, args); + va_end(args); +} + +void _mi_verbose_message(const char* fmt, ...) { + if (!mi_option_is_enabled(mi_option_verbose)) return; + va_list args; + va_start(args,fmt); + mi_vfprintf(NULL, NULL, "mimalloc: ", fmt, args); + va_end(args); +} + +static void mi_show_error_message(const char* fmt, va_list args) { + if (!mi_option_is_enabled(mi_option_verbose)) { + if (!mi_option_is_enabled(mi_option_show_errors)) return; + if (mi_max_error_count >= 0 && (long)mi_atomic_increment_acq_rel(&error_count) > mi_max_error_count) return; + } + mi_vfprintf_thread(NULL, NULL, "mimalloc: error: ", fmt, args); +} + +void _mi_warning_message(const char* fmt, ...) { + if (!mi_option_is_enabled(mi_option_verbose)) { + if (!mi_option_is_enabled(mi_option_show_errors)) return; + if (mi_max_warning_count >= 0 && (long)mi_atomic_increment_acq_rel(&warning_count) > mi_max_warning_count) return; + } + va_list args; + va_start(args,fmt); + mi_vfprintf_thread(NULL, NULL, "mimalloc: warning: ", fmt, args); + va_end(args); +} + + +#if MI_DEBUG +mi_decl_noreturn mi_decl_cold void _mi_assert_fail(const char* assertion, const char* fname, unsigned line, const char* func ) mi_attr_noexcept { + _mi_fprintf(NULL, NULL, "mimalloc: assertion failed: at \"%s\":%u, %s\n assertion: \"%s\"\n", fname, line, (func==NULL?"":func), assertion); + abort(); +} +#endif + +// -------------------------------------------------------- +// Errors +// -------------------------------------------------------- + +static mi_error_fun* volatile mi_error_handler; // = NULL +static _Atomic(void*) mi_error_arg; // = NULL + +static void mi_error_default(int err) { + MI_UNUSED(err); +#if (MI_DEBUG>0) + if (err==EFAULT) { + #ifdef _MSC_VER + __debugbreak(); + #endif + abort(); + } +#endif +#if (MI_SECURE>0) + if (err==EFAULT) { // abort on serious errors in secure mode (corrupted meta-data) + abort(); + } +#endif +#if defined(MI_XMALLOC) + if (err==ENOMEM || err==EOVERFLOW) { // abort on memory allocation fails in xmalloc mode + abort(); + } +#endif +} + +void mi_register_error(mi_error_fun* fun, void* arg) { + mi_error_handler = fun; // can be NULL + mi_atomic_store_ptr_release(void,&mi_error_arg, arg); +} + +void _mi_error_message(int err, const char* fmt, ...) { + // show detailed error message + va_list args; + va_start(args, fmt); + mi_show_error_message(fmt, args); + va_end(args); + // and call the error handler which may abort (or return normally) + if (mi_error_handler != NULL) { + mi_error_handler(err, mi_atomic_load_ptr_acquire(void,&mi_error_arg)); + } + else { + mi_error_default(err); + } +} + +// -------------------------------------------------------- +// Initialize options by checking the environment +// -------------------------------------------------------- + +// TODO: implement ourselves to reduce dependencies on the C runtime +#include // strtol +#include // strstr + + +static void mi_option_init(mi_option_desc_t* desc) { + // Read option value from the environment + char s[64 + 1]; + char buf[64+1]; + _mi_strlcpy(buf, "mimalloc_", sizeof(buf)); + _mi_strlcat(buf, desc->name, sizeof(buf)); + bool found = _mi_getenv(buf, s, sizeof(s)); + if (!found && desc->legacy_name != NULL) { + _mi_strlcpy(buf, "mimalloc_", sizeof(buf)); + _mi_strlcat(buf, desc->legacy_name, sizeof(buf)); + found = _mi_getenv(buf, s, sizeof(s)); + if (found) { + _mi_warning_message("environment option \"mimalloc_%s\" is deprecated -- use \"mimalloc_%s\" instead.\n", desc->legacy_name, desc->name); + } + } + + if (found) { + size_t len = _mi_strnlen(s, sizeof(buf) - 1); + for (size_t i = 0; i < len; i++) { + buf[i] = _mi_toupper(s[i]); + } + buf[len] = 0; + if (buf[0] == 0 || strstr("1;TRUE;YES;ON", buf) != NULL) { + desc->value = 1; + desc->init = INITIALIZED; + } + else if (strstr("0;FALSE;NO;OFF", buf) != NULL) { + desc->value = 0; + desc->init = INITIALIZED; + } + else { + char* end = buf; + long value = strtol(buf, &end, 10); + if (mi_option_has_size_in_kib(desc->option)) { + // this option is interpreted in KiB to prevent overflow of `long` for large allocations + // (long is 32-bit on 64-bit windows, which allows for 4TiB max.) + size_t size = (value < 0 ? 0 : (size_t)value); + bool overflow = false; + if (*end == 'K') { end++; } + else if (*end == 'M') { overflow = mi_mul_overflow(size,MI_KiB,&size); end++; } + else if (*end == 'G') { overflow = mi_mul_overflow(size,MI_MiB,&size); end++; } + else if (*end == 'T') { overflow = mi_mul_overflow(size,MI_GiB,&size); end++; } + else { size = (size + MI_KiB - 1) / MI_KiB; } + if (end[0] == 'I' && end[1] == 'B') { end += 2; } // KiB, MiB, GiB, TiB + else if (*end == 'B') { end++; } // Kb, Mb, Gb, Tb + if (overflow || size > MI_MAX_ALLOC_SIZE) { size = (MI_MAX_ALLOC_SIZE / MI_KiB); } + value = (size > LONG_MAX ? LONG_MAX : (long)size); + } + if (*end == 0) { + mi_option_set(desc->option, value); + } + else { + // set `init` first to avoid recursion through _mi_warning_message on mimalloc_verbose. + desc->init = DEFAULTED; + if (desc->option == mi_option_verbose && desc->value == 0) { + // if the 'mimalloc_verbose' env var has a bogus value we'd never know + // (since the value defaults to 'off') so in that case briefly enable verbose + desc->value = 1; + _mi_warning_message("environment option mimalloc_%s has an invalid value.\n", desc->name); + desc->value = 0; + } + else { + _mi_warning_message("environment option mimalloc_%s has an invalid value.\n", desc->name); + } + } + } + mi_assert_internal(desc->init != UNINIT); + } + else if (!_mi_preloading()) { + desc->init = DEFAULTED; + } +} diff --git a/compat/mimalloc/os.c b/compat/mimalloc/os.c new file mode 100644 index 00000000000000..9b1b4b460775f1 --- /dev/null +++ b/compat/mimalloc/os.c @@ -0,0 +1,770 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2025, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "mimalloc/prim.h" + +#define mi_os_stat_increase(stat,amount) _mi_stat_increase(&_mi_stats_main.stat, amount) +#define mi_os_stat_decrease(stat,amount) _mi_stat_decrease(&_mi_stats_main.stat, amount) +#define mi_os_stat_counter_increase(stat,inc) _mi_stat_counter_increase(&_mi_stats_main.stat, inc) + +/* ----------------------------------------------------------- + Initialization. +----------------------------------------------------------- */ +#ifndef MI_DEFAULT_VIRTUAL_ADDRESS_BITS +#if MI_INTPTR_SIZE < 8 +#define MI_DEFAULT_VIRTUAL_ADDRESS_BITS 32 +#else +#define MI_DEFAULT_VIRTUAL_ADDRESS_BITS 48 +#endif +#endif + +#ifndef MI_DEFAULT_PHYSICAL_MEMORY_IN_KIB +#if MI_INTPTR_SIZE < 8 +#define MI_DEFAULT_PHYSICAL_MEMORY_IN_KIB 4*MI_MiB // 4 GiB +#else +#define MI_DEFAULT_PHYSICAL_MEMORY_IN_KIB 32*MI_MiB // 32 GiB +#endif +#endif + +static mi_os_mem_config_t mi_os_mem_config = { + 4096, // page size + 0, // large page size (usually 2MiB) + 4096, // allocation granularity + MI_DEFAULT_PHYSICAL_MEMORY_IN_KIB, + MI_DEFAULT_VIRTUAL_ADDRESS_BITS, + true, // has overcommit? (if true we use MAP_NORESERVE on mmap systems) + false, // can we partially free allocated blocks? (on mmap systems we can free anywhere in a mapped range, but on Windows we must free the entire span) + true // has virtual reserve? (if true we can reserve virtual address space without using commit or physical memory) +}; + +bool _mi_os_has_overcommit(void) { + return mi_os_mem_config.has_overcommit; +} + +bool _mi_os_has_virtual_reserve(void) { + return mi_os_mem_config.has_virtual_reserve; +} + + +// OS (small) page size +size_t _mi_os_page_size(void) { + return mi_os_mem_config.page_size; +} + +// if large OS pages are supported (2 or 4MiB), then return the size, otherwise return the small page size (4KiB) +size_t _mi_os_large_page_size(void) { + return (mi_os_mem_config.large_page_size != 0 ? mi_os_mem_config.large_page_size : _mi_os_page_size()); +} + +bool _mi_os_use_large_page(size_t size, size_t alignment) { + // if we have access, check the size and alignment requirements + if (mi_os_mem_config.large_page_size == 0 || !mi_option_is_enabled(mi_option_allow_large_os_pages)) return false; + return ((size % mi_os_mem_config.large_page_size) == 0 && (alignment % mi_os_mem_config.large_page_size) == 0); +} + +// round to a good OS allocation size (bounded by max 12.5% waste) +size_t _mi_os_good_alloc_size(size_t size) { + size_t align_size; + if (size < 512*MI_KiB) align_size = _mi_os_page_size(); + else if (size < 2*MI_MiB) align_size = 64*MI_KiB; + else if (size < 8*MI_MiB) align_size = 256*MI_KiB; + else if (size < 32*MI_MiB) align_size = 1*MI_MiB; + else align_size = 4*MI_MiB; + if mi_unlikely(size >= (SIZE_MAX - align_size)) return size; // possible overflow? + return _mi_align_up(size, align_size); +} + +void _mi_os_init(void) { + _mi_prim_mem_init(&mi_os_mem_config); +} + + +/* ----------------------------------------------------------- + Util +-------------------------------------------------------------- */ +bool _mi_os_decommit(void* addr, size_t size); +bool _mi_os_commit(void* addr, size_t size, bool* is_zero); + + +/* ----------------------------------------------------------- + aligned hinting +-------------------------------------------------------------- */ + +// On systems with enough virtual address bits, we can do efficient aligned allocation by using +// the 2TiB to 30TiB area to allocate those. If we have at least 46 bits of virtual address +// space (64TiB) we use this technique. (but see issue #939) +#if (MI_INTPTR_SIZE >= 8) && !defined(MI_NO_ALIGNED_HINT) +static mi_decl_cache_align _Atomic(uintptr_t)aligned_base; + +// Return a MI_SEGMENT_SIZE aligned address that is probably available. +// If this returns NULL, the OS will determine the address but on some OS's that may not be +// properly aligned which can be more costly as it needs to be adjusted afterwards. +// For a size > 1GiB this always returns NULL in order to guarantee good ASLR randomization; +// (otherwise an initial large allocation of say 2TiB has a 50% chance to include (known) addresses +// in the middle of the 2TiB - 6TiB address range (see issue #372)) + +#define MI_HINT_BASE ((uintptr_t)2 << 40) // 2TiB start +#define MI_HINT_AREA ((uintptr_t)4 << 40) // upto 6TiB (since before win8 there is "only" 8TiB available to processes) +#define MI_HINT_MAX ((uintptr_t)30 << 40) // wrap after 30TiB (area after 32TiB is used for huge OS pages) + +void* _mi_os_get_aligned_hint(size_t try_alignment, size_t size) +{ + if (try_alignment <= 1 || try_alignment > MI_SEGMENT_SIZE) return NULL; + if (mi_os_mem_config.virtual_address_bits < 46) return NULL; // < 64TiB virtual address space + size = _mi_align_up(size, MI_SEGMENT_SIZE); + if (size > 1*MI_GiB) return NULL; // guarantee the chance of fixed valid address is at most 1/(MI_HINT_AREA / 1<<30) = 1/4096. + #if (MI_SECURE>0) + size += MI_SEGMENT_SIZE; // put in `MI_SEGMENT_SIZE` virtual gaps between hinted blocks; this splits VLA's but increases guarded areas. + #endif + + uintptr_t hint = mi_atomic_add_acq_rel(&aligned_base, size); + if (hint == 0 || hint > MI_HINT_MAX) { // wrap or initialize + uintptr_t init = MI_HINT_BASE; + #if (MI_SECURE>0 || MI_DEBUG==0) // security: randomize start of aligned allocations unless in debug mode + uintptr_t r = _mi_heap_random_next(mi_prim_get_default_heap()); + init = init + ((MI_SEGMENT_SIZE * ((r>>17) & 0xFFFFF)) % MI_HINT_AREA); // (randomly 20 bits)*4MiB == 0 to 4TiB + #endif + uintptr_t expected = hint + size; + mi_atomic_cas_strong_acq_rel(&aligned_base, &expected, init); + hint = mi_atomic_add_acq_rel(&aligned_base, size); // this may still give 0 or > MI_HINT_MAX but that is ok, it is a hint after all + } + if (hint%try_alignment != 0) return NULL; + return (void*)hint; +} +#else +void* _mi_os_get_aligned_hint(size_t try_alignment, size_t size) { + MI_UNUSED(try_alignment); MI_UNUSED(size); + return NULL; +} +#endif + +/* ----------------------------------------------------------- + Free memory +-------------------------------------------------------------- */ + +static void mi_os_free_huge_os_pages(void* p, size_t size); + +static void mi_os_prim_free(void* addr, size_t size, size_t commit_size) { + mi_assert_internal((size % _mi_os_page_size()) == 0); + if (addr == NULL) return; // || _mi_os_is_huge_reserved(addr) + int err = _mi_prim_free(addr, size); // allow size==0 (issue #1041) + if (err != 0) { + _mi_warning_message("unable to free OS memory (error: %d (0x%x), size: 0x%zx bytes, address: %p)\n", err, err, size, addr); + } + if (commit_size > 0) { + mi_os_stat_decrease(committed, commit_size); + } + mi_os_stat_decrease(reserved, size); +} + +void _mi_os_free_ex(void* addr, size_t size, bool still_committed, mi_memid_t memid) { + if (mi_memkind_is_os(memid.memkind)) { + size_t csize = memid.mem.os.size; + if (csize==0) { csize = _mi_os_good_alloc_size(size); } + mi_assert_internal(csize >= size); + size_t commit_size = (still_committed ? csize : 0); + void* base = addr; + // different base? (due to alignment) + if (memid.mem.os.base != base) { + mi_assert(memid.mem.os.base <= addr); + base = memid.mem.os.base; + const size_t diff = (uint8_t*)addr - (uint8_t*)memid.mem.os.base; + if (memid.mem.os.size==0) { + csize += diff; + } + if (still_committed) { + commit_size -= diff; // the (addr-base) part was already un-committed + } + } + // free it + if (memid.memkind == MI_MEM_OS_HUGE) { + mi_assert(memid.is_pinned); + mi_os_free_huge_os_pages(base, csize); + } + else { + mi_os_prim_free(base, csize, (still_committed ? commit_size : 0)); + } + } + else { + // nothing to do + mi_assert(memid.memkind < MI_MEM_OS); + } +} + +void _mi_os_free(void* p, size_t size, mi_memid_t memid) { + _mi_os_free_ex(p, size, true, memid); +} + + +/* ----------------------------------------------------------- + Primitive allocation from the OS. +-------------------------------------------------------------- */ + +// Note: the `try_alignment` is just a hint and the returned pointer is not guaranteed to be aligned. +// Also `hint_addr` is a hint and may be ignored. +static void* mi_os_prim_alloc_at(void* hint_addr, size_t size, size_t try_alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero) { + mi_assert_internal(size > 0 && (size % _mi_os_page_size()) == 0); + mi_assert_internal(is_zero != NULL); + mi_assert_internal(is_large != NULL); + if (size == 0) return NULL; + if (!commit) { allow_large = false; } + if (try_alignment == 0) { try_alignment = 1; } // avoid 0 to ensure there will be no divide by zero when aligning + *is_zero = false; + void* p = NULL; + int err = _mi_prim_alloc(hint_addr, size, try_alignment, commit, allow_large, is_large, is_zero, &p); + if (err != 0) { + _mi_warning_message("unable to allocate OS memory (error: %d (0x%x), addr: %p, size: 0x%zx bytes, align: 0x%zx, commit: %d, allow large: %d)\n", err, err, hint_addr, size, try_alignment, commit, allow_large); + } + + + + mi_os_stat_counter_increase(mmap_calls, 1); + if (p != NULL) { + mi_os_stat_increase(reserved, size); + if (commit) { + mi_os_stat_increase(committed, size); + // seems needed for asan (or `mimalloc-test-api` fails) + #ifdef MI_TRACK_ASAN + if (*is_zero) { mi_track_mem_defined(p,size); } + else { mi_track_mem_undefined(p,size); } + #endif + } + } + return p; +} + +static void* mi_os_prim_alloc(size_t size, size_t try_alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero) { + return mi_os_prim_alloc_at(NULL, size, try_alignment, commit, allow_large, is_large, is_zero); +} + + +// Primitive aligned allocation from the OS. +// This function guarantees the allocated memory is aligned. +static void* mi_os_prim_alloc_aligned(size_t size, size_t alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero, void** base) { + mi_assert_internal(alignment >= _mi_os_page_size() && ((alignment & (alignment - 1)) == 0)); + mi_assert_internal(size > 0 && (size % _mi_os_page_size()) == 0); + mi_assert_internal(is_large != NULL); + mi_assert_internal(is_zero != NULL); + mi_assert_internal(base != NULL); + if (!commit) allow_large = false; + if (!(alignment >= _mi_os_page_size() && ((alignment & (alignment - 1)) == 0))) return NULL; + size = _mi_align_up(size, _mi_os_page_size()); + + // try first with a requested alignment hint (this will usually be aligned directly on Win 10+ or BSD) + void* p = mi_os_prim_alloc(size, alignment, commit, allow_large, is_large, is_zero); + if (p == NULL) return NULL; + + // aligned already? + if (((uintptr_t)p % alignment) == 0) { + *base = p; + } + else { + // if not aligned, free it, overallocate, and unmap around it + #if !MI_TRACK_ASAN + _mi_warning_message("unable to allocate aligned OS memory directly, fall back to over-allocation (size: 0x%zx bytes, address: %p, alignment: 0x%zx, commit: %d)\n", size, p, alignment, commit); + #endif + if (p != NULL) { mi_os_prim_free(p, size, (commit ? size : 0)); } + if (size >= (SIZE_MAX - alignment)) return NULL; // overflow + const size_t over_size = size + alignment; + + if (!mi_os_mem_config.has_partial_free) { // win32 virtualAlloc cannot free parts of an allocated block + // over-allocate uncommitted (virtual) memory + p = mi_os_prim_alloc(over_size, 1 /*alignment*/, false /* commit? */, false /* allow_large */, is_large, is_zero); + if (p == NULL) return NULL; + + // set p to the aligned part in the full region + // note: this is dangerous on Windows as VirtualFree needs the actual base pointer + // this is handled though by having the `base` field in the memid's + *base = p; // remember the base + p = mi_align_up_ptr(p, alignment); + + // explicitly commit only the aligned part + if (commit) { + if (!_mi_os_commit(p, size, NULL)) { + mi_os_prim_free(*base, over_size, 0); + return NULL; + } + } + } + else { // mmap can free inside an allocation + // overallocate... + p = mi_os_prim_alloc(over_size, 1, commit, false, is_large, is_zero); + if (p == NULL) return NULL; + + // and selectively unmap parts around the over-allocated area. + void* aligned_p = mi_align_up_ptr(p, alignment); + size_t pre_size = (uint8_t*)aligned_p - (uint8_t*)p; + size_t mid_size = _mi_align_up(size, _mi_os_page_size()); + size_t post_size = over_size - pre_size - mid_size; + mi_assert_internal(pre_size < over_size&& post_size < over_size&& mid_size >= size); + if (pre_size > 0) { mi_os_prim_free(p, pre_size, (commit ? pre_size : 0)); } + if (post_size > 0) { mi_os_prim_free((uint8_t*)aligned_p + mid_size, post_size, (commit ? post_size : 0)); } + // we can return the aligned pointer on `mmap` systems + p = aligned_p; + *base = aligned_p; // since we freed the pre part, `*base == p`. + } + } + + mi_assert_internal(p == NULL || (p != NULL && *base != NULL && ((uintptr_t)p % alignment) == 0)); + return p; +} + + +/* ----------------------------------------------------------- + OS API: alloc and alloc_aligned +----------------------------------------------------------- */ + +void* _mi_os_alloc(size_t size, mi_memid_t* memid) { + *memid = _mi_memid_none(); + if (size == 0) return NULL; + size = _mi_os_good_alloc_size(size); + bool os_is_large = false; + bool os_is_zero = false; + void* p = mi_os_prim_alloc(size, 0, true, false, &os_is_large, &os_is_zero); + if (p == NULL) return NULL; + + *memid = _mi_memid_create_os(p, size, true, os_is_zero, os_is_large); + mi_assert_internal(memid->mem.os.size >= size); + mi_assert_internal(memid->initially_committed); + return p; +} + +void* _mi_os_alloc_aligned(size_t size, size_t alignment, bool commit, bool allow_large, mi_memid_t* memid) +{ + MI_UNUSED(&_mi_os_get_aligned_hint); // suppress unused warnings + *memid = _mi_memid_none(); + if (size == 0) return NULL; + size = _mi_os_good_alloc_size(size); + alignment = _mi_align_up(alignment, _mi_os_page_size()); + + bool os_is_large = false; + bool os_is_zero = false; + void* os_base = NULL; + void* p = mi_os_prim_alloc_aligned(size, alignment, commit, allow_large, &os_is_large, &os_is_zero, &os_base ); + if (p == NULL) return NULL; + + *memid = _mi_memid_create_os(p, size, commit, os_is_zero, os_is_large); + memid->mem.os.base = os_base; + memid->mem.os.size += ((uint8_t*)p - (uint8_t*)os_base); // todo: return from prim_alloc_aligned? + + mi_assert_internal(memid->mem.os.size >= size); + mi_assert_internal(_mi_is_aligned(p,alignment)); + if (commit) { mi_assert_internal(memid->initially_committed); } + return p; +} + + +mi_decl_nodiscard static void* mi_os_ensure_zero(void* p, size_t size, mi_memid_t* memid) { + if (p==NULL || size==0) return p; + // ensure committed + if (!memid->initially_committed) { + bool is_zero = false; + if (!_mi_os_commit(p, size, &is_zero)) { + _mi_os_free(p, size, *memid); + return NULL; + } + memid->initially_committed = true; + } + // ensure zero'd + if (memid->initially_zero) return p; + _mi_memzero_aligned(p,size); + memid->initially_zero = true; + return p; +} + +void* _mi_os_zalloc(size_t size, mi_memid_t* memid) { + void* p = _mi_os_alloc(size,memid); + return mi_os_ensure_zero(p, size, memid); +} + +/* ----------------------------------------------------------- + OS aligned allocation with an offset. This is used + for large alignments > MI_BLOCK_ALIGNMENT_MAX. We use a large mimalloc + page where the object can be aligned at an offset from the start of the segment. + As we may need to overallocate, we need to free such pointers using `mi_free_aligned` + to use the actual start of the memory region. +----------------------------------------------------------- */ + +void* _mi_os_alloc_aligned_at_offset(size_t size, size_t alignment, size_t offset, bool commit, bool allow_large, mi_memid_t* memid) { + mi_assert(offset <= MI_SEGMENT_SIZE); + mi_assert(offset <= size); + mi_assert((alignment % _mi_os_page_size()) == 0); + *memid = _mi_memid_none(); + if (offset > MI_SEGMENT_SIZE) return NULL; + if (offset == 0) { + // regular aligned allocation + return _mi_os_alloc_aligned(size, alignment, commit, allow_large, memid); + } + else { + // overallocate to align at an offset + const size_t extra = _mi_align_up(offset, alignment) - offset; + const size_t oversize = size + extra; + void* const start = _mi_os_alloc_aligned(oversize, alignment, commit, allow_large, memid); + if (start == NULL) return NULL; + + void* const p = (uint8_t*)start + extra; + mi_assert(_mi_is_aligned((uint8_t*)p + offset, alignment)); + // decommit the overallocation at the start + if (commit && extra > _mi_os_page_size()) { + _mi_os_decommit(start, extra); + } + return p; + } +} + +/* ----------------------------------------------------------- + OS memory API: reset, commit, decommit, protect, unprotect. +----------------------------------------------------------- */ + +// OS page align within a given area, either conservative (pages inside the area only), +// or not (straddling pages outside the area is possible) +static void* mi_os_page_align_areax(bool conservative, void* addr, size_t size, size_t* newsize) { + mi_assert(addr != NULL && size > 0); + if (newsize != NULL) *newsize = 0; + if (size == 0 || addr == NULL) return NULL; + + // page align conservatively within the range + void* start = (conservative ? mi_align_up_ptr(addr, _mi_os_page_size()) + : mi_align_down_ptr(addr, _mi_os_page_size())); + void* end = (conservative ? mi_align_down_ptr((uint8_t*)addr + size, _mi_os_page_size()) + : mi_align_up_ptr((uint8_t*)addr + size, _mi_os_page_size())); + ptrdiff_t diff = (uint8_t*)end - (uint8_t*)start; + if (diff <= 0) return NULL; + + mi_assert_internal((conservative && (size_t)diff <= size) || (!conservative && (size_t)diff >= size)); + if (newsize != NULL) *newsize = (size_t)diff; + return start; +} + +static void* mi_os_page_align_area_conservative(void* addr, size_t size, size_t* newsize) { + return mi_os_page_align_areax(true, addr, size, newsize); +} + +bool _mi_os_commit_ex(void* addr, size_t size, bool* is_zero, size_t stat_size) { + if (is_zero != NULL) { *is_zero = false; } + mi_os_stat_increase(committed, stat_size); // use size for precise commit vs. decommit + mi_os_stat_counter_increase(commit_calls, 1); + + // page align range + size_t csize; + void* start = mi_os_page_align_areax(false /* conservative? */, addr, size, &csize); + if (csize == 0) return true; + + // commit + bool os_is_zero = false; + int err = _mi_prim_commit(start, csize, &os_is_zero); + if (err != 0) { + _mi_warning_message("cannot commit OS memory (error: %d (0x%x), address: %p, size: 0x%zx bytes)\n", err, err, start, csize); + return false; + } + if (os_is_zero && is_zero != NULL) { + *is_zero = true; + mi_assert_expensive(mi_mem_is_zero(start, csize)); + } + // note: the following seems required for asan (otherwise `mimalloc-test-stress` fails) + #ifdef MI_TRACK_ASAN + if (os_is_zero) { mi_track_mem_defined(start,csize); } + else { mi_track_mem_undefined(start,csize); } + #endif + return true; +} + +bool _mi_os_commit(void* addr, size_t size, bool* is_zero) { + return _mi_os_commit_ex(addr, size, is_zero, size); +} + +static bool mi_os_decommit_ex(void* addr, size_t size, bool* needs_recommit, size_t stat_size) { + mi_assert_internal(needs_recommit!=NULL); + mi_os_stat_decrease(committed, stat_size); + + // page align + size_t csize; + void* start = mi_os_page_align_area_conservative(addr, size, &csize); + if (csize == 0) return true; + + // decommit + *needs_recommit = true; + int err = _mi_prim_decommit(start,csize,needs_recommit); + if (err != 0) { + _mi_warning_message("cannot decommit OS memory (error: %d (0x%x), address: %p, size: 0x%zx bytes)\n", err, err, start, csize); + } + mi_assert_internal(err == 0); + return (err == 0); +} + +bool _mi_os_decommit(void* addr, size_t size) { + bool needs_recommit; + return mi_os_decommit_ex(addr, size, &needs_recommit, size); +} + + +// Signal to the OS that the address range is no longer in use +// but may be used later again. This will release physical memory +// pages and reduce swapping while keeping the memory committed. +// We page align to a conservative area inside the range to reset. +bool _mi_os_reset(void* addr, size_t size) { + // page align conservatively within the range + size_t csize; + void* start = mi_os_page_align_area_conservative(addr, size, &csize); + if (csize == 0) return true; // || _mi_os_is_huge_reserved(addr) + mi_os_stat_increase(reset, csize); + mi_os_stat_counter_increase(reset_calls, 1); + + #if (MI_DEBUG>1) && !MI_SECURE && !MI_TRACK_ENABLED // && !MI_TSAN + memset(start, 0, csize); // pretend it is eagerly reset + #endif + + int err = _mi_prim_reset(start, csize); + if (err != 0) { + _mi_warning_message("cannot reset OS memory (error: %d (0x%x), address: %p, size: 0x%zx bytes)\n", err, err, start, csize); + } + return (err == 0); +} + + +void _mi_os_reuse( void* addr, size_t size ) { + // page align conservatively within the range + size_t csize = 0; + void* const start = mi_os_page_align_area_conservative(addr, size, &csize); + if (csize == 0) return; + const int err = _mi_prim_reuse(start, csize); + if (err != 0) { + _mi_warning_message("cannot reuse OS memory (error: %d (0x%x), address: %p, size: 0x%zx bytes)\n", err, err, start, csize); + } +} + +// either resets or decommits memory, returns true if the memory needs +// to be recommitted if it is to be re-used later on. +bool _mi_os_purge_ex(void* p, size_t size, bool allow_reset, size_t stat_size) +{ + if (mi_option_get(mi_option_purge_delay) < 0) return false; // is purging allowed? + mi_os_stat_counter_increase(purge_calls, 1); + mi_os_stat_increase(purged, size); + + if (mi_option_is_enabled(mi_option_purge_decommits) && // should decommit? + !_mi_preloading()) // don't decommit during preloading (unsafe) + { + bool needs_recommit = true; + mi_os_decommit_ex(p, size, &needs_recommit, stat_size); + return needs_recommit; + } + else { + if (allow_reset) { // this can sometimes be not allowed if the range is not fully committed + _mi_os_reset(p, size); + } + return false; // needs no recommit + } +} + +// either resets or decommits memory, returns true if the memory needs +// to be recommitted if it is to be re-used later on. +bool _mi_os_purge(void* p, size_t size) { + return _mi_os_purge_ex(p, size, true, size); +} + +// Protect a region in memory to be not accessible. +static bool mi_os_protectx(void* addr, size_t size, bool protect) { + // page align conservatively within the range + size_t csize = 0; + void* start = mi_os_page_align_area_conservative(addr, size, &csize); + if (csize == 0) return false; + /* + if (_mi_os_is_huge_reserved(addr)) { + _mi_warning_message("cannot mprotect memory allocated in huge OS pages\n"); + } + */ + int err = _mi_prim_protect(start,csize,protect); + if (err != 0) { + _mi_warning_message("cannot %s OS memory (error: %d (0x%x), address: %p, size: 0x%zx bytes)\n", (protect ? "protect" : "unprotect"), err, err, start, csize); + } + return (err == 0); +} + +bool _mi_os_protect(void* addr, size_t size) { + return mi_os_protectx(addr, size, true); +} + +bool _mi_os_unprotect(void* addr, size_t size) { + return mi_os_protectx(addr, size, false); +} + + + +/* ---------------------------------------------------------------------------- +Support for allocating huge OS pages (1Gib) that are reserved up-front +and possibly associated with a specific NUMA node. (use `numa_node>=0`) +-----------------------------------------------------------------------------*/ +#define MI_HUGE_OS_PAGE_SIZE (MI_GiB) + + +#if (MI_INTPTR_SIZE >= 8) +// To ensure proper alignment, use our own area for huge OS pages +static mi_decl_cache_align _Atomic(uintptr_t) mi_huge_start; // = 0 + +// Claim an aligned address range for huge pages +static uint8_t* mi_os_claim_huge_pages(size_t pages, size_t* total_size) { + if (total_size != NULL) *total_size = 0; + const size_t size = pages * MI_HUGE_OS_PAGE_SIZE; + + uintptr_t start = 0; + uintptr_t end = 0; + uintptr_t huge_start = mi_atomic_load_relaxed(&mi_huge_start); + do { + start = huge_start; + if (start == 0) { + // Initialize the start address after the 32TiB area + start = ((uintptr_t)32 << 40); // 32TiB virtual start address + #if (MI_SECURE>0 || MI_DEBUG==0) // security: randomize start of huge pages unless in debug mode + uintptr_t r = _mi_heap_random_next(mi_prim_get_default_heap()); + start = start + ((uintptr_t)MI_HUGE_OS_PAGE_SIZE * ((r>>17) & 0x0FFF)); // (randomly 12bits)*1GiB == between 0 to 4TiB + #endif + } + end = start + size; + mi_assert_internal(end % MI_SEGMENT_SIZE == 0); + } while (!mi_atomic_cas_strong_acq_rel(&mi_huge_start, &huge_start, end)); + + if (total_size != NULL) *total_size = size; + return (uint8_t*)start; +} +#else +static uint8_t* mi_os_claim_huge_pages(size_t pages, size_t* total_size) { + MI_UNUSED(pages); + if (total_size != NULL) *total_size = 0; + return NULL; +} +#endif + +// Allocate MI_SEGMENT_SIZE aligned huge pages +void* _mi_os_alloc_huge_os_pages(size_t pages, int numa_node, mi_msecs_t max_msecs, size_t* pages_reserved, size_t* psize, mi_memid_t* memid) { + *memid = _mi_memid_none(); + if (psize != NULL) *psize = 0; + if (pages_reserved != NULL) *pages_reserved = 0; + size_t size = 0; + uint8_t* const start = mi_os_claim_huge_pages(pages, &size); + if (start == NULL) return NULL; // or 32-bit systems + + // Allocate one page at the time but try to place them contiguously + // We allocate one page at the time to be able to abort if it takes too long + // or to at least allocate as many as available on the system. + mi_msecs_t start_t = _mi_clock_start(); + size_t page = 0; + bool all_zero = true; + while (page < pages) { + // allocate a page + bool is_zero = false; + void* addr = start + (page * MI_HUGE_OS_PAGE_SIZE); + void* p = NULL; + int err = _mi_prim_alloc_huge_os_pages(addr, MI_HUGE_OS_PAGE_SIZE, numa_node, &is_zero, &p); + if (!is_zero) { all_zero = false; } + if (err != 0) { + _mi_warning_message("unable to allocate huge OS page (error: %d (0x%x), address: %p, size: %zx bytes)\n", err, err, addr, MI_HUGE_OS_PAGE_SIZE); + break; + } + + // Did we succeed at a contiguous address? + if (p != addr) { + // no success, issue a warning and break + if (p != NULL) { + _mi_warning_message("could not allocate contiguous huge OS page %zu at %p\n", page, addr); + mi_os_prim_free(p, MI_HUGE_OS_PAGE_SIZE, MI_HUGE_OS_PAGE_SIZE); + } + break; + } + + // success, record it + page++; // increase before timeout check (see issue #711) + mi_os_stat_increase(committed, MI_HUGE_OS_PAGE_SIZE); + mi_os_stat_increase(reserved, MI_HUGE_OS_PAGE_SIZE); + + // check for timeout + if (max_msecs > 0) { + mi_msecs_t elapsed = _mi_clock_end(start_t); + if (page >= 1) { + mi_msecs_t estimate = ((elapsed / (page+1)) * pages); + if (estimate > 2*max_msecs) { // seems like we are going to timeout, break + elapsed = max_msecs + 1; + } + } + if (elapsed > max_msecs) { + _mi_warning_message("huge OS page allocation timed out (after allocating %zu page(s))\n", page); + break; + } + } + } + mi_assert_internal(page*MI_HUGE_OS_PAGE_SIZE <= size); + if (pages_reserved != NULL) { *pages_reserved = page; } + if (psize != NULL) { *psize = page * MI_HUGE_OS_PAGE_SIZE; } + if (page != 0) { + mi_assert(start != NULL); + *memid = _mi_memid_create_os(start, size, true /* is committed */, all_zero, true /* is_large */); + memid->memkind = MI_MEM_OS_HUGE; + mi_assert(memid->is_pinned); + #ifdef MI_TRACK_ASAN + if (all_zero) { mi_track_mem_defined(start,size); } + #endif + } + return (page == 0 ? NULL : start); +} + +// free every huge page in a range individually (as we allocated per page) +// note: needed with VirtualAlloc but could potentially be done in one go on mmap'd systems. +static void mi_os_free_huge_os_pages(void* p, size_t size) { + if (p==NULL || size==0) return; + uint8_t* base = (uint8_t*)p; + while (size >= MI_HUGE_OS_PAGE_SIZE) { + mi_os_prim_free(base, MI_HUGE_OS_PAGE_SIZE, MI_HUGE_OS_PAGE_SIZE); + size -= MI_HUGE_OS_PAGE_SIZE; + base += MI_HUGE_OS_PAGE_SIZE; + } +} + + +/* ---------------------------------------------------------------------------- +Support NUMA aware allocation +-----------------------------------------------------------------------------*/ + +static _Atomic(size_t) mi_numa_node_count; // = 0 // cache the node count + +int _mi_os_numa_node_count(void) { + size_t count = mi_atomic_load_acquire(&mi_numa_node_count); + if mi_unlikely(count == 0) { + long ncount = mi_option_get(mi_option_use_numa_nodes); // given explicitly? + if (ncount > 0 && ncount < INT_MAX) { + count = (size_t)ncount; + } + else { + const size_t n = _mi_prim_numa_node_count(); // or detect dynamically + if (n == 0 || n > INT_MAX) { count = 1; } + else { count = n; } + } + mi_atomic_store_release(&mi_numa_node_count, count); // save it + _mi_verbose_message("using %zd numa regions\n", count); + } + mi_assert_internal(count > 0 && count <= INT_MAX); + return (int)count; +} + +static int mi_os_numa_node_get(void) { + int numa_count = _mi_os_numa_node_count(); + if (numa_count<=1) return 0; // optimize on single numa node systems: always node 0 + // never more than the node count and >= 0 + const size_t n = _mi_prim_numa_node(); + int numa_node = (n < INT_MAX ? (int)n : 0); + if (numa_node >= numa_count) { numa_node = numa_node % numa_count; } + return numa_node; +} + +int _mi_os_numa_node(void) { + if mi_likely(mi_atomic_load_relaxed(&mi_numa_node_count) == 1) { + return 0; + } + else { + return mi_os_numa_node_get(); + } +} diff --git a/compat/mimalloc/page-queue.c b/compat/mimalloc/page-queue.c new file mode 100644 index 00000000000000..c719b6265afe54 --- /dev/null +++ b/compat/mimalloc/page-queue.c @@ -0,0 +1,390 @@ +/*---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ----------------------------------------------------------- + Definition of page queues for each block size +----------------------------------------------------------- */ + +#ifndef MI_IN_PAGE_C +#error "this file should be included from 'page.c'" +// include to help an IDE +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#endif + +/* ----------------------------------------------------------- + Minimal alignment in machine words (i.e. `sizeof(void*)`) +----------------------------------------------------------- */ + +#if (MI_MAX_ALIGN_SIZE > 4*MI_INTPTR_SIZE) + #error "define alignment for more than 4x word size for this platform" +#elif (MI_MAX_ALIGN_SIZE > 2*MI_INTPTR_SIZE) + #define MI_ALIGN4W // 4 machine words minimal alignment +#elif (MI_MAX_ALIGN_SIZE > MI_INTPTR_SIZE) + #define MI_ALIGN2W // 2 machine words minimal alignment +#else + // ok, default alignment is 1 word +#endif + + +/* ----------------------------------------------------------- + Queue query +----------------------------------------------------------- */ + + +static inline bool mi_page_queue_is_huge(const mi_page_queue_t* pq) { + return (pq->block_size == (MI_MEDIUM_OBJ_SIZE_MAX+sizeof(uintptr_t))); +} + +static inline bool mi_page_queue_is_full(const mi_page_queue_t* pq) { + return (pq->block_size == (MI_MEDIUM_OBJ_SIZE_MAX+(2*sizeof(uintptr_t)))); +} + +static inline bool mi_page_queue_is_special(const mi_page_queue_t* pq) { + return (pq->block_size > MI_MEDIUM_OBJ_SIZE_MAX); +} + +/* ----------------------------------------------------------- + Bins +----------------------------------------------------------- */ + +// Return the bin for a given field size. +// Returns MI_BIN_HUGE if the size is too large. +// We use `wsize` for the size in "machine word sizes", +// i.e. byte size == `wsize*sizeof(void*)`. +static inline size_t mi_bin(size_t size) { + size_t wsize = _mi_wsize_from_size(size); +#if defined(MI_ALIGN4W) + if mi_likely(wsize <= 4) { + return (wsize <= 1 ? 1 : (wsize+1)&~1); // round to double word sizes + } +#elif defined(MI_ALIGN2W) + if mi_likely(wsize <= 8) { + return (wsize <= 1 ? 1 : (wsize+1)&~1); // round to double word sizes + } +#else + if mi_likely(wsize <= 8) { + return (wsize == 0 ? 1 : wsize); + } +#endif + else if mi_unlikely(wsize > MI_MEDIUM_OBJ_WSIZE_MAX) { + return MI_BIN_HUGE; + } + else { + #if defined(MI_ALIGN4W) + if (wsize <= 16) { wsize = (wsize+3)&~3; } // round to 4x word sizes + #endif + wsize--; + // find the highest bit + const size_t b = (MI_SIZE_BITS - 1 - mi_clz(wsize)); // note: wsize != 0 + // and use the top 3 bits to determine the bin (~12.5% worst internal fragmentation). + // - adjust with 3 because we use do not round the first 8 sizes + // which each get an exact bin + const size_t bin = ((b << 2) + ((wsize >> (b - 2)) & 0x03)) - 3; + mi_assert_internal(bin > 0 && bin < MI_BIN_HUGE); + return bin; + } +} + + + +/* ----------------------------------------------------------- + Queue of pages with free blocks +----------------------------------------------------------- */ + +size_t _mi_bin(size_t size) { + return mi_bin(size); +} + +size_t _mi_bin_size(size_t bin) { + return _mi_heap_empty.pages[bin].block_size; +} + +// Good size for allocation +size_t mi_good_size(size_t size) mi_attr_noexcept { + if (size <= MI_MEDIUM_OBJ_SIZE_MAX) { + return _mi_bin_size(mi_bin(size + MI_PADDING_SIZE)); + } + else { + return _mi_align_up(size + MI_PADDING_SIZE,_mi_os_page_size()); + } +} + +#if (MI_DEBUG>1) +static bool mi_page_queue_contains(mi_page_queue_t* queue, const mi_page_t* page) { + mi_assert_internal(page != NULL); + mi_page_t* list = queue->first; + while (list != NULL) { + mi_assert_internal(list->next == NULL || list->next->prev == list); + mi_assert_internal(list->prev == NULL || list->prev->next == list); + if (list == page) break; + list = list->next; + } + return (list == page); +} + +#endif + +#if (MI_DEBUG>1) +static bool mi_heap_contains_queue(const mi_heap_t* heap, const mi_page_queue_t* pq) { + return (pq >= &heap->pages[0] && pq <= &heap->pages[MI_BIN_FULL]); +} +#endif + +static inline bool mi_page_is_large_or_huge(const mi_page_t* page) { + return (mi_page_block_size(page) > MI_MEDIUM_OBJ_SIZE_MAX || mi_page_is_huge(page)); +} + +size_t _mi_page_bin(const mi_page_t* page) { + const size_t bin = (mi_page_is_in_full(page) ? MI_BIN_FULL : (mi_page_is_huge(page) ? MI_BIN_HUGE : mi_bin(mi_page_block_size(page)))); + mi_assert_internal(bin <= MI_BIN_FULL); + return bin; +} + +static mi_page_queue_t* mi_heap_page_queue_of(mi_heap_t* heap, const mi_page_t* page) { + mi_assert_internal(heap!=NULL); + const size_t bin = _mi_page_bin(page); + mi_page_queue_t* pq = &heap->pages[bin]; + mi_assert_internal((mi_page_block_size(page) == pq->block_size) || + (mi_page_is_large_or_huge(page) && mi_page_queue_is_huge(pq)) || + (mi_page_is_in_full(page) && mi_page_queue_is_full(pq))); + return pq; +} + +static mi_page_queue_t* mi_page_queue_of(const mi_page_t* page) { + mi_heap_t* heap = mi_page_heap(page); + mi_page_queue_t* pq = mi_heap_page_queue_of(heap, page); + mi_assert_expensive(mi_page_queue_contains(pq, page)); + return pq; +} + +// The current small page array is for efficiency and for each +// small size (up to 256) it points directly to the page for that +// size without having to compute the bin. This means when the +// current free page queue is updated for a small bin, we need to update a +// range of entries in `_mi_page_small_free`. +static inline void mi_heap_queue_first_update(mi_heap_t* heap, const mi_page_queue_t* pq) { + mi_assert_internal(mi_heap_contains_queue(heap,pq)); + size_t size = pq->block_size; + if (size > MI_SMALL_SIZE_MAX) return; + + mi_page_t* page = pq->first; + if (pq->first == NULL) page = (mi_page_t*)&_mi_page_empty; + + // find index in the right direct page array + size_t start; + size_t idx = _mi_wsize_from_size(size); + mi_page_t** pages_free = heap->pages_free_direct; + + if (pages_free[idx] == page) return; // already set + + // find start slot + if (idx<=1) { + start = 0; + } + else { + // find previous size; due to minimal alignment upto 3 previous bins may need to be skipped + size_t bin = mi_bin(size); + const mi_page_queue_t* prev = pq - 1; + while( bin == mi_bin(prev->block_size) && prev > &heap->pages[0]) { + prev--; + } + start = 1 + _mi_wsize_from_size(prev->block_size); + if (start > idx) start = idx; + } + + // set size range to the right page + mi_assert(start <= idx); + for (size_t sz = start; sz <= idx; sz++) { + pages_free[sz] = page; + } +} + +/* +static bool mi_page_queue_is_empty(mi_page_queue_t* queue) { + return (queue->first == NULL); +} +*/ + +static void mi_page_queue_remove(mi_page_queue_t* queue, mi_page_t* page) { + mi_assert_internal(page != NULL); + mi_assert_expensive(mi_page_queue_contains(queue, page)); + mi_assert_internal(mi_page_block_size(page) == queue->block_size || + (mi_page_is_large_or_huge(page) && mi_page_queue_is_huge(queue)) || + (mi_page_is_in_full(page) && mi_page_queue_is_full(queue))); + mi_heap_t* heap = mi_page_heap(page); + + if (page->prev != NULL) page->prev->next = page->next; + if (page->next != NULL) page->next->prev = page->prev; + if (page == queue->last) queue->last = page->prev; + if (page == queue->first) { + queue->first = page->next; + // update first + mi_assert_internal(mi_heap_contains_queue(heap, queue)); + mi_heap_queue_first_update(heap,queue); + } + heap->page_count--; + page->next = NULL; + page->prev = NULL; + // mi_atomic_store_ptr_release(mi_atomic_cast(void*, &page->heap), NULL); + mi_page_set_in_full(page,false); +} + + +static void mi_page_queue_push(mi_heap_t* heap, mi_page_queue_t* queue, mi_page_t* page) { + mi_assert_internal(mi_page_heap(page) == heap); + mi_assert_internal(!mi_page_queue_contains(queue, page)); + #if MI_HUGE_PAGE_ABANDON + mi_assert_internal(_mi_page_segment(page)->kind != MI_SEGMENT_HUGE); + #endif + mi_assert_internal(mi_page_block_size(page) == queue->block_size || + (mi_page_is_large_or_huge(page) && mi_page_queue_is_huge(queue)) || + (mi_page_is_in_full(page) && mi_page_queue_is_full(queue))); + + mi_page_set_in_full(page, mi_page_queue_is_full(queue)); + // mi_atomic_store_ptr_release(mi_atomic_cast(void*, &page->heap), heap); + page->next = queue->first; + page->prev = NULL; + if (queue->first != NULL) { + mi_assert_internal(queue->first->prev == NULL); + queue->first->prev = page; + queue->first = page; + } + else { + queue->first = queue->last = page; + } + + // update direct + mi_heap_queue_first_update(heap, queue); + heap->page_count++; +} + +static void mi_page_queue_move_to_front(mi_heap_t* heap, mi_page_queue_t* queue, mi_page_t* page) { + mi_assert_internal(mi_page_heap(page) == heap); + mi_assert_internal(mi_page_queue_contains(queue, page)); + if (queue->first == page) return; + mi_page_queue_remove(queue, page); + mi_page_queue_push(heap, queue, page); + mi_assert_internal(queue->first == page); +} + +static void mi_page_queue_enqueue_from_ex(mi_page_queue_t* to, mi_page_queue_t* from, bool enqueue_at_end, mi_page_t* page) { + mi_assert_internal(page != NULL); + mi_assert_expensive(mi_page_queue_contains(from, page)); + mi_assert_expensive(!mi_page_queue_contains(to, page)); + const size_t bsize = mi_page_block_size(page); + MI_UNUSED(bsize); + mi_assert_internal((bsize == to->block_size && bsize == from->block_size) || + (bsize == to->block_size && mi_page_queue_is_full(from)) || + (bsize == from->block_size && mi_page_queue_is_full(to)) || + (mi_page_is_large_or_huge(page) && mi_page_queue_is_huge(to)) || + (mi_page_is_large_or_huge(page) && mi_page_queue_is_full(to))); + + mi_heap_t* heap = mi_page_heap(page); + + // delete from `from` + if (page->prev != NULL) page->prev->next = page->next; + if (page->next != NULL) page->next->prev = page->prev; + if (page == from->last) from->last = page->prev; + if (page == from->first) { + from->first = page->next; + // update first + mi_assert_internal(mi_heap_contains_queue(heap, from)); + mi_heap_queue_first_update(heap, from); + } + + // insert into `to` + if (enqueue_at_end) { + // enqueue at the end + page->prev = to->last; + page->next = NULL; + if (to->last != NULL) { + mi_assert_internal(heap == mi_page_heap(to->last)); + to->last->next = page; + to->last = page; + } + else { + to->first = page; + to->last = page; + mi_heap_queue_first_update(heap, to); + } + } + else { + if (to->first != NULL) { + // enqueue at 2nd place + mi_assert_internal(heap == mi_page_heap(to->first)); + mi_page_t* next = to->first->next; + page->prev = to->first; + page->next = next; + to->first->next = page; + if (next != NULL) { + next->prev = page; + } + else { + to->last = page; + } + } + else { + // enqueue at the head (singleton list) + page->prev = NULL; + page->next = NULL; + to->first = page; + to->last = page; + mi_heap_queue_first_update(heap, to); + } + } + + mi_page_set_in_full(page, mi_page_queue_is_full(to)); +} + +static void mi_page_queue_enqueue_from(mi_page_queue_t* to, mi_page_queue_t* from, mi_page_t* page) { + mi_page_queue_enqueue_from_ex(to, from, true /* enqueue at the end */, page); +} + +static void mi_page_queue_enqueue_from_full(mi_page_queue_t* to, mi_page_queue_t* from, mi_page_t* page) { + // note: we could insert at the front to increase reuse, but it slows down certain benchmarks (like `alloc-test`) + mi_page_queue_enqueue_from_ex(to, from, true /* enqueue at the end of the `to` queue? */, page); +} + +// Only called from `mi_heap_absorb`. +size_t _mi_page_queue_append(mi_heap_t* heap, mi_page_queue_t* pq, mi_page_queue_t* append) { + mi_assert_internal(mi_heap_contains_queue(heap,pq)); + mi_assert_internal(pq->block_size == append->block_size); + + if (append->first==NULL) return 0; + + // set append pages to new heap and count + size_t count = 0; + for (mi_page_t* page = append->first; page != NULL; page = page->next) { + // inline `mi_page_set_heap` to avoid wrong assertion during absorption; + // in this case it is ok to be delayed freeing since both "to" and "from" heap are still alive. + mi_atomic_store_release(&page->xheap, (uintptr_t)heap); + // set the flag to delayed free (not overriding NEVER_DELAYED_FREE) which has as a + // side effect that it spins until any DELAYED_FREEING is finished. This ensures + // that after appending only the new heap will be used for delayed free operations. + _mi_page_use_delayed_free(page, MI_USE_DELAYED_FREE, false); + count++; + } + + if (pq->last==NULL) { + // take over afresh + mi_assert_internal(pq->first==NULL); + pq->first = append->first; + pq->last = append->last; + mi_heap_queue_first_update(heap, pq); + } + else { + // append to end + mi_assert_internal(pq->last!=NULL); + mi_assert_internal(append->first!=NULL); + pq->last->next = append->first; + append->first->prev = pq->last; + pq->last = append->last; + } + return count; +} diff --git a/compat/mimalloc/page.c b/compat/mimalloc/page.c new file mode 100644 index 00000000000000..a5a10503248622 --- /dev/null +++ b/compat/mimalloc/page.c @@ -0,0 +1,1049 @@ +/*---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ----------------------------------------------------------- + The core of the allocator. Every segment contains + pages of a certain block size. The main function + exported is `mi_malloc_generic`. +----------------------------------------------------------- */ + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" + +/* ----------------------------------------------------------- + Definition of page queues for each block size +----------------------------------------------------------- */ + +#define MI_IN_PAGE_C +#include "page-queue.c" +#undef MI_IN_PAGE_C + + +/* ----------------------------------------------------------- + Page helpers +----------------------------------------------------------- */ + +// Index a block in a page +static inline mi_block_t* mi_page_block_at(const mi_page_t* page, void* page_start, size_t block_size, size_t i) { + MI_UNUSED(page); + mi_assert_internal(page != NULL); + mi_assert_internal(i <= page->reserved); + return (mi_block_t*)((uint8_t*)page_start + (i * block_size)); +} + +static void mi_page_init(mi_heap_t* heap, mi_page_t* page, size_t size, mi_tld_t* tld); +static bool mi_page_extend_free(mi_heap_t* heap, mi_page_t* page, mi_tld_t* tld); + +#if (MI_DEBUG>=3) +static size_t mi_page_list_count(mi_page_t* page, mi_block_t* head) { + size_t count = 0; + while (head != NULL) { + mi_assert_internal(page == _mi_ptr_page(head)); + count++; + head = mi_block_next(page, head); + } + return count; +} + +/* +// Start of the page available memory +static inline uint8_t* mi_page_area(const mi_page_t* page) { + return _mi_page_start(_mi_page_segment(page), page, NULL); +} +*/ + +static bool mi_page_list_is_valid(mi_page_t* page, mi_block_t* p) { + size_t psize; + uint8_t* page_area = _mi_segment_page_start(_mi_page_segment(page), page, &psize); + mi_block_t* start = (mi_block_t*)page_area; + mi_block_t* end = (mi_block_t*)(page_area + psize); + while(p != NULL) { + if (p < start || p >= end) return false; + p = mi_block_next(page, p); + } +#if MI_DEBUG>3 // generally too expensive to check this + if (page->free_is_zero) { + const size_t ubsize = mi_page_usable_block_size(page); + for (mi_block_t* block = page->free; block != NULL; block = mi_block_next(page, block)) { + mi_assert_expensive(mi_mem_is_zero(block + 1, ubsize - sizeof(mi_block_t))); + } + } +#endif + return true; +} + +static bool mi_page_is_valid_init(mi_page_t* page) { + mi_assert_internal(mi_page_block_size(page) > 0); + mi_assert_internal(page->used <= page->capacity); + mi_assert_internal(page->capacity <= page->reserved); + + uint8_t* start = mi_page_start(page); + mi_assert_internal(start == _mi_segment_page_start(_mi_page_segment(page), page, NULL)); + mi_assert_internal(page->is_huge == (_mi_page_segment(page)->kind == MI_SEGMENT_HUGE)); + //mi_assert_internal(start + page->capacity*page->block_size == page->top); + + mi_assert_internal(mi_page_list_is_valid(page,page->free)); + mi_assert_internal(mi_page_list_is_valid(page,page->local_free)); + + #if MI_DEBUG>3 // generally too expensive to check this + if (page->free_is_zero) { + const size_t ubsize = mi_page_usable_block_size(page); + for(mi_block_t* block = page->free; block != NULL; block = mi_block_next(page,block)) { + mi_assert_expensive(mi_mem_is_zero(block + 1, ubsize - sizeof(mi_block_t))); + } + } + #endif + + #if !MI_TRACK_ENABLED && !MI_TSAN + mi_block_t* tfree = mi_page_thread_free(page); + mi_assert_internal(mi_page_list_is_valid(page, tfree)); + //size_t tfree_count = mi_page_list_count(page, tfree); + //mi_assert_internal(tfree_count <= page->thread_freed + 1); + #endif + + size_t free_count = mi_page_list_count(page, page->free) + mi_page_list_count(page, page->local_free); + mi_assert_internal(page->used + free_count == page->capacity); + + return true; +} + +extern mi_decl_hidden bool _mi_process_is_initialized; // has mi_process_init been called? + +bool _mi_page_is_valid(mi_page_t* page) { + mi_assert_internal(mi_page_is_valid_init(page)); + #if MI_SECURE + mi_assert_internal(page->keys[0] != 0); + #endif + if (mi_page_heap(page)!=NULL) { + mi_segment_t* segment = _mi_page_segment(page); + + mi_assert_internal(!_mi_process_is_initialized || segment->thread_id==0 || segment->thread_id == mi_page_heap(page)->thread_id); + #if MI_HUGE_PAGE_ABANDON + if (segment->kind != MI_SEGMENT_HUGE) + #endif + { + mi_page_queue_t* pq = mi_page_queue_of(page); + mi_assert_internal(mi_page_queue_contains(pq, page)); + mi_assert_internal(pq->block_size==mi_page_block_size(page) || mi_page_block_size(page) > MI_MEDIUM_OBJ_SIZE_MAX || mi_page_is_in_full(page)); + mi_assert_internal(mi_heap_contains_queue(mi_page_heap(page),pq)); + } + } + return true; +} +#endif + +void _mi_page_use_delayed_free(mi_page_t* page, mi_delayed_t delay, bool override_never) { + while (!_mi_page_try_use_delayed_free(page, delay, override_never)) { + mi_atomic_yield(); + } +} + +bool _mi_page_try_use_delayed_free(mi_page_t* page, mi_delayed_t delay, bool override_never) { + mi_thread_free_t tfreex; + mi_delayed_t old_delay; + mi_thread_free_t tfree; + size_t yield_count = 0; + do { + tfree = mi_atomic_load_acquire(&page->xthread_free); // note: must acquire as we can break/repeat this loop and not do a CAS; + tfreex = mi_tf_set_delayed(tfree, delay); + old_delay = mi_tf_delayed(tfree); + if mi_unlikely(old_delay == MI_DELAYED_FREEING) { + if (yield_count >= 4) return false; // give up after 4 tries + yield_count++; + mi_atomic_yield(); // delay until outstanding MI_DELAYED_FREEING are done. + // tfree = mi_tf_set_delayed(tfree, MI_NO_DELAYED_FREE); // will cause CAS to busy fail + } + else if (delay == old_delay) { + break; // avoid atomic operation if already equal + } + else if (!override_never && old_delay == MI_NEVER_DELAYED_FREE) { + break; // leave never-delayed flag set + } + } while ((old_delay == MI_DELAYED_FREEING) || + !mi_atomic_cas_weak_release(&page->xthread_free, &tfree, tfreex)); + + return true; // success +} + +/* ----------------------------------------------------------- + Page collect the `local_free` and `thread_free` lists +----------------------------------------------------------- */ + +// Collect the local `thread_free` list using an atomic exchange. +// Note: The exchange must be done atomically as this is used right after +// moving to the full list in `mi_page_collect_ex` and we need to +// ensure that there was no race where the page became unfull just before the move. +static void _mi_page_thread_free_collect(mi_page_t* page) +{ + mi_block_t* head; + mi_thread_free_t tfreex; + mi_thread_free_t tfree = mi_atomic_load_relaxed(&page->xthread_free); + do { + head = mi_tf_block(tfree); + tfreex = mi_tf_set_block(tfree,NULL); + } while (!mi_atomic_cas_weak_acq_rel(&page->xthread_free, &tfree, tfreex)); + + // return if the list is empty + if (head == NULL) return; + + // find the tail -- also to get a proper count (without data races) + size_t max_count = page->capacity; // cannot collect more than capacity + size_t count = 1; + mi_block_t* tail = head; + mi_block_t* next; + while ((next = mi_block_next(page,tail)) != NULL && count <= max_count) { + count++; + tail = next; + } + // if `count > max_count` there was a memory corruption (possibly infinite list due to double multi-threaded free) + if (count > max_count) { + _mi_error_message(EFAULT, "corrupted thread-free list\n"); + return; // the thread-free items cannot be freed + } + + // and append the current local free list + mi_block_set_next(page,tail, page->local_free); + page->local_free = head; + + // update counts now + page->used -= (uint16_t)count; +} + +void _mi_page_free_collect(mi_page_t* page, bool force) { + mi_assert_internal(page!=NULL); + + // collect the thread free list + if (force || mi_page_thread_free(page) != NULL) { // quick test to avoid an atomic operation + _mi_page_thread_free_collect(page); + } + + // and the local free list + if (page->local_free != NULL) { + if mi_likely(page->free == NULL) { + // usual case + page->free = page->local_free; + page->local_free = NULL; + page->free_is_zero = false; + } + else if (force) { + // append -- only on shutdown (force) as this is a linear operation + mi_block_t* tail = page->local_free; + mi_block_t* next; + while ((next = mi_block_next(page, tail)) != NULL) { + tail = next; + } + mi_block_set_next(page, tail, page->free); + page->free = page->local_free; + page->local_free = NULL; + page->free_is_zero = false; + } + } + + mi_assert_internal(!force || page->local_free == NULL); +} + + + +/* ----------------------------------------------------------- + Page fresh and retire +----------------------------------------------------------- */ + +// called from segments when reclaiming abandoned pages +void _mi_page_reclaim(mi_heap_t* heap, mi_page_t* page) { + mi_assert_expensive(mi_page_is_valid_init(page)); + + mi_assert_internal(mi_page_heap(page) == heap); + mi_assert_internal(mi_page_thread_free_flag(page) != MI_NEVER_DELAYED_FREE); + #if MI_HUGE_PAGE_ABANDON + mi_assert_internal(_mi_page_segment(page)->kind != MI_SEGMENT_HUGE); + #endif + + // TODO: push on full queue immediately if it is full? + mi_page_queue_t* pq = mi_page_queue(heap, mi_page_block_size(page)); + mi_page_queue_push(heap, pq, page); + mi_assert_expensive(_mi_page_is_valid(page)); +} + +// allocate a fresh page from a segment +static mi_page_t* mi_page_fresh_alloc(mi_heap_t* heap, mi_page_queue_t* pq, size_t block_size, size_t page_alignment) { + #if !MI_HUGE_PAGE_ABANDON + mi_assert_internal(pq != NULL); + mi_assert_internal(mi_heap_contains_queue(heap, pq)); + mi_assert_internal(page_alignment > 0 || block_size > MI_MEDIUM_OBJ_SIZE_MAX || block_size == pq->block_size); + #endif + mi_page_t* page = _mi_segment_page_alloc(heap, block_size, page_alignment, &heap->tld->segments); + if (page == NULL) { + // this may be out-of-memory, or an abandoned page was reclaimed (and in our queue) + return NULL; + } + #if MI_HUGE_PAGE_ABANDON + mi_assert_internal(pq==NULL || _mi_page_segment(page)->page_kind != MI_PAGE_HUGE); + #endif + mi_assert_internal(page_alignment >0 || block_size > MI_MEDIUM_OBJ_SIZE_MAX || _mi_page_segment(page)->kind != MI_SEGMENT_HUGE); + mi_assert_internal(pq!=NULL || mi_page_block_size(page) >= block_size); + // a fresh page was found, initialize it + const size_t full_block_size = (pq == NULL || mi_page_is_huge(page) ? mi_page_block_size(page) : block_size); // see also: mi_segment_huge_page_alloc + mi_assert_internal(full_block_size >= block_size); + mi_page_init(heap, page, full_block_size, heap->tld); + mi_heap_stat_increase(heap, pages, 1); + mi_heap_stat_increase(heap, page_bins[_mi_page_bin(page)], 1); + if (pq != NULL) { mi_page_queue_push(heap, pq, page); } + mi_assert_expensive(_mi_page_is_valid(page)); + return page; +} + +// Get a fresh page to use +static mi_page_t* mi_page_fresh(mi_heap_t* heap, mi_page_queue_t* pq) { + mi_assert_internal(mi_heap_contains_queue(heap, pq)); + mi_page_t* page = mi_page_fresh_alloc(heap, pq, pq->block_size, 0); + if (page==NULL) return NULL; + mi_assert_internal(pq->block_size==mi_page_block_size(page)); + mi_assert_internal(pq==mi_page_queue(heap, mi_page_block_size(page))); + return page; +} + +/* ----------------------------------------------------------- + Do any delayed frees + (put there by other threads if they deallocated in a full page) +----------------------------------------------------------- */ +void _mi_heap_delayed_free_all(mi_heap_t* heap) { + while (!_mi_heap_delayed_free_partial(heap)) { + mi_atomic_yield(); + } +} + +// returns true if all delayed frees were processed +bool _mi_heap_delayed_free_partial(mi_heap_t* heap) { + // take over the list (note: no atomic exchange since it is often NULL) + mi_block_t* block = mi_atomic_load_ptr_relaxed(mi_block_t, &heap->thread_delayed_free); + while (block != NULL && !mi_atomic_cas_ptr_weak_acq_rel(mi_block_t, &heap->thread_delayed_free, &block, NULL)) { /* nothing */ }; + bool all_freed = true; + + // and free them all + while(block != NULL) { + mi_block_t* next = mi_block_nextx(heap,block, heap->keys); + // use internal free instead of regular one to keep stats etc correct + if (!_mi_free_delayed_block(block)) { + // we might already start delayed freeing while another thread has not yet + // reset the delayed_freeing flag; in that case delay it further by reinserting the current block + // into the delayed free list + all_freed = false; + mi_block_t* dfree = mi_atomic_load_ptr_relaxed(mi_block_t, &heap->thread_delayed_free); + do { + mi_block_set_nextx(heap, block, dfree, heap->keys); + } while (!mi_atomic_cas_ptr_weak_release(mi_block_t,&heap->thread_delayed_free, &dfree, block)); + } + block = next; + } + return all_freed; +} + +/* ----------------------------------------------------------- + Unfull, abandon, free and retire +----------------------------------------------------------- */ + +// Move a page from the full list back to a regular list +void _mi_page_unfull(mi_page_t* page) { + mi_assert_internal(page != NULL); + mi_assert_expensive(_mi_page_is_valid(page)); + mi_assert_internal(mi_page_is_in_full(page)); + if (!mi_page_is_in_full(page)) return; + + mi_heap_t* heap = mi_page_heap(page); + mi_page_queue_t* pqfull = &heap->pages[MI_BIN_FULL]; + mi_page_set_in_full(page, false); // to get the right queue + mi_page_queue_t* pq = mi_heap_page_queue_of(heap, page); + mi_page_set_in_full(page, true); + mi_page_queue_enqueue_from_full(pq, pqfull, page); +} + +static void mi_page_to_full(mi_page_t* page, mi_page_queue_t* pq) { + mi_assert_internal(pq == mi_page_queue_of(page)); + mi_assert_internal(!mi_page_immediate_available(page)); + mi_assert_internal(!mi_page_is_in_full(page)); + + if (mi_page_is_in_full(page)) return; + mi_page_queue_enqueue_from(&mi_page_heap(page)->pages[MI_BIN_FULL], pq, page); + _mi_page_free_collect(page,false); // try to collect right away in case another thread freed just before MI_USE_DELAYED_FREE was set +} + + +// Abandon a page with used blocks at the end of a thread. +// Note: only call if it is ensured that no references exist from +// the `page->heap->thread_delayed_free` into this page. +// Currently only called through `mi_heap_collect_ex` which ensures this. +void _mi_page_abandon(mi_page_t* page, mi_page_queue_t* pq) { + mi_assert_internal(page != NULL); + mi_assert_expensive(_mi_page_is_valid(page)); + mi_assert_internal(pq == mi_page_queue_of(page)); + mi_assert_internal(mi_page_heap(page) != NULL); + + mi_heap_t* pheap = mi_page_heap(page); + + // remove from our page list + mi_segments_tld_t* segments_tld = &pheap->tld->segments; + mi_page_queue_remove(pq, page); + + // page is no longer associated with our heap + mi_assert_internal(mi_page_thread_free_flag(page)==MI_NEVER_DELAYED_FREE); + mi_page_set_heap(page, NULL); + +#if (MI_DEBUG>1) && !MI_TRACK_ENABLED + // check there are no references left.. + for (mi_block_t* block = (mi_block_t*)pheap->thread_delayed_free; block != NULL; block = mi_block_nextx(pheap, block, pheap->keys)) { + mi_assert_internal(_mi_ptr_page(block) != page); + } +#endif + + // and abandon it + mi_assert_internal(mi_page_heap(page) == NULL); + _mi_segment_page_abandon(page,segments_tld); +} + +// force abandon a page +void _mi_page_force_abandon(mi_page_t* page) { + mi_heap_t* heap = mi_page_heap(page); + // mark page as not using delayed free + _mi_page_use_delayed_free(page, MI_NEVER_DELAYED_FREE, false); + + // ensure this page is no longer in the heap delayed free list + _mi_heap_delayed_free_all(heap); + // We can still access the page meta-info even if it is freed as we ensure + // in `mi_segment_force_abandon` that the segment is not freed (yet) + if (page->capacity == 0) return; // it may have been freed now + + // and now unlink it from the page queue and abandon (or free) + mi_page_queue_t* pq = mi_heap_page_queue_of(heap, page); + if (mi_page_all_free(page)) { + _mi_page_free(page, pq, false); + } + else { + _mi_page_abandon(page, pq); + } +} + + +// Free a page with no more free blocks +void _mi_page_free(mi_page_t* page, mi_page_queue_t* pq, bool force) { + mi_assert_internal(page != NULL); + mi_assert_expensive(_mi_page_is_valid(page)); + mi_assert_internal(pq == mi_page_queue_of(page)); + mi_assert_internal(mi_page_all_free(page)); + mi_assert_internal(mi_page_thread_free_flag(page)!=MI_DELAYED_FREEING); + + // no more aligned blocks in here + mi_page_set_has_aligned(page, false); + + // remove from the page list + // (no need to do _mi_heap_delayed_free first as all blocks are already free) + mi_heap_t* heap = mi_page_heap(page); + mi_segments_tld_t* segments_tld = &heap->tld->segments; + mi_page_queue_remove(pq, page); + + // and free it + mi_page_set_heap(page,NULL); + _mi_segment_page_free(page, force, segments_tld); +} + +#define MI_MAX_RETIRE_SIZE MI_MEDIUM_OBJ_SIZE_MAX // should be less than size for MI_BIN_HUGE +#define MI_RETIRE_CYCLES (16) + +// Retire a page with no more used blocks +// Important to not retire too quickly though as new +// allocations might coming. +// Note: called from `mi_free` and benchmarks often +// trigger this due to freeing everything and then +// allocating again so careful when changing this. +void _mi_page_retire(mi_page_t* page) mi_attr_noexcept { + mi_assert_internal(page != NULL); + mi_assert_expensive(_mi_page_is_valid(page)); + mi_assert_internal(mi_page_all_free(page)); + + mi_page_set_has_aligned(page, false); + + // don't retire too often.. + // (or we end up retiring and re-allocating most of the time) + // NOTE: refine this more: we should not retire if this + // is the only page left with free blocks. It is not clear + // how to check this efficiently though... + // for now, we don't retire if it is the only page left of this size class. + mi_page_queue_t* pq = mi_page_queue_of(page); + #if MI_RETIRE_CYCLES > 0 + const size_t bsize = mi_page_block_size(page); + if mi_likely( /* bsize < MI_MAX_RETIRE_SIZE && */ !mi_page_queue_is_special(pq)) { // not full or huge queue? + if (pq->last==page && pq->first==page) { // the only page in the queue? + mi_stat_counter_increase(_mi_stats_main.pages_retire,1); + page->retire_expire = (bsize <= MI_SMALL_OBJ_SIZE_MAX ? MI_RETIRE_CYCLES : MI_RETIRE_CYCLES/4); + mi_heap_t* heap = mi_page_heap(page); + mi_assert_internal(pq >= heap->pages); + const size_t index = pq - heap->pages; + mi_assert_internal(index < MI_BIN_FULL && index < MI_BIN_HUGE); + if (index < heap->page_retired_min) heap->page_retired_min = index; + if (index > heap->page_retired_max) heap->page_retired_max = index; + mi_assert_internal(mi_page_all_free(page)); + return; // don't free after all + } + } + #endif + _mi_page_free(page, pq, false); +} + +// free retired pages: we don't need to look at the entire queues +// since we only retire pages that are at the head position in a queue. +void _mi_heap_collect_retired(mi_heap_t* heap, bool force) { + size_t min = MI_BIN_FULL; + size_t max = 0; + for(size_t bin = heap->page_retired_min; bin <= heap->page_retired_max; bin++) { + mi_page_queue_t* pq = &heap->pages[bin]; + mi_page_t* page = pq->first; + if (page != NULL && page->retire_expire != 0) { + if (mi_page_all_free(page)) { + page->retire_expire--; + if (force || page->retire_expire == 0) { + _mi_page_free(pq->first, pq, force); + } + else { + // keep retired, update min/max + if (bin < min) min = bin; + if (bin > max) max = bin; + } + } + else { + page->retire_expire = 0; + } + } + } + heap->page_retired_min = min; + heap->page_retired_max = max; +} + + +/* ----------------------------------------------------------- + Initialize the initial free list in a page. + In secure mode we initialize a randomized list by + alternating between slices. +----------------------------------------------------------- */ + +#define MI_MAX_SLICE_SHIFT (6) // at most 64 slices +#define MI_MAX_SLICES (1UL << MI_MAX_SLICE_SHIFT) +#define MI_MIN_SLICES (2) + +static void mi_page_free_list_extend_secure(mi_heap_t* const heap, mi_page_t* const page, const size_t bsize, const size_t extend, mi_stats_t* const stats) { + MI_UNUSED(stats); + #if (MI_SECURE<=2) + mi_assert_internal(page->free == NULL); + mi_assert_internal(page->local_free == NULL); + #endif + mi_assert_internal(page->capacity + extend <= page->reserved); + mi_assert_internal(bsize == mi_page_block_size(page)); + void* const page_area = mi_page_start(page); + + // initialize a randomized free list + // set up `slice_count` slices to alternate between + size_t shift = MI_MAX_SLICE_SHIFT; + while ((extend >> shift) == 0) { + shift--; + } + const size_t slice_count = (size_t)1U << shift; + const size_t slice_extend = extend / slice_count; + mi_assert_internal(slice_extend >= 1); + mi_block_t* blocks[MI_MAX_SLICES]; // current start of the slice + size_t counts[MI_MAX_SLICES]; // available objects in the slice + for (size_t i = 0; i < slice_count; i++) { + blocks[i] = mi_page_block_at(page, page_area, bsize, page->capacity + i*slice_extend); + counts[i] = slice_extend; + } + counts[slice_count-1] += (extend % slice_count); // final slice holds the modulus too (todo: distribute evenly?) + + // and initialize the free list by randomly threading through them + // set up first element + const uintptr_t r = _mi_heap_random_next(heap); + size_t current = r % slice_count; + counts[current]--; + mi_block_t* const free_start = blocks[current]; + // and iterate through the rest; use `random_shuffle` for performance + uintptr_t rnd = _mi_random_shuffle(r|1); // ensure not 0 + for (size_t i = 1; i < extend; i++) { + // call random_shuffle only every INTPTR_SIZE rounds + const size_t round = i%MI_INTPTR_SIZE; + if (round == 0) rnd = _mi_random_shuffle(rnd); + // select a random next slice index + size_t next = ((rnd >> 8*round) & (slice_count-1)); + while (counts[next]==0) { // ensure it still has space + next++; + if (next==slice_count) next = 0; + } + // and link the current block to it + counts[next]--; + mi_block_t* const block = blocks[current]; + blocks[current] = (mi_block_t*)((uint8_t*)block + bsize); // bump to the following block + mi_block_set_next(page, block, blocks[next]); // and set next; note: we may have `current == next` + current = next; + } + // prepend to the free list (usually NULL) + mi_block_set_next(page, blocks[current], page->free); // end of the list + page->free = free_start; +} + +static mi_decl_noinline void mi_page_free_list_extend( mi_page_t* const page, const size_t bsize, const size_t extend, mi_stats_t* const stats) +{ + MI_UNUSED(stats); + #if (MI_SECURE <= 2) + mi_assert_internal(page->free == NULL); + mi_assert_internal(page->local_free == NULL); + #endif + mi_assert_internal(page->capacity + extend <= page->reserved); + mi_assert_internal(bsize == mi_page_block_size(page)); + void* const page_area = mi_page_start(page); + + mi_block_t* const start = mi_page_block_at(page, page_area, bsize, page->capacity); + + // initialize a sequential free list + mi_block_t* const last = mi_page_block_at(page, page_area, bsize, page->capacity + extend - 1); + mi_block_t* block = start; + while(block <= last) { + mi_block_t* next = (mi_block_t*)((uint8_t*)block + bsize); + mi_block_set_next(page,block,next); + block = next; + } + // prepend to free list (usually `NULL`) + mi_block_set_next(page, last, page->free); + page->free = start; +} + +/* ----------------------------------------------------------- + Page initialize and extend the capacity +----------------------------------------------------------- */ + +#define MI_MAX_EXTEND_SIZE (4*1024) // heuristic, one OS page seems to work well. +#if (MI_SECURE>0) +#define MI_MIN_EXTEND (8*MI_SECURE) // extend at least by this many +#else +#define MI_MIN_EXTEND (4) +#endif + +// Extend the capacity (up to reserved) by initializing a free list +// We do at most `MI_MAX_EXTEND` to avoid touching too much memory +// Note: we also experimented with "bump" allocation on the first +// allocations but this did not speed up any benchmark (due to an +// extra test in malloc? or cache effects?) +static bool mi_page_extend_free(mi_heap_t* heap, mi_page_t* page, mi_tld_t* tld) { + mi_assert_expensive(mi_page_is_valid_init(page)); + #if (MI_SECURE<=2) + mi_assert(page->free == NULL); + mi_assert(page->local_free == NULL); + if (page->free != NULL) return true; + #endif + if (page->capacity >= page->reserved) return true; + + mi_stat_counter_increase(tld->stats.pages_extended, 1); + + // calculate the extend count + const size_t bsize = mi_page_block_size(page); + size_t extend = page->reserved - page->capacity; + mi_assert_internal(extend > 0); + + size_t max_extend = (bsize >= MI_MAX_EXTEND_SIZE ? MI_MIN_EXTEND : MI_MAX_EXTEND_SIZE/bsize); + if (max_extend < MI_MIN_EXTEND) { max_extend = MI_MIN_EXTEND; } + mi_assert_internal(max_extend > 0); + + if (extend > max_extend) { + // ensure we don't touch memory beyond the page to reduce page commit. + // the `lean` benchmark tests this. Going from 1 to 8 increases rss by 50%. + extend = max_extend; + } + + mi_assert_internal(extend > 0 && extend + page->capacity <= page->reserved); + mi_assert_internal(extend < (1UL<<16)); + + // and append the extend the free list + if (extend < MI_MIN_SLICES || MI_SECURE==0) { //!mi_option_is_enabled(mi_option_secure)) { + mi_page_free_list_extend(page, bsize, extend, &tld->stats ); + } + else { + mi_page_free_list_extend_secure(heap, page, bsize, extend, &tld->stats); + } + // enable the new free list + page->capacity += (uint16_t)extend; + mi_stat_increase(tld->stats.page_committed, extend * bsize); + mi_assert_expensive(mi_page_is_valid_init(page)); + return true; +} + +// Initialize a fresh page +static void mi_page_init(mi_heap_t* heap, mi_page_t* page, size_t block_size, mi_tld_t* tld) { + mi_assert(page != NULL); + mi_segment_t* segment = _mi_page_segment(page); + mi_assert(segment != NULL); + mi_assert_internal(block_size > 0); + // set fields + mi_page_set_heap(page, heap); + page->block_size = block_size; + size_t page_size; + page->page_start = _mi_segment_page_start(segment, page, &page_size); + mi_track_mem_noaccess(page->page_start,page_size); + mi_assert_internal(mi_page_block_size(page) <= page_size); + mi_assert_internal(page_size <= page->slice_count*MI_SEGMENT_SLICE_SIZE); + mi_assert_internal(page_size / block_size < (1L<<16)); + page->reserved = (uint16_t)(page_size / block_size); + mi_assert_internal(page->reserved > 0); + #if (MI_PADDING || MI_ENCODE_FREELIST) + page->keys[0] = _mi_heap_random_next(heap); + page->keys[1] = _mi_heap_random_next(heap); + #endif + page->free_is_zero = page->is_zero_init; + #if MI_DEBUG>2 + if (page->is_zero_init) { + mi_track_mem_defined(page->page_start, page_size); + mi_assert_expensive(mi_mem_is_zero(page->page_start, page_size)); + } + #endif + mi_assert_internal(page->is_committed); + if (block_size > 0 && _mi_is_power_of_two(block_size)) { + page->block_size_shift = (uint8_t)(mi_ctz((uintptr_t)block_size)); + } + else { + page->block_size_shift = 0; + } + + mi_assert_internal(page->capacity == 0); + mi_assert_internal(page->free == NULL); + mi_assert_internal(page->used == 0); + mi_assert_internal(page->xthread_free == 0); + mi_assert_internal(page->next == NULL); + mi_assert_internal(page->prev == NULL); + mi_assert_internal(page->retire_expire == 0); + mi_assert_internal(!mi_page_has_aligned(page)); + #if (MI_PADDING || MI_ENCODE_FREELIST) + mi_assert_internal(page->keys[0] != 0); + mi_assert_internal(page->keys[1] != 0); + #endif + mi_assert_internal(page->block_size_shift == 0 || (block_size == ((size_t)1 << page->block_size_shift))); + mi_assert_expensive(mi_page_is_valid_init(page)); + + // initialize an initial free list + if (mi_page_extend_free(heap,page,tld)) { + mi_assert(mi_page_immediate_available(page)); + } + return; +} + + +/* ----------------------------------------------------------- + Find pages with free blocks +-------------------------------------------------------------*/ + +// search for a best next page to use for at most N pages (often cut short if immediate blocks are available) +#define MI_MAX_CANDIDATE_SEARCH (4) + +// is the page not yet used up to its reserved space? +static bool mi_page_is_expandable(const mi_page_t* page) { + mi_assert_internal(page != NULL); + mi_assert_internal(page->capacity <= page->reserved); + return (page->capacity < page->reserved); +} + + +// Find a page with free blocks of `page->block_size`. +static mi_page_t* mi_page_queue_find_free_ex(mi_heap_t* heap, mi_page_queue_t* pq, bool first_try) +{ + // search through the pages in "next fit" order + #if MI_STAT + size_t count = 0; + #endif + size_t candidate_count = 0; // we reset this on the first candidate to limit the search + mi_page_t* page_candidate = NULL; // a page with free space + mi_page_t* page = pq->first; + + while (page != NULL) + { + mi_page_t* next = page->next; // remember next + #if MI_STAT + count++; + #endif + candidate_count++; + + // collect freed blocks by us and other threads + _mi_page_free_collect(page, false); + + #if MI_MAX_CANDIDATE_SEARCH > 1 + // search up to N pages for a best candidate + + // is the local free list non-empty? + const bool immediate_available = mi_page_immediate_available(page); + + // if the page is completely full, move it to the `mi_pages_full` + // queue so we don't visit long-lived pages too often. + if (!immediate_available && !mi_page_is_expandable(page)) { + mi_assert_internal(!mi_page_is_in_full(page) && !mi_page_immediate_available(page)); + mi_page_to_full(page, pq); + } + else { + // the page has free space, make it a candidate + // we prefer non-expandable pages with high usage as candidates (to reduce commit, and increase chances of free-ing up pages) + if (page_candidate == NULL) { + page_candidate = page; + candidate_count = 0; + } + // prefer to reuse fuller pages (in the hope the less used page gets freed) + else if (page->used >= page_candidate->used && !mi_page_is_mostly_used(page) && !mi_page_is_expandable(page)) { + page_candidate = page; + } + // if we find a non-expandable candidate, or searched for N pages, return with the best candidate + if (immediate_available || candidate_count > MI_MAX_CANDIDATE_SEARCH) { + mi_assert_internal(page_candidate!=NULL); + break; + } + } + #else + // first-fit algorithm + // If the page contains free blocks, we are done + if (mi_page_immediate_available(page) || mi_page_is_expandable(page)) { + break; // pick this one + } + + // If the page is completely full, move it to the `mi_pages_full` + // queue so we don't visit long-lived pages too often. + mi_assert_internal(!mi_page_is_in_full(page) && !mi_page_immediate_available(page)); + mi_page_to_full(page, pq); + #endif + + page = next; + } // for each page + + mi_heap_stat_counter_increase(heap, page_searches, count); + + // set the page to the best candidate + if (page_candidate != NULL) { + page = page_candidate; + } + if (page != NULL) { + if (!mi_page_immediate_available(page)) { + mi_assert_internal(mi_page_is_expandable(page)); + if (!mi_page_extend_free(heap, page, heap->tld)) { + page = NULL; // failed to extend + } + } + mi_assert_internal(page == NULL || mi_page_immediate_available(page)); + } + + if (page == NULL) { + _mi_heap_collect_retired(heap, false); // perhaps make a page available? + page = mi_page_fresh(heap, pq); + if (page == NULL && first_try) { + // out-of-memory _or_ an abandoned page with free blocks was reclaimed, try once again + page = mi_page_queue_find_free_ex(heap, pq, false); + } + } + else { + // move the page to the front of the queue + mi_page_queue_move_to_front(heap, pq, page); + page->retire_expire = 0; + // _mi_heap_collect_retired(heap, false); // update retire counts; note: increases rss on MemoryLoad bench so don't do this + } + mi_assert_internal(page == NULL || mi_page_immediate_available(page)); + + + return page; +} + + + +// Find a page with free blocks of `size`. +static inline mi_page_t* mi_find_free_page(mi_heap_t* heap, size_t size) { + mi_page_queue_t* pq = mi_page_queue(heap, size); + + // check the first page: we even do this with candidate search or otherwise we re-search every time + mi_page_t* page = pq->first; + if (page != NULL) { + #if (MI_SECURE>=3) // in secure mode, we extend half the time to increase randomness + if (page->capacity < page->reserved && ((_mi_heap_random_next(heap) & 1) == 1)) { + mi_page_extend_free(heap, page, heap->tld); + mi_assert_internal(mi_page_immediate_available(page)); + } + else + #endif + { + _mi_page_free_collect(page,false); + } + + if (mi_page_immediate_available(page)) { + page->retire_expire = 0; + return page; // fast path + } + } + + return mi_page_queue_find_free_ex(heap, pq, true); +} + + +/* ----------------------------------------------------------- + Users can register a deferred free function called + when the `free` list is empty. Since the `local_free` + is separate this is deterministically called after + a certain number of allocations. +----------------------------------------------------------- */ + +static mi_deferred_free_fun* volatile deferred_free = NULL; +static _Atomic(void*) deferred_arg; // = NULL + +void _mi_deferred_free(mi_heap_t* heap, bool force) { + heap->tld->heartbeat++; + if (deferred_free != NULL && !heap->tld->recurse) { + heap->tld->recurse = true; + deferred_free(force, heap->tld->heartbeat, mi_atomic_load_ptr_relaxed(void,&deferred_arg)); + heap->tld->recurse = false; + } +} + +void mi_register_deferred_free(mi_deferred_free_fun* fn, void* arg) mi_attr_noexcept { + deferred_free = fn; + mi_atomic_store_ptr_release(void,&deferred_arg, arg); +} + + +/* ----------------------------------------------------------- + General allocation +----------------------------------------------------------- */ + +// Large and huge page allocation. +// Huge pages contain just one block, and the segment contains just that page (as `MI_SEGMENT_HUGE`). +// Huge pages are also use if the requested alignment is very large (> MI_BLOCK_ALIGNMENT_MAX) +// so their size is not always `> MI_LARGE_OBJ_SIZE_MAX`. +static mi_page_t* mi_large_huge_page_alloc(mi_heap_t* heap, size_t size, size_t page_alignment) { + size_t block_size = _mi_os_good_alloc_size(size); + mi_assert_internal(mi_bin(block_size) == MI_BIN_HUGE || page_alignment > 0); + bool is_huge = (block_size > MI_LARGE_OBJ_SIZE_MAX || page_alignment > 0); + #if MI_HUGE_PAGE_ABANDON + mi_page_queue_t* pq = (is_huge ? NULL : mi_page_queue(heap, block_size)); + #else + mi_page_queue_t* pq = mi_page_queue(heap, is_huge ? MI_LARGE_OBJ_SIZE_MAX+1 : block_size); + mi_assert_internal(!is_huge || mi_page_queue_is_huge(pq)); + #endif + mi_page_t* page = mi_page_fresh_alloc(heap, pq, block_size, page_alignment); + if (page != NULL) { + mi_assert_internal(mi_page_immediate_available(page)); + + if (is_huge) { + mi_assert_internal(mi_page_is_huge(page)); + mi_assert_internal(_mi_page_segment(page)->kind == MI_SEGMENT_HUGE); + mi_assert_internal(_mi_page_segment(page)->used==1); + #if MI_HUGE_PAGE_ABANDON + mi_assert_internal(_mi_page_segment(page)->thread_id==0); // abandoned, not in the huge queue + mi_page_set_heap(page, NULL); + #endif + } + else { + mi_assert_internal(!mi_page_is_huge(page)); + } + + const size_t bsize = mi_page_usable_block_size(page); // note: not `mi_page_block_size` to account for padding + /*if (bsize <= MI_LARGE_OBJ_SIZE_MAX) { + mi_heap_stat_increase(heap, malloc_large, bsize); + mi_heap_stat_counter_increase(heap, malloc_large_count, 1); + } + else */ + { + _mi_stat_increase(&heap->tld->stats.malloc_huge, bsize); + _mi_stat_counter_increase(&heap->tld->stats.malloc_huge_count, 1); + } + } + return page; +} + + +// Allocate a page +// Note: in debug mode the size includes MI_PADDING_SIZE and might have overflowed. +static mi_page_t* mi_find_page(mi_heap_t* heap, size_t size, size_t huge_alignment) mi_attr_noexcept { + // huge allocation? + const size_t req_size = size - MI_PADDING_SIZE; // correct for padding_size in case of an overflow on `size` + if mi_unlikely(req_size > (MI_MEDIUM_OBJ_SIZE_MAX - MI_PADDING_SIZE) || huge_alignment > 0) { + if mi_unlikely(req_size > MI_MAX_ALLOC_SIZE) { + _mi_error_message(EOVERFLOW, "allocation request is too large (%zu bytes)\n", req_size); + return NULL; + } + else { + return mi_large_huge_page_alloc(heap,size,huge_alignment); + } + } + else { + // otherwise find a page with free blocks in our size segregated queues + #if MI_PADDING + mi_assert_internal(size >= MI_PADDING_SIZE); + #endif + return mi_find_free_page(heap, size); + } +} + +// Generic allocation routine if the fast path (`alloc.c:mi_page_malloc`) does not succeed. +// Note: in debug mode the size includes MI_PADDING_SIZE and might have overflowed. +// The `huge_alignment` is normally 0 but is set to a multiple of MI_SEGMENT_SIZE for +// very large requested alignments in which case we use a huge segment. +void* _mi_malloc_generic(mi_heap_t* heap, size_t size, bool zero, size_t huge_alignment) mi_attr_noexcept +{ + mi_assert_internal(heap != NULL); + + // initialize if necessary + if mi_unlikely(!mi_heap_is_initialized(heap)) { + heap = mi_heap_get_default(); // calls mi_thread_init + if mi_unlikely(!mi_heap_is_initialized(heap)) { return NULL; } + } + mi_assert_internal(mi_heap_is_initialized(heap)); + + // do administrative tasks every N generic mallocs + if mi_unlikely(++heap->generic_count >= 100) { + heap->generic_collect_count += heap->generic_count; + heap->generic_count = 0; + // call potential deferred free routines + _mi_deferred_free(heap, false); + + // free delayed frees from other threads (but skip contended ones) + _mi_heap_delayed_free_partial(heap); + + // collect every once in a while (10000 by default) + const long generic_collect = mi_option_get_clamp(mi_option_generic_collect, 1, 1000000L); + if (heap->generic_collect_count >= generic_collect) { + heap->generic_collect_count = 0; + mi_heap_collect(heap, false /* force? */); + } + } + + // find (or allocate) a page of the right size + mi_page_t* page = mi_find_page(heap, size, huge_alignment); + if mi_unlikely(page == NULL) { // first time out of memory, try to collect and retry the allocation once more + mi_heap_collect(heap, true /* force */); + page = mi_find_page(heap, size, huge_alignment); + } + + if mi_unlikely(page == NULL) { // out of memory + const size_t req_size = size - MI_PADDING_SIZE; // correct for padding_size in case of an overflow on `size` + _mi_error_message(ENOMEM, "unable to allocate memory (%zu bytes)\n", req_size); + return NULL; + } + + mi_assert_internal(mi_page_immediate_available(page)); + mi_assert_internal(mi_page_block_size(page) >= size); + + // and try again, this time succeeding! (i.e. this should never recurse through _mi_page_malloc) + void* p; + if mi_unlikely(zero && mi_page_is_huge(page)) { + // note: we cannot call _mi_page_malloc with zeroing for huge blocks; we zero it afterwards in that case. + p = _mi_page_malloc(heap, page, size); + mi_assert_internal(p != NULL); + _mi_memzero_aligned(p, mi_page_usable_block_size(page)); + } + else { + p = _mi_page_malloc_zero(heap, page, size, zero); + mi_assert_internal(p != NULL); + } + // move singleton pages to the full queue + if (page->reserved == page->used) { + mi_page_to_full(page, mi_page_queue_of(page)); + } + return p; +} diff --git a/compat/mimalloc/prim/osx/prim.c b/compat/mimalloc/prim/osx/prim.c new file mode 100644 index 00000000000000..8a2f4e8aa47316 --- /dev/null +++ b/compat/mimalloc/prim/osx/prim.c @@ -0,0 +1,9 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +// We use the unix/prim.c with the mmap API on macOSX +#include "../unix/prim.c" diff --git a/compat/mimalloc/prim/prim.c b/compat/mimalloc/prim/prim.c new file mode 100644 index 00000000000000..5147bae81feaaf --- /dev/null +++ b/compat/mimalloc/prim/prim.c @@ -0,0 +1,76 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +// Select the implementation of the primitives +// depending on the OS. + +#if defined(_WIN32) +#include "windows/prim.c" // VirtualAlloc (Windows) + +#elif defined(__APPLE__) +#include "osx/prim.c" // macOSX (actually defers to mmap in unix/prim.c) + +#elif defined(__wasi__) +#define MI_USE_SBRK +#include "wasi/prim.c" // memory-grow or sbrk (Wasm) + +#elif defined(__EMSCRIPTEN__) +#include "emscripten/prim.c" // emmalloc_*, + pthread support + +#else +#include "unix/prim.c" // mmap() (Linux, macOSX, BSD, Illumnos, Haiku, DragonFly, etc.) + +#endif + +// Generic process initialization +#ifndef MI_PRIM_HAS_PROCESS_ATTACH +#if defined(__GNUC__) || defined(__clang__) + // gcc,clang: use the constructor/destructor attribute + // which for both seem to run before regular constructors/destructors + #if defined(__clang__) + #define mi_attr_constructor __attribute__((constructor(101))) + #define mi_attr_destructor __attribute__((destructor(101))) + #else + #define mi_attr_constructor __attribute__((constructor)) + #define mi_attr_destructor __attribute__((destructor)) + #endif + static void mi_attr_constructor mi_process_attach(void) { + _mi_auto_process_init(); + } + static void mi_attr_destructor mi_process_detach(void) { + _mi_auto_process_done(); + } +#elif defined(__cplusplus) + // C++: use static initialization to detect process start/end + // This is not guaranteed to be first/last but the best we can generally do? + struct mi_init_done_t { + mi_init_done_t() { + _mi_auto_process_init(); + } + ~mi_init_done_t() { + _mi_auto_process_done(); + } + }; + static mi_init_done_t mi_init_done; + #else + #pragma message("define a way to call _mi_auto_process_init/done on your platform") +#endif +#endif + +// Generic allocator init/done callback +#ifndef MI_PRIM_HAS_ALLOCATOR_INIT +bool _mi_is_redirected(void) { + return false; +} +bool _mi_allocator_init(const char** message) { + if (message != NULL) { *message = NULL; } + return true; +} +void _mi_allocator_done(void) { + // nothing to do +} +#endif diff --git a/compat/mimalloc/prim/unix/prim.c b/compat/mimalloc/prim/unix/prim.c new file mode 100644 index 00000000000000..650aa657b9eb18 --- /dev/null +++ b/compat/mimalloc/prim/unix/prim.c @@ -0,0 +1,934 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2025, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +// This file is included in `src/prim/prim.c` + +#ifndef _DEFAULT_SOURCE +#define _DEFAULT_SOURCE // ensure mmap flags and syscall are defined +#endif + +#if defined(__sun) +// illumos provides new mman.h api when any of these are defined +// otherwise the old api based on caddr_t which predates the void pointers one. +// stock solaris provides only the former, chose to atomically to discard those +// flags only here rather than project wide tough. +#undef _XOPEN_SOURCE +#undef _POSIX_C_SOURCE +#endif + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" + +#include // mmap +#include // sysconf +#include // open, close, read, access +#include // getenv, arc4random_buf + +#if defined(__linux__) + #include + #include // THP disable, PR_SET_VMA + #if defined(__GLIBC__) && !defined(PR_SET_VMA) + #include + #endif + #if defined(__GLIBC__) + #include // linux mmap flags + #else + #include + #endif +#elif defined(__APPLE__) + #include + #include + #if !defined(TARGET_OS_OSX) || TARGET_OS_OSX // see issue #879, used to be (!TARGET_IOS_IPHONE && !TARGET_IOS_SIMULATOR) + #include // VM_MAKE_TAG, VM_FLAGS_SUPERPAGE_SIZE_2MB, etc. + #endif + #if !defined(MAC_OS_X_VERSION_10_7) + #define MAC_OS_X_VERSION_10_7 1070 + #endif +#elif defined(__FreeBSD__) || defined(__DragonFly__) + #include + #if __FreeBSD_version >= 1200000 + #include + #include + #endif + #include +#endif + +#if (defined(__linux__) && !defined(__ANDROID__)) || defined(__FreeBSD__) + #define MI_HAS_SYSCALL_H + #include +#endif + +#if !defined(MADV_DONTNEED) && defined(POSIX_MADV_DONTNEED) // QNX +#define MADV_DONTNEED POSIX_MADV_DONTNEED +#endif +#if !defined(MADV_FREE) && defined(POSIX_MADV_FREE) // QNX +#define MADV_FREE POSIX_MADV_FREE +#endif + +#define MI_UNIX_LARGE_PAGE_SIZE (2*MI_MiB) // TODO: can we query the OS for this? + +//------------------------------------------------------------------------------------ +// Use syscalls for some primitives to allow for libraries that override open/read/close etc. +// and do allocation themselves; using syscalls prevents recursion when mimalloc is +// still initializing (issue #713) +// Declare inline to avoid unused function warnings. +//------------------------------------------------------------------------------------ + +#if defined(MI_HAS_SYSCALL_H) && defined(SYS_open) && defined(SYS_close) && defined(SYS_read) && defined(SYS_access) + +static inline int mi_prim_open(const char* fpath, int open_flags) { + return syscall(SYS_open,fpath,open_flags,0); +} +static inline ssize_t mi_prim_read(int fd, void* buf, size_t bufsize) { + return syscall(SYS_read,fd,buf,bufsize); +} +static inline int mi_prim_close(int fd) { + return syscall(SYS_close,fd); +} +static inline int mi_prim_access(const char *fpath, int mode) { + return syscall(SYS_access,fpath,mode); +} + +#else + +static inline int mi_prim_open(const char* fpath, int open_flags) { + return open(fpath,open_flags); +} +static inline ssize_t mi_prim_read(int fd, void* buf, size_t bufsize) { + return read(fd,buf,bufsize); +} +static inline int mi_prim_close(int fd) { + return close(fd); +} +static inline int mi_prim_access(const char *fpath, int mode) { + return access(fpath,mode); +} + +#endif + + + +//--------------------------------------------- +// init +//--------------------------------------------- + +static bool unix_detect_overcommit(void) { + bool os_overcommit = true; +#if defined(__linux__) + int fd = mi_prim_open("/proc/sys/vm/overcommit_memory", O_RDONLY); + if (fd >= 0) { + char buf[32]; + ssize_t nread = mi_prim_read(fd, &buf, sizeof(buf)); + mi_prim_close(fd); + // + // 0: heuristic overcommit, 1: always overcommit, 2: never overcommit (ignore NORESERVE) + if (nread >= 1) { + os_overcommit = (buf[0] == '0' || buf[0] == '1'); + } + } +#elif defined(__FreeBSD__) + int val = 0; + size_t olen = sizeof(val); + if (sysctlbyname("vm.overcommit", &val, &olen, NULL, 0) == 0) { + os_overcommit = (val != 0); + } +#else + // default: overcommit is true +#endif + return os_overcommit; +} + +void _mi_prim_mem_init( mi_os_mem_config_t* config ) +{ + long psize = sysconf(_SC_PAGESIZE); + if (psize > 0) { + config->page_size = (size_t)psize; + config->alloc_granularity = (size_t)psize; + #if defined(_SC_PHYS_PAGES) + long pphys = sysconf(_SC_PHYS_PAGES); + const size_t psize_in_kib = (size_t)psize / MI_KiB; + if (psize_in_kib > 0 && pphys > 0 && (size_t)pphys <= (SIZE_MAX/psize_in_kib)) { + config->physical_memory_in_kib = (size_t)pphys * psize_in_kib; + } + #endif + } + config->large_page_size = MI_UNIX_LARGE_PAGE_SIZE; + config->has_overcommit = unix_detect_overcommit(); + config->has_partial_free = true; // mmap can free in parts + config->has_virtual_reserve = true; // todo: check if this true for NetBSD? (for anonymous mmap with PROT_NONE) + + // disable transparent huge pages for this process? + #if (defined(__linux__) || defined(__ANDROID__)) && defined(PR_GET_THP_DISABLE) + #if defined(MI_NO_THP) + if (true) + #else + if (!mi_option_is_enabled(mi_option_allow_large_os_pages)) // disable THP also if large OS pages are not allowed in the options + #endif + { + int val = 0; + if (prctl(PR_GET_THP_DISABLE, &val, 0, 0, 0) != 0) { + // Most likely since distros often come with always/madvise settings. + val = 1; + // Disabling only for mimalloc process rather than touching system wide settings + (void)prctl(PR_SET_THP_DISABLE, &val, 0, 0, 0); + } + } + #endif +} + + +//--------------------------------------------- +// free +//--------------------------------------------- + +int _mi_prim_free(void* addr, size_t size ) { + if (size==0) return 0; + bool err = (munmap(addr, size) == -1); + return (err ? errno : 0); +} + + +//--------------------------------------------- +// mmap +//--------------------------------------------- + +static int unix_madvise(void* addr, size_t size, int advice) { + #if defined(__sun) + int res = madvise((caddr_t)addr, size, advice); // Solaris needs cast (issue #520) + #elif defined(__QNX__) + int res = posix_madvise(addr, size, advice); + #else + int res = madvise(addr, size, advice); + #endif + return (res==0 ? 0 : errno); +} + +static void* unix_mmap_prim(void* addr, size_t size, int protect_flags, int flags, int fd) { + void* p = mmap(addr, size, protect_flags, flags, fd, 0 /* offset */); + #if defined(__linux__) && defined(PR_SET_VMA) + if (p!=MAP_FAILED && p!=NULL) { + prctl(PR_SET_VMA, PR_SET_VMA_ANON_NAME, p, size, "mimalloc"); + } + #endif + return p; +} + +static void* unix_mmap_prim_aligned(void* addr, size_t size, size_t try_alignment, int protect_flags, int flags, int fd) { + MI_UNUSED(try_alignment); + void* p = NULL; + #if defined(MAP_ALIGNED) // BSD + if (addr == NULL && try_alignment > 1 && (try_alignment % _mi_os_page_size()) == 0) { + size_t n = mi_bsr(try_alignment); + if (((size_t)1 << n) == try_alignment && n >= 12 && n <= 30) { // alignment is a power of 2 and 4096 <= alignment <= 1GiB + p = unix_mmap_prim(addr, size, protect_flags, flags | MAP_ALIGNED(n), fd); + if (p==MAP_FAILED || !_mi_is_aligned(p,try_alignment)) { + int err = errno; + _mi_trace_message("unable to directly request aligned OS memory (error: %d (0x%x), size: 0x%zx bytes, alignment: 0x%zx, hint address: %p)\n", err, err, size, try_alignment, addr); + } + if (p!=MAP_FAILED) return p; + // fall back to regular mmap + } + } + #elif defined(MAP_ALIGN) // Solaris + if (addr == NULL && try_alignment > 1 && (try_alignment % _mi_os_page_size()) == 0) { + p = unix_mmap_prim((void*)try_alignment, size, protect_flags, flags | MAP_ALIGN, fd); // addr parameter is the required alignment + if (p!=MAP_FAILED) return p; + // fall back to regular mmap + } + #endif + #if (MI_INTPTR_SIZE >= 8) && !defined(MAP_ALIGNED) + // on 64-bit systems, use the virtual address area after 2TiB for 4MiB aligned allocations + if (addr == NULL) { + void* hint = _mi_os_get_aligned_hint(try_alignment, size); + if (hint != NULL) { + p = unix_mmap_prim(hint, size, protect_flags, flags, fd); + if (p==MAP_FAILED || !_mi_is_aligned(p,try_alignment)) { + #if MI_TRACK_ENABLED // asan sometimes does not instrument errno correctly? + int err = 0; + #else + int err = errno; + #endif + _mi_trace_message("unable to directly request hinted aligned OS memory (error: %d (0x%x), size: 0x%zx bytes, alignment: 0x%zx, hint address: %p)\n", err, err, size, try_alignment, hint); + } + if (p!=MAP_FAILED) return p; + // fall back to regular mmap + } + } + #endif + // regular mmap + p = unix_mmap_prim(addr, size, protect_flags, flags, fd); + if (p!=MAP_FAILED) return p; + // failed to allocate + return NULL; +} + +static int unix_mmap_fd(void) { + #if defined(VM_MAKE_TAG) + // macOS: tracking anonymous page with a specific ID. (All up to 98 are taken officially but LLVM sanitizers had taken 99) + int os_tag = (int)mi_option_get(mi_option_os_tag); + if (os_tag < 100 || os_tag > 255) { os_tag = 254; } + return VM_MAKE_TAG(os_tag); + #else + return -1; + #endif +} + +static void* unix_mmap(void* addr, size_t size, size_t try_alignment, int protect_flags, bool large_only, bool allow_large, bool* is_large) { + #if !defined(MAP_ANONYMOUS) + #define MAP_ANONYMOUS MAP_ANON + #endif + #if !defined(MAP_NORESERVE) + #define MAP_NORESERVE 0 + #endif + void* p = NULL; + const int fd = unix_mmap_fd(); + int flags = MAP_PRIVATE | MAP_ANONYMOUS; + if (_mi_os_has_overcommit()) { + flags |= MAP_NORESERVE; + } + #if defined(PROT_MAX) + protect_flags |= PROT_MAX(PROT_READ | PROT_WRITE); // BSD + #endif + // huge page allocation + if (allow_large && (large_only || (_mi_os_use_large_page(size, try_alignment) && mi_option_get(mi_option_allow_large_os_pages) == 1))) { + static _Atomic(size_t) large_page_try_ok; // = 0; + size_t try_ok = mi_atomic_load_acquire(&large_page_try_ok); + if (!large_only && try_ok > 0) { + // If the OS is not configured for large OS pages, or the user does not have + // enough permission, the `mmap` will always fail (but it might also fail for other reasons). + // Therefore, once a large page allocation failed, we don't try again for `large_page_try_ok` times + // to avoid too many failing calls to mmap. + mi_atomic_cas_strong_acq_rel(&large_page_try_ok, &try_ok, try_ok - 1); + } + else { + int lflags = flags & ~MAP_NORESERVE; // using NORESERVE on huge pages seems to fail on Linux + int lfd = fd; + #ifdef MAP_ALIGNED_SUPER + lflags |= MAP_ALIGNED_SUPER; + #endif + #ifdef MAP_HUGETLB + lflags |= MAP_HUGETLB; + #endif + #ifdef MAP_HUGE_1GB + static bool mi_huge_pages_available = true; + if (large_only && (size % MI_GiB) == 0 && mi_huge_pages_available) { + lflags |= MAP_HUGE_1GB; + } + else + #endif + { + #ifdef MAP_HUGE_2MB + lflags |= MAP_HUGE_2MB; + #endif + } + #ifdef VM_FLAGS_SUPERPAGE_SIZE_2MB + lfd |= VM_FLAGS_SUPERPAGE_SIZE_2MB; + #endif + if (large_only || lflags != flags) { + // try large OS page allocation + *is_large = true; + p = unix_mmap_prim_aligned(addr, size, try_alignment, protect_flags, lflags, lfd); + #ifdef MAP_HUGE_1GB + if (p == NULL && (lflags & MAP_HUGE_1GB) == MAP_HUGE_1GB) { + mi_huge_pages_available = false; // don't try huge 1GiB pages again + if (large_only) { + _mi_warning_message("unable to allocate huge (1GiB) page, trying large (2MiB) pages instead (errno: %i)\n", errno); + } + lflags = ((lflags & ~MAP_HUGE_1GB) | MAP_HUGE_2MB); + p = unix_mmap_prim_aligned(addr, size, try_alignment, protect_flags, lflags, lfd); + } + #endif + if (large_only) return p; + if (p == NULL) { + mi_atomic_store_release(&large_page_try_ok, (size_t)8); // on error, don't try again for the next N allocations + } + } + } + } + // regular allocation + if (p == NULL) { + *is_large = false; + p = unix_mmap_prim_aligned(addr, size, try_alignment, protect_flags, flags, fd); + if (p != NULL) { + #if defined(MADV_HUGEPAGE) + // Many Linux systems don't allow MAP_HUGETLB but they support instead + // transparent huge pages (THP). Generally, it is not required to call `madvise` with MADV_HUGE + // though since properly aligned allocations will already use large pages if available + // in that case -- in particular for our large regions (in `memory.c`). + // However, some systems only allow THP if called with explicit `madvise`, so + // when large OS pages are enabled for mimalloc, we call `madvise` anyways. + if (allow_large && _mi_os_use_large_page(size, try_alignment)) { + if (unix_madvise(p, size, MADV_HUGEPAGE) == 0) { + // *is_large = true; // possibly + }; + } + #elif defined(__sun) + if (allow_large && _mi_os_use_large_page(size, try_alignment)) { + struct memcntl_mha cmd = {0}; + cmd.mha_pagesize = _mi_os_large_page_size(); + cmd.mha_cmd = MHA_MAPSIZE_VA; + if (memcntl((caddr_t)p, size, MC_HAT_ADVISE, (caddr_t)&cmd, 0, 0) == 0) { + // *is_large = true; // possibly + } + } + #endif + } + } + return p; +} + +// Note: the `try_alignment` is just a hint and the returned pointer is not guaranteed to be aligned. +int _mi_prim_alloc(void* hint_addr, size_t size, size_t try_alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero, void** addr) { + mi_assert_internal(size > 0 && (size % _mi_os_page_size()) == 0); + mi_assert_internal(commit || !allow_large); + mi_assert_internal(try_alignment > 0); + if (hint_addr == NULL && size >= 8*MI_UNIX_LARGE_PAGE_SIZE && try_alignment > 1 && _mi_is_power_of_two(try_alignment) && try_alignment < MI_UNIX_LARGE_PAGE_SIZE) { + try_alignment = MI_UNIX_LARGE_PAGE_SIZE; // try to align along large page size for larger allocations + } + + *is_zero = true; + int protect_flags = (commit ? (PROT_WRITE | PROT_READ) : PROT_NONE); + *addr = unix_mmap(hint_addr, size, try_alignment, protect_flags, false, allow_large, is_large); + return (*addr != NULL ? 0 : errno); +} + + +//--------------------------------------------- +// Commit/Reset +//--------------------------------------------- + +static void unix_mprotect_hint(int err) { + #if defined(__linux__) && (MI_SECURE>=2) // guard page around every mimalloc page + if (err == ENOMEM) { + _mi_warning_message("The next warning may be caused by a low memory map limit.\n" + " On Linux this is controlled by the vm.max_map_count -- maybe increase it?\n" + " For example: sudo sysctl -w vm.max_map_count=262144\n"); + } + #else + MI_UNUSED(err); + #endif +} + +int _mi_prim_commit(void* start, size_t size, bool* is_zero) { + // commit: ensure we can access the area + // note: we may think that *is_zero can be true since the memory + // was either from mmap PROT_NONE, or from decommit MADV_DONTNEED, but + // we sometimes call commit on a range with still partially committed + // memory and `mprotect` does not zero the range. + *is_zero = false; + int err = mprotect(start, size, (PROT_READ | PROT_WRITE)); + if (err != 0) { + err = errno; + unix_mprotect_hint(err); + } + return err; +} + +int _mi_prim_reuse(void* start, size_t size) { + MI_UNUSED(start); MI_UNUSED(size); + #if defined(__APPLE__) && defined(MADV_FREE_REUSE) + return unix_madvise(start, size, MADV_FREE_REUSE); + #endif + return 0; +} + +int _mi_prim_decommit(void* start, size_t size, bool* needs_recommit) { + int err = 0; + #if defined(__APPLE__) && defined(MADV_FREE_REUSABLE) + // decommit on macOS: use MADV_FREE_REUSABLE as it does immediate rss accounting (issue #1097) + err = unix_madvise(start, size, MADV_FREE_REUSABLE); + if (err) { err = unix_madvise(start, size, MADV_DONTNEED); } + #else + // decommit: use MADV_DONTNEED as it decreases rss immediately (unlike MADV_FREE) + err = unix_madvise(start, size, MADV_DONTNEED); + #endif + #if !MI_DEBUG && MI_SECURE<=2 + *needs_recommit = false; + #else + *needs_recommit = true; + mprotect(start, size, PROT_NONE); + #endif + /* + // decommit: use mmap with MAP_FIXED and PROT_NONE to discard the existing memory (and reduce rss) + *needs_recommit = true; + const int fd = unix_mmap_fd(); + void* p = mmap(start, size, PROT_NONE, (MAP_FIXED | MAP_PRIVATE | MAP_ANONYMOUS | MAP_NORESERVE), fd, 0); + if (p != start) { err = errno; } + */ + return err; +} + +int _mi_prim_reset(void* start, size_t size) { + int err = 0; + + // on macOS can use MADV_FREE_REUSABLE (but we disable this for now as it seems slower) + #if 0 && defined(__APPLE__) && defined(MADV_FREE_REUSABLE) + err = unix_madvise(start, size, MADV_FREE_REUSABLE); + if (err==0) return 0; + // fall through + #endif + + #if defined(MADV_FREE) + // Otherwise, we try to use `MADV_FREE` as that is the fastest. A drawback though is that it + // will not reduce the `rss` stats in tools like `top` even though the memory is available + // to other processes. With the default `MIMALLOC_PURGE_DECOMMITS=1` we ensure that by + // default `MADV_DONTNEED` is used though. + static _Atomic(size_t) advice = MI_ATOMIC_VAR_INIT(MADV_FREE); + int oadvice = (int)mi_atomic_load_relaxed(&advice); + while ((err = unix_madvise(start, size, oadvice)) != 0 && errno == EAGAIN) { errno = 0; }; + if (err != 0 && errno == EINVAL && oadvice == MADV_FREE) { + // if MADV_FREE is not supported, fall back to MADV_DONTNEED from now on + mi_atomic_store_release(&advice, (size_t)MADV_DONTNEED); + err = unix_madvise(start, size, MADV_DONTNEED); + } + #else + err = unix_madvise(start, size, MADV_DONTNEED); + #endif + return err; +} + +int _mi_prim_protect(void* start, size_t size, bool protect) { + int err = mprotect(start, size, protect ? PROT_NONE : (PROT_READ | PROT_WRITE)); + if (err != 0) { err = errno; } + unix_mprotect_hint(err); + return err; +} + + + +//--------------------------------------------- +// Huge page allocation +//--------------------------------------------- + +#if (MI_INTPTR_SIZE >= 8) && !defined(__HAIKU__) && !defined(__CYGWIN__) + +#ifndef MPOL_PREFERRED +#define MPOL_PREFERRED 1 +#endif + +#if defined(MI_HAS_SYSCALL_H) && defined(SYS_mbind) +static long mi_prim_mbind(void* start, unsigned long len, unsigned long mode, const unsigned long* nmask, unsigned long maxnode, unsigned flags) { + return syscall(SYS_mbind, start, len, mode, nmask, maxnode, flags); +} +#else +static long mi_prim_mbind(void* start, unsigned long len, unsigned long mode, const unsigned long* nmask, unsigned long maxnode, unsigned flags) { + MI_UNUSED(start); MI_UNUSED(len); MI_UNUSED(mode); MI_UNUSED(nmask); MI_UNUSED(maxnode); MI_UNUSED(flags); + return 0; +} +#endif + +int _mi_prim_alloc_huge_os_pages(void* hint_addr, size_t size, int numa_node, bool* is_zero, void** addr) { + bool is_large = true; + *is_zero = true; + *addr = unix_mmap(hint_addr, size, MI_SEGMENT_SIZE, PROT_READ | PROT_WRITE, true, true, &is_large); + if (*addr != NULL && numa_node >= 0 && numa_node < 8*MI_INTPTR_SIZE) { // at most 64 nodes + unsigned long numa_mask = (1UL << numa_node); + // TODO: does `mbind` work correctly for huge OS pages? should we + // use `set_mempolicy` before calling mmap instead? + // see: + long err = mi_prim_mbind(*addr, size, MPOL_PREFERRED, &numa_mask, 8*MI_INTPTR_SIZE, 0); + if (err != 0) { + err = errno; + _mi_warning_message("failed to bind huge (1GiB) pages to numa node %d (error: %d (0x%x))\n", numa_node, err, err); + } + } + return (*addr != NULL ? 0 : errno); +} + +#else + +int _mi_prim_alloc_huge_os_pages(void* hint_addr, size_t size, int numa_node, bool* is_zero, void** addr) { + MI_UNUSED(hint_addr); MI_UNUSED(size); MI_UNUSED(numa_node); + *is_zero = false; + *addr = NULL; + return ENOMEM; +} + +#endif + +//--------------------------------------------- +// NUMA nodes +//--------------------------------------------- + +#if defined(__linux__) + +size_t _mi_prim_numa_node(void) { + #if defined(MI_HAS_SYSCALL_H) && defined(SYS_getcpu) + unsigned long node = 0; + unsigned long ncpu = 0; + long err = syscall(SYS_getcpu, &ncpu, &node, NULL); + if (err != 0) return 0; + return node; + #else + return 0; + #endif +} + +size_t _mi_prim_numa_node_count(void) { + char buf[128]; + unsigned node = 0; + for(node = 0; node < 256; node++) { + // enumerate node entries -- todo: it there a more efficient way to do this? (but ensure there is no allocation) + _mi_snprintf(buf, 127, "/sys/devices/system/node/node%u", node + 1); + if (mi_prim_access(buf,R_OK) != 0) break; + } + return (node+1); +} + +#elif defined(__FreeBSD__) && __FreeBSD_version >= 1200000 + +size_t _mi_prim_numa_node(void) { + domainset_t dom; + size_t node; + int policy; + if (cpuset_getdomain(CPU_LEVEL_CPUSET, CPU_WHICH_PID, -1, sizeof(dom), &dom, &policy) == -1) return 0ul; + for (node = 0; node < MAXMEMDOM; node++) { + if (DOMAINSET_ISSET(node, &dom)) return node; + } + return 0ul; +} + +size_t _mi_prim_numa_node_count(void) { + size_t ndomains = 0; + size_t len = sizeof(ndomains); + if (sysctlbyname("vm.ndomains", &ndomains, &len, NULL, 0) == -1) return 0ul; + return ndomains; +} + +#elif defined(__DragonFly__) + +size_t _mi_prim_numa_node(void) { + // TODO: DragonFly does not seem to provide any userland means to get this information. + return 0ul; +} + +size_t _mi_prim_numa_node_count(void) { + size_t ncpus = 0, nvirtcoresperphys = 0; + size_t len = sizeof(size_t); + if (sysctlbyname("hw.ncpu", &ncpus, &len, NULL, 0) == -1) return 0ul; + if (sysctlbyname("hw.cpu_topology_ht_ids", &nvirtcoresperphys, &len, NULL, 0) == -1) return 0ul; + return nvirtcoresperphys * ncpus; +} + +#else + +size_t _mi_prim_numa_node(void) { + return 0; +} + +size_t _mi_prim_numa_node_count(void) { + return 1; +} + +#endif + +// ---------------------------------------------------------------- +// Clock +// ---------------------------------------------------------------- + +#include + +#if defined(CLOCK_REALTIME) || defined(CLOCK_MONOTONIC) + +mi_msecs_t _mi_prim_clock_now(void) { + struct timespec t; + #ifdef CLOCK_MONOTONIC + clock_gettime(CLOCK_MONOTONIC, &t); + #else + clock_gettime(CLOCK_REALTIME, &t); + #endif + return ((mi_msecs_t)t.tv_sec * 1000) + ((mi_msecs_t)t.tv_nsec / 1000000); +} + +#else + +// low resolution timer +mi_msecs_t _mi_prim_clock_now(void) { + #if !defined(CLOCKS_PER_SEC) || (CLOCKS_PER_SEC == 1000) || (CLOCKS_PER_SEC == 0) + return (mi_msecs_t)clock(); + #elif (CLOCKS_PER_SEC < 1000) + return (mi_msecs_t)clock() * (1000 / (mi_msecs_t)CLOCKS_PER_SEC); + #else + return (mi_msecs_t)clock() / ((mi_msecs_t)CLOCKS_PER_SEC / 1000); + #endif +} + +#endif + + + + +//---------------------------------------------------------------- +// Process info +//---------------------------------------------------------------- + +#if defined(__unix__) || defined(__unix) || defined(unix) || defined(__APPLE__) || defined(__HAIKU__) +#include +#include +#include + +#if defined(__APPLE__) +#include +#endif + +#if defined(__HAIKU__) +#include +#endif + +static mi_msecs_t timeval_secs(const struct timeval* tv) { + return ((mi_msecs_t)tv->tv_sec * 1000L) + ((mi_msecs_t)tv->tv_usec / 1000L); +} + +void _mi_prim_process_info(mi_process_info_t* pinfo) +{ + struct rusage rusage; + getrusage(RUSAGE_SELF, &rusage); + pinfo->utime = timeval_secs(&rusage.ru_utime); + pinfo->stime = timeval_secs(&rusage.ru_stime); +#if !defined(__HAIKU__) + pinfo->page_faults = rusage.ru_majflt; +#endif +#if defined(__HAIKU__) + // Haiku does not have (yet?) a way to + // get these stats per process + thread_info tid; + area_info mem; + ssize_t c; + get_thread_info(find_thread(0), &tid); + while (get_next_area_info(tid.team, &c, &mem) == B_OK) { + pinfo->peak_rss += mem.ram_size; + } + pinfo->page_faults = 0; +#elif defined(__APPLE__) + pinfo->peak_rss = rusage.ru_maxrss; // macos reports in bytes + #ifdef MACH_TASK_BASIC_INFO + struct mach_task_basic_info info; + mach_msg_type_number_t infoCount = MACH_TASK_BASIC_INFO_COUNT; + if (task_info(mach_task_self(), MACH_TASK_BASIC_INFO, (task_info_t)&info, &infoCount) == KERN_SUCCESS) { + pinfo->current_rss = (size_t)info.resident_size; + } + #else + struct task_basic_info info; + mach_msg_type_number_t infoCount = TASK_BASIC_INFO_COUNT; + if (task_info(mach_task_self(), TASK_BASIC_INFO, (task_info_t)&info, &infoCount) == KERN_SUCCESS) { + pinfo->current_rss = (size_t)info.resident_size; + } + #endif +#else + pinfo->peak_rss = rusage.ru_maxrss * 1024; // Linux/BSD report in KiB +#endif + // use defaults for commit +} + +#else + +#ifndef __wasi__ +// WebAssembly instances are not processes +#pragma message("define a way to get process info") +#endif + +void _mi_prim_process_info(mi_process_info_t* pinfo) +{ + // use defaults + MI_UNUSED(pinfo); +} + +#endif + + +//---------------------------------------------------------------- +// Output +//---------------------------------------------------------------- + +void _mi_prim_out_stderr( const char* msg ) { + fputs(msg,stderr); +} + + +//---------------------------------------------------------------- +// Environment +//---------------------------------------------------------------- + +#if !defined(MI_USE_ENVIRON) || (MI_USE_ENVIRON!=0) +// On Posix systemsr use `environ` to access environment variables +// even before the C runtime is initialized. +#if defined(__APPLE__) && defined(__has_include) && __has_include() +#include +static char** mi_get_environ(void) { + return (*_NSGetEnviron()); +} +#else +extern char** environ; +static char** mi_get_environ(void) { + return environ; +} +#endif +bool _mi_prim_getenv(const char* name, char* result, size_t result_size) { + if (name==NULL) return false; + const size_t len = _mi_strlen(name); + if (len == 0) return false; + char** env = mi_get_environ(); + if (env == NULL) return false; + // compare up to 10000 entries + for (int i = 0; i < 10000 && env[i] != NULL; i++) { + const char* s = env[i]; + if (_mi_strnicmp(name, s, len) == 0 && s[len] == '=') { // case insensitive + // found it + _mi_strlcpy(result, s + len + 1, result_size); + return true; + } + } + return false; +} +#else +// fallback: use standard C `getenv` but this cannot be used while initializing the C runtime +bool _mi_prim_getenv(const char* name, char* result, size_t result_size) { + // cannot call getenv() when still initializing the C runtime. + if (_mi_preloading()) return false; + const char* s = getenv(name); + if (s == NULL) { + // we check the upper case name too. + char buf[64+1]; + size_t len = _mi_strnlen(name,sizeof(buf)-1); + for (size_t i = 0; i < len; i++) { + buf[i] = _mi_toupper(name[i]); + } + buf[len] = 0; + s = getenv(buf); + } + if (s == NULL || _mi_strnlen(s,result_size) >= result_size) return false; + _mi_strlcpy(result, s, result_size); + return true; +} +#endif // !MI_USE_ENVIRON + + +//---------------------------------------------------------------- +// Random +//---------------------------------------------------------------- + +#if defined(__APPLE__) && defined(MAC_OS_X_VERSION_10_15) && (MAC_OS_X_VERSION_MIN_REQUIRED >= MAC_OS_X_VERSION_10_15) +#include +#include + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + // We prefer CCRandomGenerateBytes as it returns an error code while arc4random_buf + // may fail silently on macOS. See PR #390, and + return (CCRandomGenerateBytes(buf, buf_len) == kCCSuccess); +} + +#elif defined(__ANDROID__) || defined(__DragonFly__) || \ + defined(__FreeBSD__) || defined(__NetBSD__) || defined(__OpenBSD__) || \ + defined(__sun) || \ + (defined(__APPLE__) && (MAC_OS_X_VERSION_MIN_REQUIRED >= MAC_OS_X_VERSION_10_7)) + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + arc4random_buf(buf, buf_len); + return true; +} + +#elif defined(__APPLE__) || defined(__linux__) || defined(__HAIKU__) // also for old apple versions < 10.7 (issue #829) + +#include +#include +#include + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + // Modern Linux provides `getrandom` but different distributions either use `sys/random.h` or `linux/random.h` + // and for the latter the actual `getrandom` call is not always defined. + // (see ) + // We therefore use a syscall directly and fall back dynamically to /dev/urandom when needed. + #if defined(MI_HAS_SYSCALL_H) && defined(SYS_getrandom) + #ifndef GRND_NONBLOCK + #define GRND_NONBLOCK (1) + #endif + static _Atomic(uintptr_t) no_getrandom; // = 0 + if (mi_atomic_load_acquire(&no_getrandom)==0) { + ssize_t ret = syscall(SYS_getrandom, buf, buf_len, GRND_NONBLOCK); + if (ret >= 0) return (buf_len == (size_t)ret); + if (errno != ENOSYS) return false; + mi_atomic_store_release(&no_getrandom, (uintptr_t)1); // don't call again, and fall back to /dev/urandom + } + #endif + int flags = O_RDONLY; + #if defined(O_CLOEXEC) + flags |= O_CLOEXEC; + #endif + int fd = mi_prim_open("/dev/urandom", flags); + if (fd < 0) return false; + size_t count = 0; + while(count < buf_len) { + ssize_t ret = mi_prim_read(fd, (char*)buf + count, buf_len - count); + if (ret<=0) { + if (errno!=EAGAIN && errno!=EINTR) break; + } + else { + count += ret; + } + } + mi_prim_close(fd); + return (count==buf_len); +} + +#else + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + return false; +} + +#endif + + +//---------------------------------------------------------------- +// Thread init/done +//---------------------------------------------------------------- + +#if defined(MI_USE_PTHREADS) + +// use pthread local storage keys to detect thread ending +// (and used with MI_TLS_PTHREADS for the default heap) +pthread_key_t _mi_heap_default_key = (pthread_key_t)(-1); + +static void mi_pthread_done(void* value) { + if (value!=NULL) { + _mi_thread_done((mi_heap_t*)value); + } +} + +void _mi_prim_thread_init_auto_done(void) { + mi_assert_internal(_mi_heap_default_key == (pthread_key_t)(-1)); + pthread_key_create(&_mi_heap_default_key, &mi_pthread_done); +} + +void _mi_prim_thread_done_auto_done(void) { + if (_mi_heap_default_key != (pthread_key_t)(-1)) { // do not leak the key, see issue #809 + pthread_key_delete(_mi_heap_default_key); + } +} + +void _mi_prim_thread_associate_default_heap(mi_heap_t* heap) { + if (_mi_heap_default_key != (pthread_key_t)(-1)) { // can happen during recursive invocation on freeBSD + pthread_setspecific(_mi_heap_default_key, heap); + } +} + +#else + +void _mi_prim_thread_init_auto_done(void) { + // nothing +} + +void _mi_prim_thread_done_auto_done(void) { + // nothing +} + +void _mi_prim_thread_associate_default_heap(mi_heap_t* heap) { + MI_UNUSED(heap); +} + +#endif diff --git a/compat/mimalloc/prim/windows/prim.c b/compat/mimalloc/prim/windows/prim.c new file mode 100644 index 00000000000000..eebdc4a67e2f9f --- /dev/null +++ b/compat/mimalloc/prim/windows/prim.c @@ -0,0 +1,878 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +// This file is included in `src/prim/prim.c` + +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" +#include // fputs, stderr + +// xbox has no console IO +#if !defined(WINAPI_FAMILY_PARTITION) || WINAPI_FAMILY_PARTITION(WINAPI_PARTITION_APP | WINAPI_PARTITION_SYSTEM) +#define MI_HAS_CONSOLE_IO +#endif + +//--------------------------------------------- +// Dynamically bind Windows API points for portability +//--------------------------------------------- + +// We use VirtualAlloc2 for aligned allocation, but it is only supported on Windows 10 and Windows Server 2016. +// So, we need to look it up dynamically to run on older systems. (use __stdcall for 32-bit compatibility) +// NtAllocateVirtualAllocEx is used for huge OS page allocation (1GiB) +// We define a minimal MEM_EXTENDED_PARAMETER ourselves in order to be able to compile with older SDK's. +typedef enum MI_MEM_EXTENDED_PARAMETER_TYPE_E { + MiMemExtendedParameterInvalidType = 0, + MiMemExtendedParameterAddressRequirements, + MiMemExtendedParameterNumaNode, + MiMemExtendedParameterPartitionHandle, + MiMemExtendedParameterUserPhysicalHandle, + MiMemExtendedParameterAttributeFlags, + MiMemExtendedParameterMax +} MI_MEM_EXTENDED_PARAMETER_TYPE; + +typedef struct DECLSPEC_ALIGN(8) MI_MEM_EXTENDED_PARAMETER_S { + struct { DWORD64 Type : 8; DWORD64 Reserved : 56; } Type; + union { DWORD64 ULong64; PVOID Pointer; SIZE_T Size; HANDLE Handle; DWORD ULong; } Arg; +} MI_MEM_EXTENDED_PARAMETER; + +typedef struct MI_MEM_ADDRESS_REQUIREMENTS_S { + PVOID LowestStartingAddress; + PVOID HighestEndingAddress; + SIZE_T Alignment; +} MI_MEM_ADDRESS_REQUIREMENTS; + +#define MI_MEM_EXTENDED_PARAMETER_NONPAGED_HUGE 0x00000010 + +#include +typedef PVOID (__stdcall *PVirtualAlloc2)(HANDLE, PVOID, SIZE_T, ULONG, ULONG, MI_MEM_EXTENDED_PARAMETER*, ULONG); +typedef LONG (__stdcall *PNtAllocateVirtualMemoryEx)(HANDLE, PVOID*, SIZE_T*, ULONG, ULONG, MI_MEM_EXTENDED_PARAMETER*, ULONG); // avoid NTSTATUS as it is not defined on xbox (pr #1084) +static PVirtualAlloc2 pVirtualAlloc2 = NULL; +static PNtAllocateVirtualMemoryEx pNtAllocateVirtualMemoryEx = NULL; + +// Similarly, GetNumaProcessorNodeEx is only supported since Windows 7 (and GetNumaNodeProcessorMask is not supported on xbox) +typedef struct MI_PROCESSOR_NUMBER_S { WORD Group; BYTE Number; BYTE Reserved; } MI_PROCESSOR_NUMBER; + +typedef VOID (__stdcall *PGetCurrentProcessorNumberEx)(MI_PROCESSOR_NUMBER* ProcNumber); +typedef BOOL (__stdcall *PGetNumaProcessorNodeEx)(MI_PROCESSOR_NUMBER* Processor, PUSHORT NodeNumber); +typedef BOOL (__stdcall* PGetNumaNodeProcessorMaskEx)(USHORT Node, PGROUP_AFFINITY ProcessorMask); +typedef BOOL (__stdcall *PGetNumaProcessorNode)(UCHAR Processor, PUCHAR NodeNumber); +typedef BOOL (__stdcall* PGetNumaNodeProcessorMask)(UCHAR Node, PULONGLONG ProcessorMask); +typedef BOOL (__stdcall* PGetNumaHighestNodeNumber)(PULONG Node); +static PGetCurrentProcessorNumberEx pGetCurrentProcessorNumberEx = NULL; +static PGetNumaProcessorNodeEx pGetNumaProcessorNodeEx = NULL; +static PGetNumaNodeProcessorMaskEx pGetNumaNodeProcessorMaskEx = NULL; +static PGetNumaProcessorNode pGetNumaProcessorNode = NULL; +static PGetNumaNodeProcessorMask pGetNumaNodeProcessorMask = NULL; +static PGetNumaHighestNodeNumber pGetNumaHighestNodeNumber = NULL; + +// Not available on xbox +typedef SIZE_T(__stdcall* PGetLargePageMinimum)(VOID); +static PGetLargePageMinimum pGetLargePageMinimum = NULL; + +// Available after Windows XP +typedef BOOL (__stdcall *PGetPhysicallyInstalledSystemMemory)( PULONGLONG TotalMemoryInKilobytes ); + +//--------------------------------------------- +// Enable large page support dynamically (if possible) +//--------------------------------------------- + +static bool win_enable_large_os_pages(size_t* large_page_size) +{ + static bool large_initialized = false; + if (large_initialized) return (_mi_os_large_page_size() > 0); + large_initialized = true; + if (pGetLargePageMinimum==NULL) return false; // no large page support (xbox etc.) + + // Try to see if large OS pages are supported + // To use large pages on Windows, we first need access permission + // Set "Lock pages in memory" permission in the group policy editor + // + unsigned long err = 0; + HANDLE token = NULL; + BOOL ok = OpenProcessToken(GetCurrentProcess(), TOKEN_ADJUST_PRIVILEGES | TOKEN_QUERY, &token); + if (ok) { + TOKEN_PRIVILEGES tp; + ok = LookupPrivilegeValue(NULL, TEXT("SeLockMemoryPrivilege"), &tp.Privileges[0].Luid); + if (ok) { + tp.PrivilegeCount = 1; + tp.Privileges[0].Attributes = SE_PRIVILEGE_ENABLED; + ok = AdjustTokenPrivileges(token, FALSE, &tp, 0, (PTOKEN_PRIVILEGES)NULL, 0); + if (ok) { + err = GetLastError(); + ok = (err == ERROR_SUCCESS); + if (ok && large_page_size != NULL && pGetLargePageMinimum != NULL) { + *large_page_size = (*pGetLargePageMinimum)(); + } + } + } + CloseHandle(token); + } + if (!ok) { + if (err == 0) err = GetLastError(); + _mi_warning_message("cannot enable large OS page support, error %lu\n", err); + } + return (ok!=0); +} + + +//--------------------------------------------- +// Initialize +//--------------------------------------------- + +void _mi_prim_mem_init( mi_os_mem_config_t* config ) +{ + config->has_overcommit = false; + config->has_partial_free = false; + config->has_virtual_reserve = true; + // get the page size + SYSTEM_INFO si; + GetSystemInfo(&si); + if (si.dwPageSize > 0) { config->page_size = si.dwPageSize; } + if (si.dwAllocationGranularity > 0) { config->alloc_granularity = si.dwAllocationGranularity; } + // get virtual address bits + if ((uintptr_t)si.lpMaximumApplicationAddress > 0) { + const size_t vbits = MI_SIZE_BITS - mi_clz((uintptr_t)si.lpMaximumApplicationAddress); + config->virtual_address_bits = vbits; + } + + // get the VirtualAlloc2 function + HINSTANCE hDll; + hDll = LoadLibrary(TEXT("kernelbase.dll")); + if (hDll != NULL) { + // use VirtualAlloc2FromApp if possible as it is available to Windows store apps + pVirtualAlloc2 = (PVirtualAlloc2)(void (*)(void))GetProcAddress(hDll, "VirtualAlloc2FromApp"); + if (pVirtualAlloc2==NULL) pVirtualAlloc2 = (PVirtualAlloc2)(void (*)(void))GetProcAddress(hDll, "VirtualAlloc2"); + FreeLibrary(hDll); + } + // NtAllocateVirtualMemoryEx is used for huge page allocation + hDll = LoadLibrary(TEXT("ntdll.dll")); + if (hDll != NULL) { + pNtAllocateVirtualMemoryEx = (PNtAllocateVirtualMemoryEx)(void (*)(void))GetProcAddress(hDll, "NtAllocateVirtualMemoryEx"); + FreeLibrary(hDll); + } + // Try to use Win7+ numa API + hDll = LoadLibrary(TEXT("kernel32.dll")); + if (hDll != NULL) { + pGetCurrentProcessorNumberEx = (PGetCurrentProcessorNumberEx)(void (*)(void))GetProcAddress(hDll, "GetCurrentProcessorNumberEx"); + pGetNumaProcessorNodeEx = (PGetNumaProcessorNodeEx)(void (*)(void))GetProcAddress(hDll, "GetNumaProcessorNodeEx"); + pGetNumaNodeProcessorMaskEx = (PGetNumaNodeProcessorMaskEx)(void (*)(void))GetProcAddress(hDll, "GetNumaNodeProcessorMaskEx"); + pGetNumaProcessorNode = (PGetNumaProcessorNode)(void (*)(void))GetProcAddress(hDll, "GetNumaProcessorNode"); + pGetNumaNodeProcessorMask = (PGetNumaNodeProcessorMask)(void (*)(void))GetProcAddress(hDll, "GetNumaNodeProcessorMask"); + pGetNumaHighestNodeNumber = (PGetNumaHighestNodeNumber)(void (*)(void))GetProcAddress(hDll, "GetNumaHighestNodeNumber"); + pGetLargePageMinimum = (PGetLargePageMinimum)(void (*)(void))GetProcAddress(hDll, "GetLargePageMinimum"); + // Get physical memory (not available on XP, so check dynamically) + PGetPhysicallyInstalledSystemMemory pGetPhysicallyInstalledSystemMemory = (PGetPhysicallyInstalledSystemMemory)(void (*)(void))GetProcAddress(hDll,"GetPhysicallyInstalledSystemMemory"); + if (pGetPhysicallyInstalledSystemMemory != NULL) { + ULONGLONG memInKiB = 0; + if ((*pGetPhysicallyInstalledSystemMemory)(&memInKiB)) { + if (memInKiB > 0 && memInKiB <= SIZE_MAX) { + config->physical_memory_in_kib = (size_t)memInKiB; + } + } + } + FreeLibrary(hDll); + } + // Enable large/huge OS page support? + if (mi_option_is_enabled(mi_option_allow_large_os_pages) || mi_option_is_enabled(mi_option_reserve_huge_os_pages)) { + win_enable_large_os_pages(&config->large_page_size); + } +} + + +//--------------------------------------------- +// Free +//--------------------------------------------- + +int _mi_prim_free(void* addr, size_t size ) { + MI_UNUSED(size); + DWORD errcode = 0; + bool err = (VirtualFree(addr, 0, MEM_RELEASE) == 0); + if (err) { errcode = GetLastError(); } + if (errcode == ERROR_INVALID_ADDRESS) { + // In mi_os_mem_alloc_aligned the fallback path may have returned a pointer inside + // the memory region returned by VirtualAlloc; in that case we need to free using + // the start of the region. + MEMORY_BASIC_INFORMATION info; _mi_memzero_var(info); + VirtualQuery(addr, &info, sizeof(info)); + if (info.AllocationBase < addr && ((uint8_t*)addr - (uint8_t*)info.AllocationBase) < (ptrdiff_t)MI_SEGMENT_SIZE) { + errcode = 0; + err = (VirtualFree(info.AllocationBase, 0, MEM_RELEASE) == 0); + if (err) { errcode = GetLastError(); } + } + } + return (int)errcode; +} + + +//--------------------------------------------- +// VirtualAlloc +//--------------------------------------------- + +static void* win_virtual_alloc_prim_once(void* addr, size_t size, size_t try_alignment, DWORD flags) { + #if (MI_INTPTR_SIZE >= 8) + // on 64-bit systems, try to use the virtual address area after 2TiB for 4MiB aligned allocations + if (addr == NULL) { + void* hint = _mi_os_get_aligned_hint(try_alignment,size); + if (hint != NULL) { + void* p = VirtualAlloc(hint, size, flags, PAGE_READWRITE); + if (p != NULL) return p; + _mi_verbose_message("warning: unable to allocate hinted aligned OS memory (%zu bytes, error code: 0x%x, address: %p, alignment: %zu, flags: 0x%x)\n", size, GetLastError(), hint, try_alignment, flags); + // fall through on error + } + } + #endif + // on modern Windows try use VirtualAlloc2 for aligned allocation + if (addr == NULL && try_alignment > 1 && (try_alignment % _mi_os_page_size()) == 0 && pVirtualAlloc2 != NULL) { + MI_MEM_ADDRESS_REQUIREMENTS reqs = { 0, 0, 0 }; + reqs.Alignment = try_alignment; + MI_MEM_EXTENDED_PARAMETER param = { {0, 0}, {0} }; + param.Type.Type = MiMemExtendedParameterAddressRequirements; + param.Arg.Pointer = &reqs; + void* p = (*pVirtualAlloc2)(GetCurrentProcess(), addr, size, flags, PAGE_READWRITE, ¶m, 1); + if (p != NULL) return p; + _mi_warning_message("unable to allocate aligned OS memory (0x%zx bytes, error code: 0x%x, address: %p, alignment: 0x%zx, flags: 0x%x)\n", size, GetLastError(), addr, try_alignment, flags); + // fall through on error + } + // last resort + return VirtualAlloc(addr, size, flags, PAGE_READWRITE); +} + +static bool win_is_out_of_memory_error(DWORD err) { + switch (err) { + case ERROR_COMMITMENT_MINIMUM: + case ERROR_COMMITMENT_LIMIT: + case ERROR_PAGEFILE_QUOTA: + case ERROR_NOT_ENOUGH_MEMORY: + return true; + default: + return false; + } +} + +static void* win_virtual_alloc_prim(void* addr, size_t size, size_t try_alignment, DWORD flags) { + long max_retry_msecs = mi_option_get_clamp(mi_option_retry_on_oom, 0, 2000); // at most 2 seconds + if (max_retry_msecs == 1) { max_retry_msecs = 100; } // if one sets the option to "true" + for (long tries = 1; tries <= 10; tries++) { // try at most 10 times (=2200ms) + void* p = win_virtual_alloc_prim_once(addr, size, try_alignment, flags); + if (p != NULL) { + // success, return the address + return p; + } + else if (max_retry_msecs > 0 && (try_alignment <= 2*MI_SEGMENT_ALIGN) && + (flags&MEM_COMMIT) != 0 && (flags&MEM_LARGE_PAGES) == 0 && + win_is_out_of_memory_error(GetLastError())) { + // if committing regular memory and being out-of-memory, + // keep trying for a bit in case memory frees up after all. See issue #894 + _mi_warning_message("out-of-memory on OS allocation, try again... (attempt %lu, 0x%zx bytes, error code: 0x%x, address: %p, alignment: 0x%zx, flags: 0x%x)\n", tries, size, GetLastError(), addr, try_alignment, flags); + long sleep_msecs = tries*40; // increasing waits + if (sleep_msecs > max_retry_msecs) { sleep_msecs = max_retry_msecs; } + max_retry_msecs -= sleep_msecs; + Sleep(sleep_msecs); + } + else { + // otherwise return with an error + break; + } + } + return NULL; +} + +static void* win_virtual_alloc(void* addr, size_t size, size_t try_alignment, DWORD flags, bool large_only, bool allow_large, bool* is_large) { + mi_assert_internal(!(large_only && !allow_large)); + static _Atomic(size_t) large_page_try_ok; // = 0; + void* p = NULL; + // Try to allocate large OS pages (2MiB) if allowed or required. + if ((large_only || _mi_os_use_large_page(size, try_alignment)) + && allow_large && (flags&MEM_COMMIT)!=0 && (flags&MEM_RESERVE)!=0) { + size_t try_ok = mi_atomic_load_acquire(&large_page_try_ok); + if (!large_only && try_ok > 0) { + // if a large page allocation fails, it seems the calls to VirtualAlloc get very expensive. + // therefore, once a large page allocation failed, we don't try again for `large_page_try_ok` times. + mi_atomic_cas_strong_acq_rel(&large_page_try_ok, &try_ok, try_ok - 1); + } + else { + // large OS pages must always reserve and commit. + *is_large = true; + p = win_virtual_alloc_prim(addr, size, try_alignment, flags | MEM_LARGE_PAGES); + if (large_only) return p; + // fall back to non-large page allocation on error (`p == NULL`). + if (p == NULL) { + mi_atomic_store_release(&large_page_try_ok,10UL); // on error, don't try again for the next N allocations + } + } + } + // Fall back to regular page allocation + if (p == NULL) { + *is_large = ((flags&MEM_LARGE_PAGES) != 0); + p = win_virtual_alloc_prim(addr, size, try_alignment, flags); + } + //if (p == NULL) { _mi_warning_message("unable to allocate OS memory (%zu bytes, error code: 0x%x, address: %p, alignment: %zu, flags: 0x%x, large only: %d, allow large: %d)\n", size, GetLastError(), addr, try_alignment, flags, large_only, allow_large); } + return p; +} + +int _mi_prim_alloc(void* hint_addr, size_t size, size_t try_alignment, bool commit, bool allow_large, bool* is_large, bool* is_zero, void** addr) { + mi_assert_internal(size > 0 && (size % _mi_os_page_size()) == 0); + mi_assert_internal(commit || !allow_large); + mi_assert_internal(try_alignment > 0); + *is_zero = true; + int flags = MEM_RESERVE; + if (commit) { flags |= MEM_COMMIT; } + *addr = win_virtual_alloc(hint_addr, size, try_alignment, flags, false, allow_large, is_large); + return (*addr != NULL ? 0 : (int)GetLastError()); +} + + +//--------------------------------------------- +// Commit/Reset/Protect +//--------------------------------------------- +#ifdef _MSC_VER +#pragma warning(disable:6250) // suppress warning calling VirtualFree without MEM_RELEASE (for decommit) +#endif + +int _mi_prim_commit(void* addr, size_t size, bool* is_zero) { + *is_zero = false; + /* + // zero'ing only happens on an initial commit... but checking upfront seems expensive.. + _MEMORY_BASIC_INFORMATION meminfo; _mi_memzero_var(meminfo); + if (VirtualQuery(addr, &meminfo, size) > 0) { + if ((meminfo.State & MEM_COMMIT) == 0) { + *is_zero = true; + } + } + */ + // commit + void* p = VirtualAlloc(addr, size, MEM_COMMIT, PAGE_READWRITE); + if (p == NULL) return (int)GetLastError(); + return 0; +} + +int _mi_prim_decommit(void* addr, size_t size, bool* needs_recommit) { + BOOL ok = VirtualFree(addr, size, MEM_DECOMMIT); + *needs_recommit = true; // for safety, assume always decommitted even in the case of an error. + return (ok ? 0 : (int)GetLastError()); +} + +int _mi_prim_reset(void* addr, size_t size) { + void* p = VirtualAlloc(addr, size, MEM_RESET, PAGE_READWRITE); + mi_assert_internal(p == addr); + #if 0 + if (p != NULL) { + VirtualUnlock(addr,size); // VirtualUnlock after MEM_RESET removes the memory directly from the working set + } + #endif + return (p != NULL ? 0 : (int)GetLastError()); +} + +int _mi_prim_reuse(void* addr, size_t size) { + MI_UNUSED(addr); MI_UNUSED(size); + return 0; +} + +int _mi_prim_protect(void* addr, size_t size, bool protect) { + DWORD oldprotect = 0; + BOOL ok = VirtualProtect(addr, size, protect ? PAGE_NOACCESS : PAGE_READWRITE, &oldprotect); + return (ok ? 0 : (int)GetLastError()); +} + + +//--------------------------------------------- +// Huge page allocation +//--------------------------------------------- + +static void* _mi_prim_alloc_huge_os_pagesx(void* hint_addr, size_t size, int numa_node) +{ + const DWORD flags = MEM_LARGE_PAGES | MEM_COMMIT | MEM_RESERVE; + + win_enable_large_os_pages(NULL); + + MI_MEM_EXTENDED_PARAMETER params[3] = { {{0,0},{0}},{{0,0},{0}},{{0,0},{0}} }; + // on modern Windows try use NtAllocateVirtualMemoryEx for 1GiB huge pages + static bool mi_huge_pages_available = true; + if (pNtAllocateVirtualMemoryEx != NULL && mi_huge_pages_available) { + params[0].Type.Type = MiMemExtendedParameterAttributeFlags; + params[0].Arg.ULong64 = MI_MEM_EXTENDED_PARAMETER_NONPAGED_HUGE; + ULONG param_count = 1; + if (numa_node >= 0) { + param_count++; + params[1].Type.Type = MiMemExtendedParameterNumaNode; + params[1].Arg.ULong = (unsigned)numa_node; + } + SIZE_T psize = size; + void* base = hint_addr; + LONG err = (*pNtAllocateVirtualMemoryEx)(GetCurrentProcess(), &base, &psize, flags, PAGE_READWRITE, params, param_count); + if (err == 0 && base != NULL) { + return base; + } + else { + // fall back to regular large pages + mi_huge_pages_available = false; // don't try further huge pages + _mi_warning_message("unable to allocate using huge (1GiB) pages, trying large (2MiB) pages instead (status 0x%lx)\n", err); + } + } + // on modern Windows try use VirtualAlloc2 for numa aware large OS page allocation + if (pVirtualAlloc2 != NULL && numa_node >= 0) { + params[0].Type.Type = MiMemExtendedParameterNumaNode; + params[0].Arg.ULong = (unsigned)numa_node; + return (*pVirtualAlloc2)(GetCurrentProcess(), hint_addr, size, flags, PAGE_READWRITE, params, 1); + } + + // otherwise use regular virtual alloc on older windows + return VirtualAlloc(hint_addr, size, flags, PAGE_READWRITE); +} + +int _mi_prim_alloc_huge_os_pages(void* hint_addr, size_t size, int numa_node, bool* is_zero, void** addr) { + *is_zero = true; + *addr = _mi_prim_alloc_huge_os_pagesx(hint_addr,size,numa_node); + return (*addr != NULL ? 0 : (int)GetLastError()); +} + + +//--------------------------------------------- +// Numa nodes +//--------------------------------------------- + +size_t _mi_prim_numa_node(void) { + USHORT numa_node = 0; + if (pGetCurrentProcessorNumberEx != NULL && pGetNumaProcessorNodeEx != NULL) { + // Extended API is supported + MI_PROCESSOR_NUMBER pnum; + (*pGetCurrentProcessorNumberEx)(&pnum); + USHORT nnode = 0; + BOOL ok = (*pGetNumaProcessorNodeEx)(&pnum, &nnode); + if (ok) { numa_node = nnode; } + } + else if (pGetNumaProcessorNode != NULL) { + // Vista or earlier, use older API that is limited to 64 processors. Issue #277 + DWORD pnum = GetCurrentProcessorNumber(); + UCHAR nnode = 0; + BOOL ok = pGetNumaProcessorNode((UCHAR)pnum, &nnode); + if (ok) { numa_node = nnode; } + } + return numa_node; +} + +size_t _mi_prim_numa_node_count(void) { + ULONG numa_max = 0; + if (pGetNumaHighestNodeNumber!=NULL) { + (*pGetNumaHighestNodeNumber)(&numa_max); + } + // find the highest node number that has actual processors assigned to it. Issue #282 + while (numa_max > 0) { + if (pGetNumaNodeProcessorMaskEx != NULL) { + // Extended API is supported + GROUP_AFFINITY affinity; + if ((*pGetNumaNodeProcessorMaskEx)((USHORT)numa_max, &affinity)) { + if (affinity.Mask != 0) break; // found the maximum non-empty node + } + } + else { + // Vista or earlier, use older API that is limited to 64 processors. + ULONGLONG mask; + if (pGetNumaNodeProcessorMask != NULL) { + if ((*pGetNumaNodeProcessorMask)((UCHAR)numa_max, &mask)) { + if (mask != 0) break; // found the maximum non-empty node + } + }; + } + // max node was invalid or had no processor assigned, try again + numa_max--; + } + return ((size_t)numa_max + 1); +} + + +//---------------------------------------------------------------- +// Clock +//---------------------------------------------------------------- + +static mi_msecs_t mi_to_msecs(LARGE_INTEGER t) { + static LARGE_INTEGER mfreq; // = 0 + if (mfreq.QuadPart == 0LL) { + LARGE_INTEGER f; + QueryPerformanceFrequency(&f); + mfreq.QuadPart = f.QuadPart/1000LL; + if (mfreq.QuadPart == 0) mfreq.QuadPart = 1; + } + return (mi_msecs_t)(t.QuadPart / mfreq.QuadPart); +} + +mi_msecs_t _mi_prim_clock_now(void) { + LARGE_INTEGER t; + QueryPerformanceCounter(&t); + return mi_to_msecs(t); +} + + +//---------------------------------------------------------------- +// Process Info +//---------------------------------------------------------------- + +#include + +static mi_msecs_t filetime_msecs(const FILETIME* ftime) { + ULARGE_INTEGER i; + i.LowPart = ftime->dwLowDateTime; + i.HighPart = ftime->dwHighDateTime; + mi_msecs_t msecs = (i.QuadPart / 10000); // FILETIME is in 100 nano seconds + return msecs; +} + +typedef BOOL (WINAPI *PGetProcessMemoryInfo)(HANDLE, PPROCESS_MEMORY_COUNTERS, DWORD); +static PGetProcessMemoryInfo pGetProcessMemoryInfo = NULL; + +void _mi_prim_process_info(mi_process_info_t* pinfo) +{ + FILETIME ct; + FILETIME ut; + FILETIME st; + FILETIME et; + GetProcessTimes(GetCurrentProcess(), &ct, &et, &st, &ut); + pinfo->utime = filetime_msecs(&ut); + pinfo->stime = filetime_msecs(&st); + + // load psapi on demand + if (pGetProcessMemoryInfo == NULL) { + HINSTANCE hDll = LoadLibrary(TEXT("psapi.dll")); + if (hDll != NULL) { + pGetProcessMemoryInfo = (PGetProcessMemoryInfo)(void (*)(void))GetProcAddress(hDll, "GetProcessMemoryInfo"); + } + } + + // get process info + PROCESS_MEMORY_COUNTERS info; _mi_memzero_var(info); + if (pGetProcessMemoryInfo != NULL) { + pGetProcessMemoryInfo(GetCurrentProcess(), &info, sizeof(info)); + } + pinfo->current_rss = (size_t)info.WorkingSetSize; + pinfo->peak_rss = (size_t)info.PeakWorkingSetSize; + pinfo->current_commit = (size_t)info.PagefileUsage; + pinfo->peak_commit = (size_t)info.PeakPagefileUsage; + pinfo->page_faults = (size_t)info.PageFaultCount; +} + +//---------------------------------------------------------------- +// Output +//---------------------------------------------------------------- + +void _mi_prim_out_stderr( const char* msg ) +{ + // on windows with redirection, the C runtime cannot handle locale dependent output + // after the main thread closes so we use direct console output. + if (!_mi_preloading()) { + // _cputs(msg); // _cputs cannot be used as it aborts when failing to lock the console + static HANDLE hcon = INVALID_HANDLE_VALUE; + static bool hconIsConsole = false; + if (hcon == INVALID_HANDLE_VALUE) { + hcon = GetStdHandle(STD_ERROR_HANDLE); + #ifdef MI_HAS_CONSOLE_IO + CONSOLE_SCREEN_BUFFER_INFO sbi; + hconIsConsole = ((hcon != INVALID_HANDLE_VALUE) && GetConsoleScreenBufferInfo(hcon, &sbi)); + #endif + } + const size_t len = _mi_strlen(msg); + if (len > 0 && len < UINT32_MAX) { + DWORD written = 0; + if (hconIsConsole) { + #ifdef MI_HAS_CONSOLE_IO + WriteConsoleA(hcon, msg, (DWORD)len, &written, NULL); + #endif + } + else if (hcon != INVALID_HANDLE_VALUE) { + // use direct write if stderr was redirected + WriteFile(hcon, msg, (DWORD)len, &written, NULL); + } + else { + // finally fall back to fputs after all + fputs(msg, stderr); + } + } + } +} + + +//---------------------------------------------------------------- +// Environment +//---------------------------------------------------------------- + +// On Windows use GetEnvironmentVariable instead of getenv to work +// reliably even when this is invoked before the C runtime is initialized. +// i.e. when `_mi_preloading() == true`. +// Note: on windows, environment names are not case sensitive. +bool _mi_prim_getenv(const char* name, char* result, size_t result_size) { + result[0] = 0; + size_t len = GetEnvironmentVariableA(name, result, (DWORD)result_size); + return (len > 0 && len < result_size); +} + + +//---------------------------------------------------------------- +// Random +//---------------------------------------------------------------- + +#if defined(MI_USE_RTLGENRANDOM) // || defined(__cplusplus) +// We prefer to use BCryptGenRandom instead of (the unofficial) RtlGenRandom but when using +// dynamic overriding, we observed it can raise an exception when compiled with C++, and +// sometimes deadlocks when also running under the VS debugger. +// In contrast, issue #623 implies that on Windows Server 2019 we need to use BCryptGenRandom. +// To be continued.. +#pragma comment (lib,"advapi32.lib") +#define RtlGenRandom SystemFunction036 +mi_decl_externc BOOLEAN NTAPI RtlGenRandom(PVOID RandomBuffer, ULONG RandomBufferLength); + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + return (RtlGenRandom(buf, (ULONG)buf_len) != 0); +} + +#else + +#ifndef BCRYPT_USE_SYSTEM_PREFERRED_RNG +#define BCRYPT_USE_SYSTEM_PREFERRED_RNG 0x00000002 +#endif + +typedef LONG (NTAPI *PBCryptGenRandom)(HANDLE, PUCHAR, ULONG, ULONG); +static PBCryptGenRandom pBCryptGenRandom = NULL; + +bool _mi_prim_random_buf(void* buf, size_t buf_len) { + if (pBCryptGenRandom == NULL) { + HINSTANCE hDll = LoadLibrary(TEXT("bcrypt.dll")); + if (hDll != NULL) { + pBCryptGenRandom = (PBCryptGenRandom)(void (*)(void))GetProcAddress(hDll, "BCryptGenRandom"); + } + if (pBCryptGenRandom == NULL) return false; + } + return (pBCryptGenRandom(NULL, (PUCHAR)buf, (ULONG)buf_len, BCRYPT_USE_SYSTEM_PREFERRED_RNG) >= 0); +} + +#endif // MI_USE_RTLGENRANDOM + + + +//---------------------------------------------------------------- +// Process & Thread Init/Done +//---------------------------------------------------------------- + +#if MI_WIN_USE_FIXED_TLS==1 +mi_decl_cache_align size_t _mi_win_tls_offset = 0; +#endif + +//static void mi_debug_out(const char* s) { +// HANDLE h = GetStdHandle(STD_ERROR_HANDLE); +// WriteConsole(h, s, (DWORD)_mi_strlen(s), NULL, NULL); +//} + +static void mi_win_tls_init(DWORD reason) { + if (reason==DLL_PROCESS_ATTACH || reason==DLL_THREAD_ATTACH) { + #if MI_WIN_USE_FIXED_TLS==1 // we must allocate a TLS slot dynamically + if (_mi_win_tls_offset == 0 && reason == DLL_PROCESS_ATTACH) { + const DWORD tls_slot = TlsAlloc(); // usually returns slot 1 + if (tls_slot == TLS_OUT_OF_INDEXES) { + _mi_error_message(EFAULT, "unable to allocate the a TLS slot (rebuild without MI_WIN_USE_FIXED_TLS?)\n"); + } + _mi_win_tls_offset = (size_t)tls_slot * sizeof(void*); + } + #endif + #if MI_HAS_TLS_SLOT >= 2 // we must initialize the TLS slot before any allocation + if (mi_prim_get_default_heap() == NULL) { + _mi_heap_set_default_direct((mi_heap_t*)&_mi_heap_empty); + #if MI_DEBUG && MI_WIN_USE_FIXED_TLS==1 + void* const p = TlsGetValue((DWORD)(_mi_win_tls_offset / sizeof(void*))); + mi_assert_internal(p == (void*)&_mi_heap_empty); + #endif + } + #endif + } +} + +static void NTAPI mi_win_main(PVOID module, DWORD reason, LPVOID reserved) { + MI_UNUSED(reserved); + MI_UNUSED(module); + mi_win_tls_init(reason); + if (reason==DLL_PROCESS_ATTACH) { + _mi_auto_process_init(); + } + else if (reason==DLL_PROCESS_DETACH) { + _mi_auto_process_done(); + } + else if (reason==DLL_THREAD_DETACH && !_mi_is_redirected()) { + _mi_thread_done(NULL); + } +} + + +#if defined(MI_SHARED_LIB) + #define MI_PRIM_HAS_PROCESS_ATTACH 1 + + // Windows DLL: easy to hook into process_init and thread_done + BOOL WINAPI DllMain(HINSTANCE inst, DWORD reason, LPVOID reserved) { + mi_win_main((PVOID)inst,reason,reserved); + return TRUE; + } + + // nothing to do since `_mi_thread_done` is handled through the DLL_THREAD_DETACH event. + void _mi_prim_thread_init_auto_done(void) { } + void _mi_prim_thread_done_auto_done(void) { } + void _mi_prim_thread_associate_default_heap(mi_heap_t* heap) { + MI_UNUSED(heap); + } + +#elif !defined(MI_WIN_USE_FLS) + #define MI_PRIM_HAS_PROCESS_ATTACH 1 + + static void NTAPI mi_win_main_attach(PVOID module, DWORD reason, LPVOID reserved) { + if (reason == DLL_PROCESS_ATTACH || reason == DLL_THREAD_ATTACH) { + mi_win_main(module, reason, reserved); + } + } + static void NTAPI mi_win_main_detach(PVOID module, DWORD reason, LPVOID reserved) { + if (reason == DLL_PROCESS_DETACH || reason == DLL_THREAD_DETACH) { + mi_win_main(module, reason, reserved); + } + } + + // Set up TLS callbacks in a statically linked library by using special data sections. + // See + // We use 2 entries to ensure we call attach events before constructors + // are called, and detach events after destructors are called. + #if defined(__cplusplus) + extern "C" { + #endif + + #if defined(_WIN64) + #pragma comment(linker, "/INCLUDE:_tls_used") + #pragma comment(linker, "/INCLUDE:_mi_tls_callback_pre") + #pragma comment(linker, "/INCLUDE:_mi_tls_callback_post") + #pragma const_seg(".CRT$XLB") + extern const PIMAGE_TLS_CALLBACK _mi_tls_callback_pre[]; + const PIMAGE_TLS_CALLBACK _mi_tls_callback_pre[] = { &mi_win_main_attach }; + #pragma const_seg() + #pragma const_seg(".CRT$XLY") + extern const PIMAGE_TLS_CALLBACK _mi_tls_callback_post[]; + const PIMAGE_TLS_CALLBACK _mi_tls_callback_post[] = { &mi_win_main_detach }; + #pragma const_seg() + #else + #pragma comment(linker, "/INCLUDE:__tls_used") + #pragma comment(linker, "/INCLUDE:__mi_tls_callback_pre") + #pragma comment(linker, "/INCLUDE:__mi_tls_callback_post") + #pragma data_seg(".CRT$XLB") + PIMAGE_TLS_CALLBACK _mi_tls_callback_pre[] = { &mi_win_main_attach }; + #pragma data_seg() + #pragma data_seg(".CRT$XLY") + PIMAGE_TLS_CALLBACK _mi_tls_callback_post[] = { &mi_win_main_detach }; + #pragma data_seg() + #endif + + #if defined(__cplusplus) + } + #endif + + // nothing to do since `_mi_thread_done` is handled through the DLL_THREAD_DETACH event. + void _mi_prim_thread_init_auto_done(void) { } + void _mi_prim_thread_done_auto_done(void) { } + void _mi_prim_thread_associate_default_heap(mi_heap_t* heap) { + MI_UNUSED(heap); + } + +#else // deprecated: statically linked, use fiber api + + #if defined(_MSC_VER) // on clang/gcc use the constructor attribute (in `src/prim/prim.c`) + // MSVC: use data section magic for static libraries + // See + #define MI_PRIM_HAS_PROCESS_ATTACH 1 + + static int mi_process_attach(void) { + mi_win_main(NULL,DLL_PROCESS_ATTACH,NULL); + atexit(&_mi_auto_process_done); + return 0; + } + typedef int(*mi_crt_callback_t)(void); + #if defined(_WIN64) + #pragma comment(linker, "/INCLUDE:_mi_tls_callback") + #pragma section(".CRT$XIU", long, read) + #else + #pragma comment(linker, "/INCLUDE:__mi_tls_callback") + #endif + #pragma data_seg(".CRT$XIU") + mi_decl_externc mi_crt_callback_t _mi_tls_callback[] = { &mi_process_attach }; + #pragma data_seg() + #endif + + // use the fiber api for calling `_mi_thread_done`. + #include + #if (_WIN32_WINNT < 0x600) // before Windows Vista + WINBASEAPI DWORD WINAPI FlsAlloc( _In_opt_ PFLS_CALLBACK_FUNCTION lpCallback ); + WINBASEAPI PVOID WINAPI FlsGetValue( _In_ DWORD dwFlsIndex ); + WINBASEAPI BOOL WINAPI FlsSetValue( _In_ DWORD dwFlsIndex, _In_opt_ PVOID lpFlsData ); + WINBASEAPI BOOL WINAPI FlsFree(_In_ DWORD dwFlsIndex); + #endif + + static DWORD mi_fls_key = (DWORD)(-1); + + static void NTAPI mi_fls_done(PVOID value) { + mi_heap_t* heap = (mi_heap_t*)value; + if (heap != NULL) { + _mi_thread_done(heap); + FlsSetValue(mi_fls_key, NULL); // prevent recursion as _mi_thread_done may set it back to the main heap, issue #672 + } + } + + void _mi_prim_thread_init_auto_done(void) { + mi_fls_key = FlsAlloc(&mi_fls_done); + } + + void _mi_prim_thread_done_auto_done(void) { + // call thread-done on all threads (except the main thread) to prevent + // dangling callback pointer if statically linked with a DLL; Issue #208 + FlsFree(mi_fls_key); + } + + void _mi_prim_thread_associate_default_heap(mi_heap_t* heap) { + mi_assert_internal(mi_fls_key != (DWORD)(-1)); + FlsSetValue(mi_fls_key, heap); + } +#endif + +// ---------------------------------------------------- +// Communicate with the redirection module on Windows +// ---------------------------------------------------- +#if defined(MI_SHARED_LIB) && !defined(MI_WIN_NOREDIRECT) + #define MI_PRIM_HAS_ALLOCATOR_INIT 1 + + static bool mi_redirected = false; // true if malloc redirects to mi_malloc + + bool _mi_is_redirected(void) { + return mi_redirected; + } + + #ifdef __cplusplus + extern "C" { + #endif + mi_decl_export void _mi_redirect_entry(DWORD reason) { + // called on redirection; careful as this may be called before DllMain + mi_win_tls_init(reason); + if (reason == DLL_PROCESS_ATTACH) { + mi_redirected = true; + } + else if (reason == DLL_PROCESS_DETACH) { + mi_redirected = false; + } + else if (reason == DLL_THREAD_DETACH) { + _mi_thread_done(NULL); + } + } + __declspec(dllimport) bool mi_cdecl mi_allocator_init(const char** message); + __declspec(dllimport) void mi_cdecl mi_allocator_done(void); + #ifdef __cplusplus + } + #endif + bool _mi_allocator_init(const char** message) { + return mi_allocator_init(message); + } + void _mi_allocator_done(void) { + mi_allocator_done(); + } +#endif diff --git a/compat/mimalloc/random.c b/compat/mimalloc/random.c new file mode 100644 index 00000000000000..f17698ba8a6d08 --- /dev/null +++ b/compat/mimalloc/random.c @@ -0,0 +1,258 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2021, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/prim.h" // _mi_prim_random_buf +#include // memset + +/* ---------------------------------------------------------------------------- +We use our own PRNG to keep predictable performance of random number generation +and to avoid implementations that use a lock. We only use the OS provided +random source to initialize the initial seeds. Since we do not need ultimate +performance but we do rely on the security (for secret cookies in secure mode) +we use a cryptographically secure generator (chacha20). +-----------------------------------------------------------------------------*/ + +#define MI_CHACHA_ROUNDS (20) // perhaps use 12 for better performance? + + +/* ---------------------------------------------------------------------------- +Chacha20 implementation as the original algorithm with a 64-bit nonce +and counter: https://en.wikipedia.org/wiki/Salsa20 +The input matrix has sixteen 32-bit values: +Position 0 to 3: constant key +Position 4 to 11: the key +Position 12 to 13: the counter. +Position 14 to 15: the nonce. + +The implementation uses regular C code which compiles very well on modern compilers. +(gcc x64 has no register spills, and clang 6+ uses SSE instructions) +-----------------------------------------------------------------------------*/ + +static inline uint32_t rotl(uint32_t x, uint32_t shift) { + return (x << shift) | (x >> (32 - shift)); +} + +static inline void qround(uint32_t x[16], size_t a, size_t b, size_t c, size_t d) { + x[a] += x[b]; x[d] = rotl(x[d] ^ x[a], 16); + x[c] += x[d]; x[b] = rotl(x[b] ^ x[c], 12); + x[a] += x[b]; x[d] = rotl(x[d] ^ x[a], 8); + x[c] += x[d]; x[b] = rotl(x[b] ^ x[c], 7); +} + +static void chacha_block(mi_random_ctx_t* ctx) +{ + // scramble into `x` + uint32_t x[16]; + for (size_t i = 0; i < 16; i++) { + x[i] = ctx->input[i]; + } + for (size_t i = 0; i < MI_CHACHA_ROUNDS; i += 2) { + qround(x, 0, 4, 8, 12); + qround(x, 1, 5, 9, 13); + qround(x, 2, 6, 10, 14); + qround(x, 3, 7, 11, 15); + qround(x, 0, 5, 10, 15); + qround(x, 1, 6, 11, 12); + qround(x, 2, 7, 8, 13); + qround(x, 3, 4, 9, 14); + } + + // add scrambled data to the initial state + for (size_t i = 0; i < 16; i++) { + ctx->output[i] = x[i] + ctx->input[i]; + } + ctx->output_available = 16; + + // increment the counter for the next round + ctx->input[12] += 1; + if (ctx->input[12] == 0) { + ctx->input[13] += 1; + if (ctx->input[13] == 0) { // and keep increasing into the nonce + ctx->input[14] += 1; + } + } +} + +static uint32_t chacha_next32(mi_random_ctx_t* ctx) { + if (ctx->output_available <= 0) { + chacha_block(ctx); + ctx->output_available = 16; // (assign again to suppress static analysis warning) + } + const uint32_t x = ctx->output[16 - ctx->output_available]; + ctx->output[16 - ctx->output_available] = 0; // reset once the data is handed out + ctx->output_available--; + return x; +} + +static inline uint32_t read32(const uint8_t* p, size_t idx32) { + const size_t i = 4*idx32; + return ((uint32_t)p[i+0] | (uint32_t)p[i+1] << 8 | (uint32_t)p[i+2] << 16 | (uint32_t)p[i+3] << 24); +} + +static void chacha_init(mi_random_ctx_t* ctx, const uint8_t key[32], uint64_t nonce) +{ + // since we only use chacha for randomness (and not encryption) we + // do not _need_ to read 32-bit values as little endian but we do anyways + // just for being compatible :-) + memset(ctx, 0, sizeof(*ctx)); + for (size_t i = 0; i < 4; i++) { + const uint8_t* sigma = (uint8_t*)"expand 32-byte k"; + ctx->input[i] = read32(sigma,i); + } + for (size_t i = 0; i < 8; i++) { + ctx->input[i + 4] = read32(key,i); + } + ctx->input[12] = 0; + ctx->input[13] = 0; + ctx->input[14] = (uint32_t)nonce; + ctx->input[15] = (uint32_t)(nonce >> 32); +} + +static void chacha_split(mi_random_ctx_t* ctx, uint64_t nonce, mi_random_ctx_t* ctx_new) { + memset(ctx_new, 0, sizeof(*ctx_new)); + _mi_memcpy(ctx_new->input, ctx->input, sizeof(ctx_new->input)); + ctx_new->input[12] = 0; + ctx_new->input[13] = 0; + ctx_new->input[14] = (uint32_t)nonce; + ctx_new->input[15] = (uint32_t)(nonce >> 32); + mi_assert_internal(ctx->input[14] != ctx_new->input[14] || ctx->input[15] != ctx_new->input[15]); // do not reuse nonces! + chacha_block(ctx_new); +} + + +/* ---------------------------------------------------------------------------- +Random interface +-----------------------------------------------------------------------------*/ + +#if MI_DEBUG>1 +static bool mi_random_is_initialized(mi_random_ctx_t* ctx) { + return (ctx != NULL && ctx->input[0] != 0); +} +#endif + +void _mi_random_split(mi_random_ctx_t* ctx, mi_random_ctx_t* ctx_new) { + mi_assert_internal(mi_random_is_initialized(ctx)); + mi_assert_internal(ctx != ctx_new); + chacha_split(ctx, (uintptr_t)ctx_new /*nonce*/, ctx_new); +} + +uintptr_t _mi_random_next(mi_random_ctx_t* ctx) { + mi_assert_internal(mi_random_is_initialized(ctx)); + uintptr_t r; + do { + #if MI_INTPTR_SIZE <= 4 + r = chacha_next32(ctx); + #elif MI_INTPTR_SIZE == 8 + r = (((uintptr_t)chacha_next32(ctx) << 32) | chacha_next32(ctx)); + #else + # error "define mi_random_next for this platform" + #endif + } while (r==0); + return r; +} + + +/* ---------------------------------------------------------------------------- +To initialize a fresh random context. +If we cannot get good randomness, we fall back to weak randomness based on a timer and ASLR. +-----------------------------------------------------------------------------*/ + +uintptr_t _mi_os_random_weak(uintptr_t extra_seed) { + uintptr_t x = (uintptr_t)&_mi_os_random_weak ^ extra_seed; // ASLR makes the address random + x ^= _mi_prim_clock_now(); + // and do a few randomization steps + uintptr_t max = ((x ^ (x >> 17)) & 0x0F) + 1; + for (uintptr_t i = 0; i < max || x==0; i++, x++) { + x = _mi_random_shuffle(x); + } + mi_assert_internal(x != 0); + return x; +} + +static void mi_random_init_ex(mi_random_ctx_t* ctx, bool use_weak) { + uint8_t key[32]; + if (use_weak || !_mi_prim_random_buf(key, sizeof(key))) { + // if we fail to get random data from the OS, we fall back to a + // weak random source based on the current time + #if !defined(__wasi__) + if (!use_weak) { _mi_warning_message("unable to use secure randomness\n"); } + #endif + uintptr_t x = _mi_os_random_weak(0); + for (size_t i = 0; i < 8; i++, x++) { // key is eight 32-bit words. + x = _mi_random_shuffle(x); + ((uint32_t*)key)[i] = (uint32_t)x; + } + ctx->weak = true; + } + else { + ctx->weak = false; + } + chacha_init(ctx, key, (uintptr_t)ctx /*nonce*/ ); +} + +void _mi_random_init(mi_random_ctx_t* ctx) { + mi_random_init_ex(ctx, false); +} + +void _mi_random_init_weak(mi_random_ctx_t * ctx) { + mi_random_init_ex(ctx, true); +} + +void _mi_random_reinit_if_weak(mi_random_ctx_t * ctx) { + if (ctx->weak) { + _mi_random_init(ctx); + } +} + +/* -------------------------------------------------------- +test vectors from +----------------------------------------------------------- */ +/* +static bool array_equals(uint32_t* x, uint32_t* y, size_t n) { + for (size_t i = 0; i < n; i++) { + if (x[i] != y[i]) return false; + } + return true; +} +static void chacha_test(void) +{ + uint32_t x[4] = { 0x11111111, 0x01020304, 0x9b8d6f43, 0x01234567 }; + uint32_t x_out[4] = { 0xea2a92f4, 0xcb1cf8ce, 0x4581472e, 0x5881c4bb }; + qround(x, 0, 1, 2, 3); + mi_assert_internal(array_equals(x, x_out, 4)); + + uint32_t y[16] = { + 0x879531e0, 0xc5ecf37d, 0x516461b1, 0xc9a62f8a, + 0x44c20ef3, 0x3390af7f, 0xd9fc690b, 0x2a5f714c, + 0x53372767, 0xb00a5631, 0x974c541a, 0x359e9963, + 0x5c971061, 0x3d631689, 0x2098d9d6, 0x91dbd320 }; + uint32_t y_out[16] = { + 0x879531e0, 0xc5ecf37d, 0xbdb886dc, 0xc9a62f8a, + 0x44c20ef3, 0x3390af7f, 0xd9fc690b, 0xcfacafd2, + 0xe46bea80, 0xb00a5631, 0x974c541a, 0x359e9963, + 0x5c971061, 0xccc07c79, 0x2098d9d6, 0x91dbd320 }; + qround(y, 2, 7, 8, 13); + mi_assert_internal(array_equals(y, y_out, 16)); + + mi_random_ctx_t r = { + { 0x61707865, 0x3320646e, 0x79622d32, 0x6b206574, + 0x03020100, 0x07060504, 0x0b0a0908, 0x0f0e0d0c, + 0x13121110, 0x17161514, 0x1b1a1918, 0x1f1e1d1c, + 0x00000001, 0x09000000, 0x4a000000, 0x00000000 }, + {0}, + 0 + }; + uint32_t r_out[16] = { + 0xe4e7f110, 0x15593bd1, 0x1fdd0f50, 0xc47120a3, + 0xc7f4d1c7, 0x0368c033, 0x9aaa2204, 0x4e6cd4c3, + 0x466482d2, 0x09aa9f07, 0x05d7c214, 0xa2028bd9, + 0xd19c12b5, 0xb94e16de, 0xe883d0cb, 0x4e3c50a2 }; + chacha_block(&r); + mi_assert_internal(array_equals(r.output, r_out, 16)); +} +*/ diff --git a/compat/mimalloc/segment-map.c b/compat/mimalloc/segment-map.c new file mode 100644 index 00000000000000..bbcea28aabc2e1 --- /dev/null +++ b/compat/mimalloc/segment-map.c @@ -0,0 +1,142 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2019-2023, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ + +/* ----------------------------------------------------------- + The following functions are to reliably find the segment or + block that encompasses any pointer p (or NULL if it is not + in any of our segments). + We maintain a bitmap of all memory with 1 bit per MI_SEGMENT_SIZE (64MiB) + set to 1 if it contains the segment meta data. +----------------------------------------------------------- */ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" + +// Reduce total address space to reduce .bss (due to the `mi_segment_map`) +#if (MI_INTPTR_SIZE > 4) && MI_TRACK_ASAN +#define MI_SEGMENT_MAP_MAX_ADDRESS (128*1024ULL*MI_GiB) // 128 TiB (see issue #881) +#elif (MI_INTPTR_SIZE > 4) +#define MI_SEGMENT_MAP_MAX_ADDRESS (48*1024ULL*MI_GiB) // 48 TiB +#else +#define MI_SEGMENT_MAP_MAX_ADDRESS (UINT32_MAX) +#endif + +#define MI_SEGMENT_MAP_PART_SIZE (MI_INTPTR_SIZE*MI_KiB - 128) // 128 > sizeof(mi_memid_t) ! +#define MI_SEGMENT_MAP_PART_BITS (8*MI_SEGMENT_MAP_PART_SIZE) +#define MI_SEGMENT_MAP_PART_ENTRIES (MI_SEGMENT_MAP_PART_SIZE / MI_INTPTR_SIZE) +#define MI_SEGMENT_MAP_PART_BIT_SPAN (MI_SEGMENT_ALIGN) // memory area covered by 1 bit + +#if (MI_SEGMENT_MAP_PART_BITS < (MI_SEGMENT_MAP_MAX_ADDRESS / MI_SEGMENT_MAP_PART_BIT_SPAN)) // prevent overflow on 32-bit (issue #1017) +#define MI_SEGMENT_MAP_PART_SPAN (MI_SEGMENT_MAP_PART_BITS * MI_SEGMENT_MAP_PART_BIT_SPAN) +#else +#define MI_SEGMENT_MAP_PART_SPAN MI_SEGMENT_MAP_MAX_ADDRESS +#endif + +#define MI_SEGMENT_MAP_MAX_PARTS ((MI_SEGMENT_MAP_MAX_ADDRESS / MI_SEGMENT_MAP_PART_SPAN) + 1) + +// A part of the segment map. +typedef struct mi_segmap_part_s { + mi_memid_t memid; + _Atomic(uintptr_t) map[MI_SEGMENT_MAP_PART_ENTRIES]; +} mi_segmap_part_t; + +// Allocate parts on-demand to reduce .bss footprint +static _Atomic(mi_segmap_part_t*) mi_segment_map[MI_SEGMENT_MAP_MAX_PARTS]; // = { NULL, .. } + +static mi_segmap_part_t* mi_segment_map_index_of(const mi_segment_t* segment, bool create_on_demand, size_t* idx, size_t* bitidx) { + // note: segment can be invalid or NULL. + mi_assert_internal(_mi_ptr_segment(segment + 1) == segment); // is it aligned on MI_SEGMENT_SIZE? + *idx = 0; + *bitidx = 0; + if ((uintptr_t)segment >= MI_SEGMENT_MAP_MAX_ADDRESS) return NULL; + const uintptr_t segindex = ((uintptr_t)segment) / MI_SEGMENT_MAP_PART_SPAN; + if (segindex >= MI_SEGMENT_MAP_MAX_PARTS) return NULL; + mi_segmap_part_t* part = mi_atomic_load_ptr_relaxed(mi_segmap_part_t, &mi_segment_map[segindex]); + + // allocate on demand to reduce .bss footprint + if mi_unlikely(part == NULL) { + if (!create_on_demand) return NULL; + mi_memid_t memid; + part = (mi_segmap_part_t*)_mi_os_zalloc(sizeof(mi_segmap_part_t), &memid); + if (part == NULL) return NULL; + part->memid = memid; + mi_segmap_part_t* expected = NULL; + if (!mi_atomic_cas_ptr_strong_release(mi_segmap_part_t, &mi_segment_map[segindex], &expected, part)) { + _mi_os_free(part, sizeof(mi_segmap_part_t), memid); + part = expected; + if (part == NULL) return NULL; + } + } + mi_assert(part != NULL); + const uintptr_t offset = ((uintptr_t)segment) % MI_SEGMENT_MAP_PART_SPAN; + const uintptr_t bitofs = offset / MI_SEGMENT_MAP_PART_BIT_SPAN; + *idx = bitofs / MI_INTPTR_BITS; + *bitidx = bitofs % MI_INTPTR_BITS; + return part; +} + +void _mi_segment_map_allocated_at(const mi_segment_t* segment) { + if (segment->memid.memkind == MI_MEM_ARENA) return; // we lookup segments first in the arena's and don't need the segment map + size_t index; + size_t bitidx; + mi_segmap_part_t* part = mi_segment_map_index_of(segment, true /* alloc map if needed */, &index, &bitidx); + if (part == NULL) return; // outside our address range.. + uintptr_t mask = mi_atomic_load_relaxed(&part->map[index]); + uintptr_t newmask; + do { + newmask = (mask | ((uintptr_t)1 << bitidx)); + } while (!mi_atomic_cas_weak_release(&part->map[index], &mask, newmask)); +} + +void _mi_segment_map_freed_at(const mi_segment_t* segment) { + if (segment->memid.memkind == MI_MEM_ARENA) return; + size_t index; + size_t bitidx; + mi_segmap_part_t* part = mi_segment_map_index_of(segment, false /* don't alloc if not present */, &index, &bitidx); + if (part == NULL) return; // outside our address range.. + uintptr_t mask = mi_atomic_load_relaxed(&part->map[index]); + uintptr_t newmask; + do { + newmask = (mask & ~((uintptr_t)1 << bitidx)); + } while (!mi_atomic_cas_weak_release(&part->map[index], &mask, newmask)); +} + +// Determine the segment belonging to a pointer or NULL if it is not in a valid segment. +static mi_segment_t* _mi_segment_of(const void* p) { + if (p == NULL) return NULL; + mi_segment_t* segment = _mi_ptr_segment(p); // segment can be NULL + size_t index; + size_t bitidx; + mi_segmap_part_t* part = mi_segment_map_index_of(segment, false /* dont alloc if not present */, &index, &bitidx); + if (part == NULL) return NULL; + const uintptr_t mask = mi_atomic_load_relaxed(&part->map[index]); + if mi_likely((mask & ((uintptr_t)1 << bitidx)) != 0) { + bool cookie_ok = (_mi_ptr_cookie(segment) == segment->cookie); + mi_assert_internal(cookie_ok); MI_UNUSED(cookie_ok); + return segment; // yes, allocated by us + } + return NULL; +} + +// Is this a valid pointer in our heap? +static bool mi_is_valid_pointer(const void* p) { + // first check if it is in an arena, then check if it is OS allocated + return (_mi_arena_contains(p) || _mi_segment_of(p) != NULL); +} + +mi_decl_nodiscard mi_decl_export bool mi_is_in_heap_region(const void* p) mi_attr_noexcept { + return mi_is_valid_pointer(p); +} + +void _mi_segment_map_unsafe_destroy(void) { + for (size_t i = 0; i < MI_SEGMENT_MAP_MAX_PARTS; i++) { + mi_segmap_part_t* part = mi_atomic_exchange_ptr_relaxed(mi_segmap_part_t, &mi_segment_map[i], NULL); + if (part != NULL) { + _mi_os_free(part, sizeof(mi_segmap_part_t), part->memid); + } + } +} diff --git a/compat/mimalloc/segment.c b/compat/mimalloc/segment.c new file mode 100644 index 00000000000000..32841e6deef20e --- /dev/null +++ b/compat/mimalloc/segment.c @@ -0,0 +1,1702 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2024, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" + +#include // memset +#include + +// ------------------------------------------------------------------- +// Segments +// mimalloc pages reside in segments. See `mi_segment_valid` for invariants. +// ------------------------------------------------------------------- + + +static void mi_segment_try_purge(mi_segment_t* segment, bool force); + + +// ------------------------------------------------------------------- +// commit mask +// ------------------------------------------------------------------- + +static bool mi_commit_mask_all_set(const mi_commit_mask_t* commit, const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + if ((commit->mask[i] & cm->mask[i]) != cm->mask[i]) return false; + } + return true; +} + +static bool mi_commit_mask_any_set(const mi_commit_mask_t* commit, const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + if ((commit->mask[i] & cm->mask[i]) != 0) return true; + } + return false; +} + +static void mi_commit_mask_create_intersect(const mi_commit_mask_t* commit, const mi_commit_mask_t* cm, mi_commit_mask_t* res) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + res->mask[i] = (commit->mask[i] & cm->mask[i]); + } +} + +static void mi_commit_mask_clear(mi_commit_mask_t* res, const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + res->mask[i] &= ~(cm->mask[i]); + } +} + +static void mi_commit_mask_set(mi_commit_mask_t* res, const mi_commit_mask_t* cm) { + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + res->mask[i] |= cm->mask[i]; + } +} + +static void mi_commit_mask_create(size_t bitidx, size_t bitcount, mi_commit_mask_t* cm) { + mi_assert_internal(bitidx < MI_COMMIT_MASK_BITS); + mi_assert_internal((bitidx + bitcount) <= MI_COMMIT_MASK_BITS); + if (bitcount == MI_COMMIT_MASK_BITS) { + mi_assert_internal(bitidx==0); + mi_commit_mask_create_full(cm); + } + else if (bitcount == 0) { + mi_commit_mask_create_empty(cm); + } + else { + mi_commit_mask_create_empty(cm); + size_t i = bitidx / MI_COMMIT_MASK_FIELD_BITS; + size_t ofs = bitidx % MI_COMMIT_MASK_FIELD_BITS; + while (bitcount > 0) { + mi_assert_internal(i < MI_COMMIT_MASK_FIELD_COUNT); + size_t avail = MI_COMMIT_MASK_FIELD_BITS - ofs; + size_t count = (bitcount > avail ? avail : bitcount); + size_t mask = (count >= MI_COMMIT_MASK_FIELD_BITS ? ~((size_t)0) : (((size_t)1 << count) - 1) << ofs); + cm->mask[i] = mask; + bitcount -= count; + ofs = 0; + i++; + } + } +} + +size_t _mi_commit_mask_committed_size(const mi_commit_mask_t* cm, size_t total) { + mi_assert_internal((total%MI_COMMIT_MASK_BITS)==0); + size_t count = 0; + for (size_t i = 0; i < MI_COMMIT_MASK_FIELD_COUNT; i++) { + size_t mask = cm->mask[i]; + if (~mask == 0) { + count += MI_COMMIT_MASK_FIELD_BITS; + } + else { + for (; mask != 0; mask >>= 1) { // todo: use popcount + if ((mask&1)!=0) count++; + } + } + } + // we use total since for huge segments each commit bit may represent a larger size + return ((total / MI_COMMIT_MASK_BITS) * count); +} + + +size_t _mi_commit_mask_next_run(const mi_commit_mask_t* cm, size_t* idx) { + size_t i = (*idx) / MI_COMMIT_MASK_FIELD_BITS; + size_t ofs = (*idx) % MI_COMMIT_MASK_FIELD_BITS; + size_t mask = 0; + // find first ones + while (i < MI_COMMIT_MASK_FIELD_COUNT) { + mask = cm->mask[i]; + mask >>= ofs; + if (mask != 0) { + while ((mask&1) == 0) { + mask >>= 1; + ofs++; + } + break; + } + i++; + ofs = 0; + } + if (i >= MI_COMMIT_MASK_FIELD_COUNT) { + // not found + *idx = MI_COMMIT_MASK_BITS; + return 0; + } + else { + // found, count ones + size_t count = 0; + *idx = (i*MI_COMMIT_MASK_FIELD_BITS) + ofs; + do { + mi_assert_internal(ofs < MI_COMMIT_MASK_FIELD_BITS && (mask&1) == 1); + do { + count++; + mask >>= 1; + } while ((mask&1) == 1); + if ((((*idx + count) % MI_COMMIT_MASK_FIELD_BITS) == 0)) { + i++; + if (i >= MI_COMMIT_MASK_FIELD_COUNT) break; + mask = cm->mask[i]; + ofs = 0; + } + } while ((mask&1) == 1); + mi_assert_internal(count > 0); + return count; + } +} + + +/* -------------------------------------------------------------------------------- + Segment allocation + We allocate pages inside bigger "segments" (32 MiB on 64-bit). This is to avoid + splitting VMA's on Linux and reduce fragmentation on other OS's. + Each thread owns its own segments. + + Currently we have: + - small pages (64KiB) + - medium pages (512KiB) + - large pages (4MiB), + - huge segments have 1 page in one segment that can be larger than `MI_SEGMENT_SIZE`. + it is used for blocks `> MI_LARGE_OBJ_SIZE_MAX` or with alignment `> MI_BLOCK_ALIGNMENT_MAX`. + + The memory for a segment is usually committed on demand. + (i.e. we are careful to not touch the memory until we actually allocate a block there) + + If a thread ends, it "abandons" pages that still contain live blocks. + Such segments are abandoned and these can be reclaimed by still running threads, + (much like work-stealing). +-------------------------------------------------------------------------------- */ + + +/* ----------------------------------------------------------- + Slices +----------------------------------------------------------- */ + + +static const mi_slice_t* mi_segment_slices_end(const mi_segment_t* segment) { + return &segment->slices[segment->slice_entries]; +} + +static uint8_t* mi_slice_start(const mi_slice_t* slice) { + mi_segment_t* segment = _mi_ptr_segment(slice); + mi_assert_internal(slice >= segment->slices && slice < mi_segment_slices_end(segment)); + return ((uint8_t*)segment + ((slice - segment->slices)*MI_SEGMENT_SLICE_SIZE)); +} + + +/* ----------------------------------------------------------- + Bins +----------------------------------------------------------- */ +// Use bit scan forward to quickly find the first zero bit if it is available + +static inline size_t mi_slice_bin8(size_t slice_count) { + if (slice_count<=1) return slice_count; + mi_assert_internal(slice_count <= MI_SLICES_PER_SEGMENT); + slice_count--; + size_t s = mi_bsr(slice_count); // slice_count > 1 + if (s <= 2) return slice_count + 1; + size_t bin = ((s << 2) | ((slice_count >> (s - 2))&0x03)) - 4; + return bin; +} + +static inline size_t mi_slice_bin(size_t slice_count) { + mi_assert_internal(slice_count*MI_SEGMENT_SLICE_SIZE <= MI_SEGMENT_SIZE); + mi_assert_internal(mi_slice_bin8(MI_SLICES_PER_SEGMENT) <= MI_SEGMENT_BIN_MAX); + size_t bin = mi_slice_bin8(slice_count); + mi_assert_internal(bin <= MI_SEGMENT_BIN_MAX); + return bin; +} + +static inline size_t mi_slice_index(const mi_slice_t* slice) { + mi_segment_t* segment = _mi_ptr_segment(slice); + ptrdiff_t index = slice - segment->slices; + mi_assert_internal(index >= 0 && index < (ptrdiff_t)segment->slice_entries); + return index; +} + + +/* ----------------------------------------------------------- + Slice span queues +----------------------------------------------------------- */ + +static void mi_span_queue_push(mi_span_queue_t* sq, mi_slice_t* slice) { + // todo: or push to the end? + mi_assert_internal(slice->prev == NULL && slice->next==NULL); + slice->prev = NULL; // paranoia + slice->next = sq->first; + sq->first = slice; + if (slice->next != NULL) slice->next->prev = slice; + else sq->last = slice; + slice->block_size = 0; // free +} + +static mi_span_queue_t* mi_span_queue_for(size_t slice_count, mi_segments_tld_t* tld) { + size_t bin = mi_slice_bin(slice_count); + mi_span_queue_t* sq = &tld->spans[bin]; + mi_assert_internal(sq->slice_count >= slice_count); + return sq; +} + +static void mi_span_queue_delete(mi_span_queue_t* sq, mi_slice_t* slice) { + mi_assert_internal(slice->block_size==0 && slice->slice_count>0 && slice->slice_offset==0); + // should work too if the queue does not contain slice (which can happen during reclaim) + if (slice->prev != NULL) slice->prev->next = slice->next; + if (slice == sq->first) sq->first = slice->next; + if (slice->next != NULL) slice->next->prev = slice->prev; + if (slice == sq->last) sq->last = slice->prev; + slice->prev = NULL; + slice->next = NULL; + slice->block_size = 1; // no more free +} + + +/* ----------------------------------------------------------- + Invariant checking +----------------------------------------------------------- */ + +static bool mi_slice_is_used(const mi_slice_t* slice) { + return (slice->block_size > 0); +} + + +#if (MI_DEBUG>=3) +static bool mi_span_queue_contains(mi_span_queue_t* sq, mi_slice_t* slice) { + for (mi_slice_t* s = sq->first; s != NULL; s = s->next) { + if (s==slice) return true; + } + return false; +} + +static bool mi_segment_is_valid(mi_segment_t* segment, mi_segments_tld_t* tld) { + mi_assert_internal(segment != NULL); + mi_assert_internal(_mi_ptr_cookie(segment) == segment->cookie); + mi_assert_internal(segment->abandoned <= segment->used); + mi_assert_internal(segment->thread_id == 0 || segment->thread_id == _mi_thread_id()); + mi_assert_internal(mi_commit_mask_all_set(&segment->commit_mask, &segment->purge_mask)); // can only decommit committed blocks + //mi_assert_internal(segment->segment_info_size % MI_SEGMENT_SLICE_SIZE == 0); + mi_slice_t* slice = &segment->slices[0]; + const mi_slice_t* end = mi_segment_slices_end(segment); + size_t used_count = 0; + mi_span_queue_t* sq; + while(slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + size_t index = mi_slice_index(slice); + size_t maxindex = (index + slice->slice_count >= segment->slice_entries ? segment->slice_entries : index + slice->slice_count) - 1; + if (mi_slice_is_used(slice)) { // a page in use, we need at least MAX_SLICE_OFFSET_COUNT valid back offsets + used_count++; + mi_assert_internal(slice->is_huge == (segment->kind == MI_SEGMENT_HUGE)); + for (size_t i = 0; i <= MI_MAX_SLICE_OFFSET_COUNT && index + i <= maxindex; i++) { + mi_assert_internal(segment->slices[index + i].slice_offset == i*sizeof(mi_slice_t)); + mi_assert_internal(i==0 || segment->slices[index + i].slice_count == 0); + mi_assert_internal(i==0 || segment->slices[index + i].block_size == 1); + } + // and the last entry as well (for coalescing) + const mi_slice_t* last = slice + slice->slice_count - 1; + if (last > slice && last < mi_segment_slices_end(segment)) { + mi_assert_internal(last->slice_offset == (slice->slice_count-1)*sizeof(mi_slice_t)); + mi_assert_internal(last->slice_count == 0); + mi_assert_internal(last->block_size == 1); + } + } + else { // free range of slices; only last slice needs a valid back offset + mi_slice_t* last = &segment->slices[maxindex]; + if (segment->kind != MI_SEGMENT_HUGE || slice->slice_count <= (segment->slice_entries - segment->segment_info_slices)) { + mi_assert_internal((uint8_t*)slice == (uint8_t*)last - last->slice_offset); + } + mi_assert_internal(slice == last || last->slice_count == 0 ); + mi_assert_internal(last->block_size == 0 || (segment->kind==MI_SEGMENT_HUGE && last->block_size==1)); + if (segment->kind != MI_SEGMENT_HUGE && segment->thread_id != 0) { // segment is not huge or abandoned + sq = mi_span_queue_for(slice->slice_count,tld); + mi_assert_internal(mi_span_queue_contains(sq,slice)); + } + } + slice = &segment->slices[maxindex+1]; + } + mi_assert_internal(slice == end); + mi_assert_internal(used_count == segment->used + 1); + return true; +} +#endif + +/* ----------------------------------------------------------- + Segment size calculations +----------------------------------------------------------- */ + +static size_t mi_segment_info_size(mi_segment_t* segment) { + return segment->segment_info_slices * MI_SEGMENT_SLICE_SIZE; +} + +static uint8_t* _mi_segment_page_start_from_slice(const mi_segment_t* segment, const mi_slice_t* slice, size_t block_size, size_t* page_size) +{ + const ptrdiff_t idx = slice - segment->slices; + const size_t psize = (size_t)slice->slice_count * MI_SEGMENT_SLICE_SIZE; + uint8_t* const pstart = (uint8_t*)segment + (idx*MI_SEGMENT_SLICE_SIZE); + // make the start not OS page aligned for smaller blocks to avoid page/cache effects + // note: the offset must always be a block_size multiple since we assume small allocations + // are aligned (see `mi_heap_malloc_aligned`). + size_t start_offset = 0; + if (block_size > 0 && block_size <= MI_MAX_ALIGN_GUARANTEE) { + // for small objects, ensure the page start is aligned with the block size (PR#66 by kickunderscore) + const size_t adjust = block_size - ((uintptr_t)pstart % block_size); + if (adjust < block_size && psize >= block_size + adjust) { + start_offset += adjust; + } + } + if (block_size >= MI_INTPTR_SIZE) { + if (block_size <= 64) { start_offset += 3*block_size; } + else if (block_size <= 512) { start_offset += block_size; } + } + start_offset = _mi_align_up(start_offset, MI_MAX_ALIGN_SIZE); + mi_assert_internal(_mi_is_aligned(pstart + start_offset, MI_MAX_ALIGN_SIZE)); + mi_assert_internal(block_size == 0 || block_size > MI_MAX_ALIGN_GUARANTEE || _mi_is_aligned(pstart + start_offset,block_size)); + if (page_size != NULL) { *page_size = psize - start_offset; } + return (pstart + start_offset); +} + +// Start of the page available memory; can be used on uninitialized pages +uint8_t* _mi_segment_page_start(const mi_segment_t* segment, const mi_page_t* page, size_t* page_size) +{ + const mi_slice_t* slice = mi_page_to_slice((mi_page_t*)page); + uint8_t* p = _mi_segment_page_start_from_slice(segment, slice, mi_page_block_size(page), page_size); + mi_assert_internal(mi_page_block_size(page) > 0 || _mi_ptr_page(p) == page); + mi_assert_internal(_mi_ptr_segment(p) == segment); + return p; +} + + +static size_t mi_segment_calculate_slices(size_t required, size_t* info_slices) { + size_t page_size = _mi_os_page_size(); + size_t isize = _mi_align_up(sizeof(mi_segment_t), page_size); + size_t guardsize = 0; + + if (MI_SECURE>0) { + // in secure mode, we set up a protected page in between the segment info + // and the page data (and one at the end of the segment) + guardsize = page_size; + if (required > 0) { + required = _mi_align_up(required, MI_SEGMENT_SLICE_SIZE) + page_size; + } + } + + isize = _mi_align_up(isize + guardsize, MI_SEGMENT_SLICE_SIZE); + if (info_slices != NULL) *info_slices = isize / MI_SEGMENT_SLICE_SIZE; + size_t segment_size = (required==0 ? MI_SEGMENT_SIZE : _mi_align_up( required + isize + guardsize, MI_SEGMENT_SLICE_SIZE) ); + mi_assert_internal(segment_size % MI_SEGMENT_SLICE_SIZE == 0); + return (segment_size / MI_SEGMENT_SLICE_SIZE); +} + + +/* ---------------------------------------------------------------------------- +Segment caches +We keep a small segment cache per thread to increase local +reuse and avoid setting/clearing guard pages in secure mode. +------------------------------------------------------------------------------- */ + +static void mi_segments_track_size(long segment_size, mi_segments_tld_t* tld) { + if (segment_size>=0) _mi_stat_increase(&tld->stats->segments,1); + else _mi_stat_decrease(&tld->stats->segments,1); + tld->count += (segment_size >= 0 ? 1 : -1); + if (tld->count > tld->peak_count) tld->peak_count = tld->count; + tld->current_size += segment_size; + if (tld->current_size > tld->peak_size) tld->peak_size = tld->current_size; +} + +static void mi_segment_os_free(mi_segment_t* segment, mi_segments_tld_t* tld) { + segment->thread_id = 0; + _mi_segment_map_freed_at(segment); + mi_segments_track_size(-((long)mi_segment_size(segment)),tld); + if (segment->was_reclaimed) { + tld->reclaim_count--; + segment->was_reclaimed = false; + } + if (MI_SECURE>0) { + // _mi_os_unprotect(segment, mi_segment_size(segment)); // ensure no more guard pages are set + // unprotect the guard pages; we cannot just unprotect the whole segment size as part may be decommitted + size_t os_pagesize = _mi_os_page_size(); + _mi_os_unprotect((uint8_t*)segment + mi_segment_info_size(segment) - os_pagesize, os_pagesize); + uint8_t* end = (uint8_t*)segment + mi_segment_size(segment) - os_pagesize; + _mi_os_unprotect(end, os_pagesize); + } + + // purge delayed decommits now? (no, leave it to the arena) + // mi_segment_try_purge(segment,true,tld->stats); + + const size_t size = mi_segment_size(segment); + const size_t csize = _mi_commit_mask_committed_size(&segment->commit_mask, size); + + _mi_arena_free(segment, mi_segment_size(segment), csize, segment->memid); +} + +/* ----------------------------------------------------------- + Commit/Decommit ranges +----------------------------------------------------------- */ + +static void mi_segment_commit_mask(mi_segment_t* segment, bool conservative, uint8_t* p, size_t size, uint8_t** start_p, size_t* full_size, mi_commit_mask_t* cm) { + mi_assert_internal(_mi_ptr_segment(p + 1) == segment); + mi_assert_internal(segment->kind != MI_SEGMENT_HUGE); + mi_commit_mask_create_empty(cm); + if (size == 0 || size > MI_SEGMENT_SIZE || segment->kind == MI_SEGMENT_HUGE) return; + const size_t segstart = mi_segment_info_size(segment); + const size_t segsize = mi_segment_size(segment); + if (p >= (uint8_t*)segment + segsize) return; + + size_t pstart = (p - (uint8_t*)segment); + mi_assert_internal(pstart + size <= segsize); + + size_t start; + size_t end; + if (conservative) { + // decommit conservative + start = _mi_align_up(pstart, MI_COMMIT_SIZE); + end = _mi_align_down(pstart + size, MI_COMMIT_SIZE); + mi_assert_internal(start >= segstart); + mi_assert_internal(end <= segsize); + } + else { + // commit liberal + start = _mi_align_down(pstart, MI_MINIMAL_COMMIT_SIZE); + end = _mi_align_up(pstart + size, MI_MINIMAL_COMMIT_SIZE); + } + if (pstart >= segstart && start < segstart) { // note: the mask is also calculated for an initial commit of the info area + start = segstart; + } + if (end > segsize) { + end = segsize; + } + + mi_assert_internal(start <= pstart && (pstart + size) <= end); + mi_assert_internal(start % MI_COMMIT_SIZE==0 && end % MI_COMMIT_SIZE == 0); + *start_p = (uint8_t*)segment + start; + *full_size = (end > start ? end - start : 0); + if (*full_size == 0) return; + + size_t bitidx = start / MI_COMMIT_SIZE; + mi_assert_internal(bitidx < MI_COMMIT_MASK_BITS); + + size_t bitcount = *full_size / MI_COMMIT_SIZE; // can be 0 + if (bitidx + bitcount > MI_COMMIT_MASK_BITS) { + _mi_warning_message("commit mask overflow: idx=%zu count=%zu start=%zx end=%zx p=0x%p size=%zu fullsize=%zu\n", bitidx, bitcount, start, end, p, size, *full_size); + } + mi_assert_internal((bitidx + bitcount) <= MI_COMMIT_MASK_BITS); + mi_commit_mask_create(bitidx, bitcount, cm); +} + +static bool mi_segment_commit(mi_segment_t* segment, uint8_t* p, size_t size) { + mi_assert_internal(mi_commit_mask_all_set(&segment->commit_mask, &segment->purge_mask)); + + // commit liberal + uint8_t* start = NULL; + size_t full_size = 0; + mi_commit_mask_t mask; + mi_segment_commit_mask(segment, false /* conservative? */, p, size, &start, &full_size, &mask); + if (mi_commit_mask_is_empty(&mask) || full_size == 0) return true; + + if (!mi_commit_mask_all_set(&segment->commit_mask, &mask)) { + // committing + bool is_zero = false; + mi_commit_mask_t cmask; + mi_commit_mask_create_intersect(&segment->commit_mask, &mask, &cmask); + _mi_stat_decrease(&_mi_stats_main.committed, _mi_commit_mask_committed_size(&cmask, MI_SEGMENT_SIZE)); // adjust for overlap + if (!_mi_os_commit(start, full_size, &is_zero)) return false; + mi_commit_mask_set(&segment->commit_mask, &mask); + } + + // increase purge expiration when using part of delayed purges -- we assume more allocations are coming soon. + if (mi_commit_mask_any_set(&segment->purge_mask, &mask)) { + segment->purge_expire = _mi_clock_now() + mi_option_get(mi_option_purge_delay); + } + + // always clear any delayed purges in our range (as they are either committed now) + mi_commit_mask_clear(&segment->purge_mask, &mask); + return true; +} + +static bool mi_segment_ensure_committed(mi_segment_t* segment, uint8_t* p, size_t size) { + mi_assert_internal(mi_commit_mask_all_set(&segment->commit_mask, &segment->purge_mask)); + // note: assumes commit_mask is always full for huge segments as otherwise the commit mask bits can overflow + if (mi_commit_mask_is_full(&segment->commit_mask) && mi_commit_mask_is_empty(&segment->purge_mask)) return true; // fully committed + mi_assert_internal(segment->kind != MI_SEGMENT_HUGE); + return mi_segment_commit(segment, p, size); +} + +static bool mi_segment_purge(mi_segment_t* segment, uint8_t* p, size_t size) { + mi_assert_internal(mi_commit_mask_all_set(&segment->commit_mask, &segment->purge_mask)); + if (!segment->allow_purge) return true; + + // purge conservative + uint8_t* start = NULL; + size_t full_size = 0; + mi_commit_mask_t mask; + mi_segment_commit_mask(segment, true /* conservative? */, p, size, &start, &full_size, &mask); + if (mi_commit_mask_is_empty(&mask) || full_size==0) return true; + + if (mi_commit_mask_any_set(&segment->commit_mask, &mask)) { + // purging + mi_assert_internal((void*)start != (void*)segment); + mi_assert_internal(segment->allow_decommit); + const bool decommitted = _mi_os_purge(start, full_size); // reset or decommit + if (decommitted) { + mi_commit_mask_t cmask; + mi_commit_mask_create_intersect(&segment->commit_mask, &mask, &cmask); + _mi_stat_increase(&_mi_stats_main.committed, full_size - _mi_commit_mask_committed_size(&cmask, MI_SEGMENT_SIZE)); // adjust for double counting + mi_commit_mask_clear(&segment->commit_mask, &mask); + } + } + + // always clear any scheduled purges in our range + mi_commit_mask_clear(&segment->purge_mask, &mask); + return true; +} + +static void mi_segment_schedule_purge(mi_segment_t* segment, uint8_t* p, size_t size) { + if (!segment->allow_purge) return; + + if (mi_option_get(mi_option_purge_delay) == 0) { + mi_segment_purge(segment, p, size); + } + else { + // register for future purge in the purge mask + uint8_t* start = NULL; + size_t full_size = 0; + mi_commit_mask_t mask; + mi_segment_commit_mask(segment, true /*conservative*/, p, size, &start, &full_size, &mask); + if (mi_commit_mask_is_empty(&mask) || full_size==0) return; + + // update delayed commit + mi_assert_internal(segment->purge_expire > 0 || mi_commit_mask_is_empty(&segment->purge_mask)); + mi_commit_mask_t cmask; + mi_commit_mask_create_intersect(&segment->commit_mask, &mask, &cmask); // only purge what is committed; span_free may try to decommit more + mi_commit_mask_set(&segment->purge_mask, &cmask); + mi_msecs_t now = _mi_clock_now(); + if (segment->purge_expire == 0) { + // no previous purgess, initialize now + segment->purge_expire = now + mi_option_get(mi_option_purge_delay); + } + else if (segment->purge_expire <= now) { + // previous purge mask already expired + if (segment->purge_expire + mi_option_get(mi_option_purge_extend_delay) <= now) { + mi_segment_try_purge(segment, true); + } + else { + segment->purge_expire = now + mi_option_get(mi_option_purge_extend_delay); // (mi_option_get(mi_option_purge_delay) / 8); // wait a tiny bit longer in case there is a series of free's + } + } + else { + // previous purge mask is not yet expired, increase the expiration by a bit. + segment->purge_expire += mi_option_get(mi_option_purge_extend_delay); + } + } +} + +static void mi_segment_try_purge(mi_segment_t* segment, bool force) { + if (!segment->allow_purge || segment->purge_expire == 0 || mi_commit_mask_is_empty(&segment->purge_mask)) return; + mi_msecs_t now = _mi_clock_now(); + if (!force && now < segment->purge_expire) return; + + mi_commit_mask_t mask = segment->purge_mask; + segment->purge_expire = 0; + mi_commit_mask_create_empty(&segment->purge_mask); + + size_t idx; + size_t count; + mi_commit_mask_foreach(&mask, idx, count) { + // if found, decommit that sequence + if (count > 0) { + uint8_t* p = (uint8_t*)segment + (idx*MI_COMMIT_SIZE); + size_t size = count * MI_COMMIT_SIZE; + mi_segment_purge(segment, p, size); + } + } + mi_commit_mask_foreach_end() + mi_assert_internal(mi_commit_mask_is_empty(&segment->purge_mask)); +} + +// called from `mi_heap_collect_ex` +// this can be called per-page so it is important that try_purge has fast exit path +void _mi_segment_collect(mi_segment_t* segment, bool force) { + mi_segment_try_purge(segment, force); +} + +/* ----------------------------------------------------------- + Span free +----------------------------------------------------------- */ + +static bool mi_segment_is_abandoned(mi_segment_t* segment) { + return (mi_atomic_load_relaxed(&segment->thread_id) == 0); +} + +// note: can be called on abandoned segments +static void mi_segment_span_free(mi_segment_t* segment, size_t slice_index, size_t slice_count, bool allow_purge, mi_segments_tld_t* tld) { + mi_assert_internal(slice_index < segment->slice_entries); + mi_span_queue_t* sq = (segment->kind == MI_SEGMENT_HUGE || mi_segment_is_abandoned(segment) + ? NULL : mi_span_queue_for(slice_count,tld)); + if (slice_count==0) slice_count = 1; + mi_assert_internal(slice_index + slice_count - 1 < segment->slice_entries); + + // set first and last slice (the intermediates can be undetermined) + mi_slice_t* slice = &segment->slices[slice_index]; + slice->slice_count = (uint32_t)slice_count; + mi_assert_internal(slice->slice_count == slice_count); // no overflow? + slice->slice_offset = 0; + if (slice_count > 1) { + mi_slice_t* last = slice + slice_count - 1; + mi_slice_t* end = (mi_slice_t*)mi_segment_slices_end(segment); + if (last > end) { last = end; } + last->slice_count = 0; + last->slice_offset = (uint32_t)(sizeof(mi_page_t)*(slice_count - 1)); + last->block_size = 0; + } + + // perhaps decommit + if (allow_purge) { + mi_segment_schedule_purge(segment, mi_slice_start(slice), slice_count * MI_SEGMENT_SLICE_SIZE); + } + + // and push it on the free page queue (if it was not a huge page) + if (sq != NULL) mi_span_queue_push( sq, slice ); + else slice->block_size = 0; // mark huge page as free anyways +} + +/* +// called from reclaim to add existing free spans +static void mi_segment_span_add_free(mi_slice_t* slice, mi_segments_tld_t* tld) { + mi_segment_t* segment = _mi_ptr_segment(slice); + mi_assert_internal(slice->xblock_size==0 && slice->slice_count>0 && slice->slice_offset==0); + size_t slice_index = mi_slice_index(slice); + mi_segment_span_free(segment,slice_index,slice->slice_count,tld); +} +*/ + +static void mi_segment_span_remove_from_queue(mi_slice_t* slice, mi_segments_tld_t* tld) { + mi_assert_internal(slice->slice_count > 0 && slice->slice_offset==0 && slice->block_size==0); + mi_assert_internal(_mi_ptr_segment(slice)->kind != MI_SEGMENT_HUGE); + mi_span_queue_t* sq = mi_span_queue_for(slice->slice_count, tld); + mi_span_queue_delete(sq, slice); +} + +// note: can be called on abandoned segments +static mi_slice_t* mi_segment_span_free_coalesce(mi_slice_t* slice, mi_segments_tld_t* tld) { + mi_assert_internal(slice != NULL && slice->slice_count > 0 && slice->slice_offset == 0); + mi_segment_t* const segment = _mi_ptr_segment(slice); + + // for huge pages, just mark as free but don't add to the queues + if (segment->kind == MI_SEGMENT_HUGE) { + // issue #691: segment->used can be 0 if the huge page block was freed while abandoned (reclaim will get here in that case) + mi_assert_internal((segment->used==0 && slice->block_size==0) || segment->used == 1); // decreased right after this call in `mi_segment_page_clear` + slice->block_size = 0; // mark as free anyways + // we should mark the last slice `xblock_size=0` now to maintain invariants but we skip it to + // avoid a possible cache miss (and the segment is about to be freed) + return slice; + } + + // otherwise coalesce the span and add to the free span queues + const bool is_abandoned = (segment->thread_id == 0); // mi_segment_is_abandoned(segment); + size_t slice_count = slice->slice_count; + mi_slice_t* next = slice + slice->slice_count; + mi_assert_internal(next <= mi_segment_slices_end(segment)); + if (next < mi_segment_slices_end(segment) && next->block_size==0) { + // free next block -- remove it from free and merge + mi_assert_internal(next->slice_count > 0 && next->slice_offset==0); + slice_count += next->slice_count; // extend + if (!is_abandoned) { mi_segment_span_remove_from_queue(next, tld); } + } + if (slice > segment->slices) { + mi_slice_t* prev = mi_slice_first(slice - 1); + mi_assert_internal(prev >= segment->slices); + if (prev->block_size==0) { + // free previous slice -- remove it from free and merge + mi_assert_internal(prev->slice_count > 0 && prev->slice_offset==0); + slice_count += prev->slice_count; + slice->slice_count = 0; + slice->slice_offset = (uint32_t)((uint8_t*)slice - (uint8_t*)prev); // set the slice offset for `segment_force_abandon` (in case the previous free block is very large). + if (!is_abandoned) { mi_segment_span_remove_from_queue(prev, tld); } + slice = prev; + } + } + + // and add the new free page + mi_segment_span_free(segment, mi_slice_index(slice), slice_count, true, tld); + return slice; +} + + + +/* ----------------------------------------------------------- + Page allocation +----------------------------------------------------------- */ + +// Note: may still return NULL if committing the memory failed +static mi_page_t* mi_segment_span_allocate(mi_segment_t* segment, size_t slice_index, size_t slice_count) { + mi_assert_internal(slice_index < segment->slice_entries); + mi_slice_t* const slice = &segment->slices[slice_index]; + mi_assert_internal(slice->block_size==0 || slice->block_size==1); + + // commit before changing the slice data + if (!mi_segment_ensure_committed(segment, _mi_segment_page_start_from_slice(segment, slice, 0, NULL), slice_count * MI_SEGMENT_SLICE_SIZE)) { + return NULL; // commit failed! + } + + // convert the slices to a page + slice->slice_offset = 0; + slice->slice_count = (uint32_t)slice_count; + mi_assert_internal(slice->slice_count == slice_count); + const size_t bsize = slice_count * MI_SEGMENT_SLICE_SIZE; + slice->block_size = bsize; + mi_page_t* page = mi_slice_to_page(slice); + mi_assert_internal(mi_page_block_size(page) == bsize); + + // set slice back pointers for the first MI_MAX_SLICE_OFFSET_COUNT entries + size_t extra = slice_count-1; + if (extra > MI_MAX_SLICE_OFFSET_COUNT) extra = MI_MAX_SLICE_OFFSET_COUNT; + if (slice_index + extra >= segment->slice_entries) extra = segment->slice_entries - slice_index - 1; // huge objects may have more slices than avaiable entries in the segment->slices + + mi_slice_t* slice_next = slice + 1; + for (size_t i = 1; i <= extra; i++, slice_next++) { + slice_next->slice_offset = (uint32_t)(sizeof(mi_slice_t)*i); + slice_next->slice_count = 0; + slice_next->block_size = 1; + } + + // and also for the last one (if not set already) (the last one is needed for coalescing and for large alignments) + // note: the cast is needed for ubsan since the index can be larger than MI_SLICES_PER_SEGMENT for huge allocations (see #543) + mi_slice_t* last = slice + slice_count - 1; + mi_slice_t* end = (mi_slice_t*)mi_segment_slices_end(segment); + if (last > end) last = end; + if (last > slice) { + last->slice_offset = (uint32_t)(sizeof(mi_slice_t) * (last - slice)); + last->slice_count = 0; + last->block_size = 1; + } + + // and initialize the page + page->is_committed = true; + page->is_huge = (segment->kind == MI_SEGMENT_HUGE); + segment->used++; + return page; +} + +static void mi_segment_slice_split(mi_segment_t* segment, mi_slice_t* slice, size_t slice_count, mi_segments_tld_t* tld) { + mi_assert_internal(_mi_ptr_segment(slice) == segment); + mi_assert_internal(slice->slice_count >= slice_count); + mi_assert_internal(slice->block_size > 0); // no more in free queue + if (slice->slice_count <= slice_count) return; + mi_assert_internal(segment->kind != MI_SEGMENT_HUGE); + size_t next_index = mi_slice_index(slice) + slice_count; + size_t next_count = slice->slice_count - slice_count; + mi_segment_span_free(segment, next_index, next_count, false /* don't purge left-over part */, tld); + slice->slice_count = (uint32_t)slice_count; +} + +static mi_page_t* mi_segments_page_find_and_allocate(size_t slice_count, mi_arena_id_t req_arena_id, mi_segments_tld_t* tld) { + mi_assert_internal(slice_count*MI_SEGMENT_SLICE_SIZE <= MI_LARGE_OBJ_SIZE_MAX); + // search from best fit up + mi_span_queue_t* sq = mi_span_queue_for(slice_count, tld); + if (slice_count == 0) slice_count = 1; + while (sq <= &tld->spans[MI_SEGMENT_BIN_MAX]) { + for (mi_slice_t* slice = sq->first; slice != NULL; slice = slice->next) { + if (slice->slice_count >= slice_count) { + // found one + mi_segment_t* segment = _mi_ptr_segment(slice); + if (_mi_arena_memid_is_suitable(segment->memid, req_arena_id)) { + // found a suitable page span + mi_span_queue_delete(sq, slice); + + if (slice->slice_count > slice_count) { + mi_segment_slice_split(segment, slice, slice_count, tld); + } + mi_assert_internal(slice != NULL && slice->slice_count == slice_count && slice->block_size > 0); + mi_page_t* page = mi_segment_span_allocate(segment, mi_slice_index(slice), slice->slice_count); + if (page == NULL) { + // commit failed; return NULL but first restore the slice + mi_segment_span_free_coalesce(slice, tld); + return NULL; + } + return page; + } + } + } + sq++; + } + // could not find a page.. + return NULL; +} + + +/* ----------------------------------------------------------- + Segment allocation +----------------------------------------------------------- */ + +static mi_segment_t* mi_segment_os_alloc( size_t required, size_t page_alignment, bool eager_delayed, mi_arena_id_t req_arena_id, + size_t* psegment_slices, size_t* pinfo_slices, + bool commit, mi_segments_tld_t* tld) + +{ + mi_memid_t memid; + bool allow_large = (!eager_delayed && (MI_SECURE == 0)); // only allow large OS pages once we are no longer lazy + size_t align_offset = 0; + size_t alignment = MI_SEGMENT_ALIGN; + + if (page_alignment > 0) { + // mi_assert_internal(huge_page != NULL); + mi_assert_internal(page_alignment >= MI_SEGMENT_ALIGN); + alignment = page_alignment; + const size_t info_size = (*pinfo_slices) * MI_SEGMENT_SLICE_SIZE; + align_offset = _mi_align_up( info_size, MI_SEGMENT_ALIGN ); + const size_t extra = align_offset - info_size; + // recalculate due to potential guard pages + *psegment_slices = mi_segment_calculate_slices(required + extra, pinfo_slices); + mi_assert_internal(*psegment_slices > 0 && *psegment_slices <= UINT32_MAX); + } + + const size_t segment_size = (*psegment_slices) * MI_SEGMENT_SLICE_SIZE; + mi_segment_t* segment = (mi_segment_t*)_mi_arena_alloc_aligned(segment_size, alignment, align_offset, commit, allow_large, req_arena_id, &memid); + if (segment == NULL) { + return NULL; // failed to allocate + } + + // ensure metadata part of the segment is committed + mi_commit_mask_t commit_mask; + if (memid.initially_committed) { + mi_commit_mask_create_full(&commit_mask); + } + else { + // at least commit the info slices + const size_t commit_needed = _mi_divide_up((*pinfo_slices)*MI_SEGMENT_SLICE_SIZE, MI_COMMIT_SIZE); + mi_assert_internal(commit_needed>0); + mi_commit_mask_create(0, commit_needed, &commit_mask); + mi_assert_internal(commit_needed*MI_COMMIT_SIZE >= (*pinfo_slices)*MI_SEGMENT_SLICE_SIZE); + if (!_mi_os_commit(segment, commit_needed*MI_COMMIT_SIZE, NULL)) { + _mi_arena_free(segment,segment_size,0,memid); + return NULL; + } + } + mi_assert_internal(segment != NULL && (uintptr_t)segment % MI_SEGMENT_SIZE == 0); + + segment->memid = memid; + segment->allow_decommit = !memid.is_pinned; + segment->allow_purge = segment->allow_decommit && (mi_option_get(mi_option_purge_delay) >= 0); + segment->segment_size = segment_size; + segment->subproc = tld->subproc; + segment->commit_mask = commit_mask; + segment->purge_expire = 0; + mi_commit_mask_create_empty(&segment->purge_mask); + + mi_segments_track_size((long)(segment_size), tld); + _mi_segment_map_allocated_at(segment); + return segment; +} + + +// Allocate a segment from the OS aligned to `MI_SEGMENT_SIZE` . +static mi_segment_t* mi_segment_alloc(size_t required, size_t page_alignment, mi_arena_id_t req_arena_id, mi_segments_tld_t* tld, mi_page_t** huge_page) +{ + mi_assert_internal((required==0 && huge_page==NULL) || (required>0 && huge_page != NULL)); + + // calculate needed sizes first + size_t info_slices; + size_t segment_slices = mi_segment_calculate_slices(required, &info_slices); + mi_assert_internal(segment_slices > 0 && segment_slices <= UINT32_MAX); + + // Commit eagerly only if not the first N lazy segments (to reduce impact of many threads that allocate just a little) + const bool eager_delay = (// !_mi_os_has_overcommit() && // never delay on overcommit systems + _mi_current_thread_count() > 1 && // do not delay for the first N threads + tld->peak_count < (size_t)mi_option_get(mi_option_eager_commit_delay)); + const bool eager = !eager_delay && mi_option_is_enabled(mi_option_eager_commit); + bool commit = eager || (required > 0); + + // Allocate the segment from the OS + mi_segment_t* segment = mi_segment_os_alloc(required, page_alignment, eager_delay, req_arena_id, + &segment_slices, &info_slices, commit, tld); + if (segment == NULL) return NULL; + + // zero the segment info? -- not always needed as it may be zero initialized from the OS + if (!segment->memid.initially_zero) { + ptrdiff_t ofs = offsetof(mi_segment_t, next); + size_t prefix = offsetof(mi_segment_t, slices) - ofs; + size_t zsize = prefix + (sizeof(mi_slice_t) * (segment_slices + 1)); // one more + _mi_memzero((uint8_t*)segment + ofs, zsize); + } + + // initialize the rest of the segment info + const size_t slice_entries = (segment_slices > MI_SLICES_PER_SEGMENT ? MI_SLICES_PER_SEGMENT : segment_slices); + segment->segment_slices = segment_slices; + segment->segment_info_slices = info_slices; + segment->thread_id = _mi_thread_id(); + segment->cookie = _mi_ptr_cookie(segment); + segment->slice_entries = slice_entries; + segment->kind = (required == 0 ? MI_SEGMENT_NORMAL : MI_SEGMENT_HUGE); + + // _mi_memzero(segment->slices, sizeof(mi_slice_t)*(info_slices+1)); + _mi_stat_increase(&tld->stats->page_committed, mi_segment_info_size(segment)); + + // set up guard pages + size_t guard_slices = 0; + if (MI_SECURE>0) { + // in secure mode, we set up a protected page in between the segment info + // and the page data, and at the end of the segment. + size_t os_pagesize = _mi_os_page_size(); + _mi_os_protect((uint8_t*)segment + mi_segment_info_size(segment) - os_pagesize, os_pagesize); + uint8_t* end = (uint8_t*)segment + mi_segment_size(segment) - os_pagesize; + mi_segment_ensure_committed(segment, end, os_pagesize); + _mi_os_protect(end, os_pagesize); + if (slice_entries == segment_slices) segment->slice_entries--; // don't use the last slice :-( + guard_slices = 1; + } + + // reserve first slices for segment info + mi_page_t* page0 = mi_segment_span_allocate(segment, 0, info_slices); + mi_assert_internal(page0!=NULL); if (page0==NULL) return NULL; // cannot fail as we always commit in advance + mi_assert_internal(segment->used == 1); + segment->used = 0; // don't count our internal slices towards usage + + // initialize initial free pages + if (segment->kind == MI_SEGMENT_NORMAL) { // not a huge page + mi_assert_internal(huge_page==NULL); + mi_segment_span_free(segment, info_slices, segment->slice_entries - info_slices, false /* don't purge */, tld); + } + else { + mi_assert_internal(huge_page!=NULL); + mi_assert_internal(mi_commit_mask_is_empty(&segment->purge_mask)); + mi_assert_internal(mi_commit_mask_is_full(&segment->commit_mask)); + *huge_page = mi_segment_span_allocate(segment, info_slices, segment_slices - info_slices - guard_slices); + mi_assert_internal(*huge_page != NULL); // cannot fail as we commit in advance + } + + mi_assert_expensive(mi_segment_is_valid(segment,tld)); + return segment; +} + + +static void mi_segment_free(mi_segment_t* segment, bool force, mi_segments_tld_t* tld) { + MI_UNUSED(force); + mi_assert_internal(segment != NULL); + mi_assert_internal(segment->next == NULL); + mi_assert_internal(segment->used == 0); + + // in `mi_segment_force_abandon` we set this to true to ensure the segment's memory stays valid + if (segment->dont_free) return; + + // Remove the free pages + mi_slice_t* slice = &segment->slices[0]; + const mi_slice_t* end = mi_segment_slices_end(segment); + #if MI_DEBUG>1 + size_t page_count = 0; + #endif + while (slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + mi_assert_internal(mi_slice_index(slice)==0 || slice->block_size == 0); // no more used pages .. + if (slice->block_size == 0 && segment->kind != MI_SEGMENT_HUGE) { + mi_segment_span_remove_from_queue(slice, tld); + } + #if MI_DEBUG>1 + page_count++; + #endif + slice = slice + slice->slice_count; + } + mi_assert_internal(page_count == 2); // first page is allocated by the segment itself + + // stats + // _mi_stat_decrease(&tld->stats->page_committed, mi_segment_info_size(segment)); + + // return it to the OS + mi_segment_os_free(segment, tld); +} + + +/* ----------------------------------------------------------- + Page Free +----------------------------------------------------------- */ + +static void mi_segment_abandon(mi_segment_t* segment, mi_segments_tld_t* tld); + +// note: can be called on abandoned pages +static mi_slice_t* mi_segment_page_clear(mi_page_t* page, mi_segments_tld_t* tld) { + mi_assert_internal(page->block_size > 0); + mi_assert_internal(mi_page_all_free(page)); + mi_segment_t* segment = _mi_ptr_segment(page); + mi_assert_internal(segment->used > 0); + + size_t inuse = page->capacity * mi_page_block_size(page); + _mi_stat_decrease(&tld->stats->page_committed, inuse); + _mi_stat_decrease(&tld->stats->pages, 1); + _mi_stat_decrease(&tld->stats->page_bins[_mi_page_bin(page)], 1); + + // reset the page memory to reduce memory pressure? + if (segment->allow_decommit && mi_option_is_enabled(mi_option_deprecated_page_reset)) { + size_t psize; + uint8_t* start = _mi_segment_page_start(segment, page, &psize); + _mi_os_reset(start, psize); + } + + // zero the page data, but not the segment fields and heap tag + page->is_zero_init = false; + uint8_t heap_tag = page->heap_tag; + ptrdiff_t ofs = offsetof(mi_page_t, capacity); + _mi_memzero((uint8_t*)page + ofs, sizeof(*page) - ofs); + page->block_size = 1; + page->heap_tag = heap_tag; + + // and free it + mi_slice_t* slice = mi_segment_span_free_coalesce(mi_page_to_slice(page), tld); + segment->used--; + // cannot assert segment valid as it is called during reclaim + // mi_assert_expensive(mi_segment_is_valid(segment, tld)); + return slice; +} + +void _mi_segment_page_free(mi_page_t* page, bool force, mi_segments_tld_t* tld) +{ + mi_assert(page != NULL); + mi_segment_t* segment = _mi_page_segment(page); + mi_assert_expensive(mi_segment_is_valid(segment,tld)); + + // mark it as free now + mi_segment_page_clear(page, tld); + mi_assert_expensive(mi_segment_is_valid(segment, tld)); + + if (segment->used == 0) { + // no more used pages; remove from the free list and free the segment + mi_segment_free(segment, force, tld); + } + else if (segment->used == segment->abandoned) { + // only abandoned pages; remove from free list and abandon + mi_segment_abandon(segment,tld); + } + else { + // perform delayed purges + mi_segment_try_purge(segment, false /* force? */); + } +} + + +/* ----------------------------------------------------------- +Abandonment + +When threads terminate, they can leave segments with +live blocks (reachable through other threads). Such segments +are "abandoned" and will be reclaimed by other threads to +reuse their pages and/or free them eventually. The +`thread_id` of such segments is 0. + +When a block is freed in an abandoned segment, the segment +is reclaimed into that thread. + +Moreover, if threads are looking for a fresh segment, they +will first consider abandoned segments -- these can be found +by scanning the arena memory +(segments outside arena memoryare only reclaimed by a free). +----------------------------------------------------------- */ + +/* ----------------------------------------------------------- + Abandon segment/page +----------------------------------------------------------- */ + +static void mi_segment_abandon(mi_segment_t* segment, mi_segments_tld_t* tld) { + mi_assert_internal(segment->used == segment->abandoned); + mi_assert_internal(segment->used > 0); + mi_assert_internal(segment->abandoned_visits == 0); + mi_assert_expensive(mi_segment_is_valid(segment,tld)); + + // remove the free pages from the free page queues + mi_slice_t* slice = &segment->slices[0]; + const mi_slice_t* end = mi_segment_slices_end(segment); + while (slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + if (slice->block_size == 0) { // a free page + mi_segment_span_remove_from_queue(slice,tld); + slice->block_size = 0; // but keep it free + } + slice = slice + slice->slice_count; + } + + // perform delayed decommits (forcing is much slower on mstress) + // Only abandoned segments in arena memory can be reclaimed without a free + // so if a segment is not from an arena we force purge here to be conservative. + const bool force_purge = (segment->memid.memkind != MI_MEM_ARENA) || mi_option_is_enabled(mi_option_abandoned_page_purge); + mi_segment_try_purge(segment, force_purge); + + // all pages in the segment are abandoned; add it to the abandoned list + _mi_stat_increase(&tld->stats->segments_abandoned, 1); + mi_segments_track_size(-((long)mi_segment_size(segment)), tld); + segment->thread_id = 0; + segment->abandoned_visits = 1; // from 0 to 1 to signify it is abandoned + if (segment->was_reclaimed) { + tld->reclaim_count--; + segment->was_reclaimed = false; + } + _mi_arena_segment_mark_abandoned(segment); +} + +void _mi_segment_page_abandon(mi_page_t* page, mi_segments_tld_t* tld) { + mi_assert(page != NULL); + mi_assert_internal(mi_page_thread_free_flag(page)==MI_NEVER_DELAYED_FREE); + mi_assert_internal(mi_page_heap(page) == NULL); + mi_segment_t* segment = _mi_page_segment(page); + + mi_assert_expensive(mi_segment_is_valid(segment,tld)); + segment->abandoned++; + + _mi_stat_increase(&tld->stats->pages_abandoned, 1); + mi_assert_internal(segment->abandoned <= segment->used); + if (segment->used == segment->abandoned) { + // all pages are abandoned, abandon the entire segment + mi_segment_abandon(segment, tld); + } +} + +/* ----------------------------------------------------------- + Reclaim abandoned pages +----------------------------------------------------------- */ + +static mi_slice_t* mi_slices_start_iterate(mi_segment_t* segment, const mi_slice_t** end) { + mi_slice_t* slice = &segment->slices[0]; + *end = mi_segment_slices_end(segment); + mi_assert_internal(slice->slice_count>0 && slice->block_size>0); // segment allocated page + slice = slice + slice->slice_count; // skip the first segment allocated page + return slice; +} + +// Possibly free pages and check if free space is available +static bool mi_segment_check_free(mi_segment_t* segment, size_t slices_needed, size_t block_size, mi_segments_tld_t* tld) +{ + mi_assert_internal(mi_segment_is_abandoned(segment)); + bool has_page = false; + + // for all slices + const mi_slice_t* end; + mi_slice_t* slice = mi_slices_start_iterate(segment, &end); + while (slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + if (mi_slice_is_used(slice)) { // used page + // ensure used count is up to date and collect potential concurrent frees + mi_page_t* const page = mi_slice_to_page(slice); + _mi_page_free_collect(page, false); + if (mi_page_all_free(page)) { + // if this page is all free now, free it without adding to any queues (yet) + mi_assert_internal(page->next == NULL && page->prev==NULL); + _mi_stat_decrease(&tld->stats->pages_abandoned, 1); + segment->abandoned--; + slice = mi_segment_page_clear(page, tld); // re-assign slice due to coalesce! + mi_assert_internal(!mi_slice_is_used(slice)); + if (slice->slice_count >= slices_needed) { + has_page = true; + } + } + else if (mi_page_block_size(page) == block_size && mi_page_has_any_available(page)) { + // a page has available free blocks of the right size + has_page = true; + } + } + else { + // empty span + if (slice->slice_count >= slices_needed) { + has_page = true; + } + } + slice = slice + slice->slice_count; + } + return has_page; +} + +// Reclaim an abandoned segment; returns NULL if the segment was freed +// set `right_page_reclaimed` to `true` if it reclaimed a page of the right `block_size` that was not full. +static mi_segment_t* mi_segment_reclaim(mi_segment_t* segment, mi_heap_t* heap, size_t requested_block_size, bool* right_page_reclaimed, mi_segments_tld_t* tld) { + if (right_page_reclaimed != NULL) { *right_page_reclaimed = false; } + // can be 0 still with abandoned_next, or already a thread id for segments outside an arena that are reclaimed on a free. + mi_assert_internal(mi_atomic_load_relaxed(&segment->thread_id) == 0 || mi_atomic_load_relaxed(&segment->thread_id) == _mi_thread_id()); + mi_assert_internal(segment->subproc == heap->tld->segments.subproc); // only reclaim within the same subprocess + mi_atomic_store_release(&segment->thread_id, _mi_thread_id()); + segment->abandoned_visits = 0; + segment->was_reclaimed = true; + tld->reclaim_count++; + mi_segments_track_size((long)mi_segment_size(segment), tld); + mi_assert_internal(segment->next == NULL); + _mi_stat_decrease(&tld->stats->segments_abandoned, 1); + + // for all slices + const mi_slice_t* end; + mi_slice_t* slice = mi_slices_start_iterate(segment, &end); + while (slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + if (mi_slice_is_used(slice)) { + // in use: reclaim the page in our heap + mi_page_t* page = mi_slice_to_page(slice); + mi_assert_internal(page->is_committed); + mi_assert_internal(mi_page_thread_free_flag(page)==MI_NEVER_DELAYED_FREE); + mi_assert_internal(mi_page_heap(page) == NULL); + mi_assert_internal(page->next == NULL && page->prev==NULL); + _mi_stat_decrease(&tld->stats->pages_abandoned, 1); + segment->abandoned--; + // get the target heap for this thread which has a matching heap tag (so we reclaim into a matching heap) + mi_heap_t* target_heap = _mi_heap_by_tag(heap, page->heap_tag); // allow custom heaps to separate objects + if (target_heap == NULL) { + target_heap = heap; + _mi_error_message(EFAULT, "page with tag %u cannot be reclaimed by a heap with the same tag (using heap tag %u instead)\n", page->heap_tag, heap->tag ); + } + // associate the heap with this page, and allow heap thread delayed free again. + mi_page_set_heap(page, target_heap); + _mi_page_use_delayed_free(page, MI_USE_DELAYED_FREE, true); // override never (after heap is set) + _mi_page_free_collect(page, false); // ensure used count is up to date + if (mi_page_all_free(page)) { + // if everything free by now, free the page + slice = mi_segment_page_clear(page, tld); // set slice again due to coalesceing + } + else { + // otherwise reclaim it into the heap + _mi_page_reclaim(target_heap, page); + if (requested_block_size == mi_page_block_size(page) && mi_page_has_any_available(page) && heap == target_heap) { + if (right_page_reclaimed != NULL) { *right_page_reclaimed = true; } + } + } + } + else { + // the span is free, add it to our page queues + slice = mi_segment_span_free_coalesce(slice, tld); // set slice again due to coalesceing + } + mi_assert_internal(slice->slice_count>0 && slice->slice_offset==0); + slice = slice + slice->slice_count; + } + + mi_assert(segment->abandoned == 0); + mi_assert_expensive(mi_segment_is_valid(segment, tld)); + if (segment->used == 0) { // due to page_clear + mi_assert_internal(right_page_reclaimed == NULL || !(*right_page_reclaimed)); + mi_segment_free(segment, false, tld); + return NULL; + } + else { + return segment; + } +} + + +// attempt to reclaim a particular segment (called from multi threaded free `alloc.c:mi_free_block_mt`) +bool _mi_segment_attempt_reclaim(mi_heap_t* heap, mi_segment_t* segment) { + if (mi_atomic_load_relaxed(&segment->thread_id) != 0) return false; // it is not abandoned + if (segment->subproc != heap->tld->segments.subproc) return false; // only reclaim within the same subprocess + if (!_mi_heap_memid_is_suitable(heap,segment->memid)) return false; // don't reclaim between exclusive and non-exclusive arena's + const long target = _mi_option_get_fast(mi_option_target_segments_per_thread); + if (target > 0 && (size_t)target <= heap->tld->segments.count) return false; // don't reclaim if going above the target count + + // don't reclaim more from a `free` call than half the current segments + // this is to prevent a pure free-ing thread to start owning too many segments + // (but not for out-of-arena segments as that is the main way to be reclaimed for those) + if (segment->memid.memkind == MI_MEM_ARENA && heap->tld->segments.reclaim_count * 2 > heap->tld->segments.count) { + return false; + } + if (_mi_arena_segment_clear_abandoned(segment)) { // atomically unabandon + mi_segment_t* res = mi_segment_reclaim(segment, heap, 0, NULL, &heap->tld->segments); + mi_assert_internal(res == segment); + return (res != NULL); + } + return false; +} + +void _mi_abandoned_reclaim_all(mi_heap_t* heap, mi_segments_tld_t* tld) { + mi_segment_t* segment; + mi_arena_field_cursor_t current; + _mi_arena_field_cursor_init(heap, tld->subproc, true /* visit all, blocking */, ¤t); + while ((segment = _mi_arena_segment_clear_abandoned_next(¤t)) != NULL) { + mi_segment_reclaim(segment, heap, 0, NULL, tld); + } + _mi_arena_field_cursor_done(¤t); +} + + +static bool segment_count_is_within_target(mi_segments_tld_t* tld, size_t* ptarget) { + const size_t target = (size_t)mi_option_get_clamp(mi_option_target_segments_per_thread, 0, 1024); + if (ptarget != NULL) { *ptarget = target; } + return (target == 0 || tld->count < target); +} + +static long mi_segment_get_reclaim_tries(mi_segments_tld_t* tld) { + // limit the tries to 10% (default) of the abandoned segments with at least 8 and at most 1024 tries. + const size_t perc = (size_t)mi_option_get_clamp(mi_option_max_segment_reclaim, 0, 100); + if (perc <= 0) return 0; + const size_t total_count = mi_atomic_load_relaxed(&tld->subproc->abandoned_count); + if (total_count == 0) return 0; + const size_t relative_count = (total_count > 10000 ? (total_count / 100) * perc : (total_count * perc) / 100); // avoid overflow + long max_tries = (long)(relative_count <= 1 ? 1 : (relative_count > 1024 ? 1024 : relative_count)); + if (max_tries < 8 && total_count > 8) { max_tries = 8; } + return max_tries; +} + +static mi_segment_t* mi_segment_try_reclaim(mi_heap_t* heap, size_t needed_slices, size_t block_size, bool* reclaimed, mi_segments_tld_t* tld) +{ + *reclaimed = false; + long max_tries = mi_segment_get_reclaim_tries(tld); + if (max_tries <= 0) return NULL; + + mi_segment_t* result = NULL; + mi_segment_t* segment = NULL; + mi_arena_field_cursor_t current; + _mi_arena_field_cursor_init(heap, tld->subproc, false /* non-blocking */, ¤t); + while (segment_count_is_within_target(tld,NULL) && (max_tries-- > 0) && ((segment = _mi_arena_segment_clear_abandoned_next(¤t)) != NULL)) + { + mi_assert(segment->subproc == heap->tld->segments.subproc); // cursor only visits segments in our sub-process + segment->abandoned_visits++; + // todo: should we respect numa affinity for abandoned reclaim? perhaps only for the first visit? + // todo: an arena exclusive heap will potentially visit many abandoned unsuitable segments and use many tries + // Perhaps we can skip non-suitable ones in a better way? + bool is_suitable = _mi_heap_memid_is_suitable(heap, segment->memid); + bool has_page = mi_segment_check_free(segment,needed_slices,block_size,tld); // try to free up pages (due to concurrent frees) + if (segment->used == 0) { + // free the segment (by forced reclaim) to make it available to other threads. + // note1: we prefer to free a segment as that might lead to reclaiming another + // segment that is still partially used. + // note2: we could in principle optimize this by skipping reclaim and directly + // freeing but that would violate some invariants temporarily) + mi_segment_reclaim(segment, heap, 0, NULL, tld); + } + else if (has_page && is_suitable) { + // found a large enough free span, or a page of the right block_size with free space + // we return the result of reclaim (which is usually `segment`) as it might free + // the segment due to concurrent frees (in which case `NULL` is returned). + result = mi_segment_reclaim(segment, heap, block_size, reclaimed, tld); + break; + } + else if (segment->abandoned_visits > 3 && is_suitable) { + // always reclaim on 3rd visit to limit the abandoned segment count. + mi_segment_reclaim(segment, heap, 0, NULL, tld); + } + else { + // otherwise, push on the visited list so it gets not looked at too quickly again + max_tries++; // don't count this as a try since it was not suitable + mi_segment_try_purge(segment, false /* true force? */); // force purge if needed as we may not visit soon again + _mi_arena_segment_mark_abandoned(segment); + } + } + _mi_arena_field_cursor_done(¤t); + return result; +} + +// collect abandoned segments +void _mi_abandoned_collect(mi_heap_t* heap, bool force, mi_segments_tld_t* tld) +{ + mi_segment_t* segment; + mi_arena_field_cursor_t current; _mi_arena_field_cursor_init(heap, tld->subproc, force /* blocking? */, ¤t); + long max_tries = (force ? (long)mi_atomic_load_relaxed(&tld->subproc->abandoned_count) : 1024); // limit latency + while ((max_tries-- > 0) && ((segment = _mi_arena_segment_clear_abandoned_next(¤t)) != NULL)) { + mi_segment_check_free(segment,0,0,tld); // try to free up pages (due to concurrent frees) + if (segment->used == 0) { + // free the segment (by forced reclaim) to make it available to other threads. + // note: we could in principle optimize this by skipping reclaim and directly + // freeing but that would violate some invariants temporarily) + mi_segment_reclaim(segment, heap, 0, NULL, tld); + } + else { + // otherwise, purge if needed and push on the visited list + // note: forced purge can be expensive if many threads are destroyed/created as in mstress. + mi_segment_try_purge(segment, force); + _mi_arena_segment_mark_abandoned(segment); + } + } + _mi_arena_field_cursor_done(¤t); +} + +/* ----------------------------------------------------------- + Force abandon a segment that is in use by our thread +----------------------------------------------------------- */ + +// force abandon a segment +static void mi_segment_force_abandon(mi_segment_t* segment, mi_segments_tld_t* tld) +{ + mi_assert_internal(!mi_segment_is_abandoned(segment)); + mi_assert_internal(!segment->dont_free); + + // ensure the segment does not get free'd underneath us (so we can check if a page has been freed in `mi_page_force_abandon`) + segment->dont_free = true; + + // for all slices + const mi_slice_t* end; + mi_slice_t* slice = mi_slices_start_iterate(segment, &end); + while (slice < end) { + mi_assert_internal(slice->slice_count > 0); + mi_assert_internal(slice->slice_offset == 0); + if (mi_slice_is_used(slice)) { + // ensure used count is up to date and collect potential concurrent frees + mi_page_t* const page = mi_slice_to_page(slice); + _mi_page_free_collect(page, false); + { + // abandon the page if it is still in-use (this will free it if possible as well) + mi_assert_internal(segment->used > 0); + if (segment->used == segment->abandoned+1) { + // the last page.. abandon and return as the segment will be abandoned after this + // and we should no longer access it. + segment->dont_free = false; + _mi_page_force_abandon(page); + return; + } + else { + // abandon and continue + _mi_page_force_abandon(page); + // it might be freed, reset the slice (note: relies on coalesce setting the slice_offset) + slice = mi_slice_first(slice); + } + } + } + slice = slice + slice->slice_count; + } + segment->dont_free = false; + mi_assert(segment->used == segment->abandoned); + mi_assert(segment->used == 0); + if (segment->used == 0) { // paranoia + // all free now + mi_segment_free(segment, false, tld); + } + else { + // perform delayed purges + mi_segment_try_purge(segment, false /* force? */); + } +} + + +// try abandon segments. +// this should be called from `reclaim_or_alloc` so we know all segments are (about) fully in use. +static void mi_segments_try_abandon_to_target(mi_heap_t* heap, size_t target, mi_segments_tld_t* tld) { + if (target <= 1) return; + const size_t min_target = (target > 4 ? (target*3)/4 : target); // 75% + // todo: we should maintain a list of segments per thread; for now, only consider segments from the heap full pages + for (int i = 0; i < 64 && tld->count >= min_target; i++) { + mi_page_t* page = heap->pages[MI_BIN_FULL].first; + while (page != NULL && mi_page_block_size(page) > MI_LARGE_OBJ_SIZE_MAX) { + page = page->next; + } + if (page==NULL) { + break; + } + mi_segment_t* segment = _mi_page_segment(page); + mi_segment_force_abandon(segment, tld); + mi_assert_internal(page != heap->pages[MI_BIN_FULL].first); // as it is just abandoned + } +} + +// try abandon segments. +// this should be called from `reclaim_or_alloc` so we know all segments are (about) fully in use. +static void mi_segments_try_abandon(mi_heap_t* heap, mi_segments_tld_t* tld) { + // we call this when we are about to add a fresh segment so we should be under our target segment count. + size_t target = 0; + if (segment_count_is_within_target(tld, &target)) return; + mi_segments_try_abandon_to_target(heap, target, tld); +} + +void mi_collect_reduce(size_t target_size) mi_attr_noexcept { + mi_collect(true); + mi_heap_t* heap = mi_heap_get_default(); + mi_segments_tld_t* tld = &heap->tld->segments; + size_t target = target_size / MI_SEGMENT_SIZE; + if (target == 0) { + target = (size_t)mi_option_get_clamp(mi_option_target_segments_per_thread, 1, 1024); + } + mi_segments_try_abandon_to_target(heap, target, tld); +} + +/* ----------------------------------------------------------- + Reclaim or allocate +----------------------------------------------------------- */ + +static mi_segment_t* mi_segment_reclaim_or_alloc(mi_heap_t* heap, size_t needed_slices, size_t block_size, mi_segments_tld_t* tld) +{ + mi_assert_internal(block_size <= MI_LARGE_OBJ_SIZE_MAX); + + // try to abandon some segments to increase reuse between threads + mi_segments_try_abandon(heap,tld); + + // 1. try to reclaim an abandoned segment + bool reclaimed; + mi_segment_t* segment = mi_segment_try_reclaim(heap, needed_slices, block_size, &reclaimed, tld); + if (reclaimed) { + // reclaimed the right page right into the heap + mi_assert_internal(segment != NULL); + return NULL; // pretend out-of-memory as the page will be in the page queue of the heap with available blocks + } + else if (segment != NULL) { + // reclaimed a segment with a large enough empty span in it + return segment; + } + // 2. otherwise allocate a fresh segment + return mi_segment_alloc(0, 0, heap->arena_id, tld, NULL); +} + + +/* ----------------------------------------------------------- + Page allocation +----------------------------------------------------------- */ + +static mi_page_t* mi_segments_page_alloc(mi_heap_t* heap, mi_page_kind_t page_kind, size_t required, size_t block_size, mi_segments_tld_t* tld) +{ + mi_assert_internal(required <= MI_LARGE_OBJ_SIZE_MAX && page_kind <= MI_PAGE_LARGE); + + // find a free page + size_t page_size = _mi_align_up(required, (required > MI_MEDIUM_PAGE_SIZE ? MI_MEDIUM_PAGE_SIZE : MI_SEGMENT_SLICE_SIZE)); + size_t slices_needed = page_size / MI_SEGMENT_SLICE_SIZE; + mi_assert_internal(slices_needed * MI_SEGMENT_SLICE_SIZE == page_size); + mi_page_t* page = mi_segments_page_find_and_allocate(slices_needed, heap->arena_id, tld); //(required <= MI_SMALL_SIZE_MAX ? 0 : slices_needed), tld); + if (page==NULL) { + // no free page, allocate a new segment and try again + if (mi_segment_reclaim_or_alloc(heap, slices_needed, block_size, tld) == NULL) { + // OOM or reclaimed a good page in the heap + return NULL; + } + else { + // otherwise try again + return mi_segments_page_alloc(heap, page_kind, required, block_size, tld); + } + } + mi_assert_internal(page != NULL && page->slice_count*MI_SEGMENT_SLICE_SIZE == page_size); + mi_assert_internal(_mi_ptr_segment(page)->thread_id == _mi_thread_id()); + mi_segment_try_purge(_mi_ptr_segment(page), false); + return page; +} + + + +/* ----------------------------------------------------------- + Huge page allocation +----------------------------------------------------------- */ + +static mi_page_t* mi_segment_huge_page_alloc(size_t size, size_t page_alignment, mi_arena_id_t req_arena_id, mi_segments_tld_t* tld) +{ + mi_page_t* page = NULL; + mi_segment_t* segment = mi_segment_alloc(size,page_alignment,req_arena_id,tld,&page); + if (segment == NULL || page==NULL) return NULL; + mi_assert_internal(segment->used==1); + mi_assert_internal(mi_page_block_size(page) >= size); + #if MI_HUGE_PAGE_ABANDON + segment->thread_id = 0; // huge segments are immediately abandoned + #endif + + // for huge pages we initialize the block_size as we may + // overallocate to accommodate large alignments. + size_t psize; + uint8_t* start = _mi_segment_page_start(segment, page, &psize); + page->block_size = psize; + mi_assert_internal(page->is_huge); + + // decommit the part of the prefix of a page that will not be used; this can be quite large (close to MI_SEGMENT_SIZE) + if (page_alignment > 0 && segment->allow_decommit) { + uint8_t* aligned_p = (uint8_t*)_mi_align_up((uintptr_t)start, page_alignment); + mi_assert_internal(_mi_is_aligned(aligned_p, page_alignment)); + mi_assert_internal(psize - (aligned_p - start) >= size); + uint8_t* decommit_start = start + sizeof(mi_block_t); // for the free list + ptrdiff_t decommit_size = aligned_p - decommit_start; + _mi_os_reset(decommit_start, decommit_size); // note: cannot use segment_decommit on huge segments + } + + return page; +} + +#if MI_HUGE_PAGE_ABANDON +// free huge block from another thread +void _mi_segment_huge_page_free(mi_segment_t* segment, mi_page_t* page, mi_block_t* block) { + // huge page segments are always abandoned and can be freed immediately by any thread + mi_assert_internal(segment->kind==MI_SEGMENT_HUGE); + mi_assert_internal(segment == _mi_page_segment(page)); + mi_assert_internal(mi_atomic_load_relaxed(&segment->thread_id)==0); + + // claim it and free + mi_heap_t* heap = mi_heap_get_default(); // issue #221; don't use the internal get_default_heap as we need to ensure the thread is initialized. + // paranoia: if this it the last reference, the cas should always succeed + size_t expected_tid = 0; + if (mi_atomic_cas_strong_acq_rel(&segment->thread_id, &expected_tid, heap->thread_id)) { + mi_block_set_next(page, block, page->free); + page->free = block; + page->used--; + page->is_zero_init = false; + mi_assert(page->used == 0); + mi_tld_t* tld = heap->tld; + _mi_segment_page_free(page, true, &tld->segments); + } +#if (MI_DEBUG!=0) + else { + mi_assert_internal(false); + } +#endif +} + +#else +// reset memory of a huge block from another thread +void _mi_segment_huge_page_reset(mi_segment_t* segment, mi_page_t* page, mi_block_t* block) { + MI_UNUSED(page); + mi_assert_internal(segment->kind == MI_SEGMENT_HUGE); + mi_assert_internal(segment == _mi_page_segment(page)); + mi_assert_internal(page->used == 1); // this is called just before the free + mi_assert_internal(page->free == NULL); + if (segment->allow_decommit) { + size_t csize = mi_usable_size(block); + if (csize > sizeof(mi_block_t)) { + csize = csize - sizeof(mi_block_t); + uint8_t* p = (uint8_t*)block + sizeof(mi_block_t); + _mi_os_reset(p, csize); // note: cannot use segment_decommit on huge segments + } + } +} +#endif + +/* ----------------------------------------------------------- + Page allocation and free +----------------------------------------------------------- */ +mi_page_t* _mi_segment_page_alloc(mi_heap_t* heap, size_t block_size, size_t page_alignment, mi_segments_tld_t* tld) { + mi_page_t* page; + if mi_unlikely(page_alignment > MI_BLOCK_ALIGNMENT_MAX) { + mi_assert_internal(_mi_is_power_of_two(page_alignment)); + mi_assert_internal(page_alignment >= MI_SEGMENT_SIZE); + if (page_alignment < MI_SEGMENT_SIZE) { page_alignment = MI_SEGMENT_SIZE; } + page = mi_segment_huge_page_alloc(block_size,page_alignment,heap->arena_id,tld); + } + else if (block_size <= MI_SMALL_OBJ_SIZE_MAX) { + page = mi_segments_page_alloc(heap,MI_PAGE_SMALL,block_size,block_size,tld); + } + else if (block_size <= MI_MEDIUM_OBJ_SIZE_MAX) { + page = mi_segments_page_alloc(heap,MI_PAGE_MEDIUM,MI_MEDIUM_PAGE_SIZE,block_size,tld); + } + else if (block_size <= MI_LARGE_OBJ_SIZE_MAX) { + page = mi_segments_page_alloc(heap,MI_PAGE_LARGE,block_size,block_size,tld); + } + else { + page = mi_segment_huge_page_alloc(block_size,page_alignment,heap->arena_id,tld); + } + mi_assert_internal(page == NULL || _mi_heap_memid_is_suitable(heap, _mi_page_segment(page)->memid)); + mi_assert_expensive(page == NULL || mi_segment_is_valid(_mi_page_segment(page),tld)); + mi_assert_internal(page == NULL || _mi_page_segment(page)->subproc == tld->subproc); + return page; +} + + +/* ----------------------------------------------------------- + Visit blocks in a segment (only used for abandoned segments) +----------------------------------------------------------- */ + +static bool mi_segment_visit_page(mi_page_t* page, bool visit_blocks, mi_block_visit_fun* visitor, void* arg) { + mi_heap_area_t area; + _mi_heap_area_init(&area, page); + if (!visitor(NULL, &area, NULL, area.block_size, arg)) return false; + if (visit_blocks) { + return _mi_heap_area_visit_blocks(&area, page, visitor, arg); + } + else { + return true; + } +} + +bool _mi_segment_visit_blocks(mi_segment_t* segment, int heap_tag, bool visit_blocks, mi_block_visit_fun* visitor, void* arg) { + const mi_slice_t* end; + mi_slice_t* slice = mi_slices_start_iterate(segment, &end); + while (slice < end) { + if (mi_slice_is_used(slice)) { + mi_page_t* const page = mi_slice_to_page(slice); + if (heap_tag < 0 || (int)page->heap_tag == heap_tag) { + if (!mi_segment_visit_page(page, visit_blocks, visitor, arg)) return false; + } + } + slice = slice + slice->slice_count; + } + return true; +} diff --git a/compat/mimalloc/stats.c b/compat/mimalloc/stats.c new file mode 100644 index 00000000000000..34b3d4e4ce44c8 --- /dev/null +++ b/compat/mimalloc/stats.c @@ -0,0 +1,635 @@ +/* ---------------------------------------------------------------------------- +Copyright (c) 2018-2021, Microsoft Research, Daan Leijen +This is free software; you can redistribute it and/or modify it under the +terms of the MIT license. A copy of the license can be found in the file +"LICENSE" at the root of this distribution. +-----------------------------------------------------------------------------*/ +#include "mimalloc.h" +#include "mimalloc/internal.h" +#include "mimalloc/atomic.h" +#include "mimalloc/prim.h" + +#include // memset + +#if defined(_MSC_VER) && (_MSC_VER < 1920) +#pragma warning(disable:4204) // non-constant aggregate initializer +#endif + +/* ----------------------------------------------------------- + Statistics operations +----------------------------------------------------------- */ + +static bool mi_is_in_main(void* stat) { + return ((uint8_t*)stat >= (uint8_t*)&_mi_stats_main + && (uint8_t*)stat < ((uint8_t*)&_mi_stats_main + sizeof(mi_stats_t))); +} + +static void mi_stat_update(mi_stat_count_t* stat, int64_t amount) { + if (amount == 0) return; + if mi_unlikely(mi_is_in_main(stat)) + { + // add atomically (for abandoned pages) + int64_t current = mi_atomic_addi64_relaxed(&stat->current, amount); + // if (stat == &_mi_stats_main.committed) { mi_assert_internal(current + amount >= 0); }; + mi_atomic_maxi64_relaxed(&stat->peak, current + amount); + if (amount > 0) { + mi_atomic_addi64_relaxed(&stat->total,amount); + } + } + else { + // add thread local + stat->current += amount; + if (stat->current > stat->peak) { stat->peak = stat->current; } + if (amount > 0) { stat->total += amount; } + } +} + +void _mi_stat_counter_increase(mi_stat_counter_t* stat, size_t amount) { + if (mi_is_in_main(stat)) { + mi_atomic_addi64_relaxed( &stat->total, (int64_t)amount ); + } + else { + stat->total += amount; + } +} + +void _mi_stat_increase(mi_stat_count_t* stat, size_t amount) { + mi_stat_update(stat, (int64_t)amount); +} + +void _mi_stat_decrease(mi_stat_count_t* stat, size_t amount) { + mi_stat_update(stat, -((int64_t)amount)); +} + + +static void mi_stat_adjust(mi_stat_count_t* stat, int64_t amount) { + if (amount == 0) return; + if mi_unlikely(mi_is_in_main(stat)) + { + // adjust atomically + mi_atomic_addi64_relaxed(&stat->current, amount); + mi_atomic_addi64_relaxed(&stat->total,amount); + } + else { + // adjust local + stat->current += amount; + stat->total += amount; + } +} + +void _mi_stat_adjust_decrease(mi_stat_count_t* stat, size_t amount) { + mi_stat_adjust(stat, -((int64_t)amount)); +} + + +// must be thread safe as it is called from stats_merge +static void mi_stat_count_add_mt(mi_stat_count_t* stat, const mi_stat_count_t* src) { + if (stat==src) return; + mi_atomic_void_addi64_relaxed(&stat->total, &src->total); + mi_atomic_void_addi64_relaxed(&stat->current, &src->current); + // peak scores do really not work across threads .. we just add them + mi_atomic_void_addi64_relaxed( &stat->peak, &src->peak); + // or, take the max? + // mi_atomic_maxi64_relaxed(&stat->peak, src->peak); +} + +static void mi_stat_counter_add_mt(mi_stat_counter_t* stat, const mi_stat_counter_t* src) { + if (stat==src) return; + mi_atomic_void_addi64_relaxed(&stat->total, &src->total); +} + +#define MI_STAT_COUNT(stat) mi_stat_count_add_mt(&stats->stat, &src->stat); +#define MI_STAT_COUNTER(stat) mi_stat_counter_add_mt(&stats->stat, &src->stat); + +// must be thread safe as it is called from stats_merge +static void mi_stats_add(mi_stats_t* stats, const mi_stats_t* src) { + if (stats==src) return; + + // copy all fields + MI_STAT_FIELDS() + + #if MI_STAT>1 + for (size_t i = 0; i <= MI_BIN_HUGE; i++) { + mi_stat_count_add_mt(&stats->malloc_bins[i], &src->malloc_bins[i]); + } + #endif + for (size_t i = 0; i <= MI_BIN_HUGE; i++) { + mi_stat_count_add_mt(&stats->page_bins[i], &src->page_bins[i]); + } +} + +#undef MI_STAT_COUNT +#undef MI_STAT_COUNTER + +/* ----------------------------------------------------------- + Display statistics +----------------------------------------------------------- */ + +// unit > 0 : size in binary bytes +// unit == 0: count as decimal +// unit < 0 : count in binary +static void mi_printf_amount(int64_t n, int64_t unit, mi_output_fun* out, void* arg, const char* fmt) { + char buf[32]; buf[0] = 0; + int len = 32; + const char* suffix = (unit <= 0 ? " " : "B"); + const int64_t base = (unit == 0 ? 1000 : 1024); + if (unit>0) n *= unit; + + const int64_t pos = (n < 0 ? -n : n); + if (pos < base) { + if (n!=1 || suffix[0] != 'B') { // skip printing 1 B for the unit column + _mi_snprintf(buf, len, "%lld %-3s", (long long)n, (n==0 ? "" : suffix)); + } + } + else { + int64_t divider = base; + const char* magnitude = "K"; + if (pos >= divider*base) { divider *= base; magnitude = "M"; } + if (pos >= divider*base) { divider *= base; magnitude = "G"; } + const int64_t tens = (n / (divider/10)); + const long whole = (long)(tens/10); + const long frac1 = (long)(tens%10); + char unitdesc[8]; + _mi_snprintf(unitdesc, 8, "%s%s%s", magnitude, (base==1024 ? "i" : ""), suffix); + _mi_snprintf(buf, len, "%ld.%ld %-3s", whole, (frac1 < 0 ? -frac1 : frac1), unitdesc); + } + _mi_fprintf(out, arg, (fmt==NULL ? "%12s" : fmt), buf); +} + + +static void mi_print_amount(int64_t n, int64_t unit, mi_output_fun* out, void* arg) { + mi_printf_amount(n,unit,out,arg,NULL); +} + +static void mi_print_count(int64_t n, int64_t unit, mi_output_fun* out, void* arg) { + if (unit==1) _mi_fprintf(out, arg, "%12s"," "); + else mi_print_amount(n,0,out,arg); +} + +static void mi_stat_print_ex(const mi_stat_count_t* stat, const char* msg, int64_t unit, mi_output_fun* out, void* arg, const char* notok ) { + _mi_fprintf(out, arg,"%10s:", msg); + if (unit != 0) { + if (unit > 0) { + mi_print_amount(stat->peak, unit, out, arg); + mi_print_amount(stat->total, unit, out, arg); + // mi_print_amount(stat->freed, unit, out, arg); + mi_print_amount(stat->current, unit, out, arg); + mi_print_amount(unit, 1, out, arg); + mi_print_count(stat->total, unit, out, arg); + } + else { + mi_print_amount(stat->peak, -1, out, arg); + mi_print_amount(stat->total, -1, out, arg); + // mi_print_amount(stat->freed, -1, out, arg); + mi_print_amount(stat->current, -1, out, arg); + if (unit == -1) { + _mi_fprintf(out, arg, "%24s", ""); + } + else { + mi_print_amount(-unit, 1, out, arg); + mi_print_count((stat->total / -unit), 0, out, arg); + } + } + if (stat->current != 0) { + _mi_fprintf(out, arg, " "); + _mi_fprintf(out, arg, (notok == NULL ? "not all freed" : notok)); + _mi_fprintf(out, arg, "\n"); + } + else { + _mi_fprintf(out, arg, " ok\n"); + } + } + else { + mi_print_amount(stat->peak, 1, out, arg); + mi_print_amount(stat->total, 1, out, arg); + _mi_fprintf(out, arg, "%11s", " "); // no freed + mi_print_amount(stat->current, 1, out, arg); + _mi_fprintf(out, arg, "\n"); + } +} + +static void mi_stat_print(const mi_stat_count_t* stat, const char* msg, int64_t unit, mi_output_fun* out, void* arg) { + mi_stat_print_ex(stat, msg, unit, out, arg, NULL); +} + +static void mi_stat_peak_print(const mi_stat_count_t* stat, const char* msg, int64_t unit, mi_output_fun* out, void* arg) { + _mi_fprintf(out, arg, "%10s:", msg); + mi_print_amount(stat->peak, unit, out, arg); + _mi_fprintf(out, arg, "\n"); +} + +#if MI_STAT>1 +static void mi_stat_total_print(const mi_stat_count_t* stat, const char* msg, int64_t unit, mi_output_fun* out, void* arg) { + _mi_fprintf(out, arg, "%10s:", msg); + _mi_fprintf(out, arg, "%12s", " "); // no peak + mi_print_amount(stat->total, unit, out, arg); + _mi_fprintf(out, arg, "\n"); +} +#endif + +static void mi_stat_counter_print(const mi_stat_counter_t* stat, const char* msg, mi_output_fun* out, void* arg ) { + _mi_fprintf(out, arg, "%10s:", msg); + mi_print_amount(stat->total, -1, out, arg); + _mi_fprintf(out, arg, "\n"); +} + + +static void mi_stat_counter_print_avg(const mi_stat_counter_t* stat, const char* msg, mi_output_fun* out, void* arg) { + const int64_t avg_tens = (stat->total == 0 ? 0 : (stat->total*10 / stat->total)); + const long avg_whole = (long)(avg_tens/10); + const long avg_frac1 = (long)(avg_tens%10); + _mi_fprintf(out, arg, "%10s: %5ld.%ld avg\n", msg, avg_whole, avg_frac1); +} + + +static void mi_print_header(mi_output_fun* out, void* arg ) { + _mi_fprintf(out, arg, "%10s: %11s %11s %11s %11s %11s\n", "heap stats", "peak ", "total ", "current ", "block ", "total# "); +} + +#if MI_STAT>1 +static void mi_stats_print_bins(const mi_stat_count_t* bins, size_t max, const char* fmt, mi_output_fun* out, void* arg) { + bool found = false; + char buf[64]; + for (size_t i = 0; i <= max; i++) { + if (bins[i].total > 0) { + found = true; + int64_t unit = _mi_bin_size((uint8_t)i); + _mi_snprintf(buf, 64, "%s %3lu", fmt, (long)i); + mi_stat_print(&bins[i], buf, unit, out, arg); + } + } + if (found) { + _mi_fprintf(out, arg, "\n"); + mi_print_header(out, arg); + } +} +#endif + + + +//------------------------------------------------------------ +// Use an output wrapper for line-buffered output +// (which is nice when using loggers etc.) +//------------------------------------------------------------ +typedef struct buffered_s { + mi_output_fun* out; // original output function + void* arg; // and state + char* buf; // local buffer of at least size `count+1` + size_t used; // currently used chars `used <= count` + size_t count; // total chars available for output +} buffered_t; + +static void mi_buffered_flush(buffered_t* buf) { + buf->buf[buf->used] = 0; + _mi_fputs(buf->out, buf->arg, NULL, buf->buf); + buf->used = 0; +} + +static void mi_cdecl mi_buffered_out(const char* msg, void* arg) { + buffered_t* buf = (buffered_t*)arg; + if (msg==NULL || buf==NULL) return; + for (const char* src = msg; *src != 0; src++) { + char c = *src; + if (buf->used >= buf->count) mi_buffered_flush(buf); + mi_assert_internal(buf->used < buf->count); + buf->buf[buf->used++] = c; + if (c == '\n') mi_buffered_flush(buf); + } +} + +//------------------------------------------------------------ +// Print statistics +//------------------------------------------------------------ + +static void _mi_stats_print(mi_stats_t* stats, mi_output_fun* out0, void* arg0) mi_attr_noexcept { + // wrap the output function to be line buffered + char buf[256]; + buffered_t buffer = { out0, arg0, NULL, 0, 255 }; + buffer.buf = buf; + mi_output_fun* out = &mi_buffered_out; + void* arg = &buffer; + + // and print using that + mi_print_header(out,arg); + #if MI_STAT>1 + mi_stats_print_bins(stats->malloc_bins, MI_BIN_HUGE, "bin",out,arg); + #endif + #if MI_STAT + mi_stat_print(&stats->malloc_normal, "binned", (stats->malloc_normal_count.total == 0 ? 1 : -1), out, arg); + // mi_stat_print(&stats->malloc_large, "large", (stats->malloc_large_count.total == 0 ? 1 : -1), out, arg); + mi_stat_print(&stats->malloc_huge, "huge", (stats->malloc_huge_count.total == 0 ? 1 : -1), out, arg); + mi_stat_count_t total = { 0,0,0 }; + mi_stat_count_add_mt(&total, &stats->malloc_normal); + // mi_stat_count_add(&total, &stats->malloc_large); + mi_stat_count_add_mt(&total, &stats->malloc_huge); + mi_stat_print_ex(&total, "total", 1, out, arg, ""); + #endif + #if MI_STAT>1 + mi_stat_total_print(&stats->malloc_requested, "malloc req", 1, out, arg); + _mi_fprintf(out, arg, "\n"); + #endif + mi_stat_print_ex(&stats->reserved, "reserved", 1, out, arg, ""); + mi_stat_print_ex(&stats->committed, "committed", 1, out, arg, ""); + mi_stat_peak_print(&stats->reset, "reset", 1, out, arg ); + mi_stat_peak_print(&stats->purged, "purged", 1, out, arg ); + mi_stat_print_ex(&stats->page_committed, "touched", 1, out, arg, ""); + mi_stat_print(&stats->segments, "segments", -1, out, arg); + mi_stat_print(&stats->segments_abandoned, "-abandoned", -1, out, arg); + mi_stat_print(&stats->segments_cache, "-cached", -1, out, arg); + mi_stat_print(&stats->pages, "pages", -1, out, arg); + mi_stat_print(&stats->pages_abandoned, "-abandoned", -1, out, arg); + mi_stat_counter_print(&stats->pages_extended, "-extended", out, arg); + mi_stat_counter_print(&stats->pages_retire, "-retire", out, arg); + mi_stat_counter_print(&stats->arena_count, "arenas", out, arg); + // mi_stat_counter_print(&stats->arena_crossover_count, "-crossover", out, arg); + mi_stat_counter_print(&stats->arena_rollback_count, "-rollback", out, arg); + mi_stat_counter_print(&stats->mmap_calls, "mmaps", out, arg); + mi_stat_counter_print(&stats->commit_calls, "commits", out, arg); + mi_stat_counter_print(&stats->reset_calls, "resets", out, arg); + mi_stat_counter_print(&stats->purge_calls, "purges", out, arg); + mi_stat_counter_print(&stats->malloc_guarded_count, "guarded", out, arg); + mi_stat_print(&stats->threads, "threads", -1, out, arg); + mi_stat_counter_print_avg(&stats->page_searches, "searches", out, arg); + _mi_fprintf(out, arg, "%10s: %5i\n", "numa nodes", _mi_os_numa_node_count()); + + size_t elapsed; + size_t user_time; + size_t sys_time; + size_t current_rss; + size_t peak_rss; + size_t current_commit; + size_t peak_commit; + size_t page_faults; + mi_process_info(&elapsed, &user_time, &sys_time, ¤t_rss, &peak_rss, ¤t_commit, &peak_commit, &page_faults); + _mi_fprintf(out, arg, "%10s: %5zu.%03zu s\n", "elapsed", elapsed/1000, elapsed%1000); + _mi_fprintf(out, arg, "%10s: user: %zu.%03zu s, system: %zu.%03zu s, faults: %zu, rss: ", "process", + user_time/1000, user_time%1000, sys_time/1000, sys_time%1000, page_faults ); + mi_printf_amount((int64_t)peak_rss, 1, out, arg, "%s"); + if (peak_commit > 0) { + _mi_fprintf(out, arg, ", commit: "); + mi_printf_amount((int64_t)peak_commit, 1, out, arg, "%s"); + } + _mi_fprintf(out, arg, "\n"); +} + +static mi_msecs_t mi_process_start; // = 0 + +static mi_stats_t* mi_stats_get_default(void) { + mi_heap_t* heap = mi_heap_get_default(); + return &heap->tld->stats; +} + +static void mi_stats_merge_from(mi_stats_t* stats) { + if (stats != &_mi_stats_main) { + mi_stats_add(&_mi_stats_main, stats); + memset(stats, 0, sizeof(mi_stats_t)); + } +} + +void mi_stats_reset(void) mi_attr_noexcept { + mi_stats_t* stats = mi_stats_get_default(); + if (stats != &_mi_stats_main) { memset(stats, 0, sizeof(mi_stats_t)); } + memset(&_mi_stats_main, 0, sizeof(mi_stats_t)); + if (mi_process_start == 0) { mi_process_start = _mi_clock_start(); }; +} + +void mi_stats_merge(void) mi_attr_noexcept { + mi_stats_merge_from( mi_stats_get_default() ); +} + +void _mi_stats_merge_thread(mi_tld_t* tld) { + mi_stats_merge_from( &tld->stats ); +} + +void _mi_stats_done(mi_stats_t* stats) { // called from `mi_thread_done` + mi_stats_merge_from(stats); +} + +void mi_stats_print_out(mi_output_fun* out, void* arg) mi_attr_noexcept { + mi_stats_merge_from(mi_stats_get_default()); + _mi_stats_print(&_mi_stats_main, out, arg); +} + +void mi_stats_print(void* out) mi_attr_noexcept { + // for compatibility there is an `out` parameter (which can be `stdout` or `stderr`) + mi_stats_print_out((mi_output_fun*)out, NULL); +} + +void mi_thread_stats_print_out(mi_output_fun* out, void* arg) mi_attr_noexcept { + _mi_stats_print(mi_stats_get_default(), out, arg); +} + + +// ---------------------------------------------------------------- +// Basic timer for convenience; use milli-seconds to avoid doubles +// ---------------------------------------------------------------- + +static mi_msecs_t mi_clock_diff; + +mi_msecs_t _mi_clock_now(void) { + return _mi_prim_clock_now(); +} + +mi_msecs_t _mi_clock_start(void) { + if (mi_clock_diff == 0.0) { + mi_msecs_t t0 = _mi_clock_now(); + mi_clock_diff = _mi_clock_now() - t0; + } + return _mi_clock_now(); +} + +mi_msecs_t _mi_clock_end(mi_msecs_t start) { + mi_msecs_t end = _mi_clock_now(); + return (end - start - mi_clock_diff); +} + + +// -------------------------------------------------------- +// Basic process statistics +// -------------------------------------------------------- + +mi_decl_export void mi_process_info(size_t* elapsed_msecs, size_t* user_msecs, size_t* system_msecs, size_t* current_rss, size_t* peak_rss, size_t* current_commit, size_t* peak_commit, size_t* page_faults) mi_attr_noexcept +{ + mi_process_info_t pinfo; + _mi_memzero_var(pinfo); + pinfo.elapsed = _mi_clock_end(mi_process_start); + pinfo.current_commit = (size_t)(mi_atomic_loadi64_relaxed((_Atomic(int64_t)*)&_mi_stats_main.committed.current)); + pinfo.peak_commit = (size_t)(mi_atomic_loadi64_relaxed((_Atomic(int64_t)*)&_mi_stats_main.committed.peak)); + pinfo.current_rss = pinfo.current_commit; + pinfo.peak_rss = pinfo.peak_commit; + pinfo.utime = 0; + pinfo.stime = 0; + pinfo.page_faults = 0; + + _mi_prim_process_info(&pinfo); + + if (elapsed_msecs!=NULL) *elapsed_msecs = (pinfo.elapsed < 0 ? 0 : (pinfo.elapsed < (mi_msecs_t)PTRDIFF_MAX ? (size_t)pinfo.elapsed : PTRDIFF_MAX)); + if (user_msecs!=NULL) *user_msecs = (pinfo.utime < 0 ? 0 : (pinfo.utime < (mi_msecs_t)PTRDIFF_MAX ? (size_t)pinfo.utime : PTRDIFF_MAX)); + if (system_msecs!=NULL) *system_msecs = (pinfo.stime < 0 ? 0 : (pinfo.stime < (mi_msecs_t)PTRDIFF_MAX ? (size_t)pinfo.stime : PTRDIFF_MAX)); + if (current_rss!=NULL) *current_rss = pinfo.current_rss; + if (peak_rss!=NULL) *peak_rss = pinfo.peak_rss; + if (current_commit!=NULL) *current_commit = pinfo.current_commit; + if (peak_commit!=NULL) *peak_commit = pinfo.peak_commit; + if (page_faults!=NULL) *page_faults = pinfo.page_faults; +} + + +// -------------------------------------------------------- +// Return statistics +// -------------------------------------------------------- + +void mi_stats_get(size_t stats_size, mi_stats_t* stats) mi_attr_noexcept { + if (stats == NULL || stats_size == 0) return; + _mi_memzero(stats, stats_size); + const size_t size = (stats_size > sizeof(mi_stats_t) ? sizeof(mi_stats_t) : stats_size); + _mi_memcpy(stats, &_mi_stats_main, size); + stats->version = MI_STAT_VERSION; +} + + +// -------------------------------------------------------- +// Statics in json format +// -------------------------------------------------------- + +typedef struct mi_heap_buf_s { + char* buf; + size_t size; + size_t used; + bool can_realloc; +} mi_heap_buf_t; + +static bool mi_heap_buf_expand(mi_heap_buf_t* hbuf) { + if (hbuf==NULL) return false; + if (hbuf->buf != NULL && hbuf->size>0) { + hbuf->buf[hbuf->size-1] = 0; + } + if (hbuf->size > SIZE_MAX/2 || !hbuf->can_realloc) return false; + const size_t newsize = (hbuf->size == 0 ? mi_good_size(12*MI_KiB) : 2*hbuf->size); + char* const newbuf = (char*)mi_rezalloc(hbuf->buf, newsize); + if (newbuf == NULL) return false; + hbuf->buf = newbuf; + hbuf->size = newsize; + return true; +} + +static void mi_heap_buf_print(mi_heap_buf_t* hbuf, const char* msg) { + if (msg==NULL || hbuf==NULL) return; + if (hbuf->used + 1 >= hbuf->size && !hbuf->can_realloc) return; + for (const char* src = msg; *src != 0; src++) { + char c = *src; + if (hbuf->used + 1 >= hbuf->size) { + if (!mi_heap_buf_expand(hbuf)) return; + } + mi_assert_internal(hbuf->used < hbuf->size); + hbuf->buf[hbuf->used++] = c; + } + mi_assert_internal(hbuf->used < hbuf->size); + hbuf->buf[hbuf->used] = 0; +} + +static void mi_heap_buf_print_count_bin(mi_heap_buf_t* hbuf, const char* prefix, mi_stat_count_t* stat, size_t bin, bool add_comma) { + const size_t binsize = _mi_bin_size(bin); + const size_t pagesize = (binsize <= MI_SMALL_OBJ_SIZE_MAX ? MI_SMALL_PAGE_SIZE : + (binsize <= MI_MEDIUM_OBJ_SIZE_MAX ? MI_MEDIUM_PAGE_SIZE : + #if MI_LARGE_PAGE_SIZE + (binsize <= MI_LARGE_OBJ_SIZE_MAX ? MI_LARGE_PAGE_SIZE : 0) + #else + 0 + #endif + )); + char buf[128]; + _mi_snprintf(buf, 128, "%s{ \"total\": %lld, \"peak\": %lld, \"current\": %lld, \"block_size\": %zu, \"page_size\": %zu }%s\n", prefix, stat->total, stat->peak, stat->current, binsize, pagesize, (add_comma ? "," : "")); + buf[127] = 0; + mi_heap_buf_print(hbuf, buf); +} + +static void mi_heap_buf_print_count(mi_heap_buf_t* hbuf, const char* prefix, mi_stat_count_t* stat, bool add_comma) { + char buf[128]; + _mi_snprintf(buf, 128, "%s{ \"total\": %lld, \"peak\": %lld, \"current\": %lld }%s\n", prefix, stat->total, stat->peak, stat->current, (add_comma ? "," : "")); + buf[127] = 0; + mi_heap_buf_print(hbuf, buf); +} + +static void mi_heap_buf_print_count_value(mi_heap_buf_t* hbuf, const char* name, mi_stat_count_t* stat) { + char buf[128]; + _mi_snprintf(buf, 128, " \"%s\": ", name); + buf[127] = 0; + mi_heap_buf_print(hbuf, buf); + mi_heap_buf_print_count(hbuf, "", stat, true); +} + +static void mi_heap_buf_print_value(mi_heap_buf_t* hbuf, const char* name, int64_t val) { + char buf[128]; + _mi_snprintf(buf, 128, " \"%s\": %lld,\n", name, val); + buf[127] = 0; + mi_heap_buf_print(hbuf, buf); +} + +static void mi_heap_buf_print_size(mi_heap_buf_t* hbuf, const char* name, size_t val, bool add_comma) { + char buf[128]; + _mi_snprintf(buf, 128, " \"%s\": %zu%s\n", name, val, (add_comma ? "," : "")); + buf[127] = 0; + mi_heap_buf_print(hbuf, buf); +} + +static void mi_heap_buf_print_counter_value(mi_heap_buf_t* hbuf, const char* name, mi_stat_counter_t* stat) { + mi_heap_buf_print_value(hbuf, name, stat->total); +} + +#define MI_STAT_COUNT(stat) mi_heap_buf_print_count_value(&hbuf, #stat, &stats->stat); +#define MI_STAT_COUNTER(stat) mi_heap_buf_print_counter_value(&hbuf, #stat, &stats->stat); + +char* mi_stats_get_json(size_t output_size, char* output_buf) mi_attr_noexcept { + mi_heap_buf_t hbuf = { NULL, 0, 0, true }; + if (output_size > 0 && output_buf != NULL) { + _mi_memzero(output_buf, output_size); + hbuf.buf = output_buf; + hbuf.size = output_size; + hbuf.can_realloc = false; + } + else { + if (!mi_heap_buf_expand(&hbuf)) return NULL; + } + mi_heap_buf_print(&hbuf, "{\n"); + mi_heap_buf_print_value(&hbuf, "version", MI_STAT_VERSION); + mi_heap_buf_print_value(&hbuf, "mimalloc_version", MI_MALLOC_VERSION); + + // process info + mi_heap_buf_print(&hbuf, " \"process\": {\n"); + size_t elapsed; + size_t user_time; + size_t sys_time; + size_t current_rss; + size_t peak_rss; + size_t current_commit; + size_t peak_commit; + size_t page_faults; + mi_process_info(&elapsed, &user_time, &sys_time, ¤t_rss, &peak_rss, ¤t_commit, &peak_commit, &page_faults); + mi_heap_buf_print_size(&hbuf, "elapsed_msecs", elapsed, true); + mi_heap_buf_print_size(&hbuf, "user_msecs", user_time, true); + mi_heap_buf_print_size(&hbuf, "system_msecs", sys_time, true); + mi_heap_buf_print_size(&hbuf, "page_faults", page_faults, true); + mi_heap_buf_print_size(&hbuf, "rss_current", current_rss, true); + mi_heap_buf_print_size(&hbuf, "rss_peak", peak_rss, true); + mi_heap_buf_print_size(&hbuf, "commit_current", current_commit, true); + mi_heap_buf_print_size(&hbuf, "commit_peak", peak_commit, false); + mi_heap_buf_print(&hbuf, " },\n"); + + // statistics + mi_stats_t* stats = &_mi_stats_main; + MI_STAT_FIELDS() + + // size bins + mi_heap_buf_print(&hbuf, " \"malloc_bins\": [\n"); + for (size_t i = 0; i <= MI_BIN_HUGE; i++) { + mi_heap_buf_print_count_bin(&hbuf, " ", &stats->malloc_bins[i], i, i!=MI_BIN_HUGE); + } + mi_heap_buf_print(&hbuf, " ],\n"); + mi_heap_buf_print(&hbuf, " \"page_bins\": [\n"); + for (size_t i = 0; i <= MI_BIN_HUGE; i++) { + mi_heap_buf_print_count_bin(&hbuf, " ", &stats->page_bins[i], i, i!=MI_BIN_HUGE); + } + mi_heap_buf_print(&hbuf, " ]\n"); + mi_heap_buf_print(&hbuf, "}\n"); + return hbuf.buf; +} From eec1c6954bcfcb75e79a177f7fe4ed09c11af746 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 24 Jun 2019 23:41:27 +0200 Subject: [PATCH 042/248] mimalloc: adjust for building inside Git MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We want to compile mimalloc's source code as part of Git, rather than requiring the code to be built as an external library: mimalloc uses a CMake-based build, which is not necessarily easy to integrate into the flavors of Git for Windows (which will be the main benefitting port). Signed-off-by: Johannes Schindelin Signed-off-by: Matthias Aßhauer --- compat/mimalloc/alloc.c | 1 - compat/mimalloc/mimalloc.h | 3 ++- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/compat/mimalloc/alloc.c b/compat/mimalloc/alloc.c index 0fed5e754c1918..f975a92b6b6d1a 100644 --- a/compat/mimalloc/alloc.c +++ b/compat/mimalloc/alloc.c @@ -17,7 +17,6 @@ terms of the MIT license. A copy of the license can be found in the file #include // malloc, abort #define MI_IN_ALLOC_C -#include "alloc-override.c" #include "free.c" #undef MI_IN_ALLOC_C diff --git a/compat/mimalloc/mimalloc.h b/compat/mimalloc/mimalloc.h index f887278a98879d..19829516d42c1f 100644 --- a/compat/mimalloc/mimalloc.h +++ b/compat/mimalloc/mimalloc.h @@ -95,7 +95,8 @@ terms of the MIT license. A copy of the license can be found in the file // Includes // ------------------------------------------------------ -#include // size_t +#include "compat/posix.h" + #include // bool #include // INTPTR_MAX From 86e86c841b9eb0a48d70272316823bca10118af7 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 24 Jun 2019 23:43:06 +0200 Subject: [PATCH 043/248] mimalloc: offer a build-time option to enable it By defining `USE_MIMALLOC`, Git can now be compiled with that nicely-fast and small allocator. Note that we have to disable a couple `DEVELOPER` options to build mimalloc's source code, as it makes heavy use of declarations after statements, among other things that disagree with Git's conventions. We even have to silence some GCC warnings in non-DEVELOPER mode. For example, the `-Wno-array-bounds` flag is needed because in `-O2` builds, trying to call `NtCurrentTeb()` (which `_mi_thread_id()` does on Windows) causes the bogus warning about a system header, likely related to https://sourceforge.net/p/mingw-w64/mailman/message/37674519/ and to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99578: C:/git-sdk-64-minimal/mingw64/include/psdk_inc/intrin-impl.h:838:1: error: array subscript 0 is outside array bounds of 'long long unsigned int[0]' [-Werror=array-bounds] 838 | __buildreadseg(__readgsqword, unsigned __int64, "gs", "q") | ^~~~~~~~~~~~~~ Also: The `mimalloc` library uses C11-style atomics, therefore we must require that standard when compiling with GCC if we want to use `mimalloc` (instead of requiring "only" C99). This is what we do in the CMake definition already, therefore this commit does not need to touch `contrib/buildsystems/`. Signed-off-by: Johannes Schindelin --- Makefile | 41 +++++++++++++++++++++++++++++++++++++++++ compat/.gitattributes | 1 + compat/posix.h | 10 ++++++++++ config.mak.dev | 2 ++ config.mak.uname | 2 +- 5 files changed, 55 insertions(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 1f444eec3c1d12..cd9ac4108a4503 100644 --- a/Makefile +++ b/Makefile @@ -1346,6 +1346,7 @@ BUILTIN_OBJS += builtin/write-tree.o # upstream unnecessarily (making merging in future changes easier). THIRD_PARTY_SOURCES += compat/inet_ntop.c THIRD_PARTY_SOURCES += compat/inet_pton.c +THIRD_PARTY_SOURCES += compat/mimalloc/% THIRD_PARTY_SOURCES += compat/nedmalloc/% THIRD_PARTY_SOURCES += compat/obstack.% THIRD_PARTY_SOURCES += compat/poll/% @@ -2145,6 +2146,46 @@ ifdef USE_NED_ALLOCATOR OVERRIDE_STRDUP = YesPlease endif +ifdef USE_MIMALLOC + MIMALLOC_OBJS = \ + compat/mimalloc/alloc-aligned.o \ + compat/mimalloc/alloc.o \ + compat/mimalloc/arena.o \ + compat/mimalloc/bitmap.o \ + compat/mimalloc/heap.o \ + compat/mimalloc/init.o \ + compat/mimalloc/libc.o \ + compat/mimalloc/options.o \ + compat/mimalloc/os.o \ + compat/mimalloc/page.o \ + compat/mimalloc/random.o \ + compat/mimalloc/prim/prim.o \ + compat/mimalloc/segment.o \ + compat/mimalloc/segment-map.o \ + compat/mimalloc/stats.o + + COMPAT_CFLAGS += -Icompat/mimalloc -DMI_DEBUG=0 -DUSE_MIMALLOC --std=gnu11 + COMPAT_OBJS += $(MIMALLOC_OBJS) + +$(MIMALLOC_OBJS): COMPAT_CFLAGS += -DBANNED_H + +$(MIMALLOC_OBJS): COMPAT_CFLAGS += \ + -DMI_WIN_USE_FLS \ + -Wno-attributes \ + -Wno-unknown-pragmas \ + -Wno-unused-function \ + -Wno-array-bounds + +ifdef DEVELOPER +$(MIMALLOC_OBJS): COMPAT_CFLAGS += \ + -Wno-pedantic \ + -Wno-declaration-after-statement \ + -Wno-old-style-definition \ + -Wno-missing-prototypes \ + -Wno-implicit-function-declaration +endif +endif + ifdef OVERRIDE_STRDUP COMPAT_CFLAGS += -DOVERRIDE_STRDUP COMPAT_OBJS += compat/strdup.o diff --git a/compat/.gitattributes b/compat/.gitattributes index 40dbfb170dabc5..2b5a66a3b34bda 100644 --- a/compat/.gitattributes +++ b/compat/.gitattributes @@ -1 +1,2 @@ /zlib-uncompress2.c whitespace=-indent-with-non-tab,-trailing-space +/mimalloc/**/* whitespace=-trailing-space diff --git a/compat/posix.h b/compat/posix.h index 626d71256f8428..106b16c49eaf62 100644 --- a/compat/posix.h +++ b/compat/posix.h @@ -176,6 +176,16 @@ typedef unsigned long uintptr_t; #define _ALL_SOURCE 1 #endif +#ifdef USE_MIMALLOC +#include "mimalloc.h" +#define malloc mi_malloc +#define calloc mi_calloc +#define realloc mi_realloc +#define free mi_free +#define strdup mi_strdup +#define strndup mi_strndup +#endif + #ifdef MKDIR_WO_TRAILING_SLASH #define mkdir(a,b) compat_mkdir_wo_trailing_slash((a),(b)) int compat_mkdir_wo_trailing_slash(const char*, mode_t); diff --git a/config.mak.dev b/config.mak.dev index e86b6e1b34a2d7..b63797ef509333 100644 --- a/config.mak.dev +++ b/config.mak.dev @@ -22,8 +22,10 @@ endif ifneq ($(uname_S),FreeBSD) ifneq ($(or $(filter gcc6,$(COMPILER_FEATURES)),$(filter clang7,$(COMPILER_FEATURES))),) +ifndef USE_MIMALLOC DEVELOPER_CFLAGS += -std=gnu99 endif +endif else # FreeBSD cannot limit to C99 because its system headers unconditionally # rely on C11 features. diff --git a/config.mak.uname b/config.mak.uname index 28588bdbc18e97..63ab63e461fca8 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -521,7 +521,7 @@ endif CC = compat/vcbuild/scripts/clink.pl AR = compat/vcbuild/scripts/lib.pl CFLAGS = - BASIC_CFLAGS = -nologo -I. -Icompat/vcbuild/include -DWIN32 -D_CONSOLE -DHAVE_STRING_H -D_CRT_SECURE_NO_WARNINGS -D_CRT_NONSTDC_NO_DEPRECATE + BASIC_CFLAGS = -nologo -I. -Icompat/vcbuild/include -DWIN32 -D_CONSOLE -DHAVE_STRING_H -D_CRT_SECURE_NO_WARNINGS -D_CRT_NONSTDC_NO_DEPRECATE -MP -std:c11 COMPAT_OBJS = compat/msvc.o compat/winansi.o \ compat/win32/flush.o \ compat/win32/path-utils.o \ From 3bd48df15be954cbe989dc588c33e87910d86e5c Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 16 Feb 2015 14:06:59 +0100 Subject: [PATCH 044/248] mingw: include the Python parts in the build While Git for Windows does not _ship_ Python (in order to save on bandwidth), MSYS2 provides very fine Python interpreters that users can easily take advantage of, by using Git for Windows within its SDK. Signed-off-by: Johannes Schindelin --- config.mak.uname | 1 + 1 file changed, 1 insertion(+) diff --git a/config.mak.uname b/config.mak.uname index 1691c6ae6e01e3..28588bdbc18e97 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -772,6 +772,7 @@ ifeq ($(uname_S),MINGW) ifneq (CLANGARM64,$(MSYSTEM)) USE_NED_ALLOCATOR = YesPlease endif + NO_PYTHON = ifeq (/mingw64,$(subst 32,64,$(subst clangarm,mingw,$(prefix)))) # Move system config into top-level /etc/ ETC_GITCONFIG = ../etc/gitconfig From e46c0845f038f066b6c52b148b22f35d1581ff93 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 24 Jun 2019 23:45:21 +0200 Subject: [PATCH 045/248] mingw: use mimalloc Thorough benchmarking with repacking a subset of linux.git (the commit history reachable from 93a6fefe2f ([PATCH] fix the SYSCTL=n compilation, 2007-02-28), to be precise) suggest that this allocator is on par, in multi-threaded situations maybe even better than nedmalloc: `git repack -adfq` with mimalloc, 8 threads: 31.166991900 27.576763800 28.712311000 27.373859000 27.163141900 `git repack -adfq` with nedmalloc, 8 threads: 31.915032900 27.149883100 28.244933700 27.240188800 28.580849500 In a different test using GitHub Actions build agents (probably single-threaded, a core-strength of nedmalloc)): `git repack -q -d -l -A --unpack-unreachable=2.weeks.ago` with mimalloc: 943.426 978.500 939.709 959.811 954.605 `git repack -q -d -l -A --unpack-unreachable=2.weeks.ago` with nedmalloc: 995.383 952.179 943.253 963.043 980.468 While these measurements were not executed with complete scientific rigor, as no hardware was set aside specifically for these benchmarks, it shows that mimalloc and nedmalloc perform almost the same, nedmalloc with a bit higher variance and also slightly higher average (further testing suggests that nedmalloc performs worse in multi-threaded situations than in single-threaded ones). In short: mimalloc seems to be slightly better suited for our purposes than nedmalloc. Seeing that mimalloc is developed actively, while nedmalloc ceased to see any updates in eight years, let's use mimalloc on Windows instead. Signed-off-by: Johannes Schindelin --- config.mak.uname | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/config.mak.uname b/config.mak.uname index 63ab63e461fca8..5cd2310b8ef17e 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -769,9 +769,7 @@ ifeq ($(uname_S),MINGW) HAVE_LIBCHARSET_H = YesPlease USE_GETTEXT_SCHEME = fallthrough USE_LIBPCRE = YesPlease - ifneq (CLANGARM64,$(MSYSTEM)) - USE_NED_ALLOCATOR = YesPlease - endif + USE_MIMALLOC = YesPlease NO_PYTHON = ifeq (/mingw64,$(subst 32,64,$(subst clangarm,mingw,$(prefix)))) # Move system config into top-level /etc/ From 642dd7003f03f372e2c444913770d6b81b0b5cfc Mon Sep 17 00:00:00 2001 From: Thomas Braun Date: Thu, 8 May 2014 21:43:24 +0200 Subject: [PATCH 046/248] transport: optionally disable side-band-64k Since commit 0c499ea60fda (send-pack: demultiplex a sideband stream with status data, 2010-02-05) the send-pack builtin uses the side-band-64k capability if advertised by the server. Unfortunately this breaks pushing over the dump git protocol if used over a network connection. The detailed reasons for this breakage are (by courtesy of Jeff Preshing, quoted from https://groups.google.com/d/msg/msysgit/at8D7J-h7mw/eaLujILGUWoJ): MinGW wraps Windows sockets in CRT file descriptors in order to mimic the functionality of POSIX sockets. This causes msvcrt.dll to treat sockets as Installable File System (IFS) handles, calling ReadFile, WriteFile, DuplicateHandle and CloseHandle on them. This approach works well in simple cases on recent versions of Windows, but does not support all usage patterns. In particular, using this approach, any attempt to read & write concurrently on the same socket (from one or more processes) will deadlock in a scenario where the read waits for a response from the server which is only invoked after the write. This is what send_pack currently attempts to do in the use_sideband codepath. The new config option `sendpack.sideband` allows to override the side-band-64k capability of the server, and thus makes the dumb git protocol work. Other transportation methods like ssh and http/https still benefit from the sideband channel, therefore the default value of `sendpack.sideband` is still true. Signed-off-by: Thomas Braun Signed-off-by: Oliver Schneider Signed-off-by: Johannes Schindelin --- Documentation/config.adoc | 2 ++ Documentation/config/sendpack.adoc | 5 +++++ send-pack.c | 6 +++--- 3 files changed, 10 insertions(+), 3 deletions(-) create mode 100644 Documentation/config/sendpack.adoc diff --git a/Documentation/config.adoc b/Documentation/config.adoc index ae6dcb6a6c0607..8a1d497de2bcf9 100644 --- a/Documentation/config.adoc +++ b/Documentation/config.adoc @@ -517,6 +517,8 @@ include::config/safe.adoc[] include::config/sendemail.adoc[] +include::config/sendpack.adoc[] + include::config/sequencer.adoc[] include::config/showbranch.adoc[] diff --git a/Documentation/config/sendpack.adoc b/Documentation/config/sendpack.adoc new file mode 100644 index 00000000000000..e306f657fba7dd --- /dev/null +++ b/Documentation/config/sendpack.adoc @@ -0,0 +1,5 @@ +sendpack.sideband:: + Allows to disable the side-band-64k capability for send-pack even + when it is advertised by the server. Makes it possible to work + around a limitation in the git for windows implementation together + with the dump git protocol. Defaults to true. diff --git a/send-pack.c b/send-pack.c index 67d6987b1ccd7e..22a1beed8d9823 100644 --- a/send-pack.c +++ b/send-pack.c @@ -501,7 +501,7 @@ int send_pack(struct repository *r, int need_pack_data = 0; int allow_deleting_refs = 0; int status_report = 0; - int use_sideband = 0; + int use_sideband = 1; int quiet_supported = 0; int agent_supported = 0; int advertise_sid = 0; @@ -525,6 +525,7 @@ int send_pack(struct repository *r, goto out; } + repo_config_get_bool(r, "sendpack.sideband", &use_sideband); repo_config_get_bool(r, "push.negotiate", &push_negotiate); if (push_negotiate) { trace2_region_enter("send_pack", "push_negotiate", r); @@ -546,8 +547,7 @@ int send_pack(struct repository *r, allow_deleting_refs = 1; if (server_supports("ofs-delta")) args->use_ofs_delta = 1; - if (server_supports("side-band-64k")) - use_sideband = 1; + use_sideband = use_sideband && server_supports("side-band-64k"); if (server_supports("quiet")) quiet_supported = 1; if (server_supports("agent")) From dde4f9ea1c45164bb16f411b5399355f73fa2120 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 31 Jan 2020 12:02:47 +0100 Subject: [PATCH 047/248] mingw: demonstrate a `git add` issue with NTFS junctions NTFS junctions are somewhat similar in spirit to Unix bind mounts: they point to a different directory and are resolved by the filesystem driver. As such, they appear to `lstat()` as if they are directories, not as if they are symbolic links. _Any_ user can create junctions, while symbolic links can only be created by non-administrators in Developer Mode on Windows 10. Hence NTFS junctions are much more common "in the wild" than NTFS symbolic links. It was reported in https://github.com/git-for-windows/git/issues/2481 that adding files via an absolute path that traverses an NTFS junction: since 1e64d18 (mingw: do resolve symlinks in `getcwd()`), we resolve not only symbolic links but also NTFS junctions when determining the absolute path of the current directory. The same is not true for `git add `, where symbolic links are resolved in ``, but not NTFS junctions. Signed-off-by: Johannes Schindelin --- t/t3700-add.sh | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/t/t3700-add.sh b/t/t3700-add.sh index df580a5806b4f1..d934f3417faa45 100755 --- a/t/t3700-add.sh +++ b/t/t3700-add.sh @@ -548,4 +548,15 @@ test_expect_success CASE_INSENSITIVE_FS 'path is case-insensitive' ' git add "$downcased" ' +test_expect_failure MINGW 'can add files via NTFS junctions' ' + test_when_finished "cmd //c rmdir junction && rm -rf target" && + test_create_repo target && + cmd //c "mklink /j junction target" && + >target/via-junction && + git -C junction add "$(pwd)/junction/via-junction" && + echo via-junction >expect && + git -C target diff --cached --name-only >actual && + test_cmp expect actual +' + test_done From ca0956cd8c4eb51e54b1c64d5b972f5d9ce9b456 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 21 Feb 2017 13:28:58 +0100 Subject: [PATCH 048/248] mingw: ensure valid CTYPE A change between versions 2.4.1 and 2.6.0 of the MSYS2 runtime modified how Cygwin's runtime (and hence Git for Windows' MSYS2 runtime derivative) handles locales: d16a56306d (Consolidate wctomb/mbtowc calls for POSIX-1.2008, 2016-07-20). An unintended side-effect is that "cold-calling" into the POSIX emulation will start with a locale based on the current code page, something that Git for Windows is very ill-prepared for, as it expects to be able to pass a command-line containing non-ASCII characters to the shell without having those characters munged. One symptom of this behavior: when `git clone` or `git fetch` shell out to call `git-upload-pack` with a path that contains non-ASCII characters, the shell tried to interpret the entire command-line (including command-line parameters) as executable path, which obviously must fail. This fixes https://github.com/git-for-windows/git/issues/1036 Signed-off-by: Johannes Schindelin --- compat/mingw.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..13753612a0d158 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2802,6 +2802,9 @@ static void setup_windows_environment(void) if (!tmp && (tmp = getenv("USERPROFILE"))) setenv("HOME", tmp, 1); } + + if (!getenv("LC_ALL") && !getenv("LC_CTYPE") && !getenv("LANG")) + setenv("LC_CTYPE", "C.UTF-8", 1); } static void get_current_user_sid(PSID *sid, HANDLE *linked_token) From 9b889648bc04b2989a2532d47c84de98b8651573 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 31 Jan 2020 11:44:31 +0100 Subject: [PATCH 049/248] strbuf_realpath(): use platform-dependent API if available Some platforms (e.g. Windows) provide API functions to resolve paths much quicker. Let's offer a way to short-cut `strbuf_realpath()` on those platforms. Signed-off-by: Johannes Schindelin --- abspath.c | 3 +++ git-compat-util.h | 4 ++++ 2 files changed, 7 insertions(+) diff --git a/abspath.c b/abspath.c index 1202cde23dbc9b..0c17e98654e4b0 100644 --- a/abspath.c +++ b/abspath.c @@ -93,6 +93,9 @@ static char *strbuf_realpath_1(struct strbuf *resolved, const char *path, goto error_out; } + if (platform_strbuf_realpath(resolved, path)) + return resolved->buf; + strbuf_addstr(&remaining, path); get_root_part(resolved, &remaining); diff --git a/git-compat-util.h b/git-compat-util.h index 398e0fac4fab60..3ed623e733f522 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -387,6 +387,10 @@ static inline int git_has_dir_sep(const char *path) #define query_user_email() NULL #endif +#ifndef platform_strbuf_realpath +#define platform_strbuf_realpath(resolved, path) NULL +#endif + #ifdef __TANDEM #include #include From e4abf1fcf3587eded97278b19683b07c4e6941a4 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sat, 1 Feb 2020 00:31:16 +0100 Subject: [PATCH 050/248] mingw: allow `git.exe` to be used instead of the "Git wrapper" Git for Windows wants to add `git.exe` to the users' `PATH`, without cluttering the latter with unnecessary executables such as `wish.exe`. To that end, it invented the concept of its "Git wrapper", i.e. a tiny executable located in `C:\Program Files\Git\cmd\git.exe` (originally a CMD script) whose sole purpose is to set up a couple of environment variables and then spawn the _actual_ `git.exe` (which nowadays lives in `C:\Program Files\Git\mingw64\bin\git.exe` for 64-bit, and the obvious equivalent for 32-bit installations). Currently, the following environment variables are set unless already initialized: - `MSYSTEM`, to make sure that the MSYS2 Bash and the MSYS2 Perl interpreter behave as expected, and - `PLINK_PROTOCOL`, to force PuTTY's `plink.exe` to use the SSH protocol instead of Telnet, - `PATH`, to make sure that the `bin` folder in the user's home directory, as well as the `/mingw64/bin` and the `/usr/bin` directories are included. The trick here is that the `/mingw64/bin/` and `/usr/bin/` directories are relative to the top-level installation directory of Git for Windows (which the included Bash interprets as `/`, i.e. as the MSYS pseudo root directory). Using the absence of `MSYSTEM` as a tell-tale, we can detect in `git.exe` whether these environment variables have been initialized properly. Therefore we can call `C:\Program Files\Git\mingw64\bin\git` in-place after this change, without having to call Git through the Git wrapper. Obviously, above-mentioned directories must be _prepended_ to the `PATH` variable, otherwise we risk picking up executables from unrelated Git installations. We do that by constructing the new `PATH` value from scratch, appending `$HOME/bin` (if `HOME` is set), then the MSYS2 system directories, and then appending the original `PATH`. Side note: this modification of the `PATH` variable is independent of the modification necessary to reach the executables and scripts in `/mingw64/libexec/git-core/`, i.e. the `GIT_EXEC_PATH`. That modification is still performed by Git, elsewhere, long after making the changes described above. While we _still_ cannot simply hard-link `mingw64\bin\git.exe` to `cmd` (because the former depends on a couple of `.dll` files that are only in `mingw64\bin`, i.e. calling `...\cmd\git.exe` would fail to load due to missing dependencies), at least we can now avoid that extra process of running the Git wrapper (which then has to wait for the spawned `git.exe` to finish) by calling `...\mingw64\bin\git.exe` directly, via its absolute path. Testing this is in Git's test suite tricky: we set up a "new" MSYS pseudo-root and copy the `git.exe` file into the appropriate location, then verify that `MSYSTEM` is set properly, and also that the `PATH` is modified so that scripts can be found in `$HOME/bin`, `/mingw64/bin/` and `/usr/bin/`. This addresses https://github.com/git-for-windows/git/issues/2283 Signed-off-by: Johannes Schindelin --- compat/mingw.c | 80 +++++++++++++++++++++++++++++++++++++++++++ config.mak.uname | 4 +-- t/t0060-path-utils.sh | 33 +++++++++++++++++- 3 files changed, 114 insertions(+), 3 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 13753612a0d158..1261191853c558 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2751,6 +2751,54 @@ int xwcstoutf(char *utf, const wchar_t *wcs, size_t utflen) return -1; } +#ifdef ENSURE_MSYSTEM_IS_SET +#if !defined(RUNTIME_PREFIX) || !defined(HAVE_WPGMPTR) +static size_t append_system_bin_dirs(char *path UNUSED, size_t size UNUSED) +{ + return 0; +} +#else +static size_t append_system_bin_dirs(char *path, size_t size) +{ + char prefix[32768]; + const char *slash; + size_t len = xwcstoutf(prefix, _wpgmptr, sizeof(prefix)), off = 0; + + if (len == 0 || len >= sizeof(prefix) || + !(slash = find_last_dir_sep(prefix))) + return 0; + /* strip trailing `git.exe` */ + len = slash - prefix; + + /* strip trailing `cmd` or `mingw64\bin` or `mingw32\bin` or `bin` or `libexec\git-core` */ + if (strip_suffix_mem(prefix, &len, "\\mingw64\\libexec\\git-core") || + strip_suffix_mem(prefix, &len, "\\mingw64\\bin")) + off += xsnprintf(path + off, size - off, + "%.*s\\mingw64\\bin;", (int)len, prefix); + else if (strip_suffix_mem(prefix, &len, "\\clangarm64\\libexec\\git-core") || + strip_suffix_mem(prefix, &len, "\\clangarm64\\bin")) + off += xsnprintf(path + off, size - off, + "%.*s\\clangarm64\\bin;", (int)len, prefix); + else if (strip_suffix_mem(prefix, &len, "\\mingw32\\libexec\\git-core") || + strip_suffix_mem(prefix, &len, "\\mingw32\\bin")) + off += xsnprintf(path + off, size - off, + "%.*s\\mingw32\\bin;", (int)len, prefix); + else if (strip_suffix_mem(prefix, &len, "\\cmd") || + strip_suffix_mem(prefix, &len, "\\bin") || + strip_suffix_mem(prefix, &len, "\\libexec\\git-core")) + off += xsnprintf(path + off, size - off, + "%.*s\\mingw%d\\bin;", (int)len, prefix, + (int)(sizeof(void *) * 8)); + else + return 0; + + off += xsnprintf(path + off, size - off, + "%.*s\\usr\\bin;", (int)len, prefix); + return off; +} +#endif +#endif + static void setup_windows_environment(void) { char *tmp = getenv("TMPDIR"); @@ -2803,6 +2851,38 @@ static void setup_windows_environment(void) setenv("HOME", tmp, 1); } + if (!getenv("PLINK_PROTOCOL")) + setenv("PLINK_PROTOCOL", "ssh", 0); + +#ifdef ENSURE_MSYSTEM_IS_SET + if (!(tmp = getenv("MSYSTEM")) || !tmp[0]) { + const char *home = getenv("HOME"), *path = getenv("PATH"); + char buf[32768]; + size_t off = 0; + +#if defined(__aarch64__) || defined(_M_ARM64) || defined(_M_ARM64EC) + setenv("MSYSTEM", "CLANGARM64", 1); +#elif defined(__MINGW64__) || defined(_M_AMD64) + setenv("MSYSTEM", "MINGW64", 1); +#else + setenv("MSYSTEM", "MINGW32", 1); +#endif + + if (home) + off += xsnprintf(buf + off, sizeof(buf) - off, + "%s\\bin;", home); + off += append_system_bin_dirs(buf + off, sizeof(buf) - off); + if (path) + off += xsnprintf(buf + off, sizeof(buf) - off, + "%s", path); + else if (off > 0) + buf[off - 1] = '\0'; + else + buf[0] = '\0'; + setenv("PATH", buf, 1); + } +#endif + if (!getenv("LC_ALL") && !getenv("LC_CTYPE") && !getenv("LANG")) setenv("LC_CTYPE", "C.UTF-8", 1); } diff --git a/config.mak.uname b/config.mak.uname index 1691c6ae6e01e3..94730676c64dbd 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -528,7 +528,7 @@ endif compat/win32/pthread.o compat/win32/syslog.o \ compat/win32/trace2_win32_process_info.o \ compat/win32/dirent.o - COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\" + COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\" BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO -ENTRY:wmainCRTStartup -SUBSYSTEM:CONSOLE # invalidcontinue.obj allows Git's source code to close the same file # handle twice, or to access the osfhandle of an already-closed stdout @@ -761,7 +761,7 @@ ifeq ($(uname_S),MINGW) endif CC = gcc COMPAT_CFLAGS += -D__USE_MINGW_ANSI_STDIO=0 -DDETECT_MSYS_TTY \ - -fstack-protector-strong + -DENSURE_MSYSTEM_IS_SET -fstack-protector-strong EXTLIBS += -lntdll EXTRA_PROGRAMS += headless-git$X INSTALL = /bin/install diff --git a/t/t0060-path-utils.sh b/t/t0060-path-utils.sh index 8545cdfab559b4..56faf5fe732ee0 100755 --- a/t/t0060-path-utils.sh +++ b/t/t0060-path-utils.sh @@ -602,7 +602,8 @@ test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD 'RUNTIME_PREFIX wor echo "echo HERE" | write_script pretend/libexec/git-core/git-here && GIT_EXEC_PATH= ./pretend/bin/git here >actual && echo HERE >expect && - test_cmp expect actual' + test_cmp expect actual +' test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD '%(prefix)/ works' ' git config yes.path "%(prefix)/yes" && @@ -611,4 +612,34 @@ test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD '%(prefix)/ works' test_cmp expect actual ' +test_expect_success MINGW,RUNTIME_PREFIX 'MSYSTEM/PATH is adjusted if necessary' ' + if test -z "$MINGW_PREFIX" + then + MINGW_PREFIX="/$(echo "${MSYSTEM:-MINGW64}" | tr A-Z a-z)" + fi && + mkdir -p "$HOME"/bin pretend"$MINGW_PREFIX"/bin \ + pretend"$MINGW_PREFIX"/libexec/git-core pretend/usr/bin && + cp "$GIT_EXEC_PATH"/git.exe pretend"$MINGW_PREFIX"/bin/ && + cp "$GIT_EXEC_PATH"/git.exe pretend"$MINGW_PREFIX"/libexec/git-core/ && + # copy the .dll files, if any (happens when building via CMake) + if test -n "$(ls "$GIT_EXEC_PATH"/*.dll 2>/dev/null)" + then + cp "$GIT_EXEC_PATH"/*.dll pretend"$MINGW_PREFIX"/bin/ && + cp "$GIT_EXEC_PATH"/*.dll pretend"$MINGW_PREFIX"/libexec/git-core/ + fi && + echo "env | grep MSYSTEM=" | write_script "$HOME"/bin/git-test-home && + echo "echo ${MINGW_PREFIX#/}" | write_script pretend"$MINGW_PREFIX"/bin/git-test-bin && + echo "echo usr" | write_script pretend/usr/bin/git-test-bin2 && + + ( + MSYSTEM= && + GIT_EXEC_PATH= && + pretend"$MINGW_PREFIX"/libexec/git-core/git.exe test-home >actual && + pretend"$MINGW_PREFIX"/libexec/git-core/git.exe test-bin >>actual && + pretend"$MINGW_PREFIX"/bin/git.exe test-bin2 >>actual + ) && + test_write_lines MSYSTEM=$MSYSTEM "${MINGW_PREFIX#/}" usr >expect && + test_cmp expect actual +' + test_done From 677fb134123001821d04ec02d604300eaded4eb6 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 1 Jan 2020 21:07:22 +0100 Subject: [PATCH 051/248] mingw: do resolve symlinks in `getcwd()` As pointed out in https://github.com/git-for-windows/git/issues/1676, the `git rev-parse --is-inside-work-tree` command currently fails when the current directory's path contains symbolic links. The underlying reason for this bug is that `getcwd()` is supposed to resolve symbolic links, but our `mingw_getcwd()` implementation did not. We do have all the building blocks for that, though: the `GetFinalPathByHandleW()` function will resolve symbolic links. However, we only called that function if `GetLongPathNameW()` failed, for historical reasons: the latter function was supported for a long time, but the former API function was introduced only with Windows Vista, and we used to support also Windows XP. With that support having been dropped, we are free to call the symbolic link-resolving function right away. Signed-off-by: Johannes Schindelin --- compat/mingw.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..73368c7d028595 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1235,18 +1235,16 @@ char *mingw_getcwd(char *pointer, int len) { wchar_t cwd[MAX_PATH], wpointer[MAX_PATH]; DWORD ret = GetCurrentDirectoryW(ARRAY_SIZE(cwd), cwd); + HANDLE hnd; if (!ret || ret >= ARRAY_SIZE(cwd)) { errno = ret ? ENAMETOOLONG : err_win_to_posix(GetLastError()); return NULL; } - ret = GetLongPathNameW(cwd, wpointer, ARRAY_SIZE(wpointer)); - if (!ret && GetLastError() == ERROR_ACCESS_DENIED) { - HANDLE hnd = CreateFileW(cwd, 0, - FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, - OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); - if (hnd == INVALID_HANDLE_VALUE) - return NULL; + hnd = CreateFileW(cwd, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); + if (hnd != INVALID_HANDLE_VALUE) { ret = GetFinalPathNameByHandleW(hnd, wpointer, ARRAY_SIZE(wpointer), 0); CloseHandle(hnd); if (!ret || ret >= ARRAY_SIZE(wpointer)) @@ -1255,13 +1253,11 @@ char *mingw_getcwd(char *pointer, int len) return NULL; return pointer; } - if (!ret || ret >= ARRAY_SIZE(wpointer)) - return NULL; - if (GetFileAttributesW(wpointer) == INVALID_FILE_ATTRIBUTES) { + if (GetFileAttributesW(cwd) == INVALID_FILE_ATTRIBUTES) { errno = ENOENT; return NULL; } - if (xwcstoutf(pointer, wpointer, len) < 0) + if (xwcstoutf(pointer, cwd, len) < 0) return NULL; convert_slashes(pointer); return pointer; From ecdad990463184f4fd0f04f93f0e54dc92d6653e Mon Sep 17 00:00:00 2001 From: Bjoern Mueller Date: Wed, 22 Jan 2020 13:49:13 +0100 Subject: [PATCH 052/248] mingw: fix fatal error working on mapped network drives on Windows In 1e64d18 (mingw: do resolve symlinks in `getcwd()`) a problem was introduced that causes git for Windows to stop working with certain mapped network drives (in particular, drives that are mapped to locations with long path names). Error message was "fatal: Unable to read current working directory: No such file or directory". Present change fixes this issue as discussed in https://github.com/git-for-windows/git/issues/2480 Signed-off-by: Bjoern Mueller --- compat/mingw.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..d1995227d77353 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1249,8 +1249,13 @@ char *mingw_getcwd(char *pointer, int len) return NULL; ret = GetFinalPathNameByHandleW(hnd, wpointer, ARRAY_SIZE(wpointer), 0); CloseHandle(hnd); - if (!ret || ret >= ARRAY_SIZE(wpointer)) - return NULL; + if (!ret || ret >= ARRAY_SIZE(wpointer)) { + ret = GetLongPathNameW(cwd, wpointer, ARRAY_SIZE(wpointer)); + if (!ret || ret >= ARRAY_SIZE(wpointer)) { + errno = ret ? ENAMETOOLONG : err_win_to_posix(GetLastError()); + return NULL; + } + } if (xwcstoutf(pointer, normalize_ntpath(wpointer), len) < 0) return NULL; return pointer; From aaa5af55446b74f2d85ed27224f1d92038506075 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Thu, 30 Jan 2020 14:22:27 -0500 Subject: [PATCH 053/248] clink.pl: fix MSVC compile script to handle libcurl-d.lib Update clink.pl to link with either libcurl.lib or libcurl-d.lib depending on whether DEBUG=1 is set. Signed-off-by: Jeff Hostetler Signed-off-by: Johannes Schindelin --- compat/vcbuild/scripts/clink.pl | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/compat/vcbuild/scripts/clink.pl b/compat/vcbuild/scripts/clink.pl index 3bd824154be381..c4c99d1a11f18c 100755 --- a/compat/vcbuild/scripts/clink.pl +++ b/compat/vcbuild/scripts/clink.pl @@ -56,7 +56,8 @@ # need to use that instead? foreach my $flag (@lflags) { if ($flag =~ /^-LIBPATH:(.*)/) { - foreach my $l ("libcurl_imp.lib", "libcurl.lib") { + my $libcurl = $is_debug ? "libcurl-d.lib" : "libcurl.lib"; + foreach my $l ("libcurl_imp.lib", $libcurl) { if (-f "$1/$l") { $lib = $l; last; From 513adfecd67c90ff52c57d27e20b8f8aea02e37d Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 31 Jan 2020 11:49:04 +0100 Subject: [PATCH 054/248] mingw: implement a platform-specific `strbuf_realpath()` There is a Win32 API function to resolve symbolic links, and we can use that instead of resolving them manually. Even better, this function also resolves NTFS junction points (which are somewhat similar to bind mounts). This fixes https://github.com/git-for-windows/git/issues/2481. Signed-off-by: Johannes Schindelin --- compat/mingw.c | 76 +++++++++++++++++++++++++++++++++++++++++++ compat/mingw.h | 3 ++ t/t0060-path-utils.sh | 8 +++++ t/t3700-add.sh | 2 +- t/t5601-clone.sh | 7 ++++ 5 files changed, 95 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..513ec686e00a19 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1231,6 +1231,82 @@ struct tm *localtime_r(const time_t *timep, struct tm *result) } #endif +char *mingw_strbuf_realpath(struct strbuf *resolved, const char *path) +{ + wchar_t wpath[MAX_PATH]; + HANDLE h; + DWORD ret; + int len; + const char *last_component = NULL; + char *append = NULL; + + if (xutftowcs_path(wpath, path) < 0) + return NULL; + + h = CreateFileW(wpath, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); + + /* + * strbuf_realpath() allows the last path component to not exist. If + * that is the case, now it's time to try without last component. + */ + if (h == INVALID_HANDLE_VALUE && + GetLastError() == ERROR_FILE_NOT_FOUND) { + /* cut last component off of `wpath` */ + wchar_t *p = wpath + wcslen(wpath); + + while (p != wpath) + if (*(--p) == L'/' || *p == L'\\') + break; /* found start of last component */ + + if (p != wpath && (last_component = find_last_dir_sep(path))) { + append = xstrdup(last_component + 1); /* skip directory separator */ + /* + * Do not strip the trailing slash at the drive root, otherwise + * the path would be e.g. `C:` (which resolves to the + * _current_ directory on that drive). + */ + if (p[-1] == L':') + p[1] = L'\0'; + else + *p = L'\0'; + h = CreateFileW(wpath, 0, FILE_SHARE_READ | + FILE_SHARE_WRITE | FILE_SHARE_DELETE, + NULL, OPEN_EXISTING, + FILE_FLAG_BACKUP_SEMANTICS, NULL); + } + } + + if (h == INVALID_HANDLE_VALUE) { +realpath_failed: + FREE_AND_NULL(append); + return NULL; + } + + ret = GetFinalPathNameByHandleW(h, wpath, ARRAY_SIZE(wpath), 0); + CloseHandle(h); + if (!ret || ret >= ARRAY_SIZE(wpath)) + goto realpath_failed; + + len = wcslen(wpath) * 3; + strbuf_grow(resolved, len); + len = xwcstoutf(resolved->buf, normalize_ntpath(wpath), len); + if (len < 0) + goto realpath_failed; + resolved->len = len; + + if (append) { + /* Use forward-slash, like `normalize_ntpath()` */ + strbuf_complete(resolved, '/'); + strbuf_addstr(resolved, append); + FREE_AND_NULL(append); + } + + return resolved->buf; + +} + char *mingw_getcwd(char *pointer, int len) { wchar_t cwd[MAX_PATH], wpointer[MAX_PATH]; diff --git a/compat/mingw.h b/compat/mingw.h index 444daedfa52469..f6daf47ee4e0a7 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -39,6 +39,9 @@ static inline void convert_slashes(char *path) #define PATH_SEP ';' char *mingw_query_user_email(void); #define query_user_email mingw_query_user_email +struct strbuf; +char *mingw_strbuf_realpath(struct strbuf *resolved, const char *path); +#define platform_strbuf_realpath mingw_strbuf_realpath /** * Verifies that the specified path is owned by the user running the diff --git a/t/t0060-path-utils.sh b/t/t0060-path-utils.sh index 8545cdfab559b4..eb2ab9d437ea8e 100755 --- a/t/t0060-path-utils.sh +++ b/t/t0060-path-utils.sh @@ -281,6 +281,14 @@ test_expect_success SYMLINKS 'real path works on symlinks' ' test_cmp expect actual ' +test_expect_success MINGW 'real path works near drive root' ' + # we need a non-existing path at the drive root; simply skip if C:/xyz exists + if test ! -e C:/xyz + then + test C:/xyz = $(test-tool path-utils real_path C:/xyz) + fi +' + test_expect_success SYMLINKS 'prefix_path works with absolute paths to work tree symlinks' ' ln -s target symlink && echo "symlink" >expect && diff --git a/t/t3700-add.sh b/t/t3700-add.sh index d934f3417faa45..10185f31f019a9 100755 --- a/t/t3700-add.sh +++ b/t/t3700-add.sh @@ -548,7 +548,7 @@ test_expect_success CASE_INSENSITIVE_FS 'path is case-insensitive' ' git add "$downcased" ' -test_expect_failure MINGW 'can add files via NTFS junctions' ' +test_expect_success MINGW 'can add files via NTFS junctions' ' test_when_finished "cmd //c rmdir junction && rm -rf target" && test_create_repo target && cmd //c "mklink /j junction target" && diff --git a/t/t5601-clone.sh b/t/t5601-clone.sh index d743d986c401a0..f70d99016ea2f7 100755 --- a/t/t5601-clone.sh +++ b/t/t5601-clone.sh @@ -78,6 +78,13 @@ test_expect_success 'clone respects GIT_WORK_TREE' ' ' +test_expect_success CASE_INSENSITIVE_FS 'core.worktree is not added due to path case' ' + + mkdir UPPERCASE && + git clone src "$(pwd)/uppercase" && + test "unset" = "$(git -C UPPERCASE config --default unset core.worktree)" +' + test_expect_success 'clone from hooks' ' test_create_repo r0 && From b5f7d4087cc49bf5cfc6da5449c37d7f0c4f80c1 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sat, 9 May 2020 16:19:06 +0200 Subject: [PATCH 055/248] t5505/t5516: allow running without `.git/branches/` in the templates When we commit the template directory as part of `make vcxproj`, the `branches/` directory is not actually commited, as it is empty. Two tests were not prepared for that situation. This developer tried to get rid of the support for `.git/branches/` a long time ago, but that effort did not bear fruit, so the best we can do is work around in these here tests. Signed-off-by: Johannes Schindelin --- t/t5505-remote.sh | 4 ++-- t/t5516-fetch-push.sh | 8 ++++---- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/t/t5505-remote.sh b/t/t5505-remote.sh index 2701eef85e9ef8..969d9cc7d567a6 100755 --- a/t/t5505-remote.sh +++ b/t/t5505-remote.sh @@ -1155,7 +1155,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'migrate a remote from named file in ( cd six && git remote rm origin && - mkdir .git/branches && + mkdir -p .git/branches && echo "$origin_url#main" >.git/branches/origin && git remote rename origin origin && test_path_is_missing .git/branches/origin && @@ -1170,7 +1170,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'migrate a remote from named file in ( cd seven && git remote rm origin && - mkdir .git/branches && + mkdir -p .git/branches && echo "quux#foom" > .git/branches/origin && git remote rename origin origin && test_path_is_missing .git/branches/origin && diff --git a/t/t5516-fetch-push.sh b/t/t5516-fetch-push.sh index 46926e7bbd3a9a..c15963c3d0b229 100755 --- a/t/t5516-fetch-push.sh +++ b/t/t5516-fetch-push.sh @@ -933,7 +933,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches' ' mk_empty testrepo && git branch second $the_first_commit && git checkout second && - mkdir testrepo/.git/branches && + mkdir -p testrepo/.git/branches && echo ".." > testrepo/.git/branches/branch1 && ( cd testrepo && @@ -947,7 +947,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches' ' test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches containing #' ' mk_empty testrepo && - mkdir testrepo/.git/branches && + mkdir -p testrepo/.git/branches && echo "..#second" > testrepo/.git/branches/branch2 && ( cd testrepo && @@ -964,7 +964,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'push with branches' ' git checkout second && test_when_finished "rm -rf .git/branches" && - mkdir .git/branches && + mkdir -p .git/branches && echo "testrepo" > .git/branches/branch1 && git push branch1 && @@ -980,7 +980,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'push with branches containing #' ' mk_empty testrepo && test_when_finished "rm -rf .git/branches" && - mkdir .git/branches && + mkdir -p .git/branches && echo "testrepo#branch3" > .git/branches/branch2 && git push branch2 && From 7cac36bf7fc5227203eeedec6f407335cc5e306e Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 25 Aug 2020 12:13:26 +0200 Subject: [PATCH 056/248] mingw: ignore HOMEDRIVE/HOMEPATH if it points to Windows' system directory Internally, Git expects the environment variable `HOME` to be set, and to point to the current user's home directory. This environment variable is not set by default on Windows, and therefore Git tries its best to construct one if it finds `HOME` unset. There are actually two different approaches Git tries: first, it looks at `HOMEDRIVE`/`HOMEPATH` because this is widely used in corporate environments with roaming profiles, and a user generally wants their global Git settings to be in a roaming profile. Only when `HOMEDRIVE`/`HOMEPATH` is either unset or does not point to a valid location, Git will fall back to using `USERPROFILE` instead. However, starting with Windows Vista, for secondary logons and services, the environment variables `HOMEDRIVE`/`HOMEPATH` point to Windows' system directory (usually `C:\Windows\system32`). That is undesirable, and that location is usually write-protected anyway. So let's verify that the `HOMEDRIVE`/`HOMEPATH` combo does not point to Windows' system directory before using it, falling back to `USERPROFILE` if it does. This fixes git-for-windows#2709 Initial-Path-by: Ivan Pozdeev Signed-off-by: Johannes Schindelin --- compat/mingw.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 1261191853c558..d36810128b2cd0 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2799,6 +2799,18 @@ static size_t append_system_bin_dirs(char *path, size_t size) #endif #endif +static int is_system32_path(const char *path) +{ + WCHAR system32[MAX_PATH], wpath[MAX_PATH]; + + if (xutftowcs_path(wpath, path) < 0 || + !GetSystemDirectoryW(system32, ARRAY_SIZE(system32)) || + _wcsicmp(system32, wpath)) + return 0; + + return 1; +} + static void setup_windows_environment(void) { char *tmp = getenv("TMPDIR"); @@ -2839,7 +2851,8 @@ static void setup_windows_environment(void) strbuf_addstr(&buf, tmp); if ((tmp = getenv("HOMEPATH"))) { strbuf_addstr(&buf, tmp); - if (is_directory(buf.buf)) + if (!is_system32_path(buf.buf) && + is_directory(buf.buf)) setenv("HOME", buf.buf, 1); else tmp = NULL; /* use $USERPROFILE */ From ee852995d5497dcabffd1aefb654f23a37d5436a Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 4 Mar 2020 21:55:28 +0100 Subject: [PATCH 057/248] http: use new "best effort" strategy for Secure Channel revoke checking MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The native Windows HTTPS backend is based on Secure Channel which lets the caller decide how to handle revocation checking problems caused by missing information in the certificate or offline CRL distribution points. Unfortunately, cURL chose to handle these problems differently than OpenSSL by default: while OpenSSL happily ignores those problems (essentially saying "¯\_(ツ)_/¯"), the Secure Channel backend will error out instead. As a remedy, the "no revoke" mode was introduced, which turns off revocation checking altogether. This is a bit heavy-handed. We support this via the `http.schannelCheckRevoke` setting. In https://github.com/curl/curl/pull/4981, we contributed an opt-in "best effort" strategy that emulates what OpenSSL seems to do. In Git for Windows, we actually want this to be the default. This patch makes it so, introducing it as a new value for the `http.schannelCheckRevoke" setting, which now becmes a tristate: it accepts the values "false", "true" or "best-effort" (defaulting to the last one). Signed-off-by: Johannes Schindelin --- Documentation/config/http.adoc | 12 +++++++----- http.c | 26 ++++++++++++++++++++++---- 2 files changed, 29 insertions(+), 9 deletions(-) diff --git a/Documentation/config/http.adoc b/Documentation/config/http.adoc index 9da5c298cc1d5e..9122c5dc23ea1a 100644 --- a/Documentation/config/http.adoc +++ b/Documentation/config/http.adoc @@ -233,11 +233,13 @@ http.sslKeyType:: http.schannelCheckRevoke:: Used to enforce or disable certificate revocation checks in cURL - when http.sslBackend is set to "schannel". Defaults to `true` if - unset. Only necessary to disable this if Git consistently errors - and the message is about checking the revocation status of a - certificate. This option is ignored if cURL lacks support for - setting the relevant SSL option at runtime. + when http.sslBackend is set to "schannel" via "true" and "false", + respectively. Another accepted value is "best-effort" (the default) + in which case revocation checks are performed, but errors due to + revocation list distribution points that are offline are silently + ignored, as well as errors due to certificates missing revocation + list distribution points. This option is ignored if cURL lacks + support for setting the relevant SSL option at runtime. http.schannelUseSSLCAInfo:: As of cURL v7.60.0, the Secure Channel backend can use the diff --git a/http.c b/http.c index a7d55dcbbab769..3d00a9804c06ea 100644 --- a/http.c +++ b/http.c @@ -148,7 +148,13 @@ static char *cached_accept_language; static char *http_ssl_backend; -static int http_schannel_check_revoke = 1; +static long http_schannel_check_revoke_mode = +#ifdef CURLSSLOPT_REVOKE_BEST_EFFORT + CURLSSLOPT_REVOKE_BEST_EFFORT; +#else + CURLSSLOPT_NO_REVOKE; +#endif + /* * With the backend being set to `schannel`, setting sslCAinfo would override * the Certificate Store in cURL v7.60.0 and later, which is not what we want @@ -423,7 +429,19 @@ static int http_options(const char *var, const char *value, } if (!strcmp("http.schannelcheckrevoke", var)) { - http_schannel_check_revoke = git_config_bool(var, value); + if (value && !strcmp(value, "best-effort")) { + http_schannel_check_revoke_mode = +#ifdef CURLSSLOPT_REVOKE_BEST_EFFORT + CURLSSLOPT_REVOKE_BEST_EFFORT; +#else + CURLSSLOPT_NO_REVOKE; + warning(_("%s=%s unsupported by current cURL"), + var, value); +#endif + } else + http_schannel_check_revoke_mode = + (git_config_bool(var, value) ? + 0 : CURLSSLOPT_NO_REVOKE); return 0; } @@ -1057,8 +1075,8 @@ static CURL *get_curl_handle(void) #endif if (http_ssl_backend && !strcmp("schannel", http_ssl_backend) && - !http_schannel_check_revoke) { - curl_easy_setopt(result, CURLOPT_SSL_OPTIONS, (long)CURLSSLOPT_NO_REVOKE); + http_schannel_check_revoke_mode) { + curl_easy_setopt(result, CURLOPT_SSL_OPTIONS, http_schannel_check_revoke_mode); } if (http_proactive_auth != PROACTIVE_AUTH_NONE) From b4c799107158d38360d82c7888f474f8843e954a Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sat, 9 May 2020 19:24:23 +0200 Subject: [PATCH 058/248] t5505/t5516: fix white-space around redirectors The convention in Git project's shell scripts is to have white-space _before_, but not _after_ the `>` (or `<`). Signed-off-by: Johannes Schindelin --- t/t5505-remote.sh | 6 +++--- t/t5516-fetch-push.sh | 10 +++++----- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/t/t5505-remote.sh b/t/t5505-remote.sh index 969d9cc7d567a6..bb9f3412d439ce 100755 --- a/t/t5505-remote.sh +++ b/t/t5505-remote.sh @@ -951,8 +951,8 @@ test_expect_success '"remote show" does not show symbolic refs' ' ( cd three && git remote show origin >output && - ! grep "^ *HEAD$" < output && - ! grep -i stale < output + ! grep "^ *HEAD$" .git/branches/origin && + echo "quux#foom" >.git/branches/origin && git remote rename origin origin && test_path_is_missing .git/branches/origin && test "$(git config remote.origin.url)" = "quux" && diff --git a/t/t5516-fetch-push.sh b/t/t5516-fetch-push.sh index c15963c3d0b229..d37cc7f486344d 100755 --- a/t/t5516-fetch-push.sh +++ b/t/t5516-fetch-push.sh @@ -934,7 +934,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches' ' git branch second $the_first_commit && git checkout second && mkdir -p testrepo/.git/branches && - echo ".." > testrepo/.git/branches/branch1 && + echo ".." >testrepo/.git/branches/branch1 && ( cd testrepo && git fetch branch1 && @@ -948,7 +948,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches' ' test_expect_success !WITH_BREAKING_CHANGES 'fetch with branches containing #' ' mk_empty testrepo && mkdir -p testrepo/.git/branches && - echo "..#second" > testrepo/.git/branches/branch2 && + echo "..#second" >testrepo/.git/branches/branch2 && ( cd testrepo && git fetch branch2 && @@ -965,7 +965,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'push with branches' ' test_when_finished "rm -rf .git/branches" && mkdir -p .git/branches && - echo "testrepo" > .git/branches/branch1 && + echo "testrepo" >.git/branches/branch1 && git push branch1 && ( @@ -981,7 +981,7 @@ test_expect_success !WITH_BREAKING_CHANGES 'push with branches containing #' ' test_when_finished "rm -rf .git/branches" && mkdir -p .git/branches && - echo "testrepo#branch3" > .git/branches/branch2 && + echo "testrepo#branch3" >.git/branches/branch2 && git push branch2 && ( @@ -1511,7 +1511,7 @@ EOF git init no-thin && git --git-dir=no-thin/.git config receive.unpacklimit 0 && git push no-thin/.git refs/heads/main:refs/heads/foo && - echo modified >> path1 && + echo modified >>path1 && git commit -am modified && git repack -adf && rcvpck="git receive-pack --reject-thin-pack-for-testing" && From 81677657b0203b0e5c0c8c5a70aed1797d1dcbac Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sat, 12 Sep 2015 12:25:47 +0200 Subject: [PATCH 059/248] t3701: verify that we can add *lots* of files interactively Signed-off-by: Johannes Schindelin --- t/t3701-add-interactive.sh | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/t/t3701-add-interactive.sh b/t/t3701-add-interactive.sh index d9fe289a7ad13a..73e358d74b4256 100755 --- a/t/t3701-add-interactive.sh +++ b/t/t3701-add-interactive.sh @@ -1204,6 +1204,27 @@ test_expect_success 'checkout -p patch editing of added file' ' ) ' +test_expect_success EXPENSIVE 'add -i with a lot of files' ' + git reset --hard && + x160=0123456789012345678901234567890123456789 && + x160=$x160$x160$x160$x160 && + y= && + i=0 && + while test $i -le 200 + do + name=$(printf "%s%03d" $x160 $i) && + echo $name >$name && + git add -N $name && + y="${y}y$LF" && + i=$(($i+1)) || + exit 1 + done && + echo "$y" | git add -p -- . && + git diff --cached >staged && + test_line_count = 1407 staged && + git reset --hard +' + test_expect_success 'show help from add--helper' ' git reset --hard && cat >expect <<-EOF && From f2403682669a99737ca20fedc52f0675895f1d4b Mon Sep 17 00:00:00 2001 From: Luke Bonanomi Date: Wed, 24 Jun 2020 07:45:52 -0400 Subject: [PATCH 060/248] commit: accept "scissors" with CR/LF line endings This change enhances `git commit --cleanup=scissors` by detecting scissors lines ending in either LF (UNIX-style) or CR/LF (DOS-style). Regression tests are included to specifically test for trailing comments after a CR/LF-terminated scissors line. Signed-off-by: Luke Bonanomi Signed-off-by: Johannes Schindelin --- t/t7502-commit-porcelain.sh | 42 +++++++++++++++++++++++++++++++++++++ wt-status.c | 13 +++++++++--- 2 files changed, 52 insertions(+), 3 deletions(-) diff --git a/t/t7502-commit-porcelain.sh b/t/t7502-commit-porcelain.sh index b37e2018a74a7b..c38b96b66cd20a 100755 --- a/t/t7502-commit-porcelain.sh +++ b/t/t7502-commit-porcelain.sh @@ -623,6 +623,48 @@ test_expect_success 'cleanup commit messages (scissors option,-F,-e, scissors on test_must_be_empty actual ' +test_expect_success 'helper-editor' ' + + write_script lf-to-crlf.sh <<-\EOF + sed "s/\$/Q/" <"$1" | tr Q "\\015" >"$1".new && + mv -f "$1".new "$1" + EOF +' + +test_expect_success 'cleanup commit messages (scissors option,-F,-e, CR/LF line endings)' ' + + test_config core.editor "\"$PWD/lf-to-crlf.sh\"" && + scissors="# ------------------------ >8 ------------------------" && + + test_write_lines >text \ + "# Keep this comment" "" " $scissors" \ + "# Keep this comment, too" "$scissors" \ + "# Remove this comment" "$scissors" \ + "Remove this comment, too" && + + test_write_lines >expect \ + "# Keep this comment" "" " $scissors" \ + "# Keep this comment, too" && + + git commit --cleanup=scissors -e -F text --allow-empty && + git cat-file -p HEAD >raw && + sed -e "1,/^\$/d" raw >actual && + test_cmp expect actual +' + +test_expect_success 'cleanup commit messages (scissors option,-F,-e, scissors on first line, CR/LF line endings)' ' + + scissors="# ------------------------ >8 ------------------------" && + test_write_lines >text \ + "$scissors" \ + "# Remove this comment and any following lines" && + cp text /tmp/test2-text && + git commit --cleanup=scissors -e -F text --allow-empty --allow-empty-message && + git cat-file -p HEAD >raw && + sed -e "1,/^\$/d" raw >actual && + test_must_be_empty actual +' + test_expect_success 'cleanup commit messages (strip option,-F)' ' echo >>negative && diff --git a/wt-status.c b/wt-status.c index d6917f0a832d80..a1f5717809f598 100644 --- a/wt-status.c +++ b/wt-status.c @@ -40,7 +40,7 @@ #define UF_DELAY_WARNING_IN_MS (2 * 1000) static const char cut_line[] = -"------------------------ >8 ------------------------\n"; +"------------------------ >8 ------------------------"; static char default_wt_status_colors[][COLOR_MAXLEN] = { GIT_COLOR_NORMAL, /* WT_STATUS_HEADER */ @@ -1096,15 +1096,22 @@ static void wt_longstatus_print_other(struct wt_status *s, status_printf_ln(s, GIT_COLOR_NORMAL, "%s", ""); } +static inline int starts_with_newline(const char *p) +{ + return *p == '\n' || (*p == '\r' && p[1] == '\n'); +} + size_t wt_status_locate_end(const char *s, size_t len) { const char *p; struct strbuf pattern = STRBUF_INIT; strbuf_addf(&pattern, "\n%s %s", comment_line_str, cut_line); - if (starts_with(s, pattern.buf + 1)) + if (starts_with(s, pattern.buf + 1) && + starts_with_newline(s + pattern.len - 1)) len = 0; - else if ((p = strstr(s, pattern.buf))) { + else if ((p = strstr(s, pattern.buf)) && + starts_with_newline(p + pattern.len)) { size_t newlen = p - s + 1; if (newlen < len) len = newlen; From 214eca2584d9f4b0e4bbea0da0072124c5b26869 Mon Sep 17 00:00:00 2001 From: Jens Glathe Date: Tue, 2 Jun 2020 12:12:25 +0200 Subject: [PATCH 061/248] t0014: fix indentation For some reason, this test case was indented with 4 spaces instead of 1 horizontal tab. The other test cases in the same test script are fine. Signed-off-by: Jens Glathe Signed-off-by: Johannes Schindelin --- t/t0014-alias.sh | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/t/t0014-alias.sh b/t/t0014-alias.sh index 07a53e7366ef4b..62b4d81db875ca 100755 --- a/t/t0014-alias.sh +++ b/t/t0014-alias.sh @@ -52,10 +52,10 @@ test_expect_success 'looping aliases - deprecated builtins' ' #' test_expect_success 'run-command formats empty args properly' ' - test_must_fail env GIT_TRACE=1 git frotz a "" b " " c 2>actual.raw && - sed -ne "/run_command:/s/.*trace: run_command: //p" actual.raw >actual && - echo "git-frotz a '\'''\'' b '\'' '\'' c" >expect && - test_cmp expect actual + test_must_fail env GIT_TRACE=1 git frotz a "" b " " c 2>actual.raw && + sed -ne "/run_command:/s/.*trace: run_command: //p" actual.raw >actual && + echo "git-frotz a '\'''\'' b '\'' '\'' c" >expect && + test_cmp expect actual ' test_expect_success 'tracing a shell alias with arguments shows trace of prepared command' ' From cab6ce7cb070fca47f889ba6504ed164366ec17c Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Tue, 30 Mar 2021 14:25:31 -0400 Subject: [PATCH 062/248] clink.pl: fix libexpatd.lib link error when using MSVC When building with `make MSVC=1 DEBUG=1`, link to `libexpatd.lib` rather than `libexpat.lib`. It appears that the `vcpkg` package for "libexpat" has changed and now creates `libexpatd.lib` for debug mode builds. Previously, both debug and release builds created a ".lib" with the same basename. Signed-off-by: Jeff Hostetler --- compat/vcbuild/scripts/clink.pl | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/compat/vcbuild/scripts/clink.pl b/compat/vcbuild/scripts/clink.pl index 3bd824154be381..2768ae15f1879f 100755 --- a/compat/vcbuild/scripts/clink.pl +++ b/compat/vcbuild/scripts/clink.pl @@ -66,7 +66,11 @@ } push(@args, $lib); } elsif ("$arg" eq "-lexpat") { + if ($is_debug) { + push(@args, "libexpatd.lib"); + } else { push(@args, "libexpat.lib"); + } } elsif ("$arg" =~ /^-L/ && "$arg" ne "-LTCG") { $arg =~ s/^-L/-LIBPATH:/; push(@lflags, $arg); From 09f5e9c584bd81d7b0b41972d70d94849787a47b Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 5 Apr 2021 15:27:38 -0400 Subject: [PATCH 063/248] Makefile: clean up .ilk files when MSVC=1 Signed-off-by: Jeff Hostetler --- Makefile | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Makefile b/Makefile index 1f444eec3c1d12..6c3eff3ffc1503 100644 --- a/Makefile +++ b/Makefile @@ -3798,12 +3798,15 @@ ifdef MSVC $(RM) $(patsubst %.o,%.o.pdb,$(OBJECTS)) $(RM) headless-git.o.pdb $(RM) $(patsubst %.exe,%.pdb,$(OTHER_PROGRAMS)) + $(RM) $(patsubst %.exe,%.ilk,$(OTHER_PROGRAMS)) $(RM) $(patsubst %.exe,%.iobj,$(OTHER_PROGRAMS)) $(RM) $(patsubst %.exe,%.ipdb,$(OTHER_PROGRAMS)) $(RM) $(patsubst %.exe,%.pdb,$(PROGRAMS)) + $(RM) $(patsubst %.exe,%.ilk,$(PROGRAMS)) $(RM) $(patsubst %.exe,%.iobj,$(PROGRAMS)) $(RM) $(patsubst %.exe,%.ipdb,$(PROGRAMS)) $(RM) $(patsubst %.exe,%.pdb,$(TEST_PROGRAMS)) + $(RM) $(patsubst %.exe,%.ilk,$(TEST_PROGRAMS)) $(RM) $(patsubst %.exe,%.iobj,$(TEST_PROGRAMS)) $(RM) $(patsubst %.exe,%.ipdb,$(TEST_PROGRAMS)) $(RM) compat/vcbuild/MSVC-DEFS-GEN From d40651ead149a2996cdfaf1caa990fe40ce68a1d Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 5 Apr 2021 14:08:22 -0400 Subject: [PATCH 064/248] vcbuild: add support for compiling Windows resource files Create a wrapper for the Windows Resource Compiler (RC.EXE) for use by the MSVC=1 builds. This is similar to the CL.EXE and LIB.EXE wrappers used for the MSVC=1 builds. Signed-off-by: Jeff Hostetler --- compat/vcbuild/find_vs_env.bat | 7 ++++++ compat/vcbuild/scripts/rc.pl | 46 ++++++++++++++++++++++++++++++++++ config.mak.uname | 3 ++- 3 files changed, 55 insertions(+), 1 deletion(-) create mode 100644 compat/vcbuild/scripts/rc.pl diff --git a/compat/vcbuild/find_vs_env.bat b/compat/vcbuild/find_vs_env.bat index b35d264c0e6bed..379b16296e09c2 100644 --- a/compat/vcbuild/find_vs_env.bat +++ b/compat/vcbuild/find_vs_env.bat @@ -99,6 +99,7 @@ REM ================================================================ SET sdk_dir=%WindowsSdkDir% SET sdk_ver=%WindowsSDKVersion% + SET sdk_ver_bin_dir=%WindowsSdkVerBinPath%%tgt% SET si=%sdk_dir%Include\%sdk_ver% SET sdk_includes=-I"%si%ucrt" -I"%si%um" -I"%si%shared" SET sl=%sdk_dir%lib\%sdk_ver% @@ -130,6 +131,7 @@ REM ================================================================ SET sdk_dir=%WindowsSdkDir% SET sdk_ver=%WindowsSDKVersion% + SET sdk_ver_bin_dir=%WindowsSdkVerBinPath%bin\amd64 SET si=%sdk_dir%Include\%sdk_ver% SET sdk_includes=-I"%si%ucrt" -I"%si%um" -I"%si%shared" -I"%si%winrt" SET sl=%sdk_dir%lib\%sdk_ver% @@ -160,6 +162,11 @@ REM ================================================================ echo msvc_includes=%msvc_includes% echo msvc_libs=%msvc_libs% + echo sdk_ver_bin_dir=%sdk_ver_bin_dir% + SET X1=%sdk_ver_bin_dir:C:=/C% + SET X2=%X1:\=/% + echo sdk_ver_bin_dir_msys=%X2% + echo sdk_includes=%sdk_includes% echo sdk_libs=%sdk_libs% diff --git a/compat/vcbuild/scripts/rc.pl b/compat/vcbuild/scripts/rc.pl new file mode 100644 index 00000000000000..7bca4cd81c6c63 --- /dev/null +++ b/compat/vcbuild/scripts/rc.pl @@ -0,0 +1,46 @@ +#!/usr/bin/perl -w +###################################################################### +# Compile Resources on Windows +# +# This is a wrapper to facilitate the compilation of Git with MSVC +# using GNU Make as the build system. So, instead of manipulating the +# Makefile into something nasty, just to support non-space arguments +# etc, we use this wrapper to fix the command line options +# +###################################################################### +use strict; +my @args = (); +my @input = (); + +while (@ARGV) { + my $arg = shift @ARGV; + if ("$arg" =~ /^-[dD]/) { + # GIT_VERSION gets passed with too many + # layers of dquote escaping. + $arg =~ s/\\"/"/g; + + push(@args, $arg); + + } elsif ("$arg" eq "-i") { + my $arg = shift @ARGV; + # TODO complain if NULL or is dashed ?? + push(@input, $arg); + + } elsif ("$arg" eq "-o") { + my $arg = shift @ARGV; + # TODO complain if NULL or is dashed ?? + push(@args, "-fo$arg"); + + } else { + push(@args, $arg); + } +} + +push(@args, "-nologo"); +push(@args, "-v"); +push(@args, @input); + +unshift(@args, "rc.exe"); +printf("**** @args\n"); + +exit (system(@args) != 0); diff --git a/config.mak.uname b/config.mak.uname index 94730676c64dbd..de609e931b10fb 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -467,7 +467,7 @@ ifeq ($(uname_S),Windows) # link.exe next to, and required by, cl.exe, we have to prepend this # onto the existing $PATH. # - SANE_TOOL_PATH ?= $(msvc_bin_dir_msys) + SANE_TOOL_PATH ?= $(msvc_bin_dir_msys):$(sdk_ver_bin_dir_msys) HAVE_ALLOCA_H = YesPlease NO_PREAD = YesPlease NEEDS_CRYPTO_WITH_SSL = YesPlease @@ -535,6 +535,7 @@ endif # See https://msdn.microsoft.com/en-us/library/ms235330.aspx EXTLIBS = user32.lib advapi32.lib shell32.lib wininet.lib ws2_32.lib invalidcontinue.obj kernel32.lib ntdll.lib PTHREAD_LIBS = + RC = compat/vcbuild/scripts/rc.pl lib = BASIC_CFLAGS += $(vcpkg_inc) $(sdk_includes) $(msvc_includes) ifndef DEBUG From c596235e207185d8b9a7aeb1eafd5f7e81473718 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 5 Apr 2021 14:12:14 -0400 Subject: [PATCH 065/248] config.mak.uname: add git.rc to MSVC builds Teach MSVC=1 builds to depend on the `git.rc` file so that the resulting executables have Windows-style resources and version number information within them. Signed-off-by: Jeff Hostetler --- config.mak.uname | 1 + 1 file changed, 1 insertion(+) diff --git a/config.mak.uname b/config.mak.uname index de609e931b10fb..ec5545a15ae2cc 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -534,6 +534,7 @@ endif # handle twice, or to access the osfhandle of an already-closed stdout # See https://msdn.microsoft.com/en-us/library/ms235330.aspx EXTLIBS = user32.lib advapi32.lib shell32.lib wininet.lib ws2_32.lib invalidcontinue.obj kernel32.lib ntdll.lib + GITLIBS += git.res PTHREAD_LIBS = RC = compat/vcbuild/scripts/rc.pl lib = From 6dbb0ecaedac1af03fb46612bf1127547f690f03 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 5 Apr 2021 14:24:52 -0400 Subject: [PATCH 066/248] clink.pl: ignore no-stack-protector arg on MSVC=1 builds Ignore the `-fno-stack-protector` compiler argument when building with MSVC. This will be used in a later commit that needs to build a Win32 GUI app. Signed-off-by: Jeff Hostetler --- compat/vcbuild/scripts/clink.pl | 2 ++ 1 file changed, 2 insertions(+) diff --git a/compat/vcbuild/scripts/clink.pl b/compat/vcbuild/scripts/clink.pl index 2768ae15f1879f..73c8a2b184f38b 100755 --- a/compat/vcbuild/scripts/clink.pl +++ b/compat/vcbuild/scripts/clink.pl @@ -122,6 +122,8 @@ push(@cflags, "-wd4996"); } elsif ("$arg" =~ /^-W[a-z]/) { # let's ignore those + } elsif ("$arg" eq "-fno-stack-protector") { + # eat this } else { push(@args, $arg); } From fbc5b0085522513a12b51bf677ab6f32affc8796 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 12 Aug 2020 15:06:17 +0000 Subject: [PATCH 067/248] git-gui: accommodate for intent-to-add files As of Git v2.28.0, the diff for files staged via `git add -N` marks them as new files. Git GUI was ill-prepared for that, and this patch teaches Git GUI about them. Please note that this will not even fix things with v2.28.0, as the `rp/apply-cached-with-i-t-a` patches are required on Git's side, too. This fixes https://github.com/git-for-windows/git/issues/2779 Signed-off-by: Johannes Schindelin Signed-off-by: Pratyush Yadav --- git-gui/git-gui.sh | 2 ++ git-gui/lib/diff.tcl | 12 ++++++++---- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/git-gui/git-gui.sh b/git-gui/git-gui.sh index a931d7f7c99029..211960c14c8262 100755 --- a/git-gui/git-gui.sh +++ b/git-gui/git-gui.sh @@ -1948,6 +1948,7 @@ set all_icons(U$ui_index) file_merge set all_icons(T$ui_index) file_statechange set all_icons(_$ui_workdir) file_plain +set all_icons(A$ui_workdir) file_plain set all_icons(M$ui_workdir) file_mod set all_icons(D$ui_workdir) file_question set all_icons(U$ui_workdir) file_merge @@ -1974,6 +1975,7 @@ foreach i { {A_ {mc "Staged for commit"}} {AM {mc "Portions staged for commit"}} {AD {mc "Staged for commit, missing"}} + {AA {mc "Intended to be added"}} {_D {mc "Missing"}} {D_ {mc "Staged for removal"}} diff --git a/git-gui/lib/diff.tcl b/git-gui/lib/diff.tcl index 442737ba4f260b..003e4613f3495b 100644 --- a/git-gui/lib/diff.tcl +++ b/git-gui/lib/diff.tcl @@ -554,7 +554,8 @@ proc apply_or_revert_hunk {x y revert} { if {$current_diff_side eq $ui_index} { set failed_msg [mc "Failed to unstage selected hunk."] lappend apply_cmd --reverse --cached - if {[string index $mi 0] ne {M}} { + set file_state [string index $mi 0] + if {$file_state ne {M} && $file_state ne {A}} { unlock_index return } @@ -567,7 +568,8 @@ proc apply_or_revert_hunk {x y revert} { lappend apply_cmd --cached } - if {[string index $mi 1] ne {M}} { + set file_state [string index $mi 1] + if {$file_state ne {M} && $file_state ne {A}} { unlock_index return } @@ -659,7 +661,8 @@ proc apply_or_revert_range_or_line {x y revert} { set failed_msg [mc "Failed to unstage selected line."] set to_context {+} lappend apply_cmd --reverse --cached - if {[string index $mi 0] ne {M}} { + set file_state [string index $mi 0] + if {$file_state ne {M} && $file_state ne {A}} { unlock_index return } @@ -674,7 +677,8 @@ proc apply_or_revert_range_or_line {x y revert} { lappend apply_cmd --cached } - if {[string index $mi 1] ne {M}} { + set file_state [string index $mi 1] + if {$file_state ne {M} && $file_state ne {A}} { unlock_index return } From 456013bb782573be5a27bbe4bd3121b4637b79f0 Mon Sep 17 00:00:00 2001 From: Ian Bearman Date: Fri, 31 Jan 2020 16:00:25 -0800 Subject: [PATCH 068/248] vcbuild: install ARM64 dependencies when building ARM64 binaries Co-authored-by: Dennis Ameling Signed-off-by: Ian Bearman Signed-off-by: Dennis Ameling Signed-off-by: Johannes Schindelin --- compat/vcbuild/README | 6 +++++- compat/vcbuild/vcpkg_copy_dlls.bat | 7 ++++++- compat/vcbuild/vcpkg_install.bat | 9 +++++++-- 3 files changed, 18 insertions(+), 4 deletions(-) diff --git a/compat/vcbuild/README b/compat/vcbuild/README index 29ec1d0f104b80..1df1cabb1ebbbd 100644 --- a/compat/vcbuild/README +++ b/compat/vcbuild/README @@ -6,7 +6,11 @@ The Steps to Build Git with VS2015 or VS2017 from the command line. Prompt or from an SDK bash window: $ cd - $ ./compat/vcbuild/vcpkg_install.bat + $ ./compat/vcbuild/vcpkg_install.bat x64-windows + + or + + $ ./compat/vcbuild/vcpkg_install.bat arm64-windows The vcpkg tools and all of the third-party sources will be installed in this folder: diff --git a/compat/vcbuild/vcpkg_copy_dlls.bat b/compat/vcbuild/vcpkg_copy_dlls.bat index 13661c14f8705c..8bea0cbf83b6cf 100644 --- a/compat/vcbuild/vcpkg_copy_dlls.bat +++ b/compat/vcbuild/vcpkg_copy_dlls.bat @@ -15,7 +15,12 @@ REM ================================================================ @FOR /F "delims=" %%D IN ("%~dp0") DO @SET cwd=%%~fD cd %cwd% - SET arch=x64-windows + SET arch=%2 + IF NOT DEFINED arch ( + echo defaulting to 'x64-windows`. Invoke %0 with 'x86-windows', 'x64-windows', or 'arm64-windows' + set arch=x64-windows + ) + SET inst=%cwd%vcpkg\installed\%arch% IF [%1]==[release] ( diff --git a/compat/vcbuild/vcpkg_install.bat b/compat/vcbuild/vcpkg_install.bat index 8330d8120fb511..cacef18c11dc79 100644 --- a/compat/vcbuild/vcpkg_install.bat +++ b/compat/vcbuild/vcpkg_install.bat @@ -31,6 +31,12 @@ REM ================================================================ SETLOCAL EnableDelayedExpansion + SET arch=%1 + IF NOT DEFINED arch ( + echo defaulting to 'x64-windows`. Invoke %0 with 'x86-windows', 'x64-windows', or 'arm64-windows' + set arch=x64-windows + ) + @FOR /F "delims=" %%D IN ("%~dp0") DO @SET cwd=%%~fD cd %cwd% @@ -55,9 +61,8 @@ REM ================================================================ echo Successfully installed %cwd%vcpkg\vcpkg.exe :install_libraries - SET arch=x64-windows - echo Installing third-party libraries... + echo Installing third-party libraries(%arch%)... FOR %%i IN (zlib expat libiconv openssl libssh2 curl) DO ( cd %cwd%vcpkg IF NOT EXIST "packages\%%i_%arch%" CALL :sub__install_one %%i From bcf4ed269cedf3db2b75aa7544f2af48057dde0c Mon Sep 17 00:00:00 2001 From: Ian Bearman Date: Tue, 4 Feb 2020 10:34:40 -0800 Subject: [PATCH 069/248] vcbuild: add an option to install individual 'features' In this context, a "feature" is a dependency combined with its own dependencies. Signed-off-by: Ian Bearman Signed-off-by: Johannes Schindelin --- compat/vcbuild/vcpkg_install.bat | 35 +++++++++++++++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-) diff --git a/compat/vcbuild/vcpkg_install.bat b/compat/vcbuild/vcpkg_install.bat index cacef18c11dc79..8da212487ae97d 100644 --- a/compat/vcbuild/vcpkg_install.bat +++ b/compat/vcbuild/vcpkg_install.bat @@ -85,14 +85,47 @@ REM ================================================================ :sub__install_one echo Installing package %1... + call :%1_features + REM vcpkg may not be reliable on slow, intermittent or proxy REM connections, see e.g. REM https://social.msdn.microsoft.com/Forums/windowsdesktop/en-US/4a8f7be5-5e15-4213-a7bb-ddf424a954e6/winhttpsendrequest-ends-with-12002-errorhttptimeout-after-21-seconds-no-matter-what-timeout?forum=windowssdk REM which explains the hidden 21 second timeout REM (last post by Dave : Microsoft - Windows Networking team) - .\vcpkg.exe install %1:%arch% + .\vcpkg.exe install %1%features%:%arch% IF ERRORLEVEL 1 ( EXIT /B 1 ) echo Finished %1 goto :EOF + +:: +:: features for each vcpkg to install +:: there should be an entry here for each package to install +:: 'set features=' means use the default otherwise +:: 'set features=[comma-delimited-feature-set]' is the syntax +:: + +:zlib_features +set features= +goto :EOF + +:expat_features +set features= +goto :EOF + +:libiconv_features +set features= +goto :EOF + +:openssl_features +set features= +goto :EOF + +:libssh2_features +set features= +goto :EOF + +:curl_features +set features=[core,openssl] +goto :EOF From 87185521c09e22adc0462727bde54c9980014929 Mon Sep 17 00:00:00 2001 From: Dennis Ameling Date: Fri, 4 Dec 2020 14:11:34 +0100 Subject: [PATCH 070/248] cmake: allow building for Windows/ARM64 Signed-off-by: Dennis Ameling Signed-off-by: Johannes Schindelin --- contrib/buildsystems/CMakeLists.txt | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index edb0fc04ad7649..093a6b7e186041 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -65,9 +65,9 @@ if(USE_VCPKG) set(VCPKG_DIR "${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg") if(NOT EXISTS ${VCPKG_DIR}) message("Initializing vcpkg and building the Git's dependencies (this will take a while...)") - execute_process(COMMAND ${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg_install.bat) + execute_process(COMMAND ${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg_install.bat ${VCPKG_ARCH}) endif() - list(APPEND CMAKE_PREFIX_PATH "${VCPKG_DIR}/installed/x64-windows") + list(APPEND CMAKE_PREFIX_PATH "${VCPKG_DIR}/installed/${VCPKG_ARCH}") # In the vcpkg edition, we need this to be able to link to libcurl set(CURL_NO_CURL_CMAKE ON) @@ -1213,7 +1213,7 @@ string(REPLACE "@USE_LIBPCRE2@" "" git_build_options "${git_build_options}") string(REPLACE "@WITH_BREAKING_CHANGES@" "" git_build_options "${git_build_options}") string(REPLACE "@X@" "${EXE_EXTENSION}" git_build_options "${git_build_options}") if(USE_VCPKG) - string(APPEND git_build_options "PATH=\"$PATH:$TEST_DIRECTORY/../compat/vcbuild/vcpkg/installed/x64-windows/bin\"\n") + string(APPEND git_build_options "PATH=\"$PATH:$TEST_DIRECTORY/../compat/vcbuild/vcpkg/installed/${VCPKG_ARCH}/bin\"\n") endif() file(WRITE ${CMAKE_BINARY_DIR}/GIT-BUILD-OPTIONS ${git_build_options}) From fbacfdc6bf440a9c3e807e323309f886be8a19ee Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Sun, 6 Oct 2019 18:40:55 +0100 Subject: [PATCH 071/248] vcpkg_install: detect lack of Git The vcpkg_install batch file depends on the availability of a working Git on the CMD path. This may not be present if the user has selected the 'bash only' option during Git-for-Windows install. Detect and tell the user about their lack of a working Git in the CMD window. Fixes #2348. A separate PR https://github.com/git-for-windows/build-extra/pull/258 now highlights the recommended path setting during install. Signed-off-by: Philip Oakley --- compat/vcbuild/vcpkg_install.bat | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/compat/vcbuild/vcpkg_install.bat b/compat/vcbuild/vcpkg_install.bat index ebd0bad242a8ca..bcbbf536af3141 100644 --- a/compat/vcbuild/vcpkg_install.bat +++ b/compat/vcbuild/vcpkg_install.bat @@ -36,6 +36,13 @@ REM ================================================================ dir vcpkg\vcpkg.exe >nul 2>nul && GOTO :install_libraries + git.exe version 2>nul + IF ERRORLEVEL 1 ( + echo "***" + echo "Git not found. Please adjust your CMD path or Git install option." + echo "***" + EXIT /B 1 ) + echo Fetching vcpkg in %cwd%vcpkg git.exe clone https://github.com/Microsoft/vcpkg vcpkg IF ERRORLEVEL 1 ( EXIT /B 1 ) From 47d8085aec43c42e309347a173a638e908acb9dc Mon Sep 17 00:00:00 2001 From: Dennis Ameling Date: Sun, 29 Nov 2020 00:12:26 +0100 Subject: [PATCH 072/248] ci(vs-build) also build Windows/ARM64 artifacts There are no Windows/ARM64 agents in GitHub Actions yet, therefore we just skip adjusting the `vs-test` job for now. Signed-off-by: Dennis Ameling Signed-off-by: Johannes Schindelin --- .github/workflows/main.yml | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index aa6bce673b4e99..c9905bdeb4139c 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -169,8 +169,11 @@ jobs: NO_PERL: 1 GIT_CONFIG_PARAMETERS: "'user.name=CI' 'user.email=ci@git'" runs-on: windows-latest + strategy: + matrix: + arch: [x64, arm64] concurrency: - group: vs-build-${{ github.ref }} + group: vs-build-${{ github.ref }}-${{ matrix.arch }} cancel-in-progress: ${{ needs.ci-config.outputs.skip_concurrent == 'yes' }} steps: - uses: actions/checkout@v5 @@ -189,14 +192,14 @@ jobs: uses: microsoft/setup-msbuild@v2 - name: copy dlls to root shell: cmd - run: compat\vcbuild\vcpkg_copy_dlls.bat release + run: compat\vcbuild\vcpkg_copy_dlls.bat release ${{ matrix.arch }}-windows - name: generate Visual Studio solution shell: bash run: | - cmake `pwd`/contrib/buildsystems/ -DCMAKE_PREFIX_PATH=`pwd`/compat/vcbuild/vcpkg/installed/x64-windows \ - -DNO_GETTEXT=YesPlease -DPERL_TESTS=OFF -DPYTHON_TESTS=OFF -DCURL_NO_CURL_CMAKE=ON + cmake `pwd`/contrib/buildsystems/ -DCMAKE_PREFIX_PATH=`pwd`/compat/vcbuild/vcpkg/installed/${{ matrix.arch }}-windows \ + -DNO_GETTEXT=YesPlease -DPERL_TESTS=OFF -DPYTHON_TESTS=OFF -DCURL_NO_CURL_CMAKE=ON -DCMAKE_GENERATOR_PLATFORM=${{ matrix.arch }} -DVCPKG_ARCH=${{ matrix.arch }}-windows - name: MSBuild - run: msbuild git.sln -property:Configuration=Release -property:Platform=x64 -maxCpuCount:4 -property:PlatformToolset=v142 + run: msbuild git.sln -property:Configuration=Release -property:Platform=${{ matrix.arch }} -maxCpuCount:4 -property:PlatformToolset=v142 - name: bundle artifact tar shell: bash env: @@ -210,7 +213,7 @@ jobs: - name: upload tracked files and build artifacts uses: actions/upload-artifact@v4 with: - name: vs-artifacts + name: vs-artifacts-${{ matrix.arch }} path: artifacts vs-test: name: win+VS test @@ -228,7 +231,7 @@ jobs: - name: download tracked files and build artifacts uses: actions/download-artifact@v5 with: - name: vs-artifacts + name: vs-artifacts-x64 path: ${{github.workspace}} - name: extract tracked files and build artifacts shell: bash From 0440cb887d4756c64c23f4dde544c1823b275d00 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 5 Apr 2021 14:39:33 -0400 Subject: [PATCH 073/248] clink.pl: move default linker options for MSVC=1 builds Move the default `-ENTRY` and `-SUBSYSTEM` arguments for MSVC=1 builds from `config.mak.uname` into `clink.pl`. These args are constant for console-mode executables. Add support to `clink.pl` for generating a Win32 GUI application using the `-mwindows` argument (to match how GCC does it). This changes the `-ENTRY` and `-SUBSYSTEM` arguments accordingly. Signed-off-by: Jeff Hostetler --- compat/vcbuild/scripts/clink.pl | 11 +++++++++++ config.mak.uname | 2 +- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/compat/vcbuild/scripts/clink.pl b/compat/vcbuild/scripts/clink.pl index 73c8a2b184f38b..a38b360015ece9 100755 --- a/compat/vcbuild/scripts/clink.pl +++ b/compat/vcbuild/scripts/clink.pl @@ -15,6 +15,7 @@ my @lflags = (); my $is_linking = 0; my $is_debug = 0; +my $is_gui = 0; while (@ARGV) { my $arg = shift @ARGV; if ("$arg" eq "-DDEBUG") { @@ -124,11 +125,21 @@ # let's ignore those } elsif ("$arg" eq "-fno-stack-protector") { # eat this + } elsif ("$arg" eq "-mwindows") { + $is_gui = 1; } else { push(@args, $arg); } } if ($is_linking) { + if ($is_gui) { + push(@args, "-ENTRY:wWinMainCRTStartup"); + push(@args, "-SUBSYSTEM:WINDOWS"); + } else { + push(@args, "-ENTRY:wmainCRTStartup"); + push(@args, "-SUBSYSTEM:CONSOLE"); + } + push(@args, @lflags); unshift(@args, "link.exe"); } else { diff --git a/config.mak.uname b/config.mak.uname index ec5545a15ae2cc..dfe201b3e8e476 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -529,7 +529,7 @@ endif compat/win32/trace2_win32_process_info.o \ compat/win32/dirent.o COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\" - BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO -ENTRY:wmainCRTStartup -SUBSYSTEM:CONSOLE + BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO # invalidcontinue.obj allows Git's source code to close the same file # handle twice, or to access the osfhandle of an already-closed stdout # See https://msdn.microsoft.com/en-us/library/ms235330.aspx From 8f5f1de4e714935136656b2397a13ee5cdb2236a Mon Sep 17 00:00:00 2001 From: Yuyi Wang Date: Sat, 11 Mar 2023 17:51:18 +0800 Subject: [PATCH 074/248] cmake: install headless-git. headless-git is a git executable without opening a console window. It is useful when other GUI executables want to call git. We should install it together with git on Windows. Signed-off-by: Yuyi Wang --- contrib/buildsystems/CMakeLists.txt | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index edb0fc04ad7649..e780ff98a37655 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -745,6 +745,7 @@ if(WIN32) endif() add_executable(headless-git ${CMAKE_SOURCE_DIR}/compat/win32/headless.c) + list(APPEND PROGRAMS_BUILT headless-git) if(CMAKE_C_COMPILER_ID STREQUAL "GNU" OR CMAKE_C_COMPILER_ID STREQUAL "Clang") target_link_options(headless-git PUBLIC -municode -Wl,-subsystem,windows) elseif(CMAKE_C_COMPILER_ID STREQUAL "MSVC") @@ -945,7 +946,7 @@ list(TRANSFORM git_perl_scripts PREPEND "${CMAKE_BINARY_DIR}/") #install foreach(program ${PROGRAMS_BUILT}) -if(program MATCHES "^(git|git-shell|scalar)$") +if(program MATCHES "^(git|git-shell|headless-git|scalar)$") install(TARGETS ${program} RUNTIME DESTINATION bin) else() From 081df17be33f5a4a4c78968f351c2616057b48f0 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Sun, 6 Oct 2019 18:43:57 +0100 Subject: [PATCH 075/248] vcpkg_install: add comment regarding slow network connections The vcpkg downloads may not succeed. Warn careful readers of the time out. A simple retry will usually resolve the issue. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- compat/vcbuild/vcpkg_install.bat | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/compat/vcbuild/vcpkg_install.bat b/compat/vcbuild/vcpkg_install.bat index bcbbf536af3141..8330d8120fb511 100644 --- a/compat/vcbuild/vcpkg_install.bat +++ b/compat/vcbuild/vcpkg_install.bat @@ -80,6 +80,12 @@ REM ================================================================ :sub__install_one echo Installing package %1... + REM vcpkg may not be reliable on slow, intermittent or proxy + REM connections, see e.g. + REM https://social.msdn.microsoft.com/Forums/windowsdesktop/en-US/4a8f7be5-5e15-4213-a7bb-ddf424a954e6/winhttpsendrequest-ends-with-12002-errorhttptimeout-after-21-seconds-no-matter-what-timeout?forum=windowssdk + REM which explains the hidden 21 second timeout + REM (last post by Dave : Microsoft - Windows Networking team) + .\vcpkg.exe install %1:%arch% IF ERRORLEVEL 1 ( EXIT /B 1 ) From bcb905bdb203e5122f4e55467e2f5826a2e91796 Mon Sep 17 00:00:00 2001 From: Dennis Ameling Date: Sun, 6 Dec 2020 18:39:26 +0100 Subject: [PATCH 076/248] Add schannel to curl installation Signed-off-by: Dennis Ameling --- compat/vcbuild/vcpkg_install.bat | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/compat/vcbuild/vcpkg_install.bat b/compat/vcbuild/vcpkg_install.bat index 8da212487ae97d..575c65c20ba307 100644 --- a/compat/vcbuild/vcpkg_install.bat +++ b/compat/vcbuild/vcpkg_install.bat @@ -127,5 +127,5 @@ set features= goto :EOF :curl_features -set features=[core,openssl] +set features=[core,openssl,schannel] goto :EOF From 6ab3af1245d58b131f0a6f1143fd5ec00c334fc4 Mon Sep 17 00:00:00 2001 From: Victoria Dye Date: Thu, 5 Aug 2021 19:04:13 -0400 Subject: [PATCH 077/248] subtree: update `contrib/subtree` `test` target The intention of this change is to align with how the top-level git `Makefile` defines its own test target (which also internally calls `$(MAKE) -C t/ all`). This change also ensures the consistency of `make -C contrib/subtree test` with other testing in CI executions (which rely on `$DEFAULT_TEST_TARGET` being defined as `prove`). Signed-off-by: Victoria Dye --- contrib/subtree/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/subtree/Makefile b/contrib/subtree/Makefile index c0c9f21cb78022..dab2dfc08ee222 100644 --- a/contrib/subtree/Makefile +++ b/contrib/subtree/Makefile @@ -95,7 +95,7 @@ $(GIT_SUBTREE_TEST): $(GIT_SUBTREE) cp $< $@ test: $(GIT_SUBTREE_TEST) - $(MAKE) -C t/ test + $(MAKE) -C t/ all clean: $(RM) $(GIT_SUBTREE) From 3b5a7c3b3b968c3ff58eea7e9743e0de65838ce9 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Fri, 2 Jul 2021 00:30:24 +0100 Subject: [PATCH 078/248] CMake: default Visual Studio generator has changed Correct some wording and inform users regarding the Visual Studio changes (from V16.6) to the default generator. Subsequent commits ensure that Git for Windows can be directly opened in modern Visual Studio without needing special configuration of the CMakeLists settings. It appeares that internally Visual Studio creates it's own version of the .sln file (etc.) for extension tools that expect them. The large number of references below document the shifting of Visual Studio default and CMake setting options. refs: https://docs.microsoft.com/en-us/search/?scope=C%2B%2B&view=msvc-150&terms=Ninja 1. https://docs.microsoft.com/en-us/cpp/linux/cmake-linux-configure?view=msvc-160 (note the linux bit) "In Visual Studio 2019 version 16.6 or later ***, Ninja is the default generator for configurations targeting a remote system or WSL. For more information, see this post on the C++ Team Blog [https://devblogs.microsoft.com/cppblog/linux-development-with-visual-studio-first-class-support-for-gdbserver-improved-build-times-with-ninja-and-updates-to-the-connection-manager/]. For more information about these settings, see CMakeSettings.json reference [https://docs.microsoft.com/en-us/cpp/build/cmakesettings-reference?view=msvc-160]." 2. https://docs.microsoft.com/en-us/cpp/build/cmake-presets-vs?view=msvc-160 "CMake supports two files that allow users to specify common configure, build, and test options and share them with others: CMakePresets.json and CMakeUserPresets.json." " Both files are supported in Visual Studio 2019 version 16.10 or later. ***" 3. https://devblogs.microsoft.com/cppblog/linux-development-with-visual-studio-first-class-support-for-gdbserver-improved-build-times-with-ninja-and-updates-to-the-connection-manager/ " Ninja has been the default generator (underlying build system) for CMake configurations targeting Windows for some time***, but in Visual Studio 2019 version 16.6 Preview 3*** we added support for Ninja on Linux." 4. https://docs.microsoft.com/en-us/cpp/build/cmakesettings-reference?view=msvc-160 " `generator`: specifies CMake generator to use for this configuration. May be one of: Visual Studio 2019 only: Visual Studio 16 2019 Visual Studio 16 2019 Win64 Visual Studio 16 2019 ARM Visual Studio 2017 and later: Visual Studio 15 2017 Visual Studio 15 2017 Win64 Visual Studio 15 2017 ARM Visual Studio 14 2015 Visual Studio 14 2015 Win64 Visual Studio 14 2015 ARM Unix Makefiles Ninja Because Ninja is designed for fast build speeds instead of flexibility and function, it is set as the default. However, some CMake projects may be unable to correctly build using Ninja. If this occurs, you can instruct CMake to generate Visual Studio projects instead. To specify a Visual Studio generator in Visual Studio 2017, open the settings editor from the main menu by choosing CMake | Change CMake Settings. Delete "Ninja" and type "V". This activates IntelliSense, which enables you to choose the generator you want." "To specify a Visual Studio generator in Visual Studio 2019, right-click on the CMakeLists.txt file in Solution Explorer and choose CMake Settings for project > Show Advanced Settings > CMake Generator. When the active configuration specifies a Visual Studio generator, by default MSBuild.exe is invoked with` -m -v:minimal` arguments." 5. https://docs.microsoft.com/en-us/cpp/build/cmake-presets-vs?view=msvc-160#enable-cmakepresetsjson-integration-in-visual-studio-2019 "Enable CMakePresets.json integration in Visual Studio 2019 CMakePresets.json integration isn't enabled by default in Visual Studio 2019. You can enable it for all CMake projects in Tools > Options > CMake > General: (tick a box)" ... see more. 6. https://docs.microsoft.com/en-us/cpp/build/cmakesettings-reference?view=msvc-140 (whichever v140 is..) "CMake projects are supported in Visual Studio 2017 and later." 7. https://docs.microsoft.com/en-us/cpp/overview/what-s-new-for-cpp-2017?view=msvc-150 "Support added for the CMake Ninja generator." 8. https://docs.microsoft.com/en-us/cpp/overview/what-s-new-for-cpp-2017?view=msvc-150#cmake-support-via-open-folder "CMake support via Open Folder Visual Studio 2017 introduces support for using CMake projects without converting to MSBuild project files (.vcxproj). For more information, see CMake projects in Visual Studio[https://docs.microsoft.com/en-us/cpp/build/cmake-projects-in-visual-studio?view=msvc-150]. Opening CMake projects with Open Folder automatically configures the environment for C++ editing, building, and debugging." ... +more! 9. https://docs.microsoft.com/en-us/cpp/build/cmake-presets-vs?view=msvc-160#supported-cmake-and-cmakepresetsjson-versions "Visual Studio reads and evaluates CMakePresets.json and CMakeUserPresets.json itself and doesn't invoke CMake directly with the --preset option. So, CMake version 3.20 or later isn't strictly required when you're building with CMakePresets.json inside Visual Studio. We recommend using CMake version 3.14 or later." 10. https://docs.microsoft.com/en-us/cpp/build/cmake-presets-vs?view=msvc-160#enable-cmakepresetsjson-integration-in-visual-studio-2019 "If you don't want to enable CMakePresets.json integration for all CMake projects, you can enable CMakePresets.json integration for a single CMake project by adding a CMakePresets.json file to the root of the open folder. You must close and reopen the folder in Visual Studio to activate the integration. 11. https://docs.microsoft.com/en-us/cpp/build/cmake-presets-vs?view=msvc-160#default-configure-presets ***(doesn't actually say which version..) "Default Configure Presets If no CMakePresets.json or CMakeUserPresets.json file exists, or if CMakePresets.json or CMakeUserPresets.json is invalid, Visual Studio will fall back*** on the following default Configure Presets: Windows example JSON { "name": "windows-default", "displayName": "Windows x64 Debug", "description": "Sets Ninja generator, compilers, x64 architecture, build and install directory, debug build type", "generator": "Ninja", "binaryDir": "${sourceDir}/out/build/${presetName}", "architecture": { "value": "x64", "strategy": "external" }, "cacheVariables": { "CMAKE_BUILD_TYPE": "Debug", "CMAKE_INSTALL_PREFIX": "${sourceDir}/out/install/${presetName}" }, "vendor": { "microsoft.com/VisualStudioSettings/CMake/1.0": { "hostOS": [ "Windows" ] } } }, " Signed-off-by: Philip Oakley --- contrib/buildsystems/CMakeLists.txt | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index 46fe6c3a5fb938..f8ef01e57702ac 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -14,6 +14,11 @@ Note: Visual Studio also has the option of opening `CMakeLists.txt` directly; Using this option, Visual Studio will not find the source code, though, therefore the `File>Open>Folder...` option is preferred. +Visual Studio does not produce a .sln solution file nor the .vcxproj files +that may be required by VS extension tools. + +To generate the .sln/.vcxproj files run CMake manually, as described below. + Instructions to run CMake manually: mkdir -p contrib/buildsystems/out @@ -22,7 +27,7 @@ Instructions to run CMake manually: This will build the git binaries in contrib/buildsystems/out directory (our top-level .gitignore file knows to ignore contents of -this directory). +this directory). The project .sln and .vcxproj files are also generated. Possible build configurations(-DCMAKE_BUILD_TYPE) with corresponding compiler flags @@ -35,17 +40,16 @@ empty(default) : NOTE: -DCMAKE_BUILD_TYPE is optional. For multi-config generators like Visual Studio this option is ignored -This process generates a Makefile(Linux/*BSD/MacOS) , Visual Studio solution(Windows) by default. +This process generates a Makefile(Linux/*BSD/MacOS), Visual Studio solution(Windows) by default. Run `make` to build Git on Linux/*BSD/MacOS. Open git.sln on Windows and build Git. -NOTE: By default CMake uses Makefile as the build tool on Linux and Visual Studio in Windows, -to use another tool say `ninja` add this to the command line when configuring. -`-G Ninja` - NOTE: By default CMake will install vcpkg locally to your source tree on configuration, to avoid this, add `-DNO_VCPKG=TRUE` to the command line when configuring. +The Visual Studio default generator changed in v16.6 from its Visual Studio +implemenation to `Ninja` This required changes to many CMake scripts. + ]] cmake_minimum_required(VERSION 3.14) From 7de755ee6e79da34e96498af6a5afce22302dbc5 Mon Sep 17 00:00:00 2001 From: Dennis Ameling Date: Mon, 19 Jul 2021 13:02:16 +0200 Subject: [PATCH 079/248] cmake(): allow setting HOST_CPU for cross-compilation Git's regular Makefile mentions that HOST_CPU should be defined when cross-compiling Git: https://github.com/git-for-windows/git/blob/37796bca76ef4180c39ee508ca3e42c0777ba444/Makefile#L438-L439 This is then used to set the GIT_HOST_CPU variable when compiling Git: https://github.com/git-for-windows/git/blob/37796bca76ef4180c39ee508ca3e42c0777ba444/Makefile#L1337-L1341 Then, when the user runs `git version --build-options`, it returns that value: https://github.com/git-for-windows/git/blob/37796bca76ef4180c39ee508ca3e42c0777ba444/help.c#L658 This commit adds the same functionality to the CMake configuration. Users can now set -DHOST_CPU= to set the target architecture. Signed-off-by: Dennis Ameling --- .github/workflows/main.yml | 2 +- contrib/buildsystems/CMakeLists.txt | 9 ++++++++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index c9905bdeb4139c..fdb7e937a57767 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -197,7 +197,7 @@ jobs: shell: bash run: | cmake `pwd`/contrib/buildsystems/ -DCMAKE_PREFIX_PATH=`pwd`/compat/vcbuild/vcpkg/installed/${{ matrix.arch }}-windows \ - -DNO_GETTEXT=YesPlease -DPERL_TESTS=OFF -DPYTHON_TESTS=OFF -DCURL_NO_CURL_CMAKE=ON -DCMAKE_GENERATOR_PLATFORM=${{ matrix.arch }} -DVCPKG_ARCH=${{ matrix.arch }}-windows + -DNO_GETTEXT=YesPlease -DPERL_TESTS=OFF -DPYTHON_TESTS=OFF -DCURL_NO_CURL_CMAKE=ON -DCMAKE_GENERATOR_PLATFORM=${{ matrix.arch }} -DVCPKG_ARCH=${{ matrix.arch }}-windows -DHOST_CPU=${{ matrix.arch }} - name: MSBuild run: msbuild git.sln -property:Configuration=Release -property:Platform=${{ matrix.arch }} -maxCpuCount:4 -property:PlatformToolset=v142 - name: bundle artifact tar diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index 093a6b7e186041..46fe6c3a5fb938 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -212,7 +212,14 @@ endif() #default behaviour include_directories(${CMAKE_SOURCE_DIR}) -add_compile_definitions(GIT_HOST_CPU="${CMAKE_SYSTEM_PROCESSOR}") + +# When cross-compiling, define HOST_CPU as the canonical name of the CPU on +# which the built Git will run (for instance "x86_64"). +if(NOT HOST_CPU) + add_compile_definitions(GIT_HOST_CPU="${CMAKE_SYSTEM_PROCESSOR}") +else() + add_compile_definitions(GIT_HOST_CPU="${HOST_CPU}") +endif() add_compile_definitions(SHA256_BLK INTERNAL_QSORT RUNTIME_PREFIX) add_compile_definitions(NO_OPENSSL SHA1_DC SHA1DC_NO_STANDARD_INCLUDES SHA1DC_INIT_SAFE_HASH_DEFAULT=0 From ad55d1431d71e20f2ef47e04f22168663b165ad4 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 2 Apr 2021 22:50:54 +0200 Subject: [PATCH 080/248] mingw: allow for longer paths in `parse_interpreter()` MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As reported in https://github.com/newren/git-filter-repo/pull/225, it looks like 99 bytes is not really sufficient to represent e.g. the full path to Python when installed via Windows Store (and this path is used in the hasb bang line when installing scripts via `pip`). Let's increase it to what is probably the maximum sensible path size: MAX_PATH. This makes `parse_interpreter()` in line with what `lookup_prog()` handles. Signed-off-by: Johannes Schindelin Signed-off-by: Vilius Šumskas --- compat/mingw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..e8871d9111d939 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1363,7 +1363,7 @@ static const char *quote_arg_msys2(const char *arg) static const char *parse_interpreter(const char *cmd) { - static char buf[100]; + static char buf[MAX_PATH]; char *p, *opt; ssize_t n; /* read() can return negative values */ int fd; From 8eeef6f873100193a35fdffe10d145387ad04c2f Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 17 May 2021 10:46:52 +0200 Subject: [PATCH 081/248] compat/vcbuild: document preferred way to build in Visual Studio We used to have that `make vcxproj` hack, but a hack it is. In the meantime, we have a much cleaner solution: using CMake, either explicitly, or even more conveniently via Visual Studio's built-in CMake support (simply open Git's top-level directory via File>Open>Folder...). Let's let the `README` reflect this. Signed-off-by: Johannes Schindelin --- compat/vcbuild/README | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/compat/vcbuild/README b/compat/vcbuild/README index 29ec1d0f104b80..5c71ea2daa4017 100644 --- a/compat/vcbuild/README +++ b/compat/vcbuild/README @@ -37,27 +37,17 @@ The Steps to Build Git with VS2015 or VS2017 from the command line. ================================================================ -Alternatively, run `make vcxproj` and then load the generated `git.sln` in -Visual Studio. The initial build will install the vcpkg system and build the +Alternatively, just open Git's top-level directory in Visual Studio, via +`File>Open>Folder...`. This will use CMake internally to generate the +project definitions. It will also install the vcpkg system and build the dependencies automatically. This will take a while. -Instead of generating the `git.sln` file yourself (which requires a full Git -for Windows SDK), you may want to consider fetching the `vs/master` branch of -https://github.com/git-for-windows/git instead (which is updated automatically -via CI running `make vcxproj`). The `vs/master` branch does not require a Git -for Windows to build, but you can run the test scripts in a regular Git Bash. - -Note that `make vcxproj` will automatically add and commit the generated `.sln` -and `.vcxproj` files to the repo. This is necessary to allow building a -fully-testable Git in Visual Studio, where a regular Git Bash can be used to -run the test scripts (as opposed to a full Git for Windows SDK): a number of -build targets, such as Git commands implemented as Unix shell scripts (where -`@@SHELL_PATH@@` and other placeholders are interpolated) require a full-blown -Git for Windows SDK (which is about 10x the size of a regular Git for Windows -installation). - -If your plan is to open a Pull Request with Git for Windows, it is a good idea -to drop this commit before submitting. +You can also generate the Visual Studio solution manually by downloading +and running CMake explicitly rather than letting Visual Studio doing +that implicitly. + +Another, deprecated option is to run `make vcxproj`. This option is +superseded by the CMake-based build, and will be removed at some point. ================================================================ The Steps of Build Git with VS2008 From cc69694f3a510c793a3c7d15fe86e226f60bd495 Mon Sep 17 00:00:00 2001 From: Pascal Muller Date: Wed, 23 Jun 2021 21:21:10 +0200 Subject: [PATCH 082/248] http: optionally send SSL client certificate This adds support for a new http.sslAutoClientCert config value. In cURL 7.77 or later the schannel backend does not automatically send client certificates from the Windows Certificate Store anymore. This config value is only used if http.sslBackend is set to "schannel", and can be used to opt in to the old behavior and force cURL to send client certificates. This fixes https://github.com/git-for-windows/git/issues/3292 Signed-off-by: Pascal Muller --- Documentation/config/http.adoc | 5 +++++ git-curl-compat.h | 8 ++++++++ http.c | 24 +++++++++++++++++++++--- 3 files changed, 34 insertions(+), 3 deletions(-) diff --git a/Documentation/config/http.adoc b/Documentation/config/http.adoc index 9122c5dc23ea1a..7fd001206ded22 100644 --- a/Documentation/config/http.adoc +++ b/Documentation/config/http.adoc @@ -249,6 +249,11 @@ http.schannelUseSSLCAInfo:: when the `schannel` backend was configured via `http.sslBackend`, unless `http.schannelUseSSLCAInfo` overrides this behavior. +http.sslAutoClientCert:: + As of cURL v7.77.0, the Secure Channel backend won't automatically + send client certificates from the Windows Certificate Store anymore. + To opt in to the old behavior, http.sslAutoClientCert can be set. + http.pinnedPubkey:: Public key of the https service. It may either be the filename of a PEM or DER encoded public key file or a string starting with diff --git a/git-curl-compat.h b/git-curl-compat.h index 659e5a3875e3d6..ecc2e742922313 100644 --- a/git-curl-compat.h +++ b/git-curl-compat.h @@ -37,6 +37,14 @@ #define GIT_CURL_NEED_TRANSFER_ENCODING_HEADER #endif +/** + * CURLSSLOPT_AUTO_CLIENT_CERT was added in 7.77.0, released in May + * 2021. + */ +#if LIBCURL_VERSION_NUM >= 0x074d00 +#define GIT_CURL_HAVE_CURLSSLOPT_AUTO_CLIENT_CERT +#endif + /** * CURLOPT_PROTOCOLS_STR and CURLOPT_REDIR_PROTOCOLS_STR were added in 7.85.0, * released in August 2022. diff --git a/http.c b/http.c index 3d00a9804c06ea..18508dd7f41068 100644 --- a/http.c +++ b/http.c @@ -162,6 +162,8 @@ static long http_schannel_check_revoke_mode = */ static int http_schannel_use_ssl_cainfo; +static int http_auto_client_cert; + static int always_auth_proactively(void) { return http_proactive_auth != PROACTIVE_AUTH_NONE && @@ -450,6 +452,11 @@ static int http_options(const char *var, const char *value, return 0; } + if (!strcmp("http.sslautoclientcert", var)) { + http_auto_client_cert = git_config_bool(var, value); + return 0; + } + if (!strcmp("http.minsessions", var)) { min_curl_sessions = git_config_int(var, value, ctx->kvi); if (min_curl_sessions > 1) @@ -1074,9 +1081,20 @@ static CURL *get_curl_handle(void) } #endif - if (http_ssl_backend && !strcmp("schannel", http_ssl_backend) && - http_schannel_check_revoke_mode) { - curl_easy_setopt(result, CURLOPT_SSL_OPTIONS, http_schannel_check_revoke_mode); + if (http_ssl_backend && !strcmp("schannel", http_ssl_backend)) { + long ssl_options = 0; + if (http_schannel_check_revoke_mode) { + ssl_options |= http_schannel_check_revoke_mode; + } + + if (http_auto_client_cert) { +#ifdef GIT_CURL_HAVE_CURLSSLOPT_AUTO_CLIENT_CERT + ssl_options |= CURLSSLOPT_AUTO_CLIENT_CERT; +#endif + } + + if (ssl_options) + curl_easy_setopt(result, CURLOPT_SSL_OPTIONS, ssl_options); } if (http_proactive_auth != PROACTIVE_AUTH_NONE) From a2fde069fb3e537f749091c1570be2cb986eda9d Mon Sep 17 00:00:00 2001 From: Victoria Dye Date: Thu, 5 Aug 2021 19:11:59 -0400 Subject: [PATCH 083/248] ci: run `contrib/subtree` tests in CI builds Because `git subtree` (unlike most other `contrib` modules) is included as part of the standard release of Git for Windows, its stability should be verified as consistently as it is for the rest of git. By including the `git subtree` tests in the CI workflow, these tests are as much of a gate to merging and indicator of stability as the standard test suite. Signed-off-by: Victoria Dye --- ci/run-build-and-tests.sh | 4 ++++ ci/run-test-slice.sh | 3 +++ 2 files changed, 7 insertions(+) diff --git a/ci/run-build-and-tests.sh b/ci/run-build-and-tests.sh index 01823fd0f140bb..5e94ff33830a16 100755 --- a/ci/run-build-and-tests.sh +++ b/ci/run-build-and-tests.sh @@ -72,5 +72,9 @@ case "$jobname" in ;; esac +case " $MAKE_TARGETS " in +*" all "*) make -C contrib/subtree test;; +esac + check_unignored_build_artifacts save_good_tree diff --git a/ci/run-test-slice.sh b/ci/run-test-slice.sh index 0444c79c023c82..6e21260e17543b 100755 --- a/ci/run-test-slice.sh +++ b/ci/run-test-slice.sh @@ -15,4 +15,7 @@ if [ "$1" == "0" ] ; then group "Run unit tests" make --quiet -C t unit-tests-test-tool fi +# Run the git subtree tests only if main tests succeeded +test 0 != "$1" || make -C contrib/subtree test + check_unignored_build_artifacts From 818e3bd1ee1356d63d199df9f0a4a3db8a9320f3 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Sun, 31 Oct 2021 23:15:13 +0000 Subject: [PATCH 084/248] hash-object: demonstrate a >4GB/LLP64 problem On LLP64 systems, such as Windows, the size of `long`, `int`, etc. is only 32 bits (for backward compatibility). Git's use of `unsigned long` for file memory sizes in many places, rather than size_t, limits the handling of large files on LLP64 systems (commonly given as `>4GB`). Provide a minimum test for handling a >4GB file. The `hash-object` command, with the `--literally` and without `-w` option avoids writing the object, either loose or packed. This avoids the code paths hitting the `bigFileThreshold` config test code, the zlib code, and the pack code. Subsequent patches will walk the test's call chain, converting types to `size_t` (which is larger in LLP64 data models) where appropriate. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- t/t1007-hash-object.sh | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh index de076293b62a76..7867fd1dbf940c 100755 --- a/t/t1007-hash-object.sh +++ b/t/t1007-hash-object.sh @@ -49,6 +49,9 @@ test_expect_success 'setup' ' example sha1:ddd3f836d3e3fbb7ae289aa9ae83536f76956399 example sha256:b44fe1fe65589848253737db859bd490453510719d7424daab03daf0767b85ae + + large5GB sha1:0be2be10a4c8764f32c4bf372a98edc731a4b204 + large5GB sha256:dc18ca621300c8d3cfa505a275641ebab00de189859e022a975056882d313e64 EOF ' @@ -258,4 +261,12 @@ test_expect_success '--stdin outside of repository (uses default hash)' ' test_cmp expect actual ' +test_expect_failure EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ + 'files over 4GB hash literally' ' + test-tool genzeros $((5*1024*1024*1024)) >big && + test_oid large5GB >expect && + git hash-object --stdin --literally actual && + test_cmp expect actual +' + test_done From c06f4280b62a9503020728e81b0938b2552e7fa6 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Fri, 12 Nov 2021 21:14:50 +0000 Subject: [PATCH 085/248] object-file.c: use size_t for header lengths Continue walking the code path for the >4GB `hash-object --literally` test. The `hash_object_file_literally()` function internally uses both `hash_object_file()` and `write_object_file_prepare()`. Both function signatures use `unsigned long` rather than `size_t` for the mem buffer sizes. Use `size_t` instead, for LLP64 compatibility. While at it, convert those function's object's header buffer length to `size_t` for consistency. The value is already upcast to `uintmax_t` for print format compatibility. Note: The hash-object test still does not pass. A subsequent commit continues to walk the call tree's lower level hash functions to identify further fixes. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- object-file.c | 14 +++++++------- object-file.h | 4 ++-- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/object-file.c b/object-file.c index 2bc36ab3ee8cbf..af24a5dc79303a 100644 --- a/object-file.c +++ b/object-file.c @@ -503,7 +503,7 @@ int loose_object_info(struct repository *r, static void hash_object_body(const struct git_hash_algo *algo, struct git_hash_ctx *c, const void *buf, unsigned long len, struct object_id *oid, - char *hdr, int *hdrlen) + char *hdr, size_t *hdrlen) { algo->init_fn(c); git_hash_update(c, hdr, *hdrlen); @@ -512,9 +512,9 @@ static void hash_object_body(const struct git_hash_algo *algo, struct git_hash_c } static void write_object_file_prepare(const struct git_hash_algo *algo, - const void *buf, unsigned long len, + const void *buf, size_t len, enum object_type type, struct object_id *oid, - char *hdr, int *hdrlen) + char *hdr, size_t *hdrlen) { struct git_hash_ctx c; @@ -657,11 +657,11 @@ int finalize_object_file_flags(struct repository *repo, } void hash_object_file(const struct git_hash_algo *algo, const void *buf, - unsigned long len, enum object_type type, + size_t len, enum object_type type, struct object_id *oid) { char hdr[MAX_HEADER_LEN]; - int hdrlen = sizeof(hdr); + size_t hdrlen = sizeof(hdr); write_object_file_prepare(algo, buf, len, type, oid, hdr, &hdrlen); } @@ -1037,7 +1037,7 @@ int stream_loose_object(struct odb_source *source, } int write_object_file(struct odb_source *source, - const void *buf, unsigned long len, + const void *buf, size_t len, enum object_type type, struct object_id *oid, struct object_id *compat_oid_in, unsigned flags) { @@ -1045,7 +1045,7 @@ int write_object_file(struct odb_source *source, const struct git_hash_algo *compat = source->odb->repo->compat_hash_algo; struct object_id compat_oid; char hdr[MAX_HEADER_LEN]; - int hdrlen = sizeof(hdr); + size_t hdrlen = sizeof(hdr); /* Generate compat_oid */ if (compat) { diff --git a/object-file.h b/object-file.h index 15d97630d3b11b..dc24393a9ebe1c 100644 --- a/object-file.h +++ b/object-file.h @@ -147,7 +147,7 @@ struct object_info; int parse_loose_header(const char *hdr, struct object_info *oi); int write_object_file(struct odb_source *source, - const void *buf, unsigned long len, + const void *buf, size_t len, enum object_type type, struct object_id *oid, struct object_id *compat_oid_in, unsigned flags); @@ -197,7 +197,7 @@ int finalize_object_file_flags(struct repository *repo, enum finalize_object_file_flags flags); void hash_object_file(const struct git_hash_algo *algo, const void *buf, - unsigned long len, enum object_type type, + size_t len, enum object_type type, struct object_id *oid); /* Helper to check and "touch" a file */ From 0d012084f69211bae60ffed383dfc0a87d37ff87 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Fri, 12 Nov 2021 21:16:51 +0000 Subject: [PATCH 086/248] hash algorithms: use size_t for section lengths Continue walking the code path for the >4GB `hash-object --literally` test to the hash algorithm step for LLP64 systems. This patch lets the SHA1DC code use `size_t`, making it compatible with LLP64 data models (as used e.g. by Windows). The interested reader of this patch will note that we adjust the signature of the `git_SHA1DCUpdate()` function without updating _any_ call site. This certainly puzzled at least one reviewer already, so here is an explanation: This function is never called directly, but always via the macro `platform_SHA1_Update`, which is usually called via the macro `git_SHA1_Update`. However, we never call `git_SHA1_Update()` directly in `struct git_hash_algo`. Instead, we call `git_hash_sha1_update()`, which is defined thusly: static void git_hash_sha1_update(git_hash_ctx *ctx, const void *data, size_t len) { git_SHA1_Update(&ctx->sha1, data, len); } i.e. it contains an implicit downcast from `size_t` to `unsigned long` (before this here patch). With this patch, there is no downcast anymore. With this patch, finally, the t1007-hash-object.sh "files over 4GB hash literally" test case is fixed. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- object-file.c | 4 ++-- sha1dc_git.c | 3 +-- sha1dc_git.h | 2 +- t/t1007-hash-object.sh | 2 +- 4 files changed, 5 insertions(+), 6 deletions(-) diff --git a/object-file.c b/object-file.c index af24a5dc79303a..10ca56e56248d3 100644 --- a/object-file.c +++ b/object-file.c @@ -501,7 +501,7 @@ int loose_object_info(struct repository *r, } static void hash_object_body(const struct git_hash_algo *algo, struct git_hash_ctx *c, - const void *buf, unsigned long len, + const void *buf, size_t len, struct object_id *oid, char *hdr, size_t *hdrlen) { @@ -521,7 +521,7 @@ static void write_object_file_prepare(const struct git_hash_algo *algo, /* Generate the header */ *hdrlen = format_object_header(hdr, *hdrlen, type, len); - /* Sha1.. */ + /* Hash (function pointers) computation */ hash_object_body(algo, &c, buf, len, oid, hdr, hdrlen); } diff --git a/sha1dc_git.c b/sha1dc_git.c index 9b675a046ee699..fe58d7962a30c9 100644 --- a/sha1dc_git.c +++ b/sha1dc_git.c @@ -27,10 +27,9 @@ void git_SHA1DCFinal(unsigned char hash[20], SHA1_CTX *ctx) /* * Same as SHA1DCUpdate, but adjust types to match git's usual interface. */ -void git_SHA1DCUpdate(SHA1_CTX *ctx, const void *vdata, unsigned long len) +void git_SHA1DCUpdate(SHA1_CTX *ctx, const void *vdata, size_t len) { const char *data = vdata; - /* We expect an unsigned long, but sha1dc only takes an int */ while (len > INT_MAX) { SHA1DCUpdate(ctx, data, INT_MAX); data += INT_MAX; diff --git a/sha1dc_git.h b/sha1dc_git.h index f6f880cabea382..0bcf1aa84b7241 100644 --- a/sha1dc_git.h +++ b/sha1dc_git.h @@ -15,7 +15,7 @@ void git_SHA1DCInit(SHA1_CTX *); #endif void git_SHA1DCFinal(unsigned char [20], SHA1_CTX *); -void git_SHA1DCUpdate(SHA1_CTX *ctx, const void *data, unsigned long len); +void git_SHA1DCUpdate(SHA1_CTX *ctx, const void *data, size_t len); #define platform_SHA_IS_SHA1DC /* used by "test-tool sha1-is-sha1dc" */ diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh index 7867fd1dbf940c..10382a815e4c14 100755 --- a/t/t1007-hash-object.sh +++ b/t/t1007-hash-object.sh @@ -261,7 +261,7 @@ test_expect_success '--stdin outside of repository (uses default hash)' ' test_cmp expect actual ' -test_expect_failure EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ +test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ 'files over 4GB hash literally' ' test-tool genzeros $((5*1024*1024*1024)) >big && test_oid large5GB >expect && From 9836977ce15fcfcca32e28df181d5f72165c5aee Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Sat, 24 Apr 2021 11:09:58 +0100 Subject: [PATCH 087/248] .gitignore: add Visual Studio CMakeSetting.json file The CMakeSettings.json file is tool generated. Developers may track it should they provide additional settings. Signed-off-by: Philip Oakley --- .gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/.gitignore b/.gitignore index 04c444404e4ba8..735bb036415bd5 100644 --- a/.gitignore +++ b/.gitignore @@ -252,5 +252,6 @@ Release/ /git.VC.db *.dSYM /contrib/buildsystems/out +CMakeSettings.json /contrib/libgit-rs/target /contrib/libgit-sys/target From 557305d16999fdba2d2366d92737444efd31de52 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Mon, 6 Dec 2021 22:26:50 +0000 Subject: [PATCH 088/248] hash-object --stdin: verify that it works with >4GB/LLP64 Just like the `hash-object --literally` code path, the `--stdin` code path also needs to use `size_t` instead of `unsigned long` to represent memory sizes, otherwise it would cause problems on platforms using the LLP64 data model (such as Windows). To limit the scope of the test case, the object is explicitly not written to the object store, nor are any filters applied. The `big` file from the previous test case is reused to save setup time; To avoid relying on that side effect, it is generated if it does not exist (e.g. when running via `sh t1007-*.sh --long --run=1,41`). Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- t/t1007-hash-object.sh | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh index 10382a815e4c14..59efee3affcff4 100755 --- a/t/t1007-hash-object.sh +++ b/t/t1007-hash-object.sh @@ -269,4 +269,12 @@ test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ test_cmp expect actual ' +test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ + 'files over 4GB hash correctly via --stdin' ' + { test -f big || test-tool genzeros $((5*1024*1024*1024)) >big; } && + test_oid large5GB >expect && + git hash-object --stdin actual && + test_cmp expect actual +' + test_done From b4d980593fd94e595ecf2e63e47bde33b0fa2292 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Thu, 22 Apr 2021 11:11:38 +0100 Subject: [PATCH 089/248] CMakeLists: add default "x64-windows" arch for Visual Studio In Git-for-Windows, work on using ARM64 has progressed. The commit 2d94b77b27 (cmake: allow building for Windows/ARM64, 2020-12-04) failed to notice that /compat/vcbuild/vcpkg_install.bat will default to using the "x64-windows" architecture for the vcpkg installation if not set, but CMake is not told of this default. Commit 635b6d99b3 (vcbuild: install ARM64 dependencies when building ARM64 binaries, 2020-01-31) later updated vcpkg_install.bat to accept an arch (%1) parameter, but retained the default. This default is neccessary for the use case where the project directory is opened directly in Visual Studio, which will find and build a CMakeLists.txt file without any parameters, thus expecting use of the default setting. Also Visual studio will generate internal .sln solution and .vcxproj project files needed for some extension tools. Inform users of the additional .sln/.vcxproj generation. ** How to test: rm -rf '.vs' # remove old visual studio settings rm -rf 'compat/vcbuild/vcpkg' # remove any vcpkg downloads rm -rf 'contrib/buildsystems/out' # remove builds & CMake artifacts with a fresh Visual Studio Community Edition, File>>Open>>(git *folder*) to load the project (which will take some time!). check for successful compilation. The implicit .sln (etc.) are in the hidden .vs directory created by Visual Studio. Signed-off-by: Philip Oakley --- contrib/buildsystems/CMakeLists.txt | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index f8ef01e57702ac..ce84857d7b285b 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -71,6 +71,10 @@ if(USE_VCPKG) message("Initializing vcpkg and building the Git's dependencies (this will take a while...)") execute_process(COMMAND ${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg_install.bat ${VCPKG_ARCH}) endif() + if(NOT EXISTS ${VCPKG_ARCH}) + message("VCPKG_ARCH: unset, using 'x64-windows'") + set(VCPKG_ARCH "x64-windows") # default from vcpkg_install.bat + endif() list(APPEND CMAKE_PREFIX_PATH "${VCPKG_DIR}/installed/${VCPKG_ARCH}") # In the vcpkg edition, we need this to be able to link to libcurl From 563c0d6360de896414454d33479c1a6e1e92af64 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Mon, 6 Dec 2021 22:42:46 +0000 Subject: [PATCH 090/248] hash-object: add another >4GB/LLP64 test case To complement the `--stdin` and `--literally` test cases that verify that we can hash files larger than 4GB on 64-bit platforms using the LLP64 data model, here is a test case that exercises `hash-object` _without_ any options. Just as before, we use the `big` file from the previous test case if it exists to save on setup time, otherwise generate it. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- t/t1007-hash-object.sh | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh index 59efee3affcff4..f2722380ee1436 100755 --- a/t/t1007-hash-object.sh +++ b/t/t1007-hash-object.sh @@ -277,4 +277,12 @@ test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ test_cmp expect actual ' +test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ + 'files over 4GB hash correctly' ' + { test -f big || test-tool genzeros $((5*1024*1024*1024)) >big; } && + test_oid large5GB >expect && + git hash-object -- big >actual && + test_cmp expect actual +' + test_done From ca1adae4d8773420d91cb1038145694293532078 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Mon, 10 May 2021 16:47:40 +0100 Subject: [PATCH 091/248] CMake: show Win32 and Generator_platform build-option values Ensure key CMake option values are part of the CMake output to facilitate user support when tool updates impact the wider CMake actions, particularly ongoing 'improvements' in Visual Studio. These CMake displays perform the same function as the build-options.txt provided in the main Git for Windows. CMake is already chatty. The setting of CMAKE_EXPORT_COMPILE_COMMANDS is also reported. Include the environment's CMAKE_EXPORT_COMPILE_COMMANDS value which may have been propogated to CMake's internal value. Testing the CMAKE_EXPORT_COMPILE_COMMANDS processing can be difficult in the Visual Studio environment, as it may be cached in many places. The 'environment' may include the OS, the user shell, CMake's own environment, along with the Visual Studio presets and caches. See previous commit for arefacts that need removing for a clean test. Signed-off-by: Philip Oakley --- contrib/buildsystems/CMakeLists.txt | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index ce84857d7b285b..0ea9c22e7fb257 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -63,10 +63,20 @@ endif() if(NOT DEFINED CMAKE_EXPORT_COMPILE_COMMANDS) set(CMAKE_EXPORT_COMPILE_COMMANDS TRUE) + message("settting CMAKE_EXPORT_COMPILE_COMMANDS: ${CMAKE_EXPORT_COMPILE_COMMANDS}") endif() if(USE_VCPKG) set(VCPKG_DIR "${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg") + message("WIN32: ${WIN32}") # show its underlying text values + message("VCPKG_DIR: ${VCPKG_DIR}") + message("VCPKG_ARCH: ${VCPKG_ARCH}") # maybe unset + message("MSVC: ${MSVC}") + message("CMAKE_GENERATOR: ${CMAKE_GENERATOR}") + message("CMAKE_CXX_COMPILER_ID: ${CMAKE_CXX_COMPILER_ID}") + message("CMAKE_GENERATOR_PLATFORM: ${CMAKE_GENERATOR_PLATFORM}") + message("CMAKE_EXPORT_COMPILE_COMMANDS: ${CMAKE_EXPORT_COMPILE_COMMANDS}") + message("ENV(CMAKE_EXPORT_COMPILE_COMMANDS): $ENV{CMAKE_EXPORT_COMPILE_COMMANDS}") if(NOT EXISTS ${VCPKG_DIR}) message("Initializing vcpkg and building the Git's dependencies (this will take a while...)") execute_process(COMMAND ${CMAKE_SOURCE_DIR}/compat/vcbuild/vcpkg_install.bat ${VCPKG_ARCH}) From 52942df6fb615f890e87ccd8e284b20b1d321bbf Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 8 Sep 2021 13:05:42 +0200 Subject: [PATCH 092/248] init: do parse _all_ core.* settings early In Git for Windows, `has_symlinks` is set to 0 by default. Therefore, we need to parse the config setting `core.symlinks` to know if it has been set to `true`. In `git init`, we must do that before copying the templates because they might contain symbolic links. Even if the support for symbolic links on Windows has not made it to upstream Git yet, we really should make sure that all the `core.*` settings are parsed before proceeding, as they might very well change the behavior of `git init` in a way the user intended. This fixes https://github.com/git-for-windows/git/issues/3414 Signed-off-by: Johannes Schindelin --- environment.c | 4 ++-- environment.h | 2 ++ setup.c | 2 +- 3 files changed, 5 insertions(+), 3 deletions(-) diff --git a/environment.c b/environment.c index ae1427bb9e8e62..921ed89a2afc2a 100644 --- a/environment.c +++ b/environment.c @@ -317,8 +317,8 @@ static enum fsync_component parse_fsync_components(const char *var, const char * return (current & ~negative) | positive; } -static int git_default_core_config(const char *var, const char *value, - const struct config_context *ctx, void *cb) +int git_default_core_config(const char *var, const char *value, + const struct config_context *ctx, void *cb) { /* This needs a better name */ if (!strcmp(var, "core.filemode")) { diff --git a/environment.h b/environment.h index 8cfce41015b3c8..862d0f50a45f52 100644 --- a/environment.h +++ b/environment.h @@ -106,6 +106,8 @@ const char *strip_namespace(const char *namespaced_ref); int git_default_config(const char *, const char *, const struct config_context *, void *); +int git_default_core_config(const char *var, const char *value, + const struct config_context *ctx, void *cb); /* * TODO: All the below state either explicitly or implicitly relies on diff --git a/setup.c b/setup.c index 98ddbf377f923b..ff512074ffc8fd 100644 --- a/setup.c +++ b/setup.c @@ -2610,7 +2610,7 @@ int init_db(const char *git_dir, const char *real_git_dir, * have set up the repository format such that we can evaluate * includeIf conditions correctly in the case of re-initialization. */ - repo_config(the_repository, platform_core_config, NULL); + repo_config(the_repository, git_default_core_config, NULL); safe_create_dir(the_repository, git_dir, 0); From 862326906a5b677c902a3d0a0f183ffb9326e801 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Wed, 13 Apr 2022 14:49:17 -0400 Subject: [PATCH 093/248] setup: properly use "%(prefix)/" when in WSL Signed-off-by: Derrick Stolee --- setup.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/setup.c b/setup.c index 98ddbf377f923b..0f5a7d6f68eb2c 100644 --- a/setup.c +++ b/setup.c @@ -1785,10 +1785,19 @@ const char *setup_git_directory_gently(int *nongit_ok) break; case GIT_DIR_INVALID_OWNERSHIP: if (!nongit_ok) { + struct strbuf prequoted = STRBUF_INIT; struct strbuf quoted = STRBUF_INIT; strbuf_complete(&report, '\n'); - sq_quote_buf_pretty("ed, dir.buf); + +#ifdef __MINGW32__ + if (dir.buf[0] == '/') + strbuf_addstr(&prequoted, "%(prefix)/"); +#endif + + strbuf_add(&prequoted, dir.buf, dir.len); + sq_quote_buf_pretty("ed, prequoted.buf); + die(_("detected dubious ownership in repository at '%s'\n" "%s" "To add an exception for this directory, call:\n" From 9405bf489a22315dbe329acdf0f14376a63cb725 Mon Sep 17 00:00:00 2001 From: Philip Oakley Date: Tue, 7 Dec 2021 09:53:41 +0000 Subject: [PATCH 094/248] hash-object: add a >4GB/LLP64 test case using filtered input To verify that the `clean` side of the `clean`/`smudge` filter code is correct with regards to LLP64 (read: to ensure that `size_t` is used instead of `unsigned long`), here is a test case using a trivial filter, specifically _not_ writing anything to the object store to limit the scope of the test case. As in previous commits, the `big` file from previous test cases is reused if available, to save setup time, otherwise re-generated. Signed-off-by: Philip Oakley Signed-off-by: Johannes Schindelin --- t/t1007-hash-object.sh | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh index f2722380ee1436..841a6671d1a3c1 100755 --- a/t/t1007-hash-object.sh +++ b/t/t1007-hash-object.sh @@ -285,4 +285,16 @@ test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ test_cmp expect actual ' +# This clean filter does nothing, other than excercising the interface. +# We ensure that cleaning doesn't mangle large files on 64-bit Windows. +test_expect_success EXPENSIVE,SIZE_T_IS_64BIT,!LONG_IS_64BIT \ + 'hash filtered files over 4GB correctly' ' + { test -f big || test-tool genzeros $((5*1024*1024*1024)) >big; } && + test_oid large5GB >expect && + test_config filter.null-filter.clean "cat" && + echo "big filter=null-filter" >.gitattributes && + git hash-object -- big >actual && + test_cmp expect actual +' + test_done From c79e7d51f8de4efb84f02bb54c5e027626f37a7b Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Wed, 13 Apr 2022 14:54:43 -0400 Subject: [PATCH 095/248] compat/mingw.c: do not warn when failing to get owner In the case of Git for Windows (say, in a Git Bash window) running in a Windows Subsystem for Linux (WSL) directory, the GetNamedSecurityInfoW() call in is_path_owned_By_current_side() returns an error code other than ERROR_SUCCESS. This is consistent behavior across this boundary. In these cases, the owner would always be different because the WSL owner is a different entity than the Windows user. The change here is to suppress the error message that looks like this: error: failed to get owner for '//wsl.localhost/...' (1) Before this change, this warning happens for every Git command, regardless of whether the directory is marked with safe.directory. Signed-off-by: Derrick Stolee --- compat/mingw.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..959b0fb7993fc5 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2917,9 +2917,7 @@ int is_path_owned_by_current_sid(const char *path, struct strbuf *report) DACL_SECURITY_INFORMATION, &sid, NULL, NULL, NULL, &descriptor); - if (err != ERROR_SUCCESS) - error(_("failed to get owner for '%s' (%ld)"), path, err); - else if (sid && IsValidSid(sid)) { + if (err == ERROR_SUCCESS && sid && IsValidSid(sid)) { /* Now, verify that the SID matches the current user's */ static PSID current_user_sid; static HANDLE linked_token; From 6aaa3bd09a68432a154277505409991907c49079 Mon Sep 17 00:00:00 2001 From: Rafael Kitover Date: Tue, 12 Apr 2022 19:53:33 +0000 Subject: [PATCH 096/248] mingw: $env:TERM="xterm-256color" for newer OSes For Windows builds >= 15063 set $env:TERM to "xterm-256color" instead of "cygwin" because they have a more capable console system that supports this. Also set $env:COLORTERM="truecolor" if unset. $env:TERM is initialized so that ANSI colors in color.c work, see 29a3963484 (Win32: patch Windows environment on startup, 2012-01-15). See git-for-windows/git#3629 regarding problems caused by always setting $env:TERM="cygwin". This is the same heuristic used by the Cygwin runtime. Signed-off-by: Rafael Kitover Signed-off-by: Johannes Schindelin --- compat/mingw.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..c93cb3db1d1186 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2775,9 +2775,20 @@ static void setup_windows_environment(void) convert_slashes(tmp); } - /* simulate TERM to enable auto-color (see color.c) */ - if (!getenv("TERM")) - setenv("TERM", "cygwin", 1); + + /* + * Make sure TERM is set up correctly to enable auto-color + * (see color.c .) Use "cygwin" for older OS releases which + * works correctly with MSYS2 utilities on older consoles. + */ + if (!getenv("TERM")) { + if ((GetVersion() >> 16) < 15063) + setenv("TERM", "cygwin", 0); + else { + setenv("TERM", "xterm-256color", 0); + setenv("COLORTERM", "truecolor", 0); + } + } /* calculate HOME if not set */ if (!getenv("HOME")) { From 7cbf187ab536eeafca4d74a62bfe33e3a952d468 Mon Sep 17 00:00:00 2001 From: Christopher Degawa Date: Sat, 28 May 2022 14:53:54 -0500 Subject: [PATCH 097/248] winansi: check result and Buffer before using Name NtQueryObject under Wine can return a success but fill out no name. In those situations, Wine will set Buffer to NULL, and set result to the sizeof(OBJECT_NAME_INFORMATION). Running a command such as echo "$(git.exe --version 2>/dev/null)" will crash due to a NULL pointer dereference when the code attempts to null terminate the buffer, although, weirdly, removing the subshell or redirecting stdout to a file will not trigger the crash. Code has been added to also check Buffer and Length to ensure the check is as robust as possible due to the current behavior being fragile at best, and could potentially change in the future This code is based on the behavior of NtQueryObject under wine and reactos. Signed-off-by: Christopher Degawa --- compat/winansi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/compat/winansi.c b/compat/winansi.c index ac2ffb78691a7d..d28137a20b0bcc 100644 --- a/compat/winansi.c +++ b/compat/winansi.c @@ -575,6 +575,9 @@ static void detect_msys_tty(int fd) if (!NT_SUCCESS(NtQueryObject(h, ObjectNameInformation, buffer, sizeof(buffer) - 2, &result))) return; + if (result < sizeof(*nameinfo) || !nameinfo->Name.Buffer || + !nameinfo->Name.Length) + return; name = nameinfo->Name.Buffer; name[nameinfo->Name.Length / sizeof(*name)] = 0; From 1097e9bec95a5a5e32511e6178451314caa4f140 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E5=AD=99=E5=8D=93=E8=AF=86?= Date: Sun, 16 Jan 2022 03:38:33 +0800 Subject: [PATCH 098/248] Add config option `windows.appendAtomically` MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Atomic append on windows is only supported on local disk files, and it may cause errors in other situations, e.g. network file system. If that is the case, this config option should be used to turn atomic append off. Co-Authored-By: Johannes Schindelin Signed-off-by: 孙卓识 Signed-off-by: Johannes Schindelin --- Documentation/config.adoc | 2 ++ Documentation/config/windows.adoc | 4 ++++ compat/mingw.c | 36 ++++++++++++++++++++++++++++--- 3 files changed, 39 insertions(+), 3 deletions(-) create mode 100644 Documentation/config/windows.adoc diff --git a/Documentation/config.adoc b/Documentation/config.adoc index ae6dcb6a6c0607..3d0a80fbd848d1 100644 --- a/Documentation/config.adoc +++ b/Documentation/config.adoc @@ -557,4 +557,6 @@ include::config/versionsort.adoc[] include::config/web.adoc[] +include::config/windows.adoc[] + include::config/worktree.adoc[] diff --git a/Documentation/config/windows.adoc b/Documentation/config/windows.adoc new file mode 100644 index 00000000000000..fdaaf1c65504f3 --- /dev/null +++ b/Documentation/config/windows.adoc @@ -0,0 +1,4 @@ +windows.appendAtomically:: + By default, append atomic API is used on windows. But it works only with + local disk files, if you're working on a network file system, you should + set it false to turn it off. diff --git a/compat/mingw.c b/compat/mingw.c index 736a07a028ab4d..761f321b5b4b9d 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -8,6 +8,7 @@ #include "dir.h" #include "environment.h" #include "gettext.h" +#include "repository.h" #include "run-command.h" #include "strbuf.h" #include "symlinks.h" @@ -621,6 +622,7 @@ static int is_local_named_pipe_path(const char *filename) int mingw_open (const char *filename, int oflags, ...) { + static int append_atomically = -1; typedef int (*open_fn_t)(wchar_t const *wfilename, int oflags, ...); va_list args; unsigned mode; @@ -639,7 +641,16 @@ int mingw_open (const char *filename, int oflags, ...) return -1; } - if ((oflags & O_APPEND) && !is_local_named_pipe_path(filename)) + /* + * Only set append_atomically to default value(1) when repo is initialized + * and fail to get config value + */ + if (append_atomically < 0 && the_repository && the_repository->commondir && + repo_config_get_bool(the_repository, "windows.appendatomically", &append_atomically)) + append_atomically = 1; + + if (append_atomically && (oflags & O_APPEND) && + !is_local_named_pipe_path(filename)) open_fn = mingw_open_append; else if (!(oflags & ~(O_ACCMODE | O_NOINHERIT))) open_fn = mingw_open_existing; @@ -805,9 +816,28 @@ ssize_t mingw_write(int fd, const void *buf, size_t len) /* check if fd is a pipe */ HANDLE h = (HANDLE) _get_osfhandle(fd); - if (GetFileType(h) != FILE_TYPE_PIPE) + if (GetFileType(h) != FILE_TYPE_PIPE) { + if (orig == EINVAL) { + wchar_t path[MAX_PATH]; + DWORD ret = GetFinalPathNameByHandleW(h, path, + ARRAY_SIZE(path), 0); + UINT drive_type = ret > 0 && ret < ARRAY_SIZE(path) ? + GetDriveTypeW(path) : DRIVE_UNKNOWN; + + /* + * The default atomic append causes such an error on + * network file systems, in such a case, it should be + * turned off via config. + * + * `drive_type` of UNC path: DRIVE_NO_ROOT_DIR + */ + if (DRIVE_NO_ROOT_DIR == drive_type || DRIVE_REMOTE == drive_type) + warning("invalid write operation detected; you may try:\n" + "\n\tgit config windows.appendAtomically false"); + } + errno = orig; - else if (orig == EINVAL) + } else if (orig == EINVAL) errno = EPIPE; else { DWORD buf_size; From bf9ba85bbce57db55a8cb6d7c81a9f504ee0003a Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 4 Sep 2017 11:59:45 +0200 Subject: [PATCH 099/248] mingw: change core.fsyncObjectFiles = 1 by default MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From the documentation of said setting: This boolean will enable fsync() when writing object files. This is a total waste of time and effort on a filesystem that orders data writes properly, but can be useful for filesystems that do not use journalling (traditional UNIX filesystems) or that only journal metadata and not file contents (OS X’s HFS+, or Linux ext3 with "data=writeback"). The most common file system on Windows (NTFS) does not guarantee that order, therefore a sudden loss of power (or any other event causing an unclean shutdown) would cause corrupt files (i.e. files filled with NULs). Therefore we need to change the default. Note that the documentation makes it sound as if this causes really bad performance. In reality, writing loose objects is something that is done only rarely, and only a handful of files at a time. Signed-off-by: Johannes Schindelin --- compat/mingw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 761f321b5b4b9d..ea975bd4716aa2 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -16,6 +16,7 @@ #include "win32.h" #include "win32/lazyload.h" #include "wrapper.h" +#include "write-or-die.h" #include #include #include @@ -3280,6 +3281,7 @@ int wmain(int argc, const wchar_t **wargv) #endif maybe_redirect_std_handles(); + fsync_object_files = 1; /* determine size of argv and environ conversion buffer */ maxlen = wcslen(wargv[0]); From 66f453b9148434c7ef1d8ccce95e6a59bde00eb0 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sat, 6 May 2023 22:26:15 +0200 Subject: [PATCH 100/248] http: optionally load libcurl lazily This compile-time option allows to ask Git to load libcurl dynamically at runtime. Together with a follow-up patch that optionally overrides the file name depending on the `http.sslBackend` setting, this kicks open the door for installing multiple libcurl flavors side by side, and load the one corresponding to the (runtime-)configured SSL/TLS backend. Signed-off-by: Johannes Schindelin --- Makefile | 28 +++- compat/lazyload-curl.c | 364 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 385 insertions(+), 7 deletions(-) create mode 100644 compat/lazyload-curl.c diff --git a/Makefile b/Makefile index 1f444eec3c1d12..1257118cc39dbb 100644 --- a/Makefile +++ b/Makefile @@ -470,6 +470,11 @@ include shared.mak # # CURL_LDFLAGS=-lcurl # +# Define LAZYLOAD_LIBCURL to dynamically load the libcurl; This can be useful +# if Multiple libcurl versions exist (with different file names) that link to +# various SSL/TLS backends, to support the `http.sslBackend` runtime switch in +# such a scenario. +# # === Optional library: libpcre2 === # # Define USE_LIBPCRE if you have and want to use libpcre. Various @@ -1664,10 +1669,19 @@ else CURL_LIBCURL = endif - ifndef CURL_LDFLAGS - CURL_LDFLAGS = $(eval CURL_LDFLAGS := $$(shell $$(CURL_CONFIG) --libs))$(CURL_LDFLAGS) + ifdef LAZYLOAD_LIBCURL + LAZYLOAD_LIBCURL_OBJ = compat/lazyload-curl.o + OBJECTS += $(LAZYLOAD_LIBCURL_OBJ) + # The `CURL_STATICLIB` constant must be defined to avoid seeing the functions + # declared as DLL imports + CURL_CFLAGS = -DCURL_STATICLIB + CURL_LIBCURL = -ldl + else + ifndef CURL_LDFLAGS + CURL_LDFLAGS = $(eval CURL_LDFLAGS := $$(shell $$(CURL_CONFIG) --libs))$(CURL_LDFLAGS) + endif + CURL_LIBCURL += $(CURL_LDFLAGS) endif - CURL_LIBCURL += $(CURL_LDFLAGS) ifndef CURL_CFLAGS CURL_CFLAGS = $(eval CURL_CFLAGS := $$(shell $$(CURL_CONFIG) --cflags))$(CURL_CFLAGS) @@ -1688,7 +1702,7 @@ else endif ifdef USE_CURL_FOR_IMAP_SEND BASIC_CFLAGS += -DUSE_CURL_FOR_IMAP_SEND - IMAP_SEND_BUILDDEPS = http.o + IMAP_SEND_BUILDDEPS = http.o $(LAZYLOAD_LIBCURL_OBJ) IMAP_SEND_LDFLAGS += $(CURL_LIBCURL) endif ifndef NO_EXPAT @@ -2894,10 +2908,10 @@ git-imap-send$X: imap-send.o $(IMAP_SEND_BUILDDEPS) GIT-LDFLAGS $(GITLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) \ $(IMAP_SEND_LDFLAGS) $(LIBS) -git-http-fetch$X: http.o http-walker.o http-fetch.o GIT-LDFLAGS $(GITLIBS) +git-http-fetch$X: http.o http-walker.o http-fetch.o $(LAZYLOAD_LIBCURL_OBJ) GIT-LDFLAGS $(GITLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) \ $(CURL_LIBCURL) $(LIBS) -git-http-push$X: http.o http-push.o GIT-LDFLAGS $(GITLIBS) +git-http-push$X: http.o http-push.o $(LAZYLOAD_LIBCURL_OBJ) GIT-LDFLAGS $(GITLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) \ $(CURL_LIBCURL) $(EXPAT_LIBEXPAT) $(LIBS) @@ -2907,7 +2921,7 @@ $(REMOTE_CURL_ALIASES): $(REMOTE_CURL_PRIMARY) ln -s $< $@ 2>/dev/null || \ cp $< $@ -$(REMOTE_CURL_PRIMARY): remote-curl.o http.o http-walker.o GIT-LDFLAGS $(GITLIBS) +$(REMOTE_CURL_PRIMARY): remote-curl.o http.o http-walker.o $(LAZYLOAD_LIBCURL_OBJ) GIT-LDFLAGS $(GITLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) \ $(CURL_LIBCURL) $(EXPAT_LIBEXPAT) $(LIBS) diff --git a/compat/lazyload-curl.c b/compat/lazyload-curl.c new file mode 100644 index 00000000000000..f4e08f76dfcd7f --- /dev/null +++ b/compat/lazyload-curl.c @@ -0,0 +1,364 @@ +#include "../git-compat-util.h" +#include "../git-curl-compat.h" +#include + +/* + * The ABI version of libcurl is encoded in its shared libraries' file names. + * This ABI version has not changed since October 2006 and is unlikely to be + * changed in the future. See https://curl.se/libcurl/abi.html for details. + */ +#define LIBCURL_ABI_VERSION "4" + +typedef void (*func_t)(void); + +#ifdef __APPLE__ +#define LIBCURL_FILE_NAME(base) base "." LIBCURL_ABI_VERSION ".dylib" +#else +#define LIBCURL_FILE_NAME(base) base ".so." LIBCURL_ABI_VERSION +#endif + +static void *load_library(const char *name) +{ + return dlopen(name, RTLD_LAZY); +} + +static func_t load_function(void *handle, const char *name) +{ + /* + * Casting the return value of `dlsym()` to a function pointer is + * explicitly allowed in recent POSIX standards, but GCC complains + * about this in pedantic mode nevertheless. For more about this issue, + * see https://stackoverflow.com/q/31526876/1860823 and + * http://stackoverflow.com/a/36385690/1905491. + */ + func_t f; + *(void **)&f = dlsym(handle, name); + return f; +} + +typedef struct curl_version_info_data *(*curl_version_info_type)(CURLversion version); +static curl_version_info_type curl_version_info_func; + +typedef char *(*curl_easy_escape_type)(CURL *handle, const char *string, int length); +static curl_easy_escape_type curl_easy_escape_func; + +typedef void (*curl_free_type)(void *p); +static curl_free_type curl_free_func; + +typedef CURLcode (*curl_global_init_type)(long flags); +static curl_global_init_type curl_global_init_func; + +typedef CURLsslset (*curl_global_sslset_type)(curl_sslbackend id, const char *name, const curl_ssl_backend ***avail); +static curl_global_sslset_type curl_global_sslset_func; + +typedef void (*curl_global_cleanup_type)(void); +static curl_global_cleanup_type curl_global_cleanup_func; + +typedef CURLcode (*curl_global_trace_type)(const char *config); +static curl_global_trace_type curl_global_trace_func; + +typedef struct curl_slist *(*curl_slist_append_type)(struct curl_slist *list, const char *data); +static curl_slist_append_type curl_slist_append_func; + +typedef void (*curl_slist_free_all_type)(struct curl_slist *list); +static curl_slist_free_all_type curl_slist_free_all_func; + +typedef const char *(*curl_easy_strerror_type)(CURLcode error); +static curl_easy_strerror_type curl_easy_strerror_func; + +typedef CURLM *(*curl_multi_init_type)(void); +static curl_multi_init_type curl_multi_init_func; + +typedef CURLMcode (*curl_multi_add_handle_type)(CURLM *multi_handle, CURL *curl_handle); +static curl_multi_add_handle_type curl_multi_add_handle_func; + +typedef CURLMcode (*curl_multi_remove_handle_type)(CURLM *multi_handle, CURL *curl_handle); +static curl_multi_remove_handle_type curl_multi_remove_handle_func; + +typedef CURLMcode (*curl_multi_fdset_type)(CURLM *multi_handle, fd_set *read_fd_set, fd_set *write_fd_set, fd_set *exc_fd_set, int *max_fd); +static curl_multi_fdset_type curl_multi_fdset_func; + +typedef CURLMcode (*curl_multi_perform_type)(CURLM *multi_handle, int *running_handles); +static curl_multi_perform_type curl_multi_perform_func; + +typedef CURLMcode (*curl_multi_cleanup_type)(CURLM *multi_handle); +static curl_multi_cleanup_type curl_multi_cleanup_func; + +typedef CURLMsg *(*curl_multi_info_read_type)(CURLM *multi_handle, int *msgs_in_queue); +static curl_multi_info_read_type curl_multi_info_read_func; + +typedef const char *(*curl_multi_strerror_type)(CURLMcode error); +static curl_multi_strerror_type curl_multi_strerror_func; + +typedef CURLMcode (*curl_multi_timeout_type)(CURLM *multi_handle, long *milliseconds); +static curl_multi_timeout_type curl_multi_timeout_func; + +typedef CURL *(*curl_easy_init_type)(void); +static curl_easy_init_type curl_easy_init_func; + +typedef CURLcode (*curl_easy_perform_type)(CURL *curl); +static curl_easy_perform_type curl_easy_perform_func; + +typedef void (*curl_easy_cleanup_type)(CURL *curl); +static curl_easy_cleanup_type curl_easy_cleanup_func; + +typedef CURL *(*curl_easy_duphandle_type)(CURL *curl); +static curl_easy_duphandle_type curl_easy_duphandle_func; + +typedef CURLcode (*curl_easy_getinfo_long_type)(CURL *curl, CURLINFO info, long *value); +static curl_easy_getinfo_long_type curl_easy_getinfo_long_func; + +typedef CURLcode (*curl_easy_getinfo_pointer_type)(CURL *curl, CURLINFO info, void **value); +static curl_easy_getinfo_pointer_type curl_easy_getinfo_pointer_func; + +typedef CURLcode (*curl_easy_getinfo_off_t_type)(CURL *curl, CURLINFO info, curl_off_t *value); +static curl_easy_getinfo_off_t_type curl_easy_getinfo_off_t_func; + +typedef CURLcode (*curl_easy_setopt_long_type)(CURL *curl, CURLoption opt, long value); +static curl_easy_setopt_long_type curl_easy_setopt_long_func; + +typedef CURLcode (*curl_easy_setopt_pointer_type)(CURL *curl, CURLoption opt, void *value); +static curl_easy_setopt_pointer_type curl_easy_setopt_pointer_func; + +typedef CURLcode (*curl_easy_setopt_off_t_type)(CURL *curl, CURLoption opt, curl_off_t value); +static curl_easy_setopt_off_t_type curl_easy_setopt_off_t_func; + +static void lazy_load_curl(void) +{ + static int initialized; + void *libcurl; + func_t curl_easy_getinfo_func, curl_easy_setopt_func; + + if (initialized) + return; + + initialized = 1; + libcurl = load_library(LIBCURL_FILE_NAME("libcurl")); + if (!libcurl) + die("failed to load library '%s'", LIBCURL_FILE_NAME("libcurl")); + + curl_version_info_func = (curl_version_info_type)load_function(libcurl, "curl_version_info"); + curl_easy_escape_func = (curl_easy_escape_type)load_function(libcurl, "curl_easy_escape"); + curl_free_func = (curl_free_type)load_function(libcurl, "curl_free"); + curl_global_init_func = (curl_global_init_type)load_function(libcurl, "curl_global_init"); + curl_global_sslset_func = (curl_global_sslset_type)load_function(libcurl, "curl_global_sslset"); + curl_global_cleanup_func = (curl_global_cleanup_type)load_function(libcurl, "curl_global_cleanup"); + curl_global_trace_func = (curl_global_trace_type)load_function(libcurl, "curl_global_trace"); + curl_slist_append_func = (curl_slist_append_type)load_function(libcurl, "curl_slist_append"); + curl_slist_free_all_func = (curl_slist_free_all_type)load_function(libcurl, "curl_slist_free_all"); + curl_easy_strerror_func = (curl_easy_strerror_type)load_function(libcurl, "curl_easy_strerror"); + curl_multi_init_func = (curl_multi_init_type)load_function(libcurl, "curl_multi_init"); + curl_multi_add_handle_func = (curl_multi_add_handle_type)load_function(libcurl, "curl_multi_add_handle"); + curl_multi_remove_handle_func = (curl_multi_remove_handle_type)load_function(libcurl, "curl_multi_remove_handle"); + curl_multi_fdset_func = (curl_multi_fdset_type)load_function(libcurl, "curl_multi_fdset"); + curl_multi_perform_func = (curl_multi_perform_type)load_function(libcurl, "curl_multi_perform"); + curl_multi_cleanup_func = (curl_multi_cleanup_type)load_function(libcurl, "curl_multi_cleanup"); + curl_multi_info_read_func = (curl_multi_info_read_type)load_function(libcurl, "curl_multi_info_read"); + curl_multi_strerror_func = (curl_multi_strerror_type)load_function(libcurl, "curl_multi_strerror"); + curl_multi_timeout_func = (curl_multi_timeout_type)load_function(libcurl, "curl_multi_timeout"); + curl_easy_init_func = (curl_easy_init_type)load_function(libcurl, "curl_easy_init"); + curl_easy_perform_func = (curl_easy_perform_type)load_function(libcurl, "curl_easy_perform"); + curl_easy_cleanup_func = (curl_easy_cleanup_type)load_function(libcurl, "curl_easy_cleanup"); + curl_easy_duphandle_func = (curl_easy_duphandle_type)load_function(libcurl, "curl_easy_duphandle"); + + curl_easy_getinfo_func = load_function(libcurl, "curl_easy_getinfo"); + curl_easy_getinfo_long_func = (curl_easy_getinfo_long_type)curl_easy_getinfo_func; + curl_easy_getinfo_pointer_func = (curl_easy_getinfo_pointer_type)curl_easy_getinfo_func; + curl_easy_getinfo_off_t_func = (curl_easy_getinfo_off_t_type)curl_easy_getinfo_func; + + curl_easy_setopt_func = load_function(libcurl, "curl_easy_setopt"); + curl_easy_setopt_long_func = (curl_easy_setopt_long_type)curl_easy_setopt_func; + curl_easy_setopt_pointer_func = (curl_easy_setopt_pointer_type)curl_easy_setopt_func; + curl_easy_setopt_off_t_func = (curl_easy_setopt_off_t_type)curl_easy_setopt_func; +} + +struct curl_version_info_data *curl_version_info(CURLversion version) +{ + lazy_load_curl(); + return curl_version_info_func(version); +} + +char *curl_easy_escape(CURL *handle, const char *string, int length) +{ + lazy_load_curl(); + return curl_easy_escape_func(handle, string, length); +} + +void curl_free(void *p) +{ + lazy_load_curl(); + curl_free_func(p); +} + +CURLcode curl_global_init(long flags) +{ + lazy_load_curl(); + return curl_global_init_func(flags); +} + +CURLsslset curl_global_sslset(curl_sslbackend id, const char *name, const curl_ssl_backend ***avail) +{ + lazy_load_curl(); + return curl_global_sslset_func(id, name, avail); +} + +void curl_global_cleanup(void) +{ + lazy_load_curl(); + curl_global_cleanup_func(); +} + +CURLcode curl_global_trace(const char *config) +{ + lazy_load_curl(); + return curl_global_trace_func(config); +} + +struct curl_slist *curl_slist_append(struct curl_slist *list, const char *data) +{ + lazy_load_curl(); + return curl_slist_append_func(list, data); +} + +void curl_slist_free_all(struct curl_slist *list) +{ + lazy_load_curl(); + curl_slist_free_all_func(list); +} + +const char *curl_easy_strerror(CURLcode error) +{ + lazy_load_curl(); + return curl_easy_strerror_func(error); +} + +CURLM *curl_multi_init(void) +{ + lazy_load_curl(); + return curl_multi_init_func(); +} + +CURLMcode curl_multi_add_handle(CURLM *multi_handle, CURL *curl_handle) +{ + lazy_load_curl(); + return curl_multi_add_handle_func(multi_handle, curl_handle); +} + +CURLMcode curl_multi_remove_handle(CURLM *multi_handle, CURL *curl_handle) +{ + lazy_load_curl(); + return curl_multi_remove_handle_func(multi_handle, curl_handle); +} + +CURLMcode curl_multi_fdset(CURLM *multi_handle, fd_set *read_fd_set, fd_set *write_fd_set, fd_set *exc_fd_set, int *max_fd) +{ + lazy_load_curl(); + return curl_multi_fdset_func(multi_handle, read_fd_set, write_fd_set, exc_fd_set, max_fd); +} + +CURLMcode curl_multi_perform(CURLM *multi_handle, int *running_handles) +{ + lazy_load_curl(); + return curl_multi_perform_func(multi_handle, running_handles); +} + +CURLMcode curl_multi_cleanup(CURLM *multi_handle) +{ + lazy_load_curl(); + return curl_multi_cleanup_func(multi_handle); +} + +CURLMsg *curl_multi_info_read(CURLM *multi_handle, int *msgs_in_queue) +{ + lazy_load_curl(); + return curl_multi_info_read_func(multi_handle, msgs_in_queue); +} + +const char *curl_multi_strerror(CURLMcode error) +{ + lazy_load_curl(); + return curl_multi_strerror_func(error); +} + +CURLMcode curl_multi_timeout(CURLM *multi_handle, long *milliseconds) +{ + lazy_load_curl(); + return curl_multi_timeout_func(multi_handle, milliseconds); +} + +CURL *curl_easy_init(void) +{ + lazy_load_curl(); + return curl_easy_init_func(); +} + +CURLcode curl_easy_perform(CURL *curl) +{ + lazy_load_curl(); + return curl_easy_perform_func(curl); +} + +void curl_easy_cleanup(CURL *curl) +{ + lazy_load_curl(); + curl_easy_cleanup_func(curl); +} + +CURL *curl_easy_duphandle(CURL *curl) +{ + lazy_load_curl(); + return curl_easy_duphandle_func(curl); +} + +#ifndef CURL_IGNORE_DEPRECATION +#define CURL_IGNORE_DEPRECATION(x) x +#endif + +#ifndef CURLOPTTYPE_BLOB +#define CURLOPTTYPE_BLOB 40000 +#endif + +#undef curl_easy_getinfo +CURLcode curl_easy_getinfo(CURL *curl, CURLINFO info, ...) +{ + va_list ap; + CURLcode res; + + va_start(ap, info); + lazy_load_curl(); + CURL_IGNORE_DEPRECATION( + if (info >= CURLINFO_LONG && info < CURLINFO_DOUBLE) + res = curl_easy_getinfo_long_func(curl, info, va_arg(ap, long *)); + else if ((info >= CURLINFO_STRING && info < CURLINFO_LONG) || + (info >= CURLINFO_SLIST && info < CURLINFO_SOCKET)) + res = curl_easy_getinfo_pointer_func(curl, info, va_arg(ap, void **)); + else if (info >= CURLINFO_OFF_T) + res = curl_easy_getinfo_off_t_func(curl, info, va_arg(ap, curl_off_t *)); + else + die("%s:%d: TODO (info: %d)!", __FILE__, __LINE__, info); + ) + va_end(ap); + return res; +} + +#undef curl_easy_setopt +CURLcode curl_easy_setopt(CURL *curl, CURLoption opt, ...) +{ + va_list ap; + CURLcode res; + + va_start(ap, opt); + lazy_load_curl(); + CURL_IGNORE_DEPRECATION( + if (opt >= CURLOPTTYPE_LONG && opt < CURLOPTTYPE_OBJECTPOINT) + res = curl_easy_setopt_long_func(curl, opt, va_arg(ap, long)); + else if (opt >= CURLOPTTYPE_OBJECTPOINT && opt < CURLOPTTYPE_OFF_T) + res = curl_easy_setopt_pointer_func(curl, opt, va_arg(ap, void *)); + else if (opt >= CURLOPTTYPE_OFF_T && opt < CURLOPTTYPE_BLOB) + res = curl_easy_setopt_off_t_func(curl, opt, va_arg(ap, curl_off_t)); + else + die("%s:%d: TODO (opt: %d)!", __FILE__, __LINE__, opt); + ) + va_end(ap); + return res; +} From f830b1e23b4aec0d0dbe3683b3013d848c01b2a9 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sun, 7 May 2023 22:51:52 +0200 Subject: [PATCH 101/248] http: support lazy-loading libcurl also on Windows This implements the Windows-specific support code, because everything is slightly different on Windows, even loading shared libraries. Note: I specifically do _not_ use the code from `compat/win32/lazyload.h` here because that code is optimized for loading individual functions from various system DLLs, while we specifically want to load _many_ functions from _one_ DLL here, and distinctly not a system DLL (we expect libcurl to be located outside `C:\Windows\system32`, something `INIT_PROC_ADDR` refuses to work with). Also, the `curl_easy_getinfo()`/`curl_easy_setopt()` functions are declared as vararg functions, which `lazyload.h` cannot handle. Finally, we are about to optionally override the exact file name that is to be loaded, which is a goal contrary to `lazyload.h`'s design. Signed-off-by: Johannes Schindelin --- Makefile | 4 ++++ compat/lazyload-curl.c | 52 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 56 insertions(+) diff --git a/Makefile b/Makefile index 1257118cc39dbb..cf27fa3cb97e6f 100644 --- a/Makefile +++ b/Makefile @@ -1675,7 +1675,11 @@ else # The `CURL_STATICLIB` constant must be defined to avoid seeing the functions # declared as DLL imports CURL_CFLAGS = -DCURL_STATICLIB +ifneq ($(uname_S),MINGW) +ifneq ($(uname_S),Windows) CURL_LIBCURL = -ldl +endif +endif else ifndef CURL_LDFLAGS CURL_LDFLAGS = $(eval CURL_LDFLAGS := $$(shell $$(CURL_CONFIG) --libs))$(CURL_LDFLAGS) diff --git a/compat/lazyload-curl.c b/compat/lazyload-curl.c index f4e08f76dfcd7f..82ab11de43a0fb 100644 --- a/compat/lazyload-curl.c +++ b/compat/lazyload-curl.c @@ -1,6 +1,8 @@ #include "../git-compat-util.h" #include "../git-curl-compat.h" +#ifndef WIN32 #include +#endif /* * The ABI version of libcurl is encoded in its shared libraries' file names. @@ -11,6 +13,7 @@ typedef void (*func_t)(void); +#ifndef WIN32 #ifdef __APPLE__ #define LIBCURL_FILE_NAME(base) base "." LIBCURL_ABI_VERSION ".dylib" #else @@ -35,6 +38,55 @@ static func_t load_function(void *handle, const char *name) *(void **)&f = dlsym(handle, name); return f; } +#else +#define LIBCURL_FILE_NAME(base) base "-" LIBCURL_ABI_VERSION ".dll" + +static void *load_library(const char *name) +{ + size_t name_size = strlen(name) + 1; + const char *path = getenv("PATH"); + char dll_path[MAX_PATH]; + + while (path && *path) { + const char *sep = strchrnul(path, ';'); + size_t len = sep - path; + + if (len && len + name_size < sizeof(dll_path)) { + memcpy(dll_path, path, len); + dll_path[len] = '/'; + memcpy(dll_path + len + 1, name, name_size); + + if (!access(dll_path, R_OK)) { + wchar_t wpath[MAX_PATH]; + int wlen = MultiByteToWideChar(CP_UTF8, 0, dll_path, -1, wpath, ARRAY_SIZE(wpath)); + void *res = wlen ? (void *)LoadLibraryExW(wpath, NULL, 0) : NULL; + if (!res) { + DWORD err = GetLastError(); + char buf[1024]; + + if (!FormatMessageA(FORMAT_MESSAGE_FROM_SYSTEM | + FORMAT_MESSAGE_ARGUMENT_ARRAY | + FORMAT_MESSAGE_IGNORE_INSERTS, + NULL, err, LANG_NEUTRAL, + buf, sizeof(buf) - 1, NULL)) + xsnprintf(buf, sizeof(buf), "last error: %ld", err); + error("LoadLibraryExW() failed with: %s", buf); + } + return res; + } + } + + path = *sep ? sep + 1 : NULL; + } + + return NULL; +} + +static func_t load_function(void *handle, const char *name) +{ + return (func_t)GetProcAddress((HANDLE)handle, name); +} +#endif typedef struct curl_version_info_data *(*curl_version_info_type)(CURLversion version); static curl_version_info_type curl_version_info_func; From fd97bab1e0d0008484667629da7873f908c7726e Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sun, 7 May 2023 22:05:33 +0200 Subject: [PATCH 102/248] http: when loading libcurl lazily, allow for multiple SSL backends The previous commits introduced a compile-time option to load libcurl lazily, but it uses the hard-coded name "libcurl-4.dll" (or equivalent on platforms other than Windows). To allow for installing multiple libcurl flavors side by side, where each supports one specific SSL/TLS backend, let's first look whether `libcurl--4.dll` exists, and only use `libcurl-4.dll` as a fall back. That will allow us to ship with a libcurl by default that only supports the Secure Channel backend for the `https://` protocol. This libcurl won't suffer from any dependency problem when upgrading OpenSSL to a new major version (which will change the DLL name, and hence break every program and library that depends on it). This is crucial because Git for Windows relies on libcurl to keep working when building and deploying a new OpenSSL package because that library is used by `git fetch` and `git clone`. Note that this feature is by no means specific to Windows. On Ubuntu, for example, a `git` built using `LAZY_LOAD_LIBCURL` will use `libcurl.so.4` for `http.sslbackend=openssl` and `libcurl-gnutls.so.4` for `http.sslbackend=gnutls`. Signed-off-by: Johannes Schindelin --- compat/lazyload-curl.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/compat/lazyload-curl.c b/compat/lazyload-curl.c index 82ab11de43a0fb..a6a3f7e3a7aeaa 100644 --- a/compat/lazyload-curl.c +++ b/compat/lazyload-curl.c @@ -175,17 +175,26 @@ static curl_easy_setopt_pointer_type curl_easy_setopt_pointer_func; typedef CURLcode (*curl_easy_setopt_off_t_type)(CURL *curl, CURLoption opt, curl_off_t value); static curl_easy_setopt_off_t_type curl_easy_setopt_off_t_func; +static char ssl_backend[64]; + static void lazy_load_curl(void) { static int initialized; - void *libcurl; + void *libcurl = NULL; func_t curl_easy_getinfo_func, curl_easy_setopt_func; if (initialized) return; initialized = 1; - libcurl = load_library(LIBCURL_FILE_NAME("libcurl")); + if (ssl_backend[0]) { + char dll_name[64 + 16]; + snprintf(dll_name, sizeof(dll_name) - 1, + LIBCURL_FILE_NAME("libcurl-%s"), ssl_backend); + libcurl = load_library(dll_name); + } + if (!libcurl) + libcurl = load_library(LIBCURL_FILE_NAME("libcurl")); if (!libcurl) die("failed to load library '%s'", LIBCURL_FILE_NAME("libcurl")); @@ -250,6 +259,9 @@ CURLcode curl_global_init(long flags) CURLsslset curl_global_sslset(curl_sslbackend id, const char *name, const curl_ssl_backend ***avail) { + if (name && strlen(name) < sizeof(ssl_backend)) + strlcpy(ssl_backend, name, sizeof(ssl_backend)); + lazy_load_curl(); return curl_global_sslset_func(id, name, avail); } From 0a8c0639391b52fd4ea375d7c90581154aeb6940 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sun, 10 Jul 2022 11:27:25 +0200 Subject: [PATCH 103/248] MinGW: link as terminal server aware MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Whith Windows 2000, Microsoft introduced a flag to the PE header to mark executables as "terminal server aware". Windows terminal servers provide a redirected Windows directory and redirected registry hives when launching legacy applications without this flag set. Since we do not use any INI files in the Windows directory and don't write to the registry, we don't need this additional preparation. Telling the OS that we don't need this should provide slightly improved startup times in terminal server environments. When building for supported Windows Versions with MSVC the /TSAWARE linker flag is automatically set, but MinGW requires us to set the --tsaware flag manually. This partially addresses https://github.com/git-for-windows/git/issues/3935. Signed-off-by: Matthias Aßhauer --- config.mak.uname | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/config.mak.uname b/config.mak.uname index 1691c6ae6e01e3..9458c11c259116 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -718,7 +718,7 @@ ifeq ($(uname_S),MINGW) DEFAULT_HELP_FORMAT = html HAVE_PLATFORM_PROCINFO = YesPlease CSPRNG_METHOD = rtlgenrandom - BASIC_LDFLAGS += -municode + BASIC_LDFLAGS += -municode -Wl,--tsaware COMPAT_CFLAGS += -DNOGDI -Icompat -Icompat/win32 COMPAT_CFLAGS += -DSTRIP_EXTENSION=\".exe\" COMPAT_OBJS += compat/mingw.o compat/winansi.o \ From 1e9c1c3b5eae9bb104a984562200370f86386ead Mon Sep 17 00:00:00 2001 From: Kiel Hurley Date: Wed, 2 Nov 2022 22:56:16 +1300 Subject: [PATCH 104/248] Fix Windows version resources Add FileVersion, which is a required field As not all required fields were present, none were being included Fixes #4090 Signed-off-by: Kiel Hurley --- git.rc.in | 1 + 1 file changed, 1 insertion(+) diff --git a/git.rc.in b/git.rc.in index e69444eef3f0c5..460ea39561b87f 100644 --- a/git.rc.in +++ b/git.rc.in @@ -12,6 +12,7 @@ BEGIN VALUE "OriginalFilename", "git.exe\0" VALUE "ProductName", "Git\0" VALUE "ProductVersion", "@GIT_VERSION@\0" + VALUE "FileVersion", "@GIT_VERSION@\0" END END From 10884da4333e30ec1a5a40813253e8cbba75035d Mon Sep 17 00:00:00 2001 From: Andrey Zabavnikov Date: Fri, 28 Oct 2022 17:12:06 +0300 Subject: [PATCH 105/248] status: fix for old-style submodules with commondir In f9b7573f6b00 (repository: free fields before overwriting them, 2017-09-05), Git was taught to release memory before overwriting it, but 357a03ebe9e0 (repository.c: move env-related setup code back to environment.c, 2018-03-03) changed the code so that it would not _always_ be overwritten. As a consequence, the `commondir` attribute would point to already-free()d memory. This seems not to cause problems in core Git, but there are add-on patches in Git for Windows where the `commondir` attribute is subsequently used and causing invalid memory accesses e.g. in setups containing old-style submodules (i.e. the ones with a `.git` directory within theirs worktrees) that have `commondir` configured. This fixes https://github.com/git-for-windows/git/pull/4083. Signed-off-by: Andrey Zabavnikov Signed-off-by: Johannes Schindelin --- repository.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/repository.c b/repository.c index ecd691181fc97d..adccc83f8613a7 100644 --- a/repository.c +++ b/repository.c @@ -136,7 +136,7 @@ static void repo_set_commondir(struct repository *repo, { struct strbuf sb = STRBUF_INIT; - free(repo->commondir); + FREE_AND_NULL(repo->commondir); if (commondir) { repo->different_commondir = 1; From 63830e418acd83aaa8992de4b5bcf15eca04d955 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Fri, 27 Jan 2023 08:55:21 +0100 Subject: [PATCH 106/248] windows: skip linking `git-` for built-ins It is merely a historical wart that, say, `git-commit` exists in the `libexec/git-core/` directory, a tribute to the original idea to let Git be essentially a bunch of Unix shell scripts revolving around very few "plumbing" (AKA low-level) commands. Git has evolved a lot from there. These days, most of Git's functionality is contained within the `git` executable, in the form of "built-in" commands. To accommodate for scripts that use the "dashed" form of Git commands, even today, Git provides hard-links that make the `git` executable available as, say, `git-commit`, just in case that an old script has not been updated to invoke `git commit`. Those hard-links do not come cheap: they take about half a minute for every build of Git on Windows, they are mistaken for taking up huge amounts of space by some Windows Explorer versions that do not understand hard-links, and therefore many a "bug" report had to be addressed. The "dashed form" has been officially deprecated in Git version 1.5.4, which was released on February 2nd, 2008, i.e. a very long time ago. This deprecation was never finalized by skipping these hard-links, but we can start the process now, in Git for Windows. Signed-off-by: Johannes Schindelin --- config.mak.uname | 2 ++ 1 file changed, 2 insertions(+) diff --git a/config.mak.uname b/config.mak.uname index 5cd2310b8ef17e..febf3ec26501cf 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -512,6 +512,7 @@ ifeq ($(uname_S),Windows) NO_POSIX_GOODIES = UnfortunatelyYes NATIVE_CRLF = YesPlease DEFAULT_HELP_FORMAT = html + SKIP_DASHED_BUILT_INS = YabbaDabbaDoo ifeq (/mingw64,$(subst 32,64,$(subst clangarm,mingw,$(prefix)))) # Move system config into top-level /etc/ ETC_GITCONFIG = ../etc/gitconfig @@ -704,6 +705,7 @@ ifeq ($(uname_S),MINGW) FSMONITOR_DAEMON_BACKEND = win32 FSMONITOR_OS_SETTINGS = win32 + SKIP_DASHED_BUILT_INS = YabbaDabbaDoo RUNTIME_PREFIX = YesPlease HAVE_WPGMPTR = YesWeDo NO_ST_BLOCKS_IN_STRUCT_STAT = YesPlease From 57654d41209a48c27b5998596b0a99df6886f480 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Sun, 7 May 2023 22:43:37 +0200 Subject: [PATCH 107/248] mingw: do load libcurl dynamically by default This will help with Git for Windows' maintenance going forward: It allows Git for Windows to switch its primary libcurl to a variant without the OpenSSL backend, while still loading an alternate when setting `http.sslBackend = openssl`. This is necessary to avoid maintenance headaches with upgrading OpenSSL: its major version name is encoded in the shared library's file name and hence major version updates (temporarily) break libraries that are linked against the OpenSSL library. Signed-off-by: Johannes Schindelin --- config.mak.uname | 1 + 1 file changed, 1 insertion(+) diff --git a/config.mak.uname b/config.mak.uname index 9458c11c259116..cf811a9b6ba9e6 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -719,6 +719,7 @@ ifeq ($(uname_S),MINGW) HAVE_PLATFORM_PROCINFO = YesPlease CSPRNG_METHOD = rtlgenrandom BASIC_LDFLAGS += -municode -Wl,--tsaware + LAZYLOAD_LIBCURL = YesDoThatPlease COMPAT_CFLAGS += -DNOGDI -Icompat -Icompat/win32 COMPAT_CFLAGS += -DSTRIP_EXTENSION=\".exe\" COMPAT_OBJS += compat/mingw.o compat/winansi.o \ From 3b785031df55eb4ad18a20a79d010e17827958dc Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 2 Nov 2022 16:23:58 +0100 Subject: [PATCH 108/248] Add a GitHub workflow to verify that Git/Scalar work in Nano Server In Git for Windows v2.39.0, we fixed a regression where `git.exe` would no longer work in Windows Nano Server (frequently used in Docker containers). This GitHub workflow can be used to verify manually that the Git/Scalar executables work in Nano Server. Signed-off-by: Johannes Schindelin --- .github/workflows/nano-server.yml | 76 +++++++++++++++++++++++++++++++ 1 file changed, 76 insertions(+) create mode 100644 .github/workflows/nano-server.yml diff --git a/.github/workflows/nano-server.yml b/.github/workflows/nano-server.yml new file mode 100644 index 00000000000000..85b3ed5f52ed4d --- /dev/null +++ b/.github/workflows/nano-server.yml @@ -0,0 +1,76 @@ +name: Windows Nano Server tests + +on: + workflow_dispatch: + +env: + DEVELOPER: 1 + +jobs: + test-nano-server: + runs-on: windows-2022 + env: + WINDBG_DIR: "C:/Program Files (x86)/Windows Kits/10/Debuggers/x64" + IMAGE: mcr.microsoft.com/powershell:nanoserver-ltsc2022 + + steps: + - uses: actions/checkout@v5 + - uses: git-for-windows/setup-git-for-windows-sdk@v1 + - name: build Git + shell: bash + run: make -j15 + - name: pull nanoserver image + shell: bash + run: docker pull $IMAGE + - name: run nano-server test + shell: bash + run: | + docker run \ + --user "ContainerAdministrator" \ + -v "$WINDBG_DIR:C:/dbg" \ + -v "$(cygpath -aw /mingw64/bin):C:/mingw64-bin" \ + -v "$(cygpath -aw .):C:/test" \ + $IMAGE pwsh.exe -Command ' + # Extend the PATH to include the `.dll` files in /mingw64/bin/ + $env:PATH += ";C:\mingw64-bin" + + # For each executable to test pick some no-operation set of + # flags/subcommands or something that should quickly result in an + # error with known exit code that is not a negative 32-bit + # number, and set the expected return code appropriately. + # + # Only test executables that could be expected to run in a UI + # less environment. + # + # ( Executable path, arguments, expected return code ) + # also note space is required before close parenthesis (a + # powershell quirk when defining nested arrays like this) + + $executables_to_test = @( + ("C:\test\git.exe", "", 1 ), + ("C:\test\scalar.exe", "version", 0 ) + ) + + foreach ($executable in $executables_to_test) + { + Write-Output "Now testing $($executable[0])" + &$executable[0] $executable[1] + if ($LASTEXITCODE -ne $executable[2]) { + # if we failed, run the debugger to find out what function + # or DLL could not be found and then exit the script with + # failure The missing DLL or EXE will be referenced near + # the end of the output + + # Set a flag to have the debugger show loader stub + # diagnostics. This requires running as administrator, + # otherwise the flag will be ignored. + C:\dbg\gflags -i $executable[0] +SLS + + C:\dbg\cdb.exe -c "g" -c "q" $executable[0] $executable[1] + + exit 1 + } + } + + exit 0 + ' From f95f4ba0b48905763f3a008c06856ddf9c9e7587 Mon Sep 17 00:00:00 2001 From: David Lomas Date: Fri, 28 Jul 2023 15:31:25 +0100 Subject: [PATCH 109/248] mingw: suggest `windows.appendAtomically` in more cases When running Git for Windows on a remote APFS filesystem, it would appear that the `mingw_open_append()`/`write()` combination would fail almost exactly like on some CIFS-mounted shares as had been reported in https://github.com/git-for-windows/git/issues/2753, albeit with a different `errno` value. Let's handle that `errno` value just the same, by suggesting to set `windows.appendAtomically=false`. Signed-off-by: David Lomas Signed-off-by: Johannes Schindelin --- compat/mingw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 761f321b5b4b9d..608a67d40f072f 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -811,7 +811,7 @@ ssize_t mingw_write(int fd, const void *buf, size_t len) { ssize_t result = write(fd, buf, len); - if (result < 0 && (errno == EINVAL || errno == ENOSPC) && buf) { + if (result < 0 && (errno == EINVAL || errno == EBADF || errno == ENOSPC) && buf) { int orig = errno; /* check if fd is a pipe */ @@ -837,7 +837,7 @@ ssize_t mingw_write(int fd, const void *buf, size_t len) } errno = orig; - } else if (orig == EINVAL) + } else if (orig == EINVAL || errno == EBADF) errno = EPIPE; else { DWORD buf_size; From 00f22b136fc6f8225381c054899fc7f3943ee828 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 22 Nov 2023 22:57:38 +0100 Subject: [PATCH 110/248] win32: use native ANSI sequence processing, if possible Windows 10 version 1511 (also known as Anniversary Update), according to https://learn.microsoft.com/en-us/windows/console/console-virtual-terminal-sequences introduced native support for ANSI sequence processing. This allows using colors from the entire 24-bit color range. All we need to do is test whether the console's "virtual processing support" can be enabled. If it can, we do not even need to start the `console_thread` to handle ANSI sequences. Or, almost all we need to do: When `console_thread()` does its work, it uses the Unicode-aware `write_console()` function to write to the Win32 Console, which supports Git for Windows' implicit convention that all text that is written is encoded in UTF-8. The same is not necessarily true if native ANSI sequence processing is used, as the output is then subject to the current code page. Let's ensure that the code page is set to `CP_UTF8` as long as Git writes to it. Signed-off-by: Johannes Schindelin --- compat/winansi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) diff --git a/compat/winansi.c b/compat/winansi.c index ac2ffb78691a7d..a83a7f47ada9b2 100644 --- a/compat/winansi.c +++ b/compat/winansi.c @@ -593,6 +593,49 @@ static void detect_msys_tty(int fd) #endif +static HANDLE std_console_handle; +static DWORD std_console_mode = ENABLE_VIRTUAL_TERMINAL_PROCESSING; +static UINT std_console_code_page = CP_UTF8; + +static void reset_std_console(void) +{ + if (std_console_mode != ENABLE_VIRTUAL_TERMINAL_PROCESSING) + SetConsoleMode(std_console_handle, std_console_mode); + if (std_console_code_page != CP_UTF8) + SetConsoleOutputCP(std_console_code_page); +} + +static int enable_virtual_processing(void) +{ + std_console_handle = GetStdHandle(STD_OUTPUT_HANDLE); + if (std_console_handle == INVALID_HANDLE_VALUE || + !GetConsoleMode(std_console_handle, &std_console_mode)) { + std_console_handle = GetStdHandle(STD_ERROR_HANDLE); + if (std_console_handle == INVALID_HANDLE_VALUE || + !GetConsoleMode(std_console_handle, &std_console_mode)) + return 0; + } + + std_console_code_page = GetConsoleOutputCP(); + if (std_console_code_page != CP_UTF8) + SetConsoleOutputCP(CP_UTF8); + if (!std_console_code_page) + std_console_code_page = CP_UTF8; + + atexit(reset_std_console); + + if (std_console_mode & ENABLE_VIRTUAL_TERMINAL_PROCESSING) + return 1; + + if (!SetConsoleMode(std_console_handle, + std_console_mode | + ENABLE_PROCESSED_OUTPUT | + ENABLE_VIRTUAL_TERMINAL_PROCESSING)) + return 0; + + return 1; +} + /* * Wrapper for isatty(). Most calls in the main git code * call isatty(1 or 2) to see if the instance is interactive @@ -631,6 +674,9 @@ void winansi_init(void) return; } + if (enable_virtual_processing()) + return; + /* create a named pipe to communicate with the console thread */ if (swprintf(name, ARRAY_SIZE(name) - 1, L"\\\\.\\pipe\\winansi%lu", GetCurrentProcessId()) < 0) From 6596645684ec1c607317243378521a574a89cde5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sat, 2 Dec 2023 12:10:00 +0100 Subject: [PATCH 111/248] git.rc: include winuser.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit winuser.h contains the definition of RT_MANIFEST that our LLVM based toolchain needs to understand that we want to embed compat/win32/git.manifest as an application manifest. It currently just embeds it as additional data that Windows doesn't understand. This also helps our GCC based toolchain understand that we only want one copy embedded. It currently embeds one working assembly manifest and one nearly identical, but useless copy as additional data. This also teaches our Visual Studio based buildsystems to pick up the manifest file from git.rc. This means we don't have to explicitly specify it in contrib/buildsystems/Generators/Vcxproj.pm anymore. Slightly counter-intuitively this also means we have to explicitly tell Cmake not to embed a default manifest. This fixes https://github.com/git-for-windows/git/issues/4707 Signed-off-by: Matthias Aßhauer Signed-off-by: Johannes Schindelin --- contrib/buildsystems/CMakeLists.txt | 1 + git.rc.in | 1 + 2 files changed, 2 insertions(+) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index edb0fc04ad7649..e054653e59b206 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -208,6 +208,7 @@ if(CMAKE_C_COMPILER_ID STREQUAL "MSVC") set(CMAKE_RUNTIME_OUTPUT_DIRECTORY_DEBUG ${CMAKE_BINARY_DIR}) set(CMAKE_RUNTIME_OUTPUT_DIRECTORY_RELEASE ${CMAKE_BINARY_DIR}) add_compile_options(/MP /std:c11) + add_link_options(/MANIFEST:NO) endif() #default behaviour diff --git a/git.rc.in b/git.rc.in index e69444eef3f0c5..1d5b627b610549 100644 --- a/git.rc.in +++ b/git.rc.in @@ -1,3 +1,4 @@ +#include 1 VERSIONINFO FILEVERSION @GIT_MAJOR_VERSION@,@GIT_MINOR_VERSION@,@GIT_MICRO_VERSION@,@GIT_PATCH_LEVEL@ PRODUCTVERSION @GIT_MAJOR_VERSION@,@GIT_MINOR_VERSION@,@GIT_MICRO_VERSION@,@GIT_PATCH_LEVEL@ From 95a077fc397f87ded28b8b87c97cb46f3002b9d5 Mon Sep 17 00:00:00 2001 From: MinarKotonoha Date: Mon, 8 Apr 2024 16:41:10 +0800 Subject: [PATCH 112/248] common-main.c: fflush stdout buffer upon exit By default, the buffer type of Windows' `stdout` is unbuffered (_IONBF), and there is no need to manually fflush `stdout`. But some programs, such as the Windows Filtering Platform driver provided by the security software, may change the buffer type of `stdout` to full buffering. This nees `fflush(stdout)` to be called manually, otherwise there will be no output to `stdout`. Signed-off-by: MinarKotonoha Signed-off-by: Johannes Schindelin --- common-exit.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/common-exit.c b/common-exit.c index 1aaa538be3ed67..609f32abed8b53 100644 --- a/common-exit.c +++ b/common-exit.c @@ -11,6 +11,13 @@ static void check_bug_if_BUG(void) /* We wrap exit() to call common_exit() in git-compat-util.h */ int common_exit(const char *file, int line, int code) { + /* + * Windows Filtering Platform driver provided by the security software + * may change buffer type of stdout from _IONBF to _IOFBF. + * It will no output without fflush manually. + */ + fflush(stdout); + /* * For non-POSIX systems: Take the lowest 8 bits of the "code" * to e.g. turn -1 into 255. On a POSIX system this is From b4295cde7f760d5c6897f39163b7ea71d85ed2aa Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 29 Apr 2024 08:55:03 -0400 Subject: [PATCH 113/248] survey: stub in new experimental 'git-survey' command Start work on a new 'git survey' command to scan the repository for monorepo performance and scaling problems. The goal is to measure the various known "dimensions of scale" and serve as a foundation for adding additional measurements as we learn more about Git monorepo scaling problems. The initial goal is to complement the scanning and analysis performed by the GO-based 'git-sizer' (https://github.com/github/git-sizer) tool. It is hoped that by creating a builtin command, we may be able to take advantage of internal Git data structures and code that is not accessible from GO to gain further insight into potential scaling problems. Co-authored-by: Derrick Stolee Signed-off-by: Jeff Hostetler Signed-off-by: Derrick Stolee --- .gitignore | 1 + Documentation/config.adoc | 2 + Documentation/config/survey.adoc | 11 +++++ Documentation/git-survey.adoc | 36 +++++++++++++++ Documentation/meson.build | 1 + Makefile | 1 + builtin.h | 1 + builtin/survey.c | 75 ++++++++++++++++++++++++++++++++ command-list.txt | 1 + git.c | 1 + meson.build | 1 + t/meson.build | 1 + t/t8100-git-survey.sh | 18 ++++++++ 13 files changed, 150 insertions(+) create mode 100644 Documentation/config/survey.adoc create mode 100644 Documentation/git-survey.adoc create mode 100644 builtin/survey.c create mode 100755 t/t8100-git-survey.sh diff --git a/.gitignore b/.gitignore index 04c444404e4ba8..078fc537292893 100644 --- a/.gitignore +++ b/.gitignore @@ -166,6 +166,7 @@ /git-submodule /git-submodule--helper /git-subtree +/git-survey /git-svn /git-switch /git-symbolic-ref diff --git a/Documentation/config.adoc b/Documentation/config.adoc index ae6dcb6a6c0607..81ad0adeec801e 100644 --- a/Documentation/config.adoc +++ b/Documentation/config.adoc @@ -535,6 +535,8 @@ include::config/status.adoc[] include::config/submodule.adoc[] +include::config/survey.adoc[] + include::config/tag.adoc[] include::config/tar.adoc[] diff --git a/Documentation/config/survey.adoc b/Documentation/config/survey.adoc new file mode 100644 index 00000000000000..c1b0f852a1250e --- /dev/null +++ b/Documentation/config/survey.adoc @@ -0,0 +1,11 @@ +survey.*:: + These variables adjust the default behavior of the `git survey` + command. The intention is that this command could be run in the + background with these options. ++ +-- + verbose:: + This boolean value implies the `--[no-]verbose` option. + progress:: + This boolean value implies the `--[no-]progress` option. +-- diff --git a/Documentation/git-survey.adoc b/Documentation/git-survey.adoc new file mode 100644 index 00000000000000..5f8ec9bfea673b --- /dev/null +++ b/Documentation/git-survey.adoc @@ -0,0 +1,36 @@ +git-survey(1) +============= + +NAME +---- +git-survey - EXPERIMENTAL: Measure various repository dimensions of scale + +SYNOPSIS +-------- +[verse] +(EXPERIMENTAL!) 'git survey' + +DESCRIPTION +----------- + +Survey the repository and measure various dimensions of scale. + +As repositories grow to "monorepo" size, certain data shapes can cause +performance problems. `git-survey` attempts to measure and report on +known problem areas. + +OPTIONS +------- + +--progress:: + Show progress. This is automatically enabled when interactive. + +OUTPUT +------ + +By default, `git survey` will print information about the repository in a +human-readable format that includes overviews and tables. + +GIT +--- +Part of the linkgit:git[1] suite diff --git a/Documentation/meson.build b/Documentation/meson.build index 4404c623f006db..2cf77dedaef6c4 100644 --- a/Documentation/meson.build +++ b/Documentation/meson.build @@ -141,6 +141,7 @@ manpages = { 'git-status.adoc' : 1, 'git-stripspace.adoc' : 1, 'git-submodule.adoc' : 1, + 'git-survey.adoc' : 1, 'git-svn.adoc' : 1, 'git-switch.adoc' : 1, 'git-symbolic-ref.adoc' : 1, diff --git a/Makefile b/Makefile index 1f444eec3c1d12..23c7805e2aa653 100644 --- a/Makefile +++ b/Makefile @@ -1323,6 +1323,7 @@ BUILTIN_OBJS += builtin/sparse-checkout.o BUILTIN_OBJS += builtin/stash.o BUILTIN_OBJS += builtin/stripspace.o BUILTIN_OBJS += builtin/submodule--helper.o +BUILTIN_OBJS += builtin/survey.o BUILTIN_OBJS += builtin/symbolic-ref.o BUILTIN_OBJS += builtin/tag.o BUILTIN_OBJS += builtin/unpack-file.o diff --git a/builtin.h b/builtin.h index bff13e3069b4af..36e14fc1d2855e 100644 --- a/builtin.h +++ b/builtin.h @@ -232,6 +232,7 @@ int cmd_sparse_checkout(int argc, const char **argv, const char *prefix, struct int cmd_status(int argc, const char **argv, const char *prefix, struct repository *repo); int cmd_stash(int argc, const char **argv, const char *prefix, struct repository *repo); int cmd_stripspace(int argc, const char **argv, const char *prefix, struct repository *repo); +int cmd_survey(int argc, const char **argv, const char *prefix, struct repository *repo); int cmd_submodule__helper(int argc, const char **argv, const char *prefix, struct repository *repo); int cmd_switch(int argc, const char **argv, const char *prefix, struct repository *repo); int cmd_symbolic_ref(int argc, const char **argv, const char *prefix, struct repository *repo); diff --git a/builtin/survey.c b/builtin/survey.c new file mode 100644 index 00000000000000..7b7214a289765c --- /dev/null +++ b/builtin/survey.c @@ -0,0 +1,75 @@ +#define USE_THE_REPOSITORY_VARIABLE + +#include "builtin.h" +#include "config.h" +#include "parse-options.h" + +static const char * const survey_usage[] = { + N_("(EXPERIMENTAL!) git survey "), + NULL, +}; + +struct survey_opts { + int verbose; + int show_progress; +}; + +struct survey_context { + struct repository *repo; + + /* Options that control what is done. */ + struct survey_opts opts; +}; + +static int survey_load_config_cb(const char *var, const char *value, + const struct config_context *cctx, void *pvoid) +{ + struct survey_context *ctx = pvoid; + + if (!strcmp(var, "survey.verbose")) { + ctx->opts.verbose = git_config_bool(var, value); + return 0; + } + if (!strcmp(var, "survey.progress")) { + ctx->opts.show_progress = git_config_bool(var, value); + return 0; + } + + return git_default_config(var, value, cctx, pvoid); +} + +static void survey_load_config(struct survey_context *ctx) +{ + repo_config(the_repository, survey_load_config_cb, ctx); +} + +int cmd_survey(int argc, const char **argv, const char *prefix, struct repository *repo) +{ + static struct survey_context ctx = { + .opts = { + .verbose = 0, + .show_progress = -1, /* defaults to isatty(2) */ + }, + }; + + static struct option survey_options[] = { + OPT__VERBOSE(&ctx.opts.verbose, N_("verbose output")), + OPT_BOOL(0, "progress", &ctx.opts.show_progress, N_("show progress")), + OPT_END(), + }; + + show_usage_with_options_if_asked(argc, argv, + survey_usage, survey_options); + + ctx.repo = repo; + + prepare_repo_settings(ctx.repo); + survey_load_config(&ctx); + + argc = parse_options(argc, argv, prefix, survey_options, survey_usage, 0); + + if (ctx.opts.show_progress < 0) + ctx.opts.show_progress = isatty(2); + + return 0; +} diff --git a/command-list.txt b/command-list.txt index b7ade3ab9f3319..8872958cf48a61 100644 --- a/command-list.txt +++ b/command-list.txt @@ -188,6 +188,7 @@ git-stash mainporcelain git-status mainporcelain info git-stripspace purehelpers git-submodule mainporcelain +git-survey mainporcelain git-svn foreignscminterface git-switch mainporcelain history git-symbolic-ref plumbingmanipulators diff --git a/git.c b/git.c index 8c85da84c30534..5e862ea693d11c 100644 --- a/git.c +++ b/git.c @@ -656,6 +656,7 @@ static struct cmd_struct commands[] = { { "status", cmd_status, RUN_SETUP | NEED_WORK_TREE }, { "stripspace", cmd_stripspace }, { "submodule--helper", cmd_submodule__helper, RUN_SETUP }, + { "survey", cmd_survey, RUN_SETUP }, { "switch", cmd_switch, RUN_SETUP | NEED_WORK_TREE }, { "symbolic-ref", cmd_symbolic_ref, RUN_SETUP }, { "tag", cmd_tag, RUN_SETUP | DELAY_PAGER_CONFIG }, diff --git a/meson.build b/meson.build index 5dd299b4962d84..617cf9b7facb8c 100644 --- a/meson.build +++ b/meson.build @@ -660,6 +660,7 @@ builtin_sources = [ 'builtin/stash.c', 'builtin/stripspace.c', 'builtin/submodule--helper.c', + 'builtin/survey.c', 'builtin/symbolic-ref.c', 'builtin/tag.c', 'builtin/unpack-file.c', diff --git a/t/meson.build b/t/meson.build index 983245501ce9a2..2b0e0c33994f3f 100644 --- a/t/meson.build +++ b/t/meson.build @@ -947,6 +947,7 @@ integration_tests = [ 't8012-blame-colors.sh', 't8013-blame-ignore-revs.sh', 't8014-blame-ignore-fuzzy.sh', + 't8100-git-survey.sh', 't9001-send-email.sh', 't9002-column.sh', 't9003-help-autocorrect.sh', diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh new file mode 100755 index 00000000000000..d9816419855d1a --- /dev/null +++ b/t/t8100-git-survey.sh @@ -0,0 +1,18 @@ +#!/bin/sh + +test_description='git survey' + +GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME=main +export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME + +TEST_PASSES_SANITIZE_LEAK=0 +export TEST_PASSES_SANITIZE_LEAK + +. ./test-lib.sh + +test_expect_success 'git survey -h shows experimental warning' ' + test_expect_code 129 git survey -h >usage && + grep "EXPERIMENTAL!" usage +' + +test_done From 7cada41879de99d3a1dbccfbd0364a605bdccacb Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 9 Apr 2024 16:50:56 +0200 Subject: [PATCH 114/248] t5601/t7406(mingw): do run tests with symlink support A long time ago, we decided to run tests in Git for Windows' SDK with the default `winsymlinks` mode: copying instead of linking. This is still the default mode of MSYS2 to this day. However, this is not how most users run Git for Windows: As the majority of Git for Windows' users seem to be on Windows 10 and newer, likely having enabled Developer Mode (which allows creating symbolic links without administrator privileges), they will run with symlink support enabled. This is the reason why it is crucial to get the fixes for CVE-2024-? to the users, and also why it is crucial to ensure that the test suite exercises the related test cases. This commit ensures the latter. Signed-off-by: Johannes Schindelin --- t/t5601-clone.sh | 10 ++++++++++ t/t7406-submodule-update.sh | 9 +++++++++ 2 files changed, 19 insertions(+) diff --git a/t/t5601-clone.sh b/t/t5601-clone.sh index d743d986c401a0..a859e09956222c 100755 --- a/t/t5601-clone.sh +++ b/t/t5601-clone.sh @@ -7,6 +7,16 @@ export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME . ./test-lib.sh +# This test script contains test cases that need to create symbolic links. To +# make sure that these test cases are exercised in Git for Windows, where (for +# historical reasons) `ln -s` creates copies by default, let's specifically ask +# for `ln -s` to create symbolic links whenever possible. +if test_have_prereq MINGW +then + MSYS=${MSYS+$MSYS }winsymlinks:nativestrict + export MSYS +fi + X= test_have_prereq !MINGW || X=.exe diff --git a/t/t7406-submodule-update.sh b/t/t7406-submodule-update.sh index 3adab12091a5f0..a3e0dc198ab646 100755 --- a/t/t7406-submodule-update.sh +++ b/t/t7406-submodule-update.sh @@ -14,6 +14,15 @@ export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME . ./test-lib.sh +# This test script contains test cases that need to create symbolic links. To +# make sure that these test cases are exercised in Git for Windows, where (for +# historical reasons) `ln -s` creates copies by default, let's specifically ask +# for `ln -s` to create symbolic links whenever possible. +if test_have_prereq MINGW +then + MSYS=${MSYS+$MSYS }winsymlinks:nativestrict + export MSYS +fi compare_head() { From 26f28f4738b44c14e6acf9e169508da4180f6635 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 21 May 2024 13:55:26 +0200 Subject: [PATCH 115/248] win32: ensure that `localtime_r()` is declared even in i686 builds The `__MINGW64__` constant is defined, surprise, surprise, only when building for a 64-bit CPU architecture. Therefore using it as a guard to define `_POSIX_C_SOURCE` (so that `localtime_r()` is declared, among other functions) is not enough, we also need to check `__MINGW32__`. Technically, the latter constant is defined even for 64-bit builds. But let's make things a bit easier to understand by testing for both constants. Making it so fixes this compile warning (turned error in GCC v14.1): archive-zip.c: In function 'dos_time': archive-zip.c:612:9: error: implicit declaration of function 'localtime_r'; did you mean 'localtime_s'? [-Wimplicit-function-declaration] 612 | localtime_r(&time, &tm); | ^~~~~~~~~~~ | localtime_s Signed-off-by: Johannes Schindelin --- compat/posix.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/compat/posix.h b/compat/posix.h index 067a00f33b83f3..2be79fdca6039c 100644 --- a/compat/posix.h +++ b/compat/posix.h @@ -45,7 +45,7 @@ #define UNUSED #endif -#ifdef __MINGW64__ +#if defined(__MINGW32__) || defined(__MINGW64__) #define _POSIX_C_SOURCE 1 #elif defined(__sun__) /* From 1cd294ff16043a9ae71926057f79dc75ea2e9dc5 Mon Sep 17 00:00:00 2001 From: Ariel Lourenco Date: Tue, 2 Jul 2024 18:09:43 -0300 Subject: [PATCH 116/248] Fallback to AppData if XDG_CONFIG_HOME is unset In order to be a better Windows citizenship, Git should save its configuration files on AppData folder. This can enables git configuration files be replicated between machines using the same Microsoft account logon which would reduce the friction of setting up Git on new systems. Therefore, if %APPDATA%\Git\config exists, we use it; otherwise $HOME/.config/git/config is used. Signed-off-by: Ariel Lourenco --- path.c | 23 ++++++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) diff --git a/path.c b/path.c index 7f56eaf9930374..4e3654dba12baa 100644 --- a/path.c +++ b/path.c @@ -1640,6 +1640,7 @@ int looks_like_command_line_option(const char *str) char *xdg_config_home_for(const char *subdir, const char *filename) { const char *home, *config_home; + char *home_config = NULL; assert(subdir); assert(filename); @@ -1648,10 +1649,26 @@ char *xdg_config_home_for(const char *subdir, const char *filename) return mkpathdup("%s/%s/%s", config_home, subdir, filename); home = getenv("HOME"); - if (home) - return mkpathdup("%s/.config/%s/%s", home, subdir, filename); + if (home && *home) + home_config = mkpathdup("%s/.config/%s/%s", home, subdir, filename); + + #ifdef WIN32 + { + const char *appdata = getenv("APPDATA"); + if (appdata && *appdata) { + char *appdata_config = mkpathdup("%s/Git/%s", appdata, filename); + if (file_exists(appdata_config)) { + if (home_config && file_exists(home_config)) + warning("'%s' was ignored because '%s' exists.", home_config, appdata_config); + free(home_config); + return appdata_config; + } + free(appdata_config); + } + } + #endif - return NULL; + return home_config; } char *xdg_config_home(const char *filename) From 778b51622b49a380f3ff2ba69792ac328b214f73 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 4 Jul 2024 22:41:56 +0200 Subject: [PATCH 117/248] run-command: be helpful with Git LFS fails on Windows 7 Git LFS is now built with Go 1.21 which no longer supports Windows 7. However, Git for Windows still wants to support Windows 7. Ideally, Git LFS would re-introduce Windows 7 support until Git for Windows drops support for Windows 7, but that's not going to happen: https://github.com/git-for-windows/git/issues/4996#issuecomment-2176152565 The next best thing we can do is to let the users know what is happening, and how to get out of their fix, at least. This is not quite as easy as it would first seem because programs compiled with Go 1.21 or newer will simply throw an exception and fail with an Access Violation on Windows 7. The only way I found to address this is to replicate the logic from Go's very own `version` command (which can determine the Go version with which a given executable was built) to detect the situation, and in that case offer a helpful error message. This addresses https://github.com/git-for-windows/git/issues/4996. Signed-off-by: Johannes Schindelin --- compat/win32/path-utils.c | 199 ++++++++++++++++++++++++++++++++++++++ compat/win32/path-utils.h | 3 + git-compat-util.h | 7 ++ run-command.c | 1 + 4 files changed, 210 insertions(+) diff --git a/compat/win32/path-utils.c b/compat/win32/path-utils.c index 966ef779b9ca9b..c4fea0301b5ecc 100644 --- a/compat/win32/path-utils.c +++ b/compat/win32/path-utils.c @@ -2,6 +2,9 @@ #include "../../git-compat-util.h" #include "../../environment.h" +#include "../../wrapper.h" +#include "../../strbuf.h" +#include "../../versioncmp.h" int win32_has_dos_drive_prefix(const char *path) { @@ -89,3 +92,199 @@ int win32_fspathcmp(const char *a, const char *b) { return win32_fspathncmp(a, b, (size_t)-1); } + +static int read_at(int fd, char *buffer, size_t offset, size_t size) +{ + if (lseek(fd, offset, SEEK_SET) < 0) { + fprintf(stderr, "could not seek to 0x%x\n", (unsigned int)offset); + return -1; + } + + return read_in_full(fd, buffer, size); +} + +static size_t le16(const char *buffer) +{ + unsigned char *u = (unsigned char *)buffer; + return u[0] | (u[1] << 8); +} + +static size_t le32(const char *buffer) +{ + return le16(buffer) | (le16(buffer + 2) << 16); +} + +/* + * Determine the Go version of a given executable, if it was built with Go. + * + * This recapitulates the logic from + * https://github.com/golang/go/blob/master/src/cmd/go/internal/version/version.go + * (without requiring the user to install `go.exe` to find out). + */ +static ssize_t get_go_version(const char *path, char *go_version, size_t go_version_size) +{ + int fd = open(path, O_RDONLY); + char buffer[1024]; + off_t offset; + size_t num_sections, opt_header_size, i; + char *p = NULL, *q; + ssize_t res = -1; + + if (fd < 0) + return -1; + + if (read_in_full(fd, buffer, 2) < 0) + goto fail; + + /* + * Parse the PE file format, for more details, see + * https://en.wikipedia.org/wiki/Portable_Executable#Layout and + * https://learn.microsoft.com/en-us/windows/win32/debug/pe-format + */ + if (buffer[0] != 'M' || buffer[1] != 'Z') + goto fail; + + if (read_at(fd, buffer, 0x3c, 4) < 0) + goto fail; + + /* Read the `PE\0\0` signature and the COFF file header */ + offset = le32(buffer); + if (read_at(fd, buffer, offset, 24) < 0) + goto fail; + + if (buffer[0] != 'P' || buffer[1] != 'E' || buffer[2] != '\0' || buffer[3] != '\0') + goto fail; + + num_sections = le16(buffer + 6); + opt_header_size = le16(buffer + 20); + offset += 24; /* skip file header */ + + /* + * Validate magic number 0x10b or 0x20b, for full details see + * https://learn.microsoft.com/en-us/windows/win32/debug/pe-format#optional-header-standard-fields-image-only + */ + if (read_at(fd, buffer, offset, 2) < 0 || + ((i = le16(buffer)) != 0x10b && i != 0x20b)) + goto fail; + + offset += opt_header_size; + + for (i = 0; i < num_sections; i++) { + if (read_at(fd, buffer, offset + i * 40, 40) < 0) + goto fail; + + /* + * For full details about the section headers, see + * https://learn.microsoft.com/en-us/windows/win32/debug/pe-format#section-table-section-headers + */ + if ((le32(buffer + 36) /* characteristics */ & ~0x600000) /* IMAGE_SCN_ALIGN_32BYTES */ == + (/* IMAGE_SCN_CNT_INITIALIZED_DATA */ 0x00000040 | + /* IMAGE_SCN_MEM_READ */ 0x40000000 | + /* IMAGE_SCN_MEM_WRITE */ 0x80000000)) { + size_t size = le32(buffer + 16); /* "SizeOfRawData " */ + size_t pointer = le32(buffer + 20); /* "PointerToRawData " */ + + /* + * Skip the section if either size or pointer is 0, see + * https://github.com/golang/go/blob/go1.21.0/src/debug/buildinfo/buildinfo.go#L333 + * for full details. + * + * Merely seeing a non-zero size will not actually do, + * though: he size must be at least `buildInfoSize`, + * i.e. 32, and we expect a UVarint (at least another + * byte) _and_ the bytes representing the string, + * which we expect to start with the letters "go" and + * continue with the Go version number. + */ + if (size < 32 + 1 + 2 + 1 || !pointer) + continue; + + p = malloc(size); + + if (!p || read_at(fd, p, pointer, size) < 0) + goto fail; + + /* + * Look for the build information embedded by Go, see + * https://github.com/golang/go/blob/go1.21.0/src/debug/buildinfo/buildinfo.go#L165-L175 + * for full details. + * + * Note: Go contains code to enforce alignment along a + * 16-byte boundary. In practice, no `.exe` has been + * observed that required any adjustment, therefore + * this here code skips that logic for simplicity. + */ + q = memmem(p, size - 18, "\xff Go buildinf:", 14); + if (!q) + goto fail; + /* + * Decode the build blob. For full details, see + * https://github.com/golang/go/blob/go1.21.0/src/debug/buildinfo/buildinfo.go#L177-L191 + * + * Note: The `endianness` values observed in practice + * were always 2, therefore the complex logic to handle + * any other value is skipped for simplicty. + */ + if ((q[14] == 8 || q[14] == 4) && q[15] == 2) { + /* + * Only handle a Go version string with fewer + * than 128 characters, so the Go UVarint at + * q[32] that indicates the string's length must + * be only one byte (without the high bit set). + */ + if ((q[32] & 0x80) || + !q[32] || + (q + 33 + q[32] - p) > (ssize_t)size || + q[32] + 1 > (ssize_t)go_version_size) + goto fail; + res = q[32]; + memcpy(go_version, q + 33, res); + go_version[res] = '\0'; + break; + } + } + } + +fail: + free(p); + close(fd); + return res; +} + +void win32_warn_about_git_lfs_on_windows7(int exit_code, const char *argv0) +{ + char buffer[128], *git_lfs = NULL; + const char *p; + + /* + * Git LFS v3.5.1 fails with an Access Violation on Windows 7; That + * would usually show up as an exit code 0xc0000005. For some reason + * (probably because at this point, we no longer have the _original_ + * HANDLE that was returned by `CreateProcess()`) we observe other + * values like 0xb00 and 0x2 instead. Since the exact exit code + * seems to be inconsistent, we check for a non-zero exit status. + */ + if (exit_code == 0) + return; + if (GetVersion() >> 16 > 7601) + return; /* Warn only on Windows 7 or older */ + if (!istarts_with(argv0, "git-lfs ") && + strcasecmp(argv0, "git-lfs")) + return; + if (!(git_lfs = locate_in_PATH("git-lfs"))) + return; + if (get_go_version(git_lfs, buffer, sizeof(buffer)) > 0 && + skip_prefix(buffer, "go", &p) && + versioncmp("1.21.0", p) <= 0) + warning("This program was built with Go v%s\n" + "i.e. without support for this Windows version:\n" + "\n\t%s\n" + "\n" + "To work around this, you can download and install a " + "working version from\n" + "\n" + "\thttps://github.com/git-lfs/git-lfs/releases/tag/" + "v3.4.1\n", + p, git_lfs); + free(git_lfs); +} diff --git a/compat/win32/path-utils.h b/compat/win32/path-utils.h index a561c700e75713..a69483c332c1a7 100644 --- a/compat/win32/path-utils.h +++ b/compat/win32/path-utils.h @@ -34,4 +34,7 @@ int win32_fspathcmp(const char *a, const char *b); int win32_fspathncmp(const char *a, const char *b, size_t count); #define fspathncmp win32_fspathncmp +void win32_warn_about_git_lfs_on_windows7(int exit_code, const char *argv0); +#define warn_about_git_lfs_on_windows7 win32_warn_about_git_lfs_on_windows7 + #endif diff --git a/git-compat-util.h b/git-compat-util.h index 398e0fac4fab60..9b5192571d074e 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -294,6 +294,13 @@ static inline int git_offset_1st_component(const char *path) #define fspathncmp git_fspathncmp #endif +#ifndef warn_about_git_lfs_on_windows7 +static inline void warn_about_git_lfs_on_windows7(int exit_code UNUSED, + const char *argv0 UNUSED) +{ +} +#endif + #ifndef is_valid_path #define is_valid_path(path) 1 #endif diff --git a/run-command.c b/run-command.c index ed9575bd6a8cbb..b4c8a089bfb4d1 100644 --- a/run-command.c +++ b/run-command.c @@ -582,6 +582,7 @@ static int wait_or_whine(pid_t pid, const char *argv0, int in_signal) */ code += 128; } else if (WIFEXITED(status)) { + warn_about_git_lfs_on_windows7(status, argv0); code = WEXITSTATUS(status); } else { if (!in_signal) From 5703eb437ac71638586429d85a8860a17019f24a Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Fri, 6 Sep 2024 14:16:13 -0400 Subject: [PATCH 118/248] revision: create mark_trees_uninteresting_dense() The sparse tree walk algorithm was created in d5d2e93577e (revision: implement sparse algorithm, 2019-01-16) and involves using the mark_trees_uninteresting_sparse() method. This method takes a repository and an oidset of tree IDs, some of which have the UNINTERESTING flag and some of which do not. Create a method that has an equivalent set of preconditions but uses a "dense" walk (recursively visits all reachable trees, as long as they have not previously been marked UNINTERESTING). This is an important difference from mark_tree_uninteresting(), which short-circuits if the given tree has the UNINTERESTING flag. A use of this method will be added in a later change, with a condition set whether the sparse or dense approach should be used. Signed-off-by: Derrick Stolee From 0274d9e75dfac4f925737174c3a9fbe64e5be81e Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 24 Sep 2024 08:47:39 +0200 Subject: [PATCH 119/248] ci: work around a problem with HTTP/2 vs libcurl v8.10.0 As reported in https://lore.kernel.org/git/ZuPKvYP9ZZ2mhb4m@pks.im/, libcurl v8.10.0 had a regression that was picked up by Git's t5559.30 "large fetch-pack requests can be sent using chunked encoding". This bug was fixed in libcurl v8.10.1. Sadly, the macos-13 runner image was updated in the brief window between these two libcurl versions, breaking each and every CI build, as reported at https://github.com/git-for-windows/git/issues/5159. This would usually not matter, we would just ignore the failing CI builds until the macos-13 runner image is rebuilt in a couple of days, and then the CI builds would succeed again. However. As has become the custom, a surprise Git version was released, and now that Git for Windows wants to follow suit, since Git for Windows has this custom of trying to never release a version with a failing CI build, we _must_ work around it. This patch implements this work-around, basically for the sake of Git for Windows v2.46.2's CI build. Signed-off-by: Johannes Schindelin --- t/t5551-http-fetch-smart.sh | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/t/t5551-http-fetch-smart.sh b/t/t5551-http-fetch-smart.sh index b0d4ea78015a25..e52bcbed347ca7 100755 --- a/t/t5551-http-fetch-smart.sh +++ b/t/t5551-http-fetch-smart.sh @@ -410,7 +410,15 @@ test_expect_success CMDLINE_LIMIT \ ) ' -test_expect_success 'large fetch-pack requests can be sent using chunked encoding' ' +# This is a temporary work-around for libcurl v8.10.0 on the macos-* runners; +# see https://github.com/git-for-windows/git/issues/5159 for full details +test_lazy_prereq UNBROKEN_HTTP2 ' + test "$HTTP_PROTO" = HTTP/2 && + test -z "$(brew info -q curl 2>/dev/null | + sed -n "/^Installed/{N;s/.*8\\.10\\.0.*/BROKEN HTTP2/p;}")" +' + +test_expect_success UNBROKEN_HTTP2 'large fetch-pack requests can be sent using chunked encoding' ' GIT_TRACE_CURL=true git -c http.postbuffer=65536 \ clone --bare "$HTTPD_URL/smart/repo.git" split.git 2>err && { From 6a4a1f17dfe5cf57bf63964c04137cc44894f0bd Mon Sep 17 00:00:00 2001 From: Jeff Hostetler Date: Mon, 29 Apr 2024 09:51:34 -0400 Subject: [PATCH 120/248] survey: add command line opts to select references By default we will scan all references in "refs/heads/", "refs/tags/" and "refs/remotes/". Add command line opts let the use ask for all refs or a subset of them and to include a detached HEAD. Signed-off-by: Jeff Hostetler Signed-off-by: Derrick Stolee Signed-off-by: Johannes Schindelin --- Documentation/git-survey.adoc | 34 +++++ builtin/survey.c | 248 ++++++++++++++++++++++++++++++++++ t/t8100-git-survey.sh | 9 ++ 3 files changed, 291 insertions(+) diff --git a/Documentation/git-survey.adoc b/Documentation/git-survey.adoc index 5f8ec9bfea673b..56060d14b5cfef 100644 --- a/Documentation/git-survey.adoc +++ b/Documentation/git-survey.adoc @@ -19,12 +19,46 @@ As repositories grow to "monorepo" size, certain data shapes can cause performance problems. `git-survey` attempts to measure and report on known problem areas. +Ref Selection and Reachable Objects +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +In this first analysis phase, `git survey` will iterate over the set of +requested branches, tags, and other refs and treewalk over all of the +reachable commits, trees, and blobs and generate various statistics. + OPTIONS ------- --progress:: Show progress. This is automatically enabled when interactive. +Ref Selection +~~~~~~~~~~~~~ + +The following options control the set of refs that `git survey` will examine. +By default, `git survey` will look at tags, local branches, and remote refs. +If any of the following options are given, the default set is cleared and +only refs for the given options are added. + +--all-refs:: + Use all refs. This includes local branches, tags, remote refs, + notes, and stashes. This option overrides all of the following. + +--branches:: + Add local branches (`refs/heads/`) to the set. + +--tags:: + Add tags (`refs/tags/`) to the set. + +--remotes:: + Add remote branches (`refs/remote/`) to the set. + +--detached:: + Add HEAD to the set. + +--other:: + Add notes (`refs/notes/`) and stashes (`refs/stash/`) to the set. + OUTPUT ------ diff --git a/builtin/survey.c b/builtin/survey.c index 7b7214a289765c..8fbc104ec7bd74 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -2,16 +2,55 @@ #include "builtin.h" #include "config.h" +#include "object.h" +#include "odb.h" #include "parse-options.h" +#include "progress.h" +#include "ref-filter.h" +#include "strvec.h" +#include "trace2.h" static const char * const survey_usage[] = { N_("(EXPERIMENTAL!) git survey "), NULL, }; +struct survey_refs_wanted { + int want_all_refs; /* special override */ + + int want_branches; + int want_tags; + int want_remotes; + int want_detached; + int want_other; /* see FILTER_REFS_OTHERS -- refs/notes/, refs/stash/ */ +}; + +static struct survey_refs_wanted default_ref_options = { + .want_all_refs = 1, +}; + struct survey_opts { int verbose; int show_progress; + struct survey_refs_wanted refs; +}; + +struct survey_report_ref_summary { + size_t refs_nr; + size_t branches_nr; + size_t remote_refs_nr; + size_t tags_nr; + size_t tags_annotated_nr; + size_t others_nr; + size_t unknown_nr; +}; + +/** + * This struct contains all of the information that needs to be printed + * at the end of the exploration of the repository and its references. + */ +struct survey_report { + struct survey_report_ref_summary refs; }; struct survey_context { @@ -19,8 +58,84 @@ struct survey_context { /* Options that control what is done. */ struct survey_opts opts; + + /* Info for output only. */ + struct survey_report report; + + /* + * The rest of the members are about enabling the activity + * of the 'git survey' command, including ref listings, object + * pointers, and progress. + */ + + struct progress *progress; + size_t progress_nr; + size_t progress_total; + + struct strvec refs; }; +static void clear_survey_context(struct survey_context *ctx) +{ + strvec_clear(&ctx->refs); +} + +/* + * After parsing the command line arguments, figure out which refs we + * should scan. + * + * If ANY were given in positive sense, then we ONLY include them and + * do not use the builtin values. + */ +static void fixup_refs_wanted(struct survey_context *ctx) +{ + struct survey_refs_wanted *rw = &ctx->opts.refs; + + /* + * `--all-refs` overrides and enables everything. + */ + if (rw->want_all_refs == 1) { + rw->want_branches = 1; + rw->want_tags = 1; + rw->want_remotes = 1; + rw->want_detached = 1; + rw->want_other = 1; + return; + } + + /* + * If none of the `--` were given, we assume all + * of the builtin unspecified values. + */ + if (rw->want_branches == -1 && + rw->want_tags == -1 && + rw->want_remotes == -1 && + rw->want_detached == -1 && + rw->want_other == -1) { + *rw = default_ref_options; + return; + } + + /* + * Since we only allow positive boolean values on the command + * line, we will only have true values where they specified + * a `--`. + * + * So anything that still has an unspecified value should be + * set to false. + */ + if (rw->want_branches == -1) + rw->want_branches = 0; + if (rw->want_tags == -1) + rw->want_tags = 0; + if (rw->want_remotes == -1) + rw->want_remotes = 0; + if (rw->want_detached == -1) + rw->want_detached = 0; + if (rw->want_other == -1) + rw->want_other = 0; +} + static int survey_load_config_cb(const char *var, const char *value, const struct config_context *cctx, void *pvoid) { @@ -43,18 +158,146 @@ static void survey_load_config(struct survey_context *ctx) repo_config(the_repository, survey_load_config_cb, ctx); } +static void do_load_refs(struct survey_context *ctx, + struct ref_array *ref_array) +{ + struct ref_filter filter = REF_FILTER_INIT; + struct ref_sorting *sorting; + struct string_list sorting_options = STRING_LIST_INIT_DUP; + + string_list_append(&sorting_options, "objectname"); + sorting = ref_sorting_options(&sorting_options); + + if (ctx->opts.refs.want_detached) + strvec_push(&ctx->refs, "HEAD"); + + if (ctx->opts.refs.want_all_refs) { + strvec_push(&ctx->refs, "refs/"); + } else { + if (ctx->opts.refs.want_branches) + strvec_push(&ctx->refs, "refs/heads/"); + if (ctx->opts.refs.want_tags) + strvec_push(&ctx->refs, "refs/tags/"); + if (ctx->opts.refs.want_remotes) + strvec_push(&ctx->refs, "refs/remotes/"); + if (ctx->opts.refs.want_other) { + strvec_push(&ctx->refs, "refs/notes/"); + strvec_push(&ctx->refs, "refs/stash/"); + } + } + + filter.name_patterns = ctx->refs.v; + filter.ignore_case = 0; + filter.match_as_path = 1; + + if (ctx->opts.show_progress) { + ctx->progress_total = 0; + ctx->progress = start_progress(ctx->repo, + _("Scanning refs..."), 0); + } + + filter_refs(ref_array, &filter, FILTER_REFS_KIND_MASK); + + if (ctx->opts.show_progress) { + ctx->progress_total = ref_array->nr; + display_progress(ctx->progress, ctx->progress_total); + } + + ref_array_sort(sorting, ref_array); + + stop_progress(&ctx->progress); + ref_filter_clear(&filter); + ref_sorting_release(sorting); +} + +/* + * The REFS phase: + * + * Load the set of requested refs and assess them for scalablity problems. + * Use that set to start a treewalk to all reachable objects and assess + * them. + * + * This data will give us insights into the repository itself (the number + * of refs, the size and shape of the DAG, the number and size of the + * objects). + * + * Theoretically, this data is independent of the on-disk representation + * (e.g. independent of packing concerns). + */ +static void survey_phase_refs(struct survey_context *ctx) +{ + struct ref_array ref_array = { 0 }; + + trace2_region_enter("survey", "phase/refs", ctx->repo); + do_load_refs(ctx, &ref_array); + + ctx->report.refs.refs_nr = ref_array.nr; + for (int i = 0; i < ref_array.nr; i++) { + unsigned long size; + struct ref_array_item *item = ref_array.items[i]; + + switch (item->kind) { + case FILTER_REFS_TAGS: + ctx->report.refs.tags_nr++; + if (odb_read_object_info(ctx->repo->objects, + &item->objectname, + &size) == OBJ_TAG) + ctx->report.refs.tags_annotated_nr++; + break; + + case FILTER_REFS_BRANCHES: + ctx->report.refs.branches_nr++; + break; + + case FILTER_REFS_REMOTES: + ctx->report.refs.remote_refs_nr++; + break; + + case FILTER_REFS_OTHERS: + ctx->report.refs.others_nr++; + break; + + default: + ctx->report.refs.unknown_nr++; + break; + } + } + + trace2_region_leave("survey", "phase/refs", ctx->repo); + + ref_array_clear(&ref_array); +} + int cmd_survey(int argc, const char **argv, const char *prefix, struct repository *repo) { static struct survey_context ctx = { .opts = { .verbose = 0, .show_progress = -1, /* defaults to isatty(2) */ + + .refs.want_all_refs = -1, + + .refs.want_branches = -1, /* default these to undefined */ + .refs.want_tags = -1, + .refs.want_remotes = -1, + .refs.want_detached = -1, + .refs.want_other = -1, }, + .refs = STRVEC_INIT, }; static struct option survey_options[] = { OPT__VERBOSE(&ctx.opts.verbose, N_("verbose output")), OPT_BOOL(0, "progress", &ctx.opts.show_progress, N_("show progress")), + + OPT_BOOL_F(0, "all-refs", &ctx.opts.refs.want_all_refs, N_("include all refs"), PARSE_OPT_NONEG), + + OPT_BOOL_F(0, "branches", &ctx.opts.refs.want_branches, N_("include branches"), PARSE_OPT_NONEG), + OPT_BOOL_F(0, "tags", &ctx.opts.refs.want_tags, N_("include tags"), PARSE_OPT_NONEG), + OPT_BOOL_F(0, "remotes", &ctx.opts.refs.want_remotes, N_("include all remotes refs"), PARSE_OPT_NONEG), + OPT_BOOL_F(0, "detached", &ctx.opts.refs.want_detached, N_("include detached HEAD"), PARSE_OPT_NONEG), + OPT_BOOL_F(0, "other", &ctx.opts.refs.want_other, N_("include notes and stashes"), PARSE_OPT_NONEG), + OPT_END(), }; @@ -71,5 +314,10 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor if (ctx.opts.show_progress < 0) ctx.opts.show_progress = isatty(2); + fixup_refs_wanted(&ctx); + + survey_phase_refs(&ctx); + + clear_survey_context(&ctx); return 0; } diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index d9816419855d1a..9bac3c2ba47e2c 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -15,4 +15,13 @@ test_expect_success 'git survey -h shows experimental warning' ' grep "EXPERIMENTAL!" usage ' +test_expect_success 'create a semi-interesting repo' ' + test_commit_bulk 10 +' + +test_expect_success 'git survey (default)' ' + git survey >out 2>err && + test_line_count = 0 err +' + test_done From ffaf469f03fe41c750d30a11360e10f2fc94f3ed Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 15:58:32 -0400 Subject: [PATCH 121/248] survey: start pretty printing data in table form When 'git survey' provides information to the user, this will be presented in one of two formats: plaintext and JSON. The JSON implementation will be delayed until the functionality is complete for the plaintext format. The most important parts of the plaintext format are headers specifying the different sections of the report and tables providing concreted data. Create a custom table data structure that allows specifying a list of strings for the row values. When printing the table, check each column for the maximum width so we can create a table of the correct size from the start. The table structure is designed to be flexible to the different kinds of output that will be implemented in future changes. Signed-off-by: Derrick Stolee --- Documentation/git-survey.adoc | 7 ++ builtin/survey.c | 157 ++++++++++++++++++++++++++++++++++ t/t8100-git-survey.sh | 18 +++- 3 files changed, 181 insertions(+), 1 deletion(-) diff --git a/Documentation/git-survey.adoc b/Documentation/git-survey.adoc index 56060d14b5cfef..120ecb9a4d49f2 100644 --- a/Documentation/git-survey.adoc +++ b/Documentation/git-survey.adoc @@ -65,6 +65,13 @@ OUTPUT By default, `git survey` will print information about the repository in a human-readable format that includes overviews and tables. +References Summary +~~~~~~~~~~~~~~~~~~ + +The references summary includes a count of each kind of reference, +including branches, remote refs, and tags (split by "all" and +"annotated"). + GIT --- Part of the linkgit:git[1] suite diff --git a/builtin/survey.c b/builtin/survey.c index 8fbc104ec7bd74..e79f97f8d75923 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -7,6 +7,7 @@ #include "parse-options.h" #include "progress.h" #include "ref-filter.h" +#include "strbuf.h" #include "strvec.h" #include "trace2.h" @@ -80,6 +81,160 @@ static void clear_survey_context(struct survey_context *ctx) strvec_clear(&ctx->refs); } +struct survey_table { + const char *table_name; + struct strvec header; + struct strvec *rows; + size_t rows_nr; + size_t rows_alloc; +}; + +#define SURVEY_TABLE_INIT { \ + .header = STRVEC_INIT, \ +} + +static void clear_table(struct survey_table *table) +{ + strvec_clear(&table->header); + for (size_t i = 0; i < table->rows_nr; i++) + strvec_clear(&table->rows[i]); + free(table->rows); +} + +static void insert_table_rowv(struct survey_table *table, ...) +{ + va_list ap; + char *arg; + ALLOC_GROW(table->rows, table->rows_nr + 1, table->rows_alloc); + + memset(&table->rows[table->rows_nr], 0, sizeof(struct strvec)); + + va_start(ap, table); + while ((arg = va_arg(ap, char *))) + strvec_push(&table->rows[table->rows_nr], arg); + va_end(ap); + + table->rows_nr++; +} + +#define SECTION_SEGMENT "========================================" +#define SECTION_SEGMENT_LEN 40 +static const char *section_line = SECTION_SEGMENT + SECTION_SEGMENT + SECTION_SEGMENT + SECTION_SEGMENT; +static const size_t section_len = 4 * SECTION_SEGMENT_LEN; + +static void print_table_title(const char *name, size_t *widths, size_t nr) +{ + size_t width = 3 * (nr - 1); + + for (size_t i = 0; i < nr; i++) + width += widths[i]; + + if (width > section_len) + width = section_len; + + printf("\n%s\n%.*s\n", name, (int)width, section_line); +} + +static void print_row_plaintext(struct strvec *row, size_t *widths) +{ + static struct strbuf line = STRBUF_INIT; + strbuf_setlen(&line, 0); + + for (size_t i = 0; i < row->nr; i++) { + const char *str = row->v[i]; + size_t len = strlen(str); + if (i) + strbuf_add(&line, " | ", 3); + strbuf_addchars(&line, ' ', widths[i] - len); + strbuf_add(&line, str, len); + } + printf("%s\n", line.buf); +} + +static void print_divider_plaintext(size_t *widths, size_t nr) +{ + static struct strbuf line = STRBUF_INIT; + strbuf_setlen(&line, 0); + + for (size_t i = 0; i < nr; i++) { + if (i) + strbuf_add(&line, "-+-", 3); + strbuf_addchars(&line, '-', widths[i]); + } + printf("%s\n", line.buf); +} + +static void print_table_plaintext(struct survey_table *table) +{ + size_t *column_widths; + size_t columns_nr = table->header.nr; + CALLOC_ARRAY(column_widths, columns_nr); + + for (size_t i = 0; i < columns_nr; i++) { + column_widths[i] = strlen(table->header.v[i]); + + for (size_t j = 0; j < table->rows_nr; j++) { + size_t rowlen = strlen(table->rows[j].v[i]); + if (column_widths[i] < rowlen) + column_widths[i] = rowlen; + } + } + + print_table_title(table->table_name, column_widths, columns_nr); + print_row_plaintext(&table->header, column_widths); + print_divider_plaintext(column_widths, columns_nr); + + for (size_t j = 0; j < table->rows_nr; j++) + print_row_plaintext(&table->rows[j], column_widths); + + free(column_widths); +} + +static void survey_report_plaintext_refs(struct survey_context *ctx) +{ + struct survey_report_ref_summary *refs = &ctx->report.refs; + struct survey_table table = SURVEY_TABLE_INIT; + + table.table_name = _("REFERENCES SUMMARY"); + + strvec_push(&table.header, _("Ref Type")); + strvec_push(&table.header, _("Count")); + + if (ctx->opts.refs.want_all_refs || ctx->opts.refs.want_branches) { + char *fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)refs->branches_nr); + insert_table_rowv(&table, _("Branches"), fmt, NULL); + free(fmt); + } + + if (ctx->opts.refs.want_all_refs || ctx->opts.refs.want_remotes) { + char *fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)refs->remote_refs_nr); + insert_table_rowv(&table, _("Remote refs"), fmt, NULL); + free(fmt); + } + + if (ctx->opts.refs.want_all_refs || ctx->opts.refs.want_tags) { + char *fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)refs->tags_nr); + insert_table_rowv(&table, _("Tags (all)"), fmt, NULL); + free(fmt); + fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)refs->tags_annotated_nr); + insert_table_rowv(&table, _("Tags (annotated)"), fmt, NULL); + free(fmt); + } + + print_table_plaintext(&table); + clear_table(&table); +} + +static void survey_report_plaintext(struct survey_context *ctx) +{ + printf("GIT SURVEY for \"%s\"\n", ctx->repo->worktree); + printf("-----------------------------------------------------\n"); + survey_report_plaintext_refs(ctx); +} + /* * After parsing the command line arguments, figure out which refs we * should scan. @@ -318,6 +473,8 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor survey_phase_refs(&ctx); + survey_report_plaintext(&ctx); + clear_survey_context(&ctx); return 0; } diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index 9bac3c2ba47e2c..e518e4844fe2d0 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -21,7 +21,23 @@ test_expect_success 'create a semi-interesting repo' ' test_expect_success 'git survey (default)' ' git survey >out 2>err && - test_line_count = 0 err + test_line_count = 0 err && + + tr , " " >expect <<-EOF && + GIT SURVEY for "$(pwd)" + ----------------------------------------------------- + + REFERENCES SUMMARY + ======================== + , Ref Type | Count + -----------------+------ + , Branches | 1 + Remote refs | 0 + Tags (all) | 0 + Tags (annotated) | 0 + EOF + + test_cmp expect out ' test_done From c8836a75b0ce1be51c06537372b72807928be72f Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 20:33:47 -0400 Subject: [PATCH 122/248] survey: add object count summary At the moment, nothing is obvious about the reason for the use of the path-walk API, but this will become more prevelant in future iterations. For now, use the path-walk API to sum up the counts of each kind of object. For example, this is the reachable object summary output for my local repo: REACHABLE OBJECT SUMMARY ======================== Object Type | Count ------------+------- Tags | 1343 Commits | 179344 Trees | 314350 Blobs | 184030 Signed-off-by: Derrick Stolee --- Documentation/git-survey.adoc | 6 ++ builtin/survey.c | 131 ++++++++++++++++++++++++++++++++-- t/t8100-git-survey.sh | 23 ++++-- 3 files changed, 149 insertions(+), 11 deletions(-) diff --git a/Documentation/git-survey.adoc b/Documentation/git-survey.adoc index 120ecb9a4d49f2..44f3a0568b7697 100644 --- a/Documentation/git-survey.adoc +++ b/Documentation/git-survey.adoc @@ -72,6 +72,12 @@ The references summary includes a count of each kind of reference, including branches, remote refs, and tags (split by "all" and "annotated"). +Reachable Object Summary +~~~~~~~~~~~~~~~~~~~~~~~~ + +The reachable object summary shows the total number of each kind of Git +object, including tags, commits, trees, and blobs. + GIT --- Part of the linkgit:git[1] suite diff --git a/builtin/survey.c b/builtin/survey.c index e79f97f8d75923..1e8b9c1e5492aa 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -2,13 +2,20 @@ #include "builtin.h" #include "config.h" +#include "environment.h" +#include "hex.h" #include "object.h" #include "odb.h" +#include "object-name.h" #include "parse-options.h" +#include "path-walk.h" #include "progress.h" #include "ref-filter.h" +#include "refs.h" +#include "revision.h" #include "strbuf.h" #include "strvec.h" +#include "tag.h" #include "trace2.h" static const char * const survey_usage[] = { @@ -46,12 +53,20 @@ struct survey_report_ref_summary { size_t unknown_nr; }; +struct survey_report_object_summary { + size_t commits_nr; + size_t tags_nr; + size_t trees_nr; + size_t blobs_nr; +}; + /** * This struct contains all of the information that needs to be printed * at the end of the exploration of the repository and its references. */ struct survey_report { struct survey_report_ref_summary refs; + struct survey_report_object_summary reachable_objects; }; struct survey_context { @@ -74,10 +89,12 @@ struct survey_context { size_t progress_total; struct strvec refs; + struct ref_array ref_array; }; static void clear_survey_context(struct survey_context *ctx) { + ref_array_clear(&ctx->ref_array); strvec_clear(&ctx->refs); } @@ -128,10 +145,14 @@ static const size_t section_len = 4 * SECTION_SEGMENT_LEN; static void print_table_title(const char *name, size_t *widths, size_t nr) { size_t width = 3 * (nr - 1); + size_t min_width = strlen(name); for (size_t i = 0; i < nr; i++) width += widths[i]; + if (width < min_width) + width = min_width; + if (width > section_len) width = section_len; @@ -228,11 +249,43 @@ static void survey_report_plaintext_refs(struct survey_context *ctx) clear_table(&table); } +static void survey_report_plaintext_reachable_object_summary(struct survey_context *ctx) +{ + struct survey_report_object_summary *objs = &ctx->report.reachable_objects; + struct survey_table table = SURVEY_TABLE_INIT; + char *fmt; + + table.table_name = _("REACHABLE OBJECT SUMMARY"); + + strvec_push(&table.header, _("Object Type")); + strvec_push(&table.header, _("Count")); + + fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)objs->tags_nr); + insert_table_rowv(&table, _("Tags"), fmt, NULL); + free(fmt); + + fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)objs->commits_nr); + insert_table_rowv(&table, _("Commits"), fmt, NULL); + free(fmt); + + fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)objs->trees_nr); + insert_table_rowv(&table, _("Trees"), fmt, NULL); + free(fmt); + + fmt = xstrfmt("%"PRIuMAX"", (uintmax_t)objs->blobs_nr); + insert_table_rowv(&table, _("Blobs"), fmt, NULL); + free(fmt); + + print_table_plaintext(&table); + clear_table(&table); +} + static void survey_report_plaintext(struct survey_context *ctx) { printf("GIT SURVEY for \"%s\"\n", ctx->repo->worktree); printf("-----------------------------------------------------\n"); survey_report_plaintext_refs(ctx); + survey_report_plaintext_reachable_object_summary(ctx); } /* @@ -381,15 +434,13 @@ static void do_load_refs(struct survey_context *ctx, */ static void survey_phase_refs(struct survey_context *ctx) { - struct ref_array ref_array = { 0 }; - trace2_region_enter("survey", "phase/refs", ctx->repo); - do_load_refs(ctx, &ref_array); + do_load_refs(ctx, &ctx->ref_array); - ctx->report.refs.refs_nr = ref_array.nr; - for (int i = 0; i < ref_array.nr; i++) { + ctx->report.refs.refs_nr = ctx->ref_array.nr; + for (int i = 0; i < ctx->ref_array.nr; i++) { unsigned long size; - struct ref_array_item *item = ref_array.items[i]; + struct ref_array_item *item = ctx->ref_array.items[i]; switch (item->kind) { case FILTER_REFS_TAGS: @@ -419,8 +470,72 @@ static void survey_phase_refs(struct survey_context *ctx) } trace2_region_leave("survey", "phase/refs", ctx->repo); +} + +static void increment_object_counts( + struct survey_report_object_summary *summary, + enum object_type type, + size_t nr) +{ + switch (type) { + case OBJ_COMMIT: + summary->commits_nr += nr; + break; - ref_array_clear(&ref_array); + case OBJ_TREE: + summary->trees_nr += nr; + break; + + case OBJ_BLOB: + summary->blobs_nr += nr; + break; + + case OBJ_TAG: + summary->tags_nr += nr; + break; + + default: + break; + } +} + +static int survey_objects_path_walk_fn(const char *path UNUSED, + struct oid_array *oids, + enum object_type type, + void *data) +{ + struct survey_context *ctx = data; + + increment_object_counts(&ctx->report.reachable_objects, + type, oids->nr); + + return 0; +} + +static void survey_phase_objects(struct survey_context *ctx) +{ + struct rev_info revs = REV_INFO_INIT; + struct path_walk_info info = PATH_WALK_INFO_INIT; + unsigned int add_flags = 0; + + trace2_region_enter("survey", "phase/objects", ctx->repo); + + info.revs = &revs; + info.path_fn = survey_objects_path_walk_fn; + info.path_fn_data = ctx; + + repo_init_revisions(ctx->repo, &revs, ""); + revs.tag_objects = 1; + + for (int i = 0; i < ctx->ref_array.nr; i++) { + struct ref_array_item *item = ctx->ref_array.items[i]; + add_pending_oid(&revs, NULL, &item->objectname, add_flags); + } + + walk_objects_by_path(&info); + + release_revisions(&revs); + trace2_region_leave("survey", "phase/objects", ctx->repo); } int cmd_survey(int argc, const char **argv, const char *prefix, struct repository *repo) @@ -473,6 +588,8 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor survey_phase_refs(&ctx); + survey_phase_objects(&ctx); + survey_report_plaintext(&ctx); clear_survey_context(&ctx); diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index e518e4844fe2d0..d3086784090352 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -16,11 +16,17 @@ test_expect_success 'git survey -h shows experimental warning' ' ' test_expect_success 'create a semi-interesting repo' ' - test_commit_bulk 10 + test_commit_bulk 10 && + git tag -a -m one one HEAD~5 && + git tag -a -m two two HEAD~3 && + git tag -a -m three three two && + git tag -a -m four four three && + git update-ref -d refs/tags/three && + git update-ref -d refs/tags/two ' test_expect_success 'git survey (default)' ' - git survey >out 2>err && + git survey --all-refs >out 2>err && test_line_count = 0 err && tr , " " >expect <<-EOF && @@ -33,8 +39,17 @@ test_expect_success 'git survey (default)' ' -----------------+------ , Branches | 1 Remote refs | 0 - Tags (all) | 0 - Tags (annotated) | 0 + Tags (all) | 2 + Tags (annotated) | 2 + + REACHABLE OBJECT SUMMARY + ======================== + Object Type | Count + ------------+------ + Tags | 4 + Commits | 10 + Trees | 10 + Blobs | 10 EOF test_cmp expect out From 20f25ddfa65b94be9471d363c1452422664892f7 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 20:58:35 -0400 Subject: [PATCH 123/248] survey: summarize total sizes by object type Now that we have explored objects by count, we can expand that a bit more to summarize the data for the on-disk and inflated size of those objects. This information is helpful for diagnosing both why disk space (and perhaps clone or fetch times) is growing but also why certain operations are slow because the inflated size of the abstract objects that must be processed is so large. Signed-off-by: Derrick Stolee --- builtin/survey.c | 133 ++++++++++++++++++++++++++++++++++++++++++ t/t8100-git-survey.sh | 29 +++++++++ 2 files changed, 162 insertions(+) diff --git a/builtin/survey.c b/builtin/survey.c index 1e8b9c1e5492aa..1d1290553250a1 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -60,6 +60,19 @@ struct survey_report_object_summary { size_t blobs_nr; }; +/** + * For some category given by 'label', count the number of objects + * that match that label along with the on-disk size and the size + * after decompressing (both with delta bases and zlib). + */ +struct survey_report_object_size_summary { + char *label; + size_t nr; + size_t disk_size; + size_t inflated_size; + size_t num_missing; +}; + /** * This struct contains all of the information that needs to be printed * at the end of the exploration of the repository and its references. @@ -67,8 +80,16 @@ struct survey_report_object_summary { struct survey_report { struct survey_report_ref_summary refs; struct survey_report_object_summary reachable_objects; + + struct survey_report_object_size_summary *by_type; }; +#define REPORT_TYPE_COMMIT 0 +#define REPORT_TYPE_TREE 1 +#define REPORT_TYPE_BLOB 2 +#define REPORT_TYPE_TAG 3 +#define REPORT_TYPE_COUNT 4 + struct survey_context { struct repository *repo; @@ -280,12 +301,48 @@ static void survey_report_plaintext_reachable_object_summary(struct survey_conte clear_table(&table); } +static void survey_report_object_sizes(const char *title, + const char *categories, + struct survey_report_object_size_summary *summary, + size_t summary_nr) +{ + struct survey_table table = SURVEY_TABLE_INIT; + table.table_name = title; + + strvec_push(&table.header, categories); + strvec_push(&table.header, _("Count")); + strvec_push(&table.header, _("Disk Size")); + strvec_push(&table.header, _("Inflated Size")); + + for (size_t i = 0; i < summary_nr; i++) { + char *label_str = xstrdup(summary[i].label); + char *nr_str = xstrfmt("%"PRIuMAX, (uintmax_t)summary[i].nr); + char *disk_str = xstrfmt("%"PRIuMAX, (uintmax_t)summary[i].disk_size); + char *inflate_str = xstrfmt("%"PRIuMAX, (uintmax_t)summary[i].inflated_size); + + insert_table_rowv(&table, label_str, nr_str, + disk_str, inflate_str, NULL); + + free(label_str); + free(nr_str); + free(disk_str); + free(inflate_str); + } + + print_table_plaintext(&table); + clear_table(&table); +} + static void survey_report_plaintext(struct survey_context *ctx) { printf("GIT SURVEY for \"%s\"\n", ctx->repo->worktree); printf("-----------------------------------------------------\n"); survey_report_plaintext_refs(ctx); survey_report_plaintext_reachable_object_summary(ctx); + survey_report_object_sizes(_("TOTAL OBJECT SIZES BY TYPE"), + _("Object Type"), + ctx->report.by_type, + REPORT_TYPE_COUNT); } /* @@ -499,6 +556,69 @@ static void increment_object_counts( } } +static void increment_totals(struct survey_context *ctx, + struct oid_array *oids, + struct survey_report_object_size_summary *summary) +{ + for (size_t i = 0; i < oids->nr; i++) { + struct object_info oi = OBJECT_INFO_INIT; + unsigned oi_flags = OBJECT_INFO_FOR_PREFETCH; + unsigned long object_length = 0; + off_t disk_sizep = 0; + enum object_type type; + + oi.typep = &type; + oi.sizep = &object_length; + oi.disk_sizep = &disk_sizep; + + if (odb_read_object_info_extended(ctx->repo->objects, + &oids->oid[i], + &oi, oi_flags) < 0) { + summary->num_missing++; + } else { + summary->nr++; + summary->disk_size += disk_sizep; + summary->inflated_size += object_length; + } + } +} + +static void increment_object_totals(struct survey_context *ctx, + struct oid_array *oids, + enum object_type type) +{ + struct survey_report_object_size_summary *total; + struct survey_report_object_size_summary summary = { 0 }; + + increment_totals(ctx, oids, &summary); + + switch (type) { + case OBJ_COMMIT: + total = &ctx->report.by_type[REPORT_TYPE_COMMIT]; + break; + + case OBJ_TREE: + total = &ctx->report.by_type[REPORT_TYPE_TREE]; + break; + + case OBJ_BLOB: + total = &ctx->report.by_type[REPORT_TYPE_BLOB]; + break; + + case OBJ_TAG: + total = &ctx->report.by_type[REPORT_TYPE_TAG]; + break; + + default: + BUG("No other type allowed"); + } + + total->nr += summary.nr; + total->disk_size += summary.disk_size; + total->inflated_size += summary.inflated_size; + total->num_missing += summary.num_missing; +} + static int survey_objects_path_walk_fn(const char *path UNUSED, struct oid_array *oids, enum object_type type, @@ -508,10 +628,20 @@ static int survey_objects_path_walk_fn(const char *path UNUSED, increment_object_counts(&ctx->report.reachable_objects, type, oids->nr); + increment_object_totals(ctx, oids, type); return 0; } +static void initialize_report(struct survey_context *ctx) +{ + CALLOC_ARRAY(ctx->report.by_type, REPORT_TYPE_COUNT); + ctx->report.by_type[REPORT_TYPE_COMMIT].label = xstrdup(_("Commits")); + ctx->report.by_type[REPORT_TYPE_TREE].label = xstrdup(_("Trees")); + ctx->report.by_type[REPORT_TYPE_BLOB].label = xstrdup(_("Blobs")); + ctx->report.by_type[REPORT_TYPE_TAG].label = xstrdup(_("Tags")); +} + static void survey_phase_objects(struct survey_context *ctx) { struct rev_info revs = REV_INFO_INIT; @@ -524,12 +654,15 @@ static void survey_phase_objects(struct survey_context *ctx) info.path_fn = survey_objects_path_walk_fn; info.path_fn_data = ctx; + initialize_report(ctx); + repo_init_revisions(ctx->repo, &revs, ""); revs.tag_objects = 1; for (int i = 0; i < ctx->ref_array.nr; i++) { struct ref_array_item *item = ctx->ref_array.items[i]; add_pending_oid(&revs, NULL, &item->objectname, add_flags); + display_progress(ctx->progress, ++(ctx->progress_nr)); } walk_objects_by_path(&info); diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index d3086784090352..e31066bd07b335 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -29,6 +29,26 @@ test_expect_success 'git survey (default)' ' git survey --all-refs >out 2>err && test_line_count = 0 err && + test_oid_cache <<-EOF && + commits_size_on_disk sha1: 1523 + commits_size_on_disk sha256: 1811 + + commits_size sha1: 2153 + commits_size sha256: 2609 + + trees_size_on_disk sha1: 495 + trees_size_on_disk sha256: 635 + + trees_size sha1: 1706 + trees_size sha256: 2366 + + tags_size sha1: 528 + tags_size sha256: 624 + + tags_size_on_disk sha1: 510 + tags_size_on_disk sha256: 569 + EOF + tr , " " >expect <<-EOF && GIT SURVEY for "$(pwd)" ----------------------------------------------------- @@ -50,6 +70,15 @@ test_expect_success 'git survey (default)' ' Commits | 10 Trees | 10 Blobs | 10 + + TOTAL OBJECT SIZES BY TYPE + =============================================== + Object Type | Count | Disk Size | Inflated Size + ------------+-------+-----------+-------------- + Commits | 10 | $(test_oid commits_size_on_disk) | $(test_oid commits_size) + Trees | 10 | $(test_oid trees_size_on_disk) | $(test_oid trees_size) + Blobs | 10 | 191 | 101 + Tags | 4 | $(test_oid tags_size_on_disk) | $(test_oid tags_size) EOF test_cmp expect out From 4632f04b9e8ccb9c750eb272966ee6b66e89e51d Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 21:21:54 -0400 Subject: [PATCH 124/248] survey: show progress during object walk Signed-off-by: Derrick Stolee --- builtin/survey.c | 16 ++++++++++++++++ t/t8100-git-survey.sh | 5 +++++ 2 files changed, 21 insertions(+) diff --git a/builtin/survey.c b/builtin/survey.c index 1d1290553250a1..c570a1470122f4 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -630,6 +630,9 @@ static int survey_objects_path_walk_fn(const char *path UNUSED, type, oids->nr); increment_object_totals(ctx, oids, type); + ctx->progress_nr += oids->nr; + display_progress(ctx->progress, ctx->progress_nr); + return 0; } @@ -659,13 +662,26 @@ static void survey_phase_objects(struct survey_context *ctx) repo_init_revisions(ctx->repo, &revs, ""); revs.tag_objects = 1; + ctx->progress_nr = 0; + ctx->progress_total = ctx->ref_array.nr; + if (ctx->opts.show_progress) + ctx->progress = start_progress(ctx->repo, + _("Preparing object walk"), + ctx->progress_total); for (int i = 0; i < ctx->ref_array.nr; i++) { struct ref_array_item *item = ctx->ref_array.items[i]; add_pending_oid(&revs, NULL, &item->objectname, add_flags); display_progress(ctx->progress, ++(ctx->progress_nr)); } + stop_progress(&ctx->progress); + ctx->progress_nr = 0; + ctx->progress_total = 0; + if (ctx->opts.show_progress) + ctx->progress = start_progress(ctx->repo, + _("Walking objects"), 0); walk_objects_by_path(&info); + stop_progress(&ctx->progress); release_revisions(&revs); trace2_region_leave("survey", "phase/objects", ctx->repo); diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index e31066bd07b335..9636f147a0b14c 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -25,6 +25,11 @@ test_expect_success 'create a semi-interesting repo' ' git update-ref -d refs/tags/two ' +test_expect_success 'git survey --progress' ' + GIT_PROGRESS_DELAY=0 git survey --all-refs --progress >out 2>err && + grep "Preparing object walk" err +' + test_expect_success 'git survey (default)' ' git survey --all-refs >out 2>err && test_line_count = 0 err && From 7ebf4b370d695cba015fd32186e81e23d584d3f3 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 22:35:06 -0400 Subject: [PATCH 125/248] survey: add ability to track prioritized lists In future changes, we will make use of these methods. The intention is to keep track of the top contributors according to some metric. We don't want to store all of the entries and do a sort at the end, so track a constant-size table and remove rows that get pushed out depending on the chosen sorting algorithm. Co-authored-by: Jeff Hostetler Signed-off-by; Jeff Hostetler Signed-off-by: Derrick Stolee --- builtin/survey.c | 113 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 113 insertions(+) diff --git a/builtin/survey.c b/builtin/survey.c index c570a1470122f4..5ff62fa4ab921c 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -73,6 +73,119 @@ struct survey_report_object_size_summary { size_t num_missing; }; +typedef int (*survey_top_cmp)(void *v1, void *v2); + +MAYBE_UNUSED +static int cmp_by_nr(void *v1, void *v2) +{ + struct survey_report_object_size_summary *s1 = v1; + struct survey_report_object_size_summary *s2 = v2; + + if (s1->nr < s2->nr) + return -1; + if (s1->nr > s2->nr) + return 1; + return 0; +} + +MAYBE_UNUSED +static int cmp_by_disk_size(void *v1, void *v2) +{ + struct survey_report_object_size_summary *s1 = v1; + struct survey_report_object_size_summary *s2 = v2; + + if (s1->disk_size < s2->disk_size) + return -1; + if (s1->disk_size > s2->disk_size) + return 1; + return 0; +} + +MAYBE_UNUSED +static int cmp_by_inflated_size(void *v1, void *v2) +{ + struct survey_report_object_size_summary *s1 = v1; + struct survey_report_object_size_summary *s2 = v2; + + if (s1->inflated_size < s2->inflated_size) + return -1; + if (s1->inflated_size > s2->inflated_size) + return 1; + return 0; +} + +/** + * Store a list of "top" categories by some sorting function. When + * inserting a new category, reorder the list and free the one that + * got ejected (if any). + */ +struct survey_report_top_table { + const char *name; + survey_top_cmp cmp_fn; + size_t nr; + size_t alloc; + + /** + * 'data' stores an array of structs and must be cast into + * the proper array type before evaluating an index. + */ + void *data; +}; + +MAYBE_UNUSED +static void init_top_sizes(struct survey_report_top_table *top, + size_t limit, const char *name, + survey_top_cmp cmp) +{ + struct survey_report_object_size_summary *sz_array; + + top->name = name; + top->cmp_fn = cmp; + top->alloc = limit; + top->nr = 0; + + CALLOC_ARRAY(sz_array, limit); + top->data = sz_array; +} + +MAYBE_UNUSED +static void clear_top_sizes(struct survey_report_top_table *top) +{ + struct survey_report_object_size_summary *sz_array = top->data; + + for (size_t i = 0; i < top->nr; i++) + free(sz_array[i].label); + free(sz_array); +} + +MAYBE_UNUSED +static void maybe_insert_into_top_size(struct survey_report_top_table *top, + struct survey_report_object_size_summary *summary) +{ + struct survey_report_object_size_summary *sz_array = top->data; + size_t pos = top->nr; + + /* Compare against list from the bottom. */ + while (pos > 0 && top->cmp_fn(&sz_array[pos - 1], summary) < 0) + pos--; + + /* Not big enough! */ + if (pos >= top->alloc) + return; + + /* We need to shift the data. */ + if (top->nr == top->alloc) + free(sz_array[top->nr - 1].label); + else + top->nr++; + + for (size_t i = top->nr - 1; i > pos; i--) + memcpy(&sz_array[i], &sz_array[i - 1], sizeof(*sz_array)); + + memcpy(&sz_array[pos], summary, sizeof(*summary)); + sz_array[pos].label = xstrdup(summary->label); +} + /** * This struct contains all of the information that needs to be printed * at the end of the exploration of the repository and its references. From f6f250ff8bf874e5af119e35c42aa563d5594132 Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Sun, 1 Sep 2024 22:35:40 -0400 Subject: [PATCH 126/248] survey: add report of "largest" paths MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since we are already walking our reachable objects using the path-walk API, let's now collect lists of the paths that contribute most to different metrics. Specifically, we care about * Number of versions. * Total size on disk. * Total inflated size (no delta or zlib compression). This information can be critical to discovering which parts of the repository are causing the most growth, especially on-disk size. Different packing strategies might help compress data more efficiently, but the toal inflated size is a representation of the raw size of all snapshots of those paths. Even when stored efficiently on disk, that size represents how much information must be processed to complete a command such as 'git blame'. The exact disk size seems to be not quite robust enough for testing, as could be seen by the `linux-musl-meson` job consistently failing, possibly because of zlib-ng deflates differently: t8100.4(git survey (default)) was failing with a symptom like this: TOTAL OBJECT SIZES BY TYPE =============================================== Object Type | Count | Disk Size | Inflated Size ------------+-------+-----------+-------------- - Commits | 10 | 1523 | 2153 + Commits | 10 | 1528 | 2153 Trees | 10 | 495 | 1706 Blobs | 10 | 191 | 101 - Tags | 4 | 510 | 528 + Tags | 4 | 547 | 528 This means: the disk size is unlikely something we can verify robustly. Since zlib-ng seems to increase the disk size of the tags from 528 to 547, we cannot even assume that the disk size is always smaller than the inflated size. We will most likely want to either skip verifying the disk size altogether, or go for some kind of fuzzy matching, say, by replacing `s/ 1[45][0-9][0-9] / ~1.5k /` and `s/ [45][0-9][0-9] / ~½k /` or something like that. Signed-off-by: Derrick Stolee Signed-off-by: Johannes Schindelin --- builtin/survey.c | 79 ++++++++++++++++++++++++++++++++++++++----- t/t8100-git-survey.sh | 13 ++++++- 2 files changed, 83 insertions(+), 9 deletions(-) diff --git a/builtin/survey.c b/builtin/survey.c index 5ff62fa4ab921c..2dd1eedfda74f1 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -75,7 +75,6 @@ struct survey_report_object_size_summary { typedef int (*survey_top_cmp)(void *v1, void *v2); -MAYBE_UNUSED static int cmp_by_nr(void *v1, void *v2) { struct survey_report_object_size_summary *s1 = v1; @@ -88,7 +87,6 @@ static int cmp_by_nr(void *v1, void *v2) return 0; } -MAYBE_UNUSED static int cmp_by_disk_size(void *v1, void *v2) { struct survey_report_object_size_summary *s1 = v1; @@ -101,7 +99,6 @@ static int cmp_by_disk_size(void *v1, void *v2) return 0; } -MAYBE_UNUSED static int cmp_by_inflated_size(void *v1, void *v2) { struct survey_report_object_size_summary *s1 = v1; @@ -132,7 +129,6 @@ struct survey_report_top_table { void *data; }; -MAYBE_UNUSED static void init_top_sizes(struct survey_report_top_table *top, size_t limit, const char *name, survey_top_cmp cmp) @@ -158,7 +154,6 @@ static void clear_top_sizes(struct survey_report_top_table *top) free(sz_array); } -MAYBE_UNUSED static void maybe_insert_into_top_size(struct survey_report_top_table *top, struct survey_report_object_size_summary *summary) { @@ -195,6 +190,10 @@ struct survey_report { struct survey_report_object_summary reachable_objects; struct survey_report_object_size_summary *by_type; + + struct survey_report_top_table *top_paths_by_count; + struct survey_report_top_table *top_paths_by_disk; + struct survey_report_top_table *top_paths_by_inflate; }; #define REPORT_TYPE_COMMIT 0 @@ -446,6 +445,13 @@ static void survey_report_object_sizes(const char *title, clear_table(&table); } +static void survey_report_plaintext_sorted_size( + struct survey_report_top_table *top) +{ + survey_report_object_sizes(top->name, _("Path"), + top->data, top->nr); +} + static void survey_report_plaintext(struct survey_context *ctx) { printf("GIT SURVEY for \"%s\"\n", ctx->repo->worktree); @@ -456,6 +462,21 @@ static void survey_report_plaintext(struct survey_context *ctx) _("Object Type"), ctx->report.by_type, REPORT_TYPE_COUNT); + + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_count[REPORT_TYPE_TREE]); + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_count[REPORT_TYPE_BLOB]); + + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_disk[REPORT_TYPE_TREE]); + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_disk[REPORT_TYPE_BLOB]); + + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_inflate[REPORT_TYPE_TREE]); + survey_report_plaintext_sorted_size( + &ctx->report.top_paths_by_inflate[REPORT_TYPE_BLOB]); } /* @@ -698,7 +719,8 @@ static void increment_totals(struct survey_context *ctx, static void increment_object_totals(struct survey_context *ctx, struct oid_array *oids, - enum object_type type) + enum object_type type, + const char *path) { struct survey_report_object_size_summary *total; struct survey_report_object_size_summary summary = { 0 }; @@ -730,9 +752,30 @@ static void increment_object_totals(struct survey_context *ctx, total->disk_size += summary.disk_size; total->inflated_size += summary.inflated_size; total->num_missing += summary.num_missing; + + if (type == OBJ_TREE || type == OBJ_BLOB) { + int index = type == OBJ_TREE ? + REPORT_TYPE_TREE : REPORT_TYPE_BLOB; + struct survey_report_top_table *top; + + /* + * Temporarily store (const char *) here, but it will + * be duped if inserted and will not be freed. + */ + summary.label = (char *)path; + + top = ctx->report.top_paths_by_count; + maybe_insert_into_top_size(&top[index], &summary); + + top = ctx->report.top_paths_by_disk; + maybe_insert_into_top_size(&top[index], &summary); + + top = ctx->report.top_paths_by_inflate; + maybe_insert_into_top_size(&top[index], &summary); + } } -static int survey_objects_path_walk_fn(const char *path UNUSED, +static int survey_objects_path_walk_fn(const char *path, struct oid_array *oids, enum object_type type, void *data) @@ -741,7 +784,7 @@ static int survey_objects_path_walk_fn(const char *path UNUSED, increment_object_counts(&ctx->report.reachable_objects, type, oids->nr); - increment_object_totals(ctx, oids, type); + increment_object_totals(ctx, oids, type, path); ctx->progress_nr += oids->nr; display_progress(ctx->progress, ctx->progress_nr); @@ -751,11 +794,31 @@ static int survey_objects_path_walk_fn(const char *path UNUSED, static void initialize_report(struct survey_context *ctx) { + const int top_limit = 100; + CALLOC_ARRAY(ctx->report.by_type, REPORT_TYPE_COUNT); ctx->report.by_type[REPORT_TYPE_COMMIT].label = xstrdup(_("Commits")); ctx->report.by_type[REPORT_TYPE_TREE].label = xstrdup(_("Trees")); ctx->report.by_type[REPORT_TYPE_BLOB].label = xstrdup(_("Blobs")); ctx->report.by_type[REPORT_TYPE_TAG].label = xstrdup(_("Tags")); + + CALLOC_ARRAY(ctx->report.top_paths_by_count, REPORT_TYPE_COUNT); + init_top_sizes(&ctx->report.top_paths_by_count[REPORT_TYPE_TREE], + top_limit, _("TOP DIRECTORIES BY COUNT"), cmp_by_nr); + init_top_sizes(&ctx->report.top_paths_by_count[REPORT_TYPE_BLOB], + top_limit, _("TOP FILES BY COUNT"), cmp_by_nr); + + CALLOC_ARRAY(ctx->report.top_paths_by_disk, REPORT_TYPE_COUNT); + init_top_sizes(&ctx->report.top_paths_by_disk[REPORT_TYPE_TREE], + top_limit, _("TOP DIRECTORIES BY DISK SIZE"), cmp_by_disk_size); + init_top_sizes(&ctx->report.top_paths_by_disk[REPORT_TYPE_BLOB], + top_limit, _("TOP FILES BY DISK SIZE"), cmp_by_disk_size); + + CALLOC_ARRAY(ctx->report.top_paths_by_inflate, REPORT_TYPE_COUNT); + init_top_sizes(&ctx->report.top_paths_by_inflate[REPORT_TYPE_TREE], + top_limit, _("TOP DIRECTORIES BY INFLATED SIZE"), cmp_by_inflated_size); + init_top_sizes(&ctx->report.top_paths_by_inflate[REPORT_TYPE_BLOB], + top_limit, _("TOP FILES BY INFLATED SIZE"), cmp_by_inflated_size); } static void survey_phase_objects(struct survey_context *ctx) diff --git a/t/t8100-git-survey.sh b/t/t8100-git-survey.sh index 9636f147a0b14c..eba9624ed5825c 100755 --- a/t/t8100-git-survey.sh +++ b/t/t8100-git-survey.sh @@ -86,7 +86,18 @@ test_expect_success 'git survey (default)' ' Tags | 4 | $(test_oid tags_size_on_disk) | $(test_oid tags_size) EOF - test_cmp expect out + lines=$(wc -l out-trimmed && + sed -e "s/ 1528 / 1523 /" -e "s/ 547 / 510 /" out-trimmed >out-edited && + test_cmp expect out-edited && + + for type in "DIRECTORIES" "FILES" + do + for metric in "COUNT" "DISK SIZE" "INFLATED SIZE" + do + grep "TOP $type BY $metric" out || return 1 + done || return 1 + done ' test_done From 16cad3911a4b7ab761a8d518c8b5019badc7108a Mon Sep 17 00:00:00 2001 From: Derrick Stolee Date: Mon, 23 Sep 2024 15:38:25 -0400 Subject: [PATCH 127/248] survey: add --top= option and config The 'git survey' builtin provides several detail tables, such as "top files by on-disk size". The size of these tables defaults to 10, currently. Allow the user to specify this number via a new --top= option or the new survey.top config key. Signed-off-by: Derrick Stolee Signed-off-by: Johannes Schindelin --- Documentation/config/survey.adoc | 3 +++ builtin/survey.c | 22 ++++++++++++++-------- 2 files changed, 17 insertions(+), 8 deletions(-) diff --git a/Documentation/config/survey.adoc b/Documentation/config/survey.adoc index c1b0f852a1250e..9e594a2092f225 100644 --- a/Documentation/config/survey.adoc +++ b/Documentation/config/survey.adoc @@ -8,4 +8,7 @@ survey.*:: This boolean value implies the `--[no-]verbose` option. progress:: This boolean value implies the `--[no-]progress` option. + top:: + This integer value implies `--top=`, specifying the + number of entries in the detail tables. -- diff --git a/builtin/survey.c b/builtin/survey.c index 2dd1eedfda74f1..c1d78222146628 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -40,6 +40,7 @@ static struct survey_refs_wanted default_ref_options = { struct survey_opts { int verbose; int show_progress; + int top_nr; struct survey_refs_wanted refs; }; @@ -548,6 +549,10 @@ static int survey_load_config_cb(const char *var, const char *value, ctx->opts.show_progress = git_config_bool(var, value); return 0; } + if (!strcmp(var, "survey.top")) { + ctx->opts.top_nr = git_config_bool(var, value); + return 0; + } return git_default_config(var, value, cctx, pvoid); } @@ -794,8 +799,6 @@ static int survey_objects_path_walk_fn(const char *path, static void initialize_report(struct survey_context *ctx) { - const int top_limit = 100; - CALLOC_ARRAY(ctx->report.by_type, REPORT_TYPE_COUNT); ctx->report.by_type[REPORT_TYPE_COMMIT].label = xstrdup(_("Commits")); ctx->report.by_type[REPORT_TYPE_TREE].label = xstrdup(_("Trees")); @@ -804,21 +807,21 @@ static void initialize_report(struct survey_context *ctx) CALLOC_ARRAY(ctx->report.top_paths_by_count, REPORT_TYPE_COUNT); init_top_sizes(&ctx->report.top_paths_by_count[REPORT_TYPE_TREE], - top_limit, _("TOP DIRECTORIES BY COUNT"), cmp_by_nr); + ctx->opts.top_nr, _("TOP DIRECTORIES BY COUNT"), cmp_by_nr); init_top_sizes(&ctx->report.top_paths_by_count[REPORT_TYPE_BLOB], - top_limit, _("TOP FILES BY COUNT"), cmp_by_nr); + ctx->opts.top_nr, _("TOP FILES BY COUNT"), cmp_by_nr); CALLOC_ARRAY(ctx->report.top_paths_by_disk, REPORT_TYPE_COUNT); init_top_sizes(&ctx->report.top_paths_by_disk[REPORT_TYPE_TREE], - top_limit, _("TOP DIRECTORIES BY DISK SIZE"), cmp_by_disk_size); + ctx->opts.top_nr, _("TOP DIRECTORIES BY DISK SIZE"), cmp_by_disk_size); init_top_sizes(&ctx->report.top_paths_by_disk[REPORT_TYPE_BLOB], - top_limit, _("TOP FILES BY DISK SIZE"), cmp_by_disk_size); + ctx->opts.top_nr, _("TOP FILES BY DISK SIZE"), cmp_by_disk_size); CALLOC_ARRAY(ctx->report.top_paths_by_inflate, REPORT_TYPE_COUNT); init_top_sizes(&ctx->report.top_paths_by_inflate[REPORT_TYPE_TREE], - top_limit, _("TOP DIRECTORIES BY INFLATED SIZE"), cmp_by_inflated_size); + ctx->opts.top_nr, _("TOP DIRECTORIES BY INFLATED SIZE"), cmp_by_inflated_size); init_top_sizes(&ctx->report.top_paths_by_inflate[REPORT_TYPE_BLOB], - top_limit, _("TOP FILES BY INFLATED SIZE"), cmp_by_inflated_size); + ctx->opts.top_nr, _("TOP FILES BY INFLATED SIZE"), cmp_by_inflated_size); } static void survey_phase_objects(struct survey_context *ctx) @@ -869,6 +872,7 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor .opts = { .verbose = 0, .show_progress = -1, /* defaults to isatty(2) */ + .top_nr = 10, .refs.want_all_refs = -1, @@ -884,6 +888,8 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor static struct option survey_options[] = { OPT__VERBOSE(&ctx.opts.verbose, N_("verbose output")), OPT_BOOL(0, "progress", &ctx.opts.show_progress, N_("show progress")), + OPT_INTEGER('n', "top", &ctx.opts.top_nr, + N_("number of entries to include in detail tables")), OPT_BOOL_F(0, "all-refs", &ctx.opts.refs.want_all_refs, N_("include all refs"), PARSE_OPT_NONEG), From 64fbfff392efc118f3dc364d9001d8cf43df9060 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 14 Nov 2019 20:09:23 +0100 Subject: [PATCH 128/248] mingw: make sure `errno` is set correctly when socket operations fail The winsock2 library provides functions that work on different data types than file descriptors, therefore we wrap them. But that is not the only difference: they also do not set `errno` but expect the callers to enquire about errors via `WSAGetLastError()`. Let's translate that into appropriate `errno` values whenever the socket operations fail so that Git's code base does not have to change its expectations. This closes https://github.com/git-for-windows/git/issues/2404 Helped-by: Jeff Hostetler Signed-off-by: Johannes Schindelin --- compat/mingw.c | 157 +++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 147 insertions(+), 10 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..322a87268814c7 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2089,18 +2089,150 @@ static void ensure_socket_initialization(void) initialized = 1; } +static int winsock_error_to_errno(DWORD err) +{ + switch (err) { + case WSAEINTR: return EINTR; + case WSAEBADF: return EBADF; + case WSAEACCES: return EACCES; + case WSAEFAULT: return EFAULT; + case WSAEINVAL: return EINVAL; + case WSAEMFILE: return EMFILE; + case WSAEWOULDBLOCK: return EWOULDBLOCK; + case WSAEINPROGRESS: return EINPROGRESS; + case WSAEALREADY: return EALREADY; + case WSAENOTSOCK: return ENOTSOCK; + case WSAEDESTADDRREQ: return EDESTADDRREQ; + case WSAEMSGSIZE: return EMSGSIZE; + case WSAEPROTOTYPE: return EPROTOTYPE; + case WSAENOPROTOOPT: return ENOPROTOOPT; + case WSAEPROTONOSUPPORT: return EPROTONOSUPPORT; + case WSAEOPNOTSUPP: return EOPNOTSUPP; + case WSAEAFNOSUPPORT: return EAFNOSUPPORT; + case WSAEADDRINUSE: return EADDRINUSE; + case WSAEADDRNOTAVAIL: return EADDRNOTAVAIL; + case WSAENETDOWN: return ENETDOWN; + case WSAENETUNREACH: return ENETUNREACH; + case WSAENETRESET: return ENETRESET; + case WSAECONNABORTED: return ECONNABORTED; + case WSAECONNRESET: return ECONNRESET; + case WSAENOBUFS: return ENOBUFS; + case WSAEISCONN: return EISCONN; + case WSAENOTCONN: return ENOTCONN; + case WSAETIMEDOUT: return ETIMEDOUT; + case WSAECONNREFUSED: return ECONNREFUSED; + case WSAELOOP: return ELOOP; + case WSAENAMETOOLONG: return ENAMETOOLONG; + case WSAEHOSTUNREACH: return EHOSTUNREACH; + case WSAENOTEMPTY: return ENOTEMPTY; + /* No errno equivalent; default to EIO */ + case WSAESOCKTNOSUPPORT: + case WSAEPFNOSUPPORT: + case WSAESHUTDOWN: + case WSAETOOMANYREFS: + case WSAEHOSTDOWN: + case WSAEPROCLIM: + case WSAEUSERS: + case WSAEDQUOT: + case WSAESTALE: + case WSAEREMOTE: + case WSASYSNOTREADY: + case WSAVERNOTSUPPORTED: + case WSANOTINITIALISED: + case WSAEDISCON: + case WSAENOMORE: + case WSAECANCELLED: + case WSAEINVALIDPROCTABLE: + case WSAEINVALIDPROVIDER: + case WSAEPROVIDERFAILEDINIT: + case WSASYSCALLFAILURE: + case WSASERVICE_NOT_FOUND: + case WSATYPE_NOT_FOUND: + case WSA_E_NO_MORE: + case WSA_E_CANCELLED: + case WSAEREFUSED: + case WSAHOST_NOT_FOUND: + case WSATRY_AGAIN: + case WSANO_RECOVERY: + case WSANO_DATA: + case WSA_QOS_RECEIVERS: + case WSA_QOS_SENDERS: + case WSA_QOS_NO_SENDERS: + case WSA_QOS_NO_RECEIVERS: + case WSA_QOS_REQUEST_CONFIRMED: + case WSA_QOS_ADMISSION_FAILURE: + case WSA_QOS_POLICY_FAILURE: + case WSA_QOS_BAD_STYLE: + case WSA_QOS_BAD_OBJECT: + case WSA_QOS_TRAFFIC_CTRL_ERROR: + case WSA_QOS_GENERIC_ERROR: + case WSA_QOS_ESERVICETYPE: + case WSA_QOS_EFLOWSPEC: + case WSA_QOS_EPROVSPECBUF: + case WSA_QOS_EFILTERSTYLE: + case WSA_QOS_EFILTERTYPE: + case WSA_QOS_EFILTERCOUNT: + case WSA_QOS_EOBJLENGTH: + case WSA_QOS_EFLOWCOUNT: +#ifndef _MSC_VER + case WSA_QOS_EUNKNOWNPSOBJ: +#endif + case WSA_QOS_EPOLICYOBJ: + case WSA_QOS_EFLOWDESC: + case WSA_QOS_EPSFLOWSPEC: + case WSA_QOS_EPSFILTERSPEC: + case WSA_QOS_ESDMODEOBJ: + case WSA_QOS_ESHAPERATEOBJ: + case WSA_QOS_RESERVED_PETYPE: + default: return EIO; + } +} + +/* + * On Windows, `errno` is a global macro to a function call. + * This makes it difficult to debug and single-step our mappings. + */ +static inline void set_wsa_errno(void) +{ + DWORD wsa = WSAGetLastError(); + int e = winsock_error_to_errno(wsa); + errno = e; + +#ifdef DEBUG_WSA_ERRNO + fprintf(stderr, "winsock error: %d -> %d\n", wsa, e); + fflush(stderr); +#endif +} + +static inline int winsock_return(int ret) +{ + if (ret < 0) + set_wsa_errno(); + + return ret; +} + +#define WINSOCK_RETURN(x) do { return winsock_return(x); } while (0) + #undef gethostname int mingw_gethostname(char *name, int namelen) { - ensure_socket_initialization(); - return gethostname(name, namelen); + ensure_socket_initialization(); + WINSOCK_RETURN(gethostname(name, namelen)); } #undef gethostbyname struct hostent *mingw_gethostbyname(const char *host) { + struct hostent *ret; + ensure_socket_initialization(); - return gethostbyname(host); + + ret = gethostbyname(host); + if (!ret) + set_wsa_errno(); + + return ret; } #undef getaddrinfo @@ -2108,7 +2240,7 @@ int mingw_getaddrinfo(const char *node, const char *service, const struct addrinfo *hints, struct addrinfo **res) { ensure_socket_initialization(); - return getaddrinfo(node, service, hints, res); + WINSOCK_RETURN(getaddrinfo(node, service, hints, res)); } int mingw_socket(int domain, int type, int protocol) @@ -2128,7 +2260,7 @@ int mingw_socket(int domain, int type, int protocol) * in errno so that _if_ someone looks up the code somewhere, * then it is at least the number that are usually listed. */ - errno = WSAGetLastError(); + set_wsa_errno(); return -1; } /* convert into a file descriptor */ @@ -2144,35 +2276,35 @@ int mingw_socket(int domain, int type, int protocol) int mingw_connect(int sockfd, struct sockaddr *sa, size_t sz) { SOCKET s = (SOCKET)_get_osfhandle(sockfd); - return connect(s, sa, sz); + WINSOCK_RETURN(connect(s, sa, sz)); } #undef bind int mingw_bind(int sockfd, struct sockaddr *sa, size_t sz) { SOCKET s = (SOCKET)_get_osfhandle(sockfd); - return bind(s, sa, sz); + WINSOCK_RETURN(bind(s, sa, sz)); } #undef setsockopt int mingw_setsockopt(int sockfd, int lvl, int optname, void *optval, int optlen) { SOCKET s = (SOCKET)_get_osfhandle(sockfd); - return setsockopt(s, lvl, optname, (const char*)optval, optlen); + WINSOCK_RETURN(setsockopt(s, lvl, optname, (const char*)optval, optlen)); } #undef shutdown int mingw_shutdown(int sockfd, int how) { SOCKET s = (SOCKET)_get_osfhandle(sockfd); - return shutdown(s, how); + WINSOCK_RETURN(shutdown(s, how)); } #undef listen int mingw_listen(int sockfd, int backlog) { SOCKET s = (SOCKET)_get_osfhandle(sockfd); - return listen(s, backlog); + WINSOCK_RETURN(listen(s, backlog)); } #undef accept @@ -2183,6 +2315,11 @@ int mingw_accept(int sockfd1, struct sockaddr *sa, socklen_t *sz) SOCKET s1 = (SOCKET)_get_osfhandle(sockfd1); SOCKET s2 = accept(s1, sa, sz); + if (s2 == INVALID_SOCKET) { + set_wsa_errno(); + return -1; + } + /* convert into a file descriptor */ if ((sockfd2 = _open_osfhandle(s2, O_RDWR|O_BINARY)) < 0) { int err = errno; From ba05f2f6cb942199f34bc3442dcafa1b18996d13 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 1 Jul 2024 23:28:45 +0200 Subject: [PATCH 129/248] survey: clearly note the experimental nature in the output While this command is definitely something we _want_, chances are that upstreaming this will require substantial changes. We still want to be able to experiment with this before that, to focus on what we need out of this command: To assist with diagnosing issues with large repositories, as well as to help monitoring the growth and the associated painpoints of such repositories. To that end, we are about to integrate this command into `microsoft/git`, to get the tool into the hands of users who need it most, with the idea to iterate in close collaboration between these users and the developers familar with Git's internals. However, we will definitely want to avoid letting anybody have the impression that this command, its exact inner workings, as well as its output format, are anywhere close to stable. To make that fact utterly clear (and thereby protect the freedom to iterate and innovate freely before upstreaming the command), let's mark its output as experimental in all-caps, as the first thing we do. Signed-off-by: Johannes Schindelin --- builtin/survey.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/builtin/survey.c b/builtin/survey.c index c1d78222146628..f40905fb2fd57a 100644 --- a/builtin/survey.c +++ b/builtin/survey.c @@ -17,6 +17,7 @@ #include "strvec.h" #include "tag.h" #include "trace2.h" +#include "color.h" static const char * const survey_usage[] = { N_("(EXPERIMENTAL!) git survey "), @@ -905,6 +906,11 @@ int cmd_survey(int argc, const char **argv, const char *prefix, struct repositor show_usage_with_options_if_asked(argc, argv, survey_usage, survey_options); + if (isatty(2)) + color_fprintf_ln(stderr, + want_color_fd(2, GIT_COLOR_AUTO) ? GIT_COLOR_YELLOW : "", + "(THIS IS EXPERIMENTAL, EXPECT THE OUTPUT FORMAT TO CHANGE!)"); + ctx.repo = repo; prepare_repo_settings(ctx.repo); From 88eb4d47bb94f7b5f3857b736935d95df77139a3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sun, 22 Dec 2024 17:15:39 +0100 Subject: [PATCH 130/248] compat/mingw: handle WSA errors in strerror MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We map WSAGetLastError() errors to errno errors in winsock_error_to_errno(), but the MSVC strerror() implementation only produces "Unknown error" for most of them. Produce some more meaningful error messages in these cases. Our builds for ARM64 link against the newer UCRT strerror() that does know these errors, so we won't change the strerror() used there. The wording of the messages is copied from glibc strerror() messages. Reported-by: M Hickford Signed-off-by: Matthias Aßhauer Signed-off-by: Johannes Schindelin --- Makefile | 1 + compat/mingw-posix.h | 5 +++ compat/mingw.c | 85 ++++++++++++++++++++++++++++++++++++++++++ t/meson.build | 1 + t/unit-tests/u-mingw.c | 72 +++++++++++++++++++++++++++++++++++ 5 files changed, 164 insertions(+) create mode 100644 t/unit-tests/u-mingw.c diff --git a/Makefile b/Makefile index 1f444eec3c1d12..35cedc0901fdce 100644 --- a/Makefile +++ b/Makefile @@ -1360,6 +1360,7 @@ CLAR_TEST_SUITES += u-example-decorate CLAR_TEST_SUITES += u-hash CLAR_TEST_SUITES += u-hashmap CLAR_TEST_SUITES += u-mem-pool +CLAR_TEST_SUITES += u-mingw CLAR_TEST_SUITES += u-oid-array CLAR_TEST_SUITES += u-oidmap CLAR_TEST_SUITES += u-oidtree diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index 631a20868489be..a0c7986fd267b0 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -293,6 +293,11 @@ int mingw_socket(int domain, int type, int protocol); int mingw_connect(int sockfd, struct sockaddr *sa, size_t sz); #define connect mingw_connect +char *mingw_strerror(int errnum); +#ifndef _UCRT +#define strerror mingw_strerror +#endif + int mingw_bind(int sockfd, struct sockaddr *sa, size_t sz); #define bind mingw_bind diff --git a/compat/mingw.c b/compat/mingw.c index 322a87268814c7..845f0898e2a86c 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2214,6 +2214,91 @@ static inline int winsock_return(int ret) #define WINSOCK_RETURN(x) do { return winsock_return(x); } while (0) +#undef strerror +char *mingw_strerror(int errnum) +{ + static char buf[41] =""; + switch (errnum) { + case EWOULDBLOCK: + xsnprintf(buf, 41, "%s", "Operation would block"); + break; + case EINPROGRESS: + xsnprintf(buf, 41, "%s", "Operation now in progress"); + break; + case EALREADY: + xsnprintf(buf, 41, "%s", "Operation already in progress"); + break; + case ENOTSOCK: + xsnprintf(buf, 41, "%s", "Socket operation on non-socket"); + break; + case EDESTADDRREQ: + xsnprintf(buf, 41, "%s", "Destination address required"); + break; + case EMSGSIZE: + xsnprintf(buf, 41, "%s", "Message too long"); + break; + case EPROTOTYPE: + xsnprintf(buf, 41, "%s", "Protocol wrong type for socket"); + break; + case ENOPROTOOPT: + xsnprintf(buf, 41, "%s", "Protocol not available"); + break; + case EPROTONOSUPPORT: + xsnprintf(buf, 41, "%s", "Protocol not supported"); + break; + case EOPNOTSUPP: + xsnprintf(buf, 41, "%s", "Operation not supported"); + break; + case EAFNOSUPPORT: + xsnprintf(buf, 41, "%s", "Address family not supported by protocol"); + break; + case EADDRINUSE: + xsnprintf(buf, 41, "%s", "Address already in use"); + break; + case EADDRNOTAVAIL: + xsnprintf(buf, 41, "%s", "Cannot assign requested address"); + break; + case ENETDOWN: + xsnprintf(buf, 41, "%s", "Network is down"); + break; + case ENETUNREACH: + xsnprintf(buf, 41, "%s", "Network is unreachable"); + break; + case ENETRESET: + xsnprintf(buf, 41, "%s", "Network dropped connection on reset"); + break; + case ECONNABORTED: + xsnprintf(buf, 41, "%s", "Software caused connection abort"); + break; + case ECONNRESET: + xsnprintf(buf, 41, "%s", "Connection reset by peer"); + break; + case ENOBUFS: + xsnprintf(buf, 41, "%s", "No buffer space available"); + break; + case EISCONN: + xsnprintf(buf, 41, "%s", "Transport endpoint is already connected"); + break; + case ENOTCONN: + xsnprintf(buf, 41, "%s", "Transport endpoint is not connected"); + break; + case ETIMEDOUT: + xsnprintf(buf, 41, "%s", "Connection timed out"); + break; + case ECONNREFUSED: + xsnprintf(buf, 41, "%s", "Connection refused"); + break; + case ELOOP: + xsnprintf(buf, 41, "%s", "Too many levels of symbolic links"); + break; + case EHOSTUNREACH: + xsnprintf(buf, 41, "%s", "No route to host"); + break; + default: return strerror(errnum); + } + return buf; +} + #undef gethostname int mingw_gethostname(char *name, int namelen) { diff --git a/t/meson.build b/t/meson.build index 983245501ce9a2..2bb496f56b2c9b 100644 --- a/t/meson.build +++ b/t/meson.build @@ -4,6 +4,7 @@ clar_test_suites = [ 'unit-tests/u-hash.c', 'unit-tests/u-hashmap.c', 'unit-tests/u-mem-pool.c', + 'unit-tests/u-mingw.c', 'unit-tests/u-oid-array.c', 'unit-tests/u-oidmap.c', 'unit-tests/u-oidtree.c', diff --git a/t/unit-tests/u-mingw.c b/t/unit-tests/u-mingw.c new file mode 100644 index 00000000000000..cb74da5e793a33 --- /dev/null +++ b/t/unit-tests/u-mingw.c @@ -0,0 +1,72 @@ +#include "unit-test.h" + +#if defined(GIT_WINDOWS_NATIVE) && !defined(_UCRT) +#undef strerror +int errnos_contains(int); +static int errnos [53]={ + /* errnos in err_win_to_posix */ + EACCES, EBUSY, EEXIST, ERANGE, EIO, ENODEV, ENXIO, ENOEXEC, EINVAL, ENOENT, + EPIPE, ENAMETOOLONG, ENOSYS, ENOTEMPTY, ENOSPC, EFAULT, EBADF, EPERM, EINTR, + E2BIG, ESPIPE, ENOMEM, EXDEV, EAGAIN, ENFILE, EMFILE, ECHILD, EROFS, + /* errnos only in winsock_error_to_errno */ + EWOULDBLOCK, EINPROGRESS, EALREADY, ENOTSOCK, EDESTADDRREQ, EMSGSIZE, + EPROTOTYPE, ENOPROTOOPT, EPROTONOSUPPORT, EOPNOTSUPP, EAFNOSUPPORT, + EADDRINUSE, EADDRNOTAVAIL, ENETDOWN, ENETUNREACH, ENETRESET, ECONNABORTED, + ECONNRESET, ENOBUFS, EISCONN, ENOTCONN, ETIMEDOUT, ECONNREFUSED, ELOOP, + EHOSTUNREACH + }; + +int errnos_contains(int errnum) +{ + for(int i=0;i<53;i++) + if(errnos[i]==errnum) + return 1; + return 0; +} +#endif + +void test_mingw__no_strerror_shim_on_ucrt(void) +{ +#if defined(GIT_WINDOWS_NATIVE) && defined(_UCRT) + cl_assert_(strerror != mingw_strerror, + "mingw_strerror is unnescessary when building against UCRT"); +#else + cl_skip(); +#endif +} + +void test_mingw__strerror(void) +{ +#if defined(GIT_WINDOWS_NATIVE) && !defined(_UCRT) + for(int i=0;i<53;i++) + { + char *crt; + char *mingw; + mingw = mingw_strerror(errnos[i]); + crt = strerror(errnos[i]); + cl_assert_(!strcasestr(mingw, "unknown error"), + "mingw_strerror should know all errno values we care about"); + if(!strcasestr(crt, "unknown error")) + cl_assert_equal_s(crt,mingw); + } +#else + cl_skip(); +#endif +} + +void test_mingw__errno_translation(void) +{ +#if defined(GIT_WINDOWS_NATIVE) && !defined(_UCRT) + /* GetLastError() return values are currently defined from 0 to 15841, + testing up to 20000 covers some room for future expansion */ + for (int i=0;i<20000;i++) + { + if(i!=ERROR_SUCCESS) + cl_assert_(errnos_contains(err_win_to_posix(i)), + "all err_win_to_posix return values should be tested against mingw_strerror"); + /* ideally we'd test the same for winsock_error_to_errno, but it's static */ + } +#else + cl_skip(); +#endif +} From 53aa240499283f3b9e0d24edb931c758afa6a171 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sun, 22 Dec 2024 17:43:45 +0100 Subject: [PATCH 131/248] compat/mingw: drop outdated comment MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This comment has been true for the longest time; The combination of the two preceding commits made it incorrect, so let's drop that comment. Signed-off-by: Matthias Aßhauer Signed-off-by: Johannes Schindelin --- compat/mingw.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 845f0898e2a86c..5613fd525a970a 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2336,15 +2336,6 @@ int mingw_socket(int domain, int type, int protocol) ensure_socket_initialization(); s = WSASocket(domain, type, protocol, NULL, 0, 0); if (s == INVALID_SOCKET) { - /* - * WSAGetLastError() values are regular BSD error codes - * biased by WSABASEERR. - * However, strerror() does not know about networking - * specific errors, which are values beginning at 38 or so. - * Therefore, we choose to leave the biased error code - * in errno so that _if_ someone looks up the code somewhere, - * then it is at least the number that are usually listed. - */ set_wsa_errno(); return -1; } From 36eccac1574eac587933c83b74b142e320c6d9b3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sun, 29 Dec 2024 11:48:34 +0100 Subject: [PATCH 132/248] t0301: actually test credential-cache on Windows MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 2406bf5 (Win32: detect unix socket support at runtime, 2024-04-03) introduced a runtime detection for whether the operating system supports unix sockets for Windows, but a mistake snuck into the tests. When building and testing Git without NO_UNIX_SOCKETS we currently skip t0301-credential-cache on Windows if unix sockets are supported and run the tests if they aren't. Flip that logic to actually work the way it was intended. Signed-off-by: Matthias Aßhauer Signed-off-by: Johannes Schindelin --- t/t0301-credential-cache.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/t0301-credential-cache.sh b/t/t0301-credential-cache.sh index dc30289f7539ee..586681c681c3cb 100755 --- a/t/t0301-credential-cache.sh +++ b/t/t0301-credential-cache.sh @@ -12,7 +12,7 @@ test -z "$NO_UNIX_SOCKETS" || { if test_have_prereq MINGW then service_running=$(sc query afunix | grep "4 RUNNING") - test -z "$service_running" || { + test -n "$service_running" || { skip_all='skipping credential-cache tests, unix sockets not available' test_done } From ec180e40840ba3abc1b5e8f5e9be63ff07aaf847 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matthias=20A=C3=9Fhauer?= Date: Sun, 22 Dec 2024 17:24:24 +0100 Subject: [PATCH 133/248] credential-cache: handle ECONNREFUSED gracefully MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In 245670c (credential-cache: check for windows specific errors, 2021-09-14) we concluded that on Windows we would always encounter ENETDOWN where we would expect ECONNREFUSED on POSIX systems, when connecting to unix sockets. As reported in [1], we do encounter ECONNREFUSED on Windows if the socket file doesn't exist, but the containing directory does and ENETDOWN if neither exists. We should handle this case like we do on non-windows systems. [1] https://github.com/git-for-windows/git/pull/4762#issuecomment-2545498245 This fixes https://github.com/git-for-windows/git/issues/5314 Helped-by: M Hickford Signed-off-by: Matthias Aßhauer Signed-off-by: Johannes Schindelin --- builtin/credential-cache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/builtin/credential-cache.c b/builtin/credential-cache.c index 7f733cb756e03c..3b8130d3d64f9c 100644 --- a/builtin/credential-cache.c +++ b/builtin/credential-cache.c @@ -23,7 +23,7 @@ static int connection_closed(int error) static int connection_fatally_broken(int error) { - return (error != ENOENT) && (error != ENETDOWN); + return (error != ENOENT) && (error != ENETDOWN) && (error != ECONNREFUSED); } #else From b345652a257701ac968e4ee8206be96b9f643b97 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 15 Apr 2025 19:55:12 +0200 Subject: [PATCH 134/248] max_tree_depth: lower it for clangarm64 on Windows Just as in b64d78ad02ca (max_tree_depth: lower it for MSVC to avoid stack overflows, 2023-11-01), I encountered the same problem with the clang builds on Windows/ARM64. The symptom is an exit code 127 when t6700 tries to verify that `git archive big` fails. This exit code is reserved on Unix/Linux to mean "command not found". Unfortunately in this case, it is the fall-back chosen by Cygwin's `pinfo::status_exit()` method when encountering the NSTATUS `STATUS_STACK_OVERFLOW`, see https://github.com/cygwin/cygwin/blob/cygwin-3.6.1/winsup/cygwin/pinfo.cc#L171 I verified manually that the stack overflow always happens somewhere around tree depth 1403, therefore 1280 should be a safe bound in these instances. Signed-off-by: Johannes Schindelin --- environment.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/environment.c b/environment.c index ae1427bb9e8e62..3c4af142174872 100644 --- a/environment.c +++ b/environment.c @@ -102,9 +102,21 @@ int max_allowed_tree_depth = * tree depth; This value seems to be low enough. */ 1280; +#else +#if defined(GIT_WINDOWS_NATIVE) && defined(__clang__) && defined(__aarch64__) + /* + * Similar to Visual C, it seems that on Windows/ARM64 the clang-based + * builds have a smaller stack space available. When running out of + * that stack space, a `STATUS_STACK_OVERFLOW` is produced. When the + * Git command was run from an MSYS2 Bash, this unfortunately results + * in an exit code 127. Let's prevent that by lowering the maximal + * tree depth; This value seems to be low enough. + */ + 1280; #else 2048; #endif +#endif #ifndef PROTECT_HFS_DEFAULT #define PROTECT_HFS_DEFAULT 0 From 3f15959d6f5f4b7546d27310a2ab975d1fa90cfc Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 6 Mar 2025 14:05:03 +0100 Subject: [PATCH 135/248] reftable: do make sure to use custom allocators The reftable library goes out of its way to use its own set of allocator functions that can be configured using `reftable_set_alloc()`. However, Git does not configure this. That is not typically a problem, except when Git uses a custom allocator via some definitions in `git-compat-util.h`, as is the case in Git for Windows (which switched away from the long-unmaintained nedmalloc to mimalloc). Then, it is quite possible that Git assigns a `strbuf` (allocated via the custom allocator) to, say, the `refname` field of a `reftable_log_record` in `write_transaction_table()`, and later on asks the reftable library function `reftable_log_record_release()` to release it, but that function was compiled without using `git-compat-util.h` and hence calls regular `free()` (i.e. _not_ the custom allocator's own function). This has been a problem for a long time and it was a matter of some sort of "luck" that 1) reftables are not commonly used on Windows, and 2) mimalloc can often ignore gracefully when it is asked to release memory that it has not allocated. However, a recent update to `seen` brought this problem to the forefront, letting t1460 fail in Git for Windows, with symptoms much in the same way as the problem I had to address in d02c37c3e6ba (t-reftable-basics: allow for `malloc` to be `#define`d, 2025-01-08) where exit code 127 was also produced in lieu of `STATUS_HEAP_CORRUPTION` (C0000374) because exit codes are only 7 bits wide. It was not possible to figure out what change in particular caused these new failures within a reasonable time frame, as there are too many changes in `seen` that conflict with Git for Windows' patches, I had to stop the investigation after spending four hours on it fruitlessly. To verify that this patch fixes the issue, I avoided using mimalloc and temporarily patched in a "custom allocator" that would more reliably point out problems, like this: diff --git a/refs/reftable-backend.c b/refs/reftable-backend.c index 68f38291f84c..9421d630b9f5 100644 --- a/refs/reftable-backend.c +++ b/refs/reftable-backend.c @@ -353,6 +353,69 @@ static int reftable_be_fsync(int fd) return fsync_component(FSYNC_COMPONENT_REFERENCE, fd); } +#define DEBUG_REFTABLE_ALLOC +#ifdef DEBUG_REFTABLE_ALLOC +#include "khash.h" + +static inline khint_t __ac_X31_hash_ptr(void *ptr) +{ + union { + void *ptr; + char s[sizeof(void *)]; + } u; + size_t i; + khint_t h; + + u.ptr = ptr; + h = (khint_t)*u.s; + for (i = 0; i < sizeof(void *); i++) + h = (h << 5) - h + (khint_t)u.s[i]; + return h; +} + +#define kh_ptr_hash_func(key) __ac_X31_hash_ptr(key) +#define kh_ptr_hash_equal(a, b) ((a) == (b)) + +KHASH_INIT(ptr, void *, int, 0, kh_ptr_hash_func, kh_ptr_hash_equal) + +static kh_ptr_t *my_malloced; + +static void *my_malloc(size_t sz) +{ + int dummy; + void *ptr = malloc(sz); + if (ptr) + kh_put_ptr(my_malloced, ptr, &dummy); + return ptr; +} + +static void *my_realloc(void *ptr, size_t sz) +{ + int dummy; + if (ptr) { + khiter_t pos = kh_get_ptr(my_malloced, ptr); + if (pos >= kh_end(my_malloced)) + die("Was not my_malloc()ed: %p", ptr); + kh_del_ptr(my_malloced, pos); + } + ptr = realloc(ptr, sz); + if (ptr) + kh_put_ptr(my_malloced, ptr, &dummy); + return ptr; +} + +static void my_free(void *ptr) +{ + if (ptr) { + khiter_t pos = kh_get_ptr(my_malloced, ptr); + if (pos >= kh_end(my_malloced)) + die("Was not my_malloc()ed: %p", ptr); + kh_del_ptr(my_malloced, pos); + } + free(ptr); +} +#endif + static struct ref_store *reftable_be_init(struct repository *repo, const char *gitdir, unsigned int store_flags) @@ -362,6 +425,11 @@ static struct ref_store *reftable_be_init(struct repository *repo, int is_worktree; mode_t mask; +#ifdef DEBUG_REFTABLE_ALLOC + my_malloced = kh_init_ptr(); + reftable_set_alloc(my_malloc, my_realloc, my_free); +#endif + mask = umask(0); umask(mask); I briefly considered contributing this "custom allocator" patch, too, but it is unwieldy (for example, it would not work at all when compiling with mimalloc support) and it would only waste space (or even time, if a compile flag was introduced and exercised as part of the CI builds). Given that it is highly unlikely that Git will lose the new `reftable_set_alloc()` call by mistake, I rejected that idea as simply too wasteful. Signed-off-by: Johannes Schindelin --- refs/reftable-backend.c | 1 + 1 file changed, 1 insertion(+) diff --git a/refs/reftable-backend.c b/refs/reftable-backend.c index c0440b4bd0e8e2..ac02b64c48c190 100644 --- a/refs/reftable-backend.c +++ b/refs/reftable-backend.c @@ -365,6 +365,7 @@ static struct ref_store *reftable_be_init(struct repository *repo, mask = umask(0); umask(mask); + reftable_set_alloc(malloc, realloc, free); base_ref_store_init(&refs->base, repo, gitdir, &refs_be_reftable); strmap_init(&refs->worktree_backends); refs->store_flags = store_flags; From 78a62deb8b297ba3cd323eb27a8a782546978d99 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Tue, 3 Jun 2025 12:45:39 +0200 Subject: [PATCH 136/248] check-whitespace: avoid alerts about upstream commits Every once in a while, whitespace errors are introduced in Git for Windows' rebases to newer Git versions, simply by virtue of integrating upstream commits that do not follow upstream Git's own whitespace rule. In Git v2.50.0-rc0, for example, 03f2915541a4 (xdiff: disable cleanup_records heuristic with --minimal, 2025-04-29) introduced a trailing space. Arguably, non-actionable alerts are worse than no alerts at all, so let's suppress those alerts that we cannot do anything about, anyway. Signed-off-by: Johannes Schindelin --- ci/check-whitespace.sh | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/ci/check-whitespace.sh b/ci/check-whitespace.sh index c40804394cb079..e590ac0dfd765e 100755 --- a/ci/check-whitespace.sh +++ b/ci/check-whitespace.sh @@ -19,6 +19,7 @@ problems=() commit= commitText= commitTextmd= +committerEmail= goodParent= if ! git rev-parse --quiet --verify "${baseCommit}" @@ -27,7 +28,7 @@ then exit 1 fi -while read dash sha etc +while read dash email sha etc do case "${dash}" in "---") # Line contains commit information. @@ -40,10 +41,14 @@ do commit="${sha}" commitText="${sha} ${etc}" commitTextmd="[${sha}](${url}/commit/${sha}) ${etc}" + committerEmail="${email}" ;; "") ;; *) # Line contains whitespace error information for current commit. + # Quod licet Iovi non licet bovi + test gitster@pobox.com != "$committerEmail" || break + if test -n "${goodParent}" then problems+=("1) --- ${commitTextmd}") @@ -64,7 +69,7 @@ do echo "${dash} ${sha} ${etc}" ;; esac -done <<< "$(git log --check --pretty=format:"---% h% s" "${baseCommit}"..)" +done <<< "$(git log --check --pretty=format:"---% ce% h% s" "${baseCommit}"..)" if test ${#problems[*]} -gt 0 then From b1053840441ed3226a6dc8a9ee3383f73d750104 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Thu, 5 Jun 2025 08:20:30 +0200 Subject: [PATCH 137/248] mingw: avoid the comma operator The pattern `return errno = ..., -1;` is observed several times in `compat/mingw.c`. It has served us well over the years, but now clang starts complaining: compat/mingw.c:723:24: error: possible misuse of comma operator here [-Werror,-Wcomma] 723 | return errno = ENOSYS, -1; | ^ See for example this failing workflow run: https://github.com/git-for-windows/git-sdk-arm64/actions/runs/15457893907/job/43513458823#step:8:201 Let's appease clang (and also reduce the use of the no longer common comma operator). Signed-off-by: Johannes Schindelin --- compat/mingw.c | 48 ++++++++++++++++++++++++++++-------------------- 1 file changed, 28 insertions(+), 20 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8538e3d1729d25..9b894d9639b642 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -491,8 +491,10 @@ static int mingw_open_append(wchar_t const *wfilename, int oflags, ...) DWORD create = (oflags & O_CREAT) ? OPEN_ALWAYS : OPEN_EXISTING; /* only these flags are supported */ - if ((oflags & ~O_CREAT) != (O_WRONLY | O_APPEND)) - return errno = ENOSYS, -1; + if ((oflags & ~O_CREAT) != (O_WRONLY | O_APPEND)) { + errno = ENOSYS; + return -1; + } /* * FILE_SHARE_WRITE is required to permit child processes @@ -2450,12 +2452,14 @@ static int start_timer_thread(void) timer_event = CreateEvent(NULL, FALSE, FALSE, NULL); if (timer_event) { timer_thread = (HANDLE) _beginthreadex(NULL, 0, ticktack, NULL, 0, NULL); - if (!timer_thread ) - return errno = ENOMEM, - error("cannot start timer thread"); - } else - return errno = ENOMEM, - error("cannot allocate resources for timer"); + if (!timer_thread ) { + errno = ENOMEM; + return error("cannot start timer thread"); + } + } else { + errno = ENOMEM; + return error("cannot allocate resources for timer"); + } return 0; } @@ -2488,13 +2492,15 @@ int setitimer(int type UNUSED, struct itimerval *in, struct itimerval *out) static const struct timeval zero; static int atexit_done; - if (out) - return errno = EINVAL, - error("setitimer param 3 != NULL not implemented"); + if (out) { + errno = EINVAL; + return error("setitimer param 3 != NULL not implemented"); + } if (!is_timeval_eq(&in->it_interval, &zero) && - !is_timeval_eq(&in->it_interval, &in->it_value)) - return errno = EINVAL, - error("setitimer: it_interval must be zero or eq it_value"); + !is_timeval_eq(&in->it_interval, &in->it_value)) { + errno = EINVAL; + return error("setitimer: it_interval must be zero or eq it_value"); + } if (timer_thread) stop_timer_thread(); @@ -2516,12 +2522,14 @@ int sigaction(int sig, struct sigaction *in, struct sigaction *out) { if (sig == SIGCHLD) return -1; - else if (sig != SIGALRM) - return errno = EINVAL, - error("sigaction only implemented for SIGALRM"); - if (out) - return errno = EINVAL, - error("sigaction: param 3 != NULL not implemented"); + else if (sig != SIGALRM) { + errno = EINVAL; + return error("sigaction only implemented for SIGALRM"); + } + if (out) { + errno = EINVAL; + return error("sigaction: param 3 != NULL not implemented"); + } timer_fn = in->sa_handler; return 0; From d905624f2c8fe4f321cf5de8ebe5ff9ea4a476d1 Mon Sep 17 00:00:00 2001 From: David Macek Date: Fri, 3 Oct 2025 10:06:40 +0200 Subject: [PATCH 138/248] wincred: Avoid memory corruption `wcsncpy_s()` wants to write the terminating null character so we need to allocate one more space for it in the target memory block. This should fix crashes when trying to read passwords. When this happened, the password/token wouldn't print out and Git would therefore ask for a new password every time. Signed-off-by: David Macek --- contrib/credential/wincred/git-credential-wincred.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/credential/wincred/git-credential-wincred.c b/contrib/credential/wincred/git-credential-wincred.c index 5683846b4b4d1f..73c2b9b72ab53e 100644 --- a/contrib/credential/wincred/git-credential-wincred.c +++ b/contrib/credential/wincred/git-credential-wincred.c @@ -165,7 +165,7 @@ static void get_credential(void) write_item("username", creds[i]->UserName, creds[i]->UserName ? wcslen(creds[i]->UserName) : 0); if (creds[i]->CredentialBlobSize > 0) { - secret = xmalloc(creds[i]->CredentialBlobSize); + secret = xmalloc(creds[i]->CredentialBlobSize + sizeof(WCHAR)); wcsncpy_s(secret, creds[i]->CredentialBlobSize, (LPCWSTR)creds[i]->CredentialBlob, creds[i]->CredentialBlobSize / sizeof(WCHAR)); line = wcstok_s(secret, L"\r\n", &remaining_lines); write_item("password", line, line ? wcslen(line) : 0); From 5760c6e8398a8446492a92cfc73f2d836452db92 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 20 Sep 2017 21:52:28 +0200 Subject: [PATCH 139/248] git-gui--askyesno: fix funny text wrapping The text wrapping seems to be aligned to the right side of the Yes button, leaving an awful lot of empty space. Let's try to counter this by using pixel units. Signed-off-by: Johannes Schindelin --- git-gui/git-gui--askyesno | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/git-gui/git-gui--askyesno b/git-gui/git-gui--askyesno index 2a6e6fd11122f5..cf9c990d0919b3 100755 --- a/git-gui/git-gui--askyesno +++ b/git-gui/git-gui--askyesno @@ -20,8 +20,8 @@ if {$argc < 1} { } ${NS}::frame .t -${NS}::label .t.m -text $prompt -justify center -width 40 -.t.m configure -wraplength 400 +${NS}::label .t.m -text $prompt -justify center -width 400px +.t.m configure -wraplength 400px pack .t.m -side top -fill x -padx 20 -pady 20 -expand 1 pack .t -side top -fill x -ipadx 20 -ipady 20 -expand 1 From 1ac75e7f243e33faff0cd2453e1454537757ca97 Mon Sep 17 00:00:00 2001 From: Heiko Voigt Date: Sun, 21 Feb 2010 21:05:04 +0100 Subject: [PATCH 140/248] git-gui: provide question helper for retry fallback on Windows Make use of the new environment variable GIT_ASK_YESNO to support the recently implemented fallback in case unlink, rename or rmdir fail for files in use on Windows. The added dialog will present a yes/no question to the the user which will currently be used by the windows compat layer to let the user retry a failed file operation. Signed-off-by: Heiko Voigt --- git-gui/Makefile | 2 ++ git-gui/git-gui--askyesno | 51 +++++++++++++++++++++++++++++++++++++++ git-gui/git-gui.sh | 3 +++ 3 files changed, 56 insertions(+) create mode 100755 git-gui/git-gui--askyesno diff --git a/git-gui/Makefile b/git-gui/Makefile index 27bbe051de87b2..69b0b844352bc1 100644 --- a/git-gui/Makefile +++ b/git-gui/Makefile @@ -186,6 +186,7 @@ install: all $(QUIET)$(INSTALL_D0)'$(DESTDIR_SQ)$(gitexecdir_SQ)' $(INSTALL_D1) $(QUIET)$(INSTALL_X0)git-gui $(INSTALL_X1) '$(DESTDIR_SQ)$(gitexecdir_SQ)' $(QUIET)$(INSTALL_X0)git-gui--askpass $(INSTALL_X1) '$(DESTDIR_SQ)$(gitexecdir_SQ)' + $(QUIET)$(INSTALL_X0)git-gui--askyesno $(INSTALL_X1) '$(DESTDIR_SQ)$(gitexecdir_SQ)' $(QUIET)$(foreach p,$(GITGUI_BUILT_INS), $(INSTALL_L0)'$(DESTDIR_SQ)$(gitexecdir_SQ)/$p' $(INSTALL_L1)'$(DESTDIR_SQ)$(gitexecdir_SQ)/git-gui' $(INSTALL_L2)'$(DESTDIR_SQ)$(gitexecdir_SQ)/$p' $(INSTALL_L3) &&) true ifdef GITGUI_WINDOWS_WRAPPER $(QUIET)$(INSTALL_R0)git-gui.tcl $(INSTALL_R1) '$(DESTDIR_SQ)$(gitexecdir_SQ)' @@ -200,6 +201,7 @@ uninstall: $(QUIET)$(CLEAN_DST) '$(DESTDIR_SQ)$(gitexecdir_SQ)' $(QUIET)$(REMOVE_F0)'$(DESTDIR_SQ)$(gitexecdir_SQ)'/git-gui $(REMOVE_F1) $(QUIET)$(REMOVE_F0)'$(DESTDIR_SQ)$(gitexecdir_SQ)'/git-gui--askpass $(REMOVE_F1) + $(QUIET)$(REMOVE_F0)'$(DESTDIR_SQ)$(gitexecdir_SQ)'/git-gui--askyesno $(REMOVE_F1) $(QUIET)$(foreach p,$(GITGUI_BUILT_INS), $(REMOVE_F0)'$(DESTDIR_SQ)$(gitexecdir_SQ)'/$p $(REMOVE_F1) &&) true ifdef GITGUI_WINDOWS_WRAPPER $(QUIET)$(REMOVE_F0)'$(DESTDIR_SQ)$(gitexecdir_SQ)'/git-gui.tcl $(REMOVE_F1) diff --git a/git-gui/git-gui--askyesno b/git-gui/git-gui--askyesno new file mode 100755 index 00000000000000..2a6e6fd11122f5 --- /dev/null +++ b/git-gui/git-gui--askyesno @@ -0,0 +1,51 @@ +#!/bin/sh +# Tcl ignores the next line -*- tcl -*- \ +exec wish "$0" -- "$@" + +# This is an implementation of a simple yes no dialog +# which is injected into the git commandline by git gui +# in case a yesno question needs to be answered. + +set NS {} +set use_ttk [package vsatisfies [package provide Tk] 8.5] +if {$use_ttk} { + set NS ttk +} + +if {$argc < 1} { + puts stderr "Usage: $argv0 " + exit 1 +} else { + set prompt [join $argv " "] +} + +${NS}::frame .t +${NS}::label .t.m -text $prompt -justify center -width 40 +.t.m configure -wraplength 400 +pack .t.m -side top -fill x -padx 20 -pady 20 -expand 1 +pack .t -side top -fill x -ipadx 20 -ipady 20 -expand 1 + +${NS}::frame .b +${NS}::frame .b.left -width 200 +${NS}::button .b.yes -text Yes -command yes +${NS}::button .b.no -text No -command no + + +pack .b.left -side left -expand 1 -fill x +pack .b.yes -side left -expand 1 +pack .b.no -side right -expand 1 -ipadx 5 +pack .b -side bottom -fill x -ipadx 20 -ipady 15 + +bind . {exit 0} +bind . {exit 1} + +proc no {} { + exit 1 +} + +proc yes {} { + exit 0 +} + +wm title . "Question?" +tk::PlaceWindow . diff --git a/git-gui/git-gui.sh b/git-gui/git-gui.sh index 211960c14c8262..7358bc2c894eb9 100755 --- a/git-gui/git-gui.sh +++ b/git-gui/git-gui.sh @@ -1130,6 +1130,9 @@ set argv0dir [file dirname [file normalize $::argv0]] if {![info exists env(SSH_ASKPASS)]} { set env(SSH_ASKPASS) [file join $argv0dir git-gui--askpass] } +if {![info exists env(GIT_ASK_YESNO)]} { + set env(GIT_ASK_YESNO) [file join $argv0dir git-gui--askyesno] +} unset argv0dir ###################################################################### From aee4f8fa0f115574a167f5168d889c24dbd48eba Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Wed, 20 Sep 2017 21:53:45 +0200 Subject: [PATCH 141/248] git-gui--askyesno: allow overriding the window title "Question?" is maybe not the most informative thing to ask. In the absence of better information, it is the best we can do, of course. However, Git for Windows' auto updater just learned the trick to use git-gui--askyesno to ask the user whether to update now or not. And in this scripted scenario, we can easily pass a command-line option to change the window title. So let's support that with the new `--title ` option. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- git-gui/git-gui--askyesno | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/git-gui/git-gui--askyesno b/git-gui/git-gui--askyesno index cf9c990d0919b3..45b0260eff8145 100755 --- a/git-gui/git-gui--askyesno +++ b/git-gui/git-gui--askyesno @@ -12,10 +12,15 @@ if {$use_ttk} { set NS ttk } +set title "Question?" if {$argc < 1} { puts stderr "Usage: $argv0 <question>" exit 1 } else { + if {$argc > 2 && [lindex $argv 0] == "--title"} { + set title [lindex $argv 1] + set argv [lreplace $argv 0 1] + } set prompt [join $argv " "] } @@ -47,5 +52,5 @@ proc yes {} { exit 0 } -wm title . "Question?" +wm title . $title tk::PlaceWindow . From f04fe9b80f815ba2a27ac2f8c4de3a09d51c30eb Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 23 Jul 2010 18:06:05 +0200 Subject: [PATCH 142/248] git gui: set GIT_ASKPASS=git-gui--askpass if not set yet Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- git-gui/git-gui.sh | 3 +++ 1 file changed, 3 insertions(+) diff --git a/git-gui/git-gui.sh b/git-gui/git-gui.sh index 7358bc2c894eb9..fa6995530bd287 100755 --- a/git-gui/git-gui.sh +++ b/git-gui/git-gui.sh @@ -1130,6 +1130,9 @@ set argv0dir [file dirname [file normalize $::argv0]] if {![info exists env(SSH_ASKPASS)]} { set env(SSH_ASKPASS) [file join $argv0dir git-gui--askpass] } +if {![info exists env(GIT_ASKPASS)]} { + set env(GIT_ASKPASS) [file join $argv0dir git-gui--askpass] +} if {![info exists env(GIT_ASK_YESNO)]} { set env(GIT_ASK_YESNO) [file join $argv0dir git-gui--askyesno] } From afe8b8ab888a00ecdc30c9c9c97dabcca4a97395 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 20 Sep 2017 21:55:45 +0200 Subject: [PATCH 143/248] git-gui--askyesno (mingw): use Git for Windows' icon, if available For additional GUI goodness. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- git-gui/git-gui--askyesno | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/git-gui/git-gui--askyesno b/git-gui/git-gui--askyesno index 45b0260eff8145..c0c82e7cbd01d6 100755 --- a/git-gui/git-gui--askyesno +++ b/git-gui/git-gui--askyesno @@ -52,5 +52,17 @@ proc yes {} { exit 0 } +if {$::tcl_platform(platform) eq {windows}} { + set icopath [file dirname [file normalize $argv0]] + if {[file tail $icopath] eq {git-core}} { + set icopath [file dirname $icopath] + } + set icopath [file dirname $icopath] + set icopath [file join $icopath share git git-for-windows.ico] + if {[file exists $icopath]} { + wm iconbitmap . -default $icopath + } +} + wm title . $title tk::PlaceWindow . From afddb28232b299951cc6b277079b1c11b2afe5e0 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 6 Jul 2013 02:09:35 +0200 Subject: [PATCH 144/248] Win32: make FILETIME conversion functions public We will use them in the upcoming "FSCache" patches (to accelerate sequential lstat() calls). Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw-posix.h | 18 ++++++++++++++++++ compat/mingw.c | 18 ------------------ 2 files changed, 18 insertions(+), 18 deletions(-) diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index a0c7986fd267b0..4ba76a6603712c 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -343,6 +343,17 @@ static inline int getrlimit(int resource, struct rlimit *rlp) return 0; } +/* + * The unit of FILETIME is 100-nanoseconds since January 1, 1601, UTC. + * Returns the 100-nanoseconds ("hekto nanoseconds") since the epoch. + */ +static inline long long filetime_to_hnsec(const FILETIME *ft) +{ + long long winTime = ((long long)ft->dwHighDateTime << 32) + ft->dwLowDateTime; + /* Windows to Unix Epoch conversion */ + return winTime - 116444736000000000LL; +} + /* * Use mingw specific stat()/lstat()/fstat() implementations on Windows, * including our own struct stat with 64 bit st_size and nanosecond-precision @@ -359,6 +370,13 @@ struct timespec { #endif #endif +static inline void filetime_to_timespec(const FILETIME *ft, struct timespec *ts) +{ + long long hnsec = filetime_to_hnsec(ft); + ts->tv_sec = (time_t)(hnsec / 10000000); + ts->tv_nsec = (hnsec % 10000000) * 100; +} + struct mingw_stat { _dev_t st_dev; _ino_t st_ino; diff --git a/compat/mingw.c b/compat/mingw.c index 30b9fc9b6cddeb..231d9c17decc21 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -883,24 +883,6 @@ int mingw_chmod(const char *filename, int mode) return _wchmod(wfilename, mode); } -/* - * The unit of FILETIME is 100-nanoseconds since January 1, 1601, UTC. - * Returns the 100-nanoseconds ("hekto nanoseconds") since the epoch. - */ -static inline long long filetime_to_hnsec(const FILETIME *ft) -{ - long long winTime = ((long long)ft->dwHighDateTime << 32) + ft->dwLowDateTime; - /* Windows to Unix Epoch conversion */ - return winTime - 116444736000000000LL; -} - -static inline void filetime_to_timespec(const FILETIME *ft, struct timespec *ts) -{ - long long hnsec = filetime_to_hnsec(ft); - ts->tv_sec = (time_t)(hnsec / 10000000); - ts->tv_nsec = (hnsec % 10000000) * 100; -} - /** * Verifies that safe_create_leading_directories() would succeed. */ From 542d6083b9c905d4c04adae01187620e15f01a27 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 8 Sep 2013 14:17:31 +0200 Subject: [PATCH 145/248] Win32: dirent.c: Move opendir down Move opendir down in preparation for the next patch. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/win32/dirent.c | 68 +++++++++++++++++++++---------------------- 1 file changed, 34 insertions(+), 34 deletions(-) diff --git a/compat/win32/dirent.c b/compat/win32/dirent.c index 52420ec7d4dad7..2603a0fa39f45a 100644 --- a/compat/win32/dirent.c +++ b/compat/win32/dirent.c @@ -18,40 +18,6 @@ static inline void finddata2dirent(struct dirent *ent, WIN32_FIND_DATAW *fdata) ent->d_type = DT_REG; } -DIR *opendir(const char *name) -{ - wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ - WIN32_FIND_DATAW fdata; - HANDLE h; - int len; - DIR *dir; - - /* convert name to UTF-16 and check length < MAX_PATH */ - if ((len = xutftowcs_path(pattern, name)) < 0) - return NULL; - - /* append optional '/' and wildcard '*' */ - if (len && !is_dir_sep(pattern[len - 1])) - pattern[len++] = '/'; - pattern[len++] = '*'; - pattern[len] = 0; - - /* open find handle */ - h = FindFirstFileW(pattern, &fdata); - if (h == INVALID_HANDLE_VALUE) { - DWORD err = GetLastError(); - errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err); - return NULL; - } - - /* initialize DIR structure and copy first dir entry */ - dir = xmalloc(sizeof(DIR)); - dir->dd_handle = h; - dir->dd_stat = 0; - finddata2dirent(&dir->dd_dir, &fdata); - return dir; -} - struct dirent *readdir(DIR *dir) { if (!dir) { @@ -90,3 +56,37 @@ int closedir(DIR *dir) free(dir); return 0; } + +DIR *opendir(const char *name) +{ + wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ + WIN32_FIND_DATAW fdata; + HANDLE h; + int len; + DIR *dir; + + /* convert name to UTF-16 and check length < MAX_PATH */ + if ((len = xutftowcs_path(pattern, name)) < 0) + return NULL; + + /* append optional '/' and wildcard '*' */ + if (len && !is_dir_sep(pattern[len - 1])) + pattern[len++] = '/'; + pattern[len++] = '*'; + pattern[len] = 0; + + /* open find handle */ + h = FindFirstFileW(pattern, &fdata); + if (h == INVALID_HANDLE_VALUE) { + DWORD err = GetLastError(); + errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err); + return NULL; + } + + /* initialize DIR structure and copy first dir entry */ + dir = xmalloc(sizeof(DIR)); + dir->dd_handle = h; + dir->dd_stat = 0; + finddata2dirent(&dir->dd_dir, &fdata); + return dir; +} From e019d5afaded8bc9bd248db595678b0ed7781dcf Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 8 Sep 2013 14:18:40 +0200 Subject: [PATCH 146/248] mingw: make the dirent implementation pluggable Emulating the POSIX `dirent` API on Windows via `FindFirstFile()`/`FindNextFile()` is pretty staightforward, however, most of the information provided in the `WIN32_FIND_DATA` structure is thrown away in the process. A more sophisticated implementation may cache this data, e.g. for later reuse in calls to `lstat()`. Make the `dirent` implementation pluggable so that it can be switched at runtime, e.g. based on a config option. Define a base DIR structure with pointers to `readdir()`/`closedir()` that match the `opendir()` implementation (similar to vtable pointers in Object-Oriented Programming). Define `readdir()`/`closedir()` so that they call the function pointers in the `DIR` structure. This allows to choose the `opendir()` implementation on a call-by-call basis. Make the fixed-size `dirent.d_name` buffer a flex array, as `d_name` may be implementation specific (e.g. a caching implementation may allocate a `struct dirent` with _just_ the size needed to hold the `d_name` in question). Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/dirent.c | 30 +++++++++++++++++++----------- compat/win32/dirent.h | 28 +++++++++++++++++++++------- 2 files changed, 40 insertions(+), 18 deletions(-) diff --git a/compat/win32/dirent.c b/compat/win32/dirent.c index 2603a0fa39f45a..139d2ba3c4da34 100644 --- a/compat/win32/dirent.c +++ b/compat/win32/dirent.c @@ -1,15 +1,21 @@ #include "../../git-compat-util.h" -struct DIR { - struct dirent dd_dir; /* includes d_type */ +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wpedantic" +typedef struct dirent_DIR { + struct DIR base_dir; /* extend base struct DIR */ HANDLE dd_handle; /* FindFirstFile handle */ int dd_stat; /* 0-based index */ -}; + struct dirent dd_dir; /* includes d_type */ +} dirent_DIR; +#pragma GCC diagnostic pop + +DIR *(*opendir)(const char *dirname) = dirent_opendir; static inline void finddata2dirent(struct dirent *ent, WIN32_FIND_DATAW *fdata) { - /* convert UTF-16 name to UTF-8 */ - xwcstoutf(ent->d_name, fdata->cFileName, sizeof(ent->d_name)); + /* convert UTF-16 name to UTF-8 (d_name points to dirent_DIR.dd_name) */ + xwcstoutf(ent->d_name, fdata->cFileName, MAX_PATH * 3); /* Set file type, based on WIN32_FIND_DATA */ if (fdata->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) @@ -18,7 +24,7 @@ static inline void finddata2dirent(struct dirent *ent, WIN32_FIND_DATAW *fdata) ent->d_type = DT_REG; } -struct dirent *readdir(DIR *dir) +static struct dirent *dirent_readdir(dirent_DIR *dir) { if (!dir) { errno = EBADF; /* No set_errno for mingw */ @@ -45,7 +51,7 @@ struct dirent *readdir(DIR *dir) return &dir->dd_dir; } -int closedir(DIR *dir) +static int dirent_closedir(dirent_DIR *dir) { if (!dir) { errno = EBADF; @@ -57,13 +63,13 @@ int closedir(DIR *dir) return 0; } -DIR *opendir(const char *name) +DIR *dirent_opendir(const char *name) { wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ WIN32_FIND_DATAW fdata; HANDLE h; int len; - DIR *dir; + dirent_DIR *dir; /* convert name to UTF-16 and check length < MAX_PATH */ if ((len = xutftowcs_path(pattern, name)) < 0) @@ -84,9 +90,11 @@ DIR *opendir(const char *name) } /* initialize DIR structure and copy first dir entry */ - dir = xmalloc(sizeof(DIR)); + dir = xmalloc(sizeof(dirent_DIR) + MAX_PATH); + dir->base_dir.preaddir = (struct dirent *(*)(DIR *dir)) dirent_readdir; + dir->base_dir.pclosedir = (int (*)(DIR *dir)) dirent_closedir; dir->dd_handle = h; dir->dd_stat = 0; finddata2dirent(&dir->dd_dir, &fdata); - return dir; + return (DIR*) dir; } diff --git a/compat/win32/dirent.h b/compat/win32/dirent.h index 058207e4bfed62..a58a8075fd70e3 100644 --- a/compat/win32/dirent.h +++ b/compat/win32/dirent.h @@ -1,20 +1,34 @@ #ifndef DIRENT_H #define DIRENT_H -typedef struct DIR DIR; - #define DT_UNKNOWN 0 #define DT_DIR 1 #define DT_REG 2 #define DT_LNK 3 struct dirent { - unsigned char d_type; /* file type to prevent lstat after readdir */ - char d_name[MAX_PATH * 3]; /* file name (* 3 for UTF-8 conversion) */ + unsigned char d_type; /* file type to prevent lstat after readdir */ + char d_name[/* FLEX_ARRAY */]; /* file name */ }; -DIR *opendir(const char *dirname); -struct dirent *readdir(DIR *dir); -int closedir(DIR *dir); +/* + * Base DIR structure, contains pointers to readdir/closedir implementations so + * that opendir may choose a concrete implementation on a call-by-call basis. + */ +typedef struct DIR { + struct dirent *(*preaddir)(struct DIR *dir); + int (*pclosedir)(struct DIR *dir); +} DIR; + +/* default dirent implementation */ +extern DIR *dirent_opendir(const char *dirname); + +#define opendir git_opendir + +/* current dirent implementation */ +extern DIR *(*opendir)(const char *dirname); + +#define readdir(dir) (dir->preaddir(dir)) +#define closedir(dir) (dir->pclosedir(dir)) #endif /* DIRENT_H */ From 75a4afcc2da9ccdc4cf4ef4cedcdebaebac6f758 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 8 Sep 2013 14:21:30 +0200 Subject: [PATCH 147/248] Win32: make the lstat implementation pluggable Emulating the POSIX lstat API on Windows via GetFileAttributes[Ex] is quite slow. Windows operating system APIs seem to be much better at scanning the status of entire directories than checking single files. A caching implementation may improve performance by bulk-reading entire directories or reusing data obtained via opendir / readdir. Make the lstat implementation pluggable so that it can be switched at runtime, e.g. based on a config option. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw-posix.h | 2 +- compat/mingw.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index 4ba76a6603712c..626e33280a1664 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -409,7 +409,7 @@ int mingw_fstat(int fd, struct stat *buf); #ifdef lstat #undef lstat #endif -#define lstat mingw_lstat +extern int (*lstat)(const char *file_name, struct stat *buf); int mingw_utime(const char *file_name, const struct utimbuf *times); diff --git a/compat/mingw.c b/compat/mingw.c index 231d9c17decc21..fa3ef48a43d825 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1022,6 +1022,8 @@ static int do_stat_internal(int follow, const char *file_name, struct stat *buf) return do_lstat(follow, alt_name, buf); } +int (*lstat)(const char *file_name, struct stat *buf) = mingw_lstat; + static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) { BY_HANDLE_FILE_INFORMATION fdata; From 02a1afee7025eeaf251491bd1ca62e15bfbb2b64 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 8 Sep 2013 14:23:27 +0200 Subject: [PATCH 148/248] mingw: add infrastructure for read-only file system level caches Add a macro to mark code sections that only read from the file system, along with a config option and documentation. This facilitates implementation of relatively simple file system level caches without the need to synchronize with the file system. Enable read-only sections for 'git status' and preload_index. Signed-off-by: Karsten Blees <blees@dcon.de> --- Documentation/config/core.adoc | 6 ++++++ builtin/commit.c | 1 + compat/mingw.c | 6 ++++++ compat/mingw.h | 2 ++ git-compat-util.h | 15 +++++++++++++++ preload-index.c | 3 +++ 6 files changed, 33 insertions(+) diff --git a/Documentation/config/core.adoc b/Documentation/config/core.adoc index 8866ed27714b4a..32a5872888c5eb 100644 --- a/Documentation/config/core.adoc +++ b/Documentation/config/core.adoc @@ -690,6 +690,12 @@ relatively high IO latencies. When enabled, Git will do the index comparison to the filesystem data in parallel, allowing overlapping IO's. Defaults to true. +core.fscache:: + Enable additional caching of file system data for some operations. ++ +Git for Windows uses this to bulk-read and cache lstat data of entire +directories (instead of doing lstat file by file). + core.unsetenvvars:: Windows-only: comma-separated list of environment variables' names that need to be unset before spawning any other process. diff --git a/builtin/commit.c b/builtin/commit.c index b5b96088133b12..39c595276fdc1a 100644 --- a/builtin/commit.c +++ b/builtin/commit.c @@ -1618,6 +1618,7 @@ struct repository *repo UNUSED) PATHSPEC_PREFER_FULL, prefix, argv); + enable_fscache(1); if (status_format != STATUS_FORMAT_PORCELAIN && status_format != STATUS_FORMAT_PORCELAIN_V2) progress_flag = REFRESH_PROGRESS; diff --git a/compat/mingw.c b/compat/mingw.c index fa3ef48a43d825..436be66f6c5fc0 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -248,6 +248,7 @@ enum hide_dotfiles_type { static enum hide_dotfiles_type hide_dotfiles = HIDE_DOTFILES_DOTGITONLY; static char *unset_environment_variables; +int core_fscache; int mingw_core_config(const char *var, const char *value, const struct config_context *ctx UNUSED, @@ -261,6 +262,11 @@ int mingw_core_config(const char *var, const char *value, return 0; } + if (!strcmp(var, "core.fscache")) { + core_fscache = git_config_bool(var, value); + return 0; + } + if (!strcmp(var, "core.unsetenvvars")) { if (!value) return config_error_nonbool(var); diff --git a/compat/mingw.h b/compat/mingw.h index 6ea53ee0d29e17..65df57d2a786e4 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -1,5 +1,7 @@ #include "mingw-posix.h" +extern int core_fscache; + struct config_context; int mingw_core_config(const char *var, const char *value, const struct config_context *ctx, void *cb); diff --git a/git-compat-util.h b/git-compat-util.h index 98d46bc3b7a767..77547563b7b082 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1048,6 +1048,21 @@ static inline int is_missing_file_error(int errno_) return (errno_ == ENOENT || errno_ == ENOTDIR); } +/* + * Enable/disable a read-only cache for file system data on platforms that + * support it. + * + * Implementing a live-cache is complicated and requires special platform + * support (inotify, ReadDirectoryChangesW...). enable_fscache shall be used + * to mark sections of git code that extensively read from the file system + * without modifying anything. Implementations can use this to cache e.g. stat + * data or even file content without the need to synchronize with the file + * system. + */ +#ifndef enable_fscache +#define enable_fscache(x) /* noop */ +#endif + int cmd_main(int, const char **); /* diff --git a/preload-index.c b/preload-index.c index b222821b448526..61e8f3a1f6ec84 100644 --- a/preload-index.c +++ b/preload-index.c @@ -141,6 +141,7 @@ void preload_index(struct index_state *index, pthread_mutex_init(&pd.mutex, NULL); } + enable_fscache(1); for (i = 0; i < threads; i++) { struct thread_data *p = data+i; int err; @@ -176,6 +177,8 @@ void preload_index(struct index_state *index, trace2_data_intmax("index", NULL, "preload/sum_lstat", t2_sum_lstat); trace2_region_leave("index", "preload", NULL); + + enable_fscache(0); } int repo_read_index_preload(struct repository *repo, From e5267cba41b9220e57c3f4671c4cd17e3dfd2651 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 1 Oct 2013 12:51:54 +0200 Subject: [PATCH 149/248] mingw: add a cache below mingw's lstat and dirent implementations Checking the work tree status is quite slow on Windows, due to slow `lstat()` emulation (git calls `lstat()` once for each file in the index). Windows operating system APIs seem to be much better at scanning the status of entire directories than checking single files. Add an `lstat()` implementation that uses a cache for lstat data. Cache misses read the entire parent directory and add it to the cache. Subsequent `lstat()` calls for the same directory are served directly from the cache. Also implement `opendir()`/`readdir()`/`closedir()` so that they create and use directory listings in the cache. The cache doesn't track file system changes and doesn't plug into any modifying file APIs, so it has to be explicitly enabled for git functions that don't modify the working copy. Note: in an earlier version of this patch, the cache was always active and tracked file system changes via ReadDirectoryChangesW. However, this was much more complex and had negative impact on the performance of modifying git commands such as 'git checkout'. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 473 ++++++++++++++++++++++++++++ compat/win32/fscache.h | 10 + config.mak.uname | 4 +- contrib/buildsystems/CMakeLists.txt | 3 +- git-compat-util.h | 2 + meson.build | 1 + 6 files changed, 490 insertions(+), 3 deletions(-) create mode 100644 compat/win32/fscache.c create mode 100644 compat/win32/fscache.h diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c new file mode 100644 index 00000000000000..dc765ddd57b5bc --- /dev/null +++ b/compat/win32/fscache.c @@ -0,0 +1,473 @@ +#include "../../git-compat-util.h" +#include "../../hashmap.h" +#include "../win32.h" +#include "fscache.h" +#include "../../dir.h" +#include "../../abspath.h" + +static int initialized; +static volatile long enabled; +static struct hashmap map; +static CRITICAL_SECTION mutex; + +/* + * An entry in the file system cache. Used for both entire directory listings + * and file entries. + */ +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wpedantic" +struct fsentry { + struct hashmap_entry ent; + mode_t st_mode; + /* Pointer to the directory listing, or NULL for the listing itself. */ + struct fsentry *list; + /* Pointer to the next file entry of the list. */ + struct fsentry *next; + + union { + /* Reference count of the directory listing. */ + volatile long refcnt; + struct { + /* More stat members (only used for file entries). */ + off64_t st_size; + struct timespec st_atim; + struct timespec st_mtim; + struct timespec st_ctim; + } s; + } u; + + /* Length of name. */ + unsigned short len; + /* + * Name of the entry. For directory listings: relative path of the + * directory, without trailing '/' (empty for cwd()). For file entries: + * name of the file. Typically points to the end of the structure if + * the fsentry is allocated on the heap (see fsentry_alloc), or to a + * local variable if on the stack (see fsentry_init). + */ + struct dirent dirent; +}; +#pragma GCC diagnostic pop + +#pragma GCC diagnostic push +#ifdef __clang__ +#pragma GCC diagnostic ignored "-Wflexible-array-extensions" +#endif +struct heap_fsentry { + union { + struct fsentry ent; + char dummy[sizeof(struct fsentry) + MAX_PATH]; + } u; +}; +#pragma GCC diagnostic pop + +/* + * Compares the paths of two fsentry structures for equality. + */ +static int fsentry_cmp(void *cmp_data UNUSED, + const struct fsentry *fse1, const struct fsentry *fse2, + void *keydata UNUSED) +{ + int res; + if (fse1 == fse2) + return 0; + + /* compare the list parts first */ + if (fse1->list != fse2->list && + (res = fsentry_cmp(NULL, fse1->list ? fse1->list : fse1, + fse2->list ? fse2->list : fse2, NULL))) + return res; + + /* if list parts are equal, compare len and name */ + if (fse1->len != fse2->len) + return fse1->len - fse2->len; + return fspathncmp(fse1->dirent.d_name, fse2->dirent.d_name, fse1->len); +} + +/* + * Calculates the hash code of an fsentry structure's path. + */ +static unsigned int fsentry_hash(const struct fsentry *fse) +{ + unsigned int hash = fse->list ? fse->list->ent.hash : 0; + return hash ^ memihash(fse->dirent.d_name, fse->len); +} + +/* + * Initialize an fsentry structure for use by fsentry_hash and fsentry_cmp. + */ +static void fsentry_init(struct fsentry *fse, struct fsentry *list, + const char *name, size_t len) +{ + fse->list = list; + if (len > MAX_PATH) + BUG("Trying to allocate fsentry for long path '%.*s'", + (int)len, name); + memcpy(fse->dirent.d_name, name, len); + fse->dirent.d_name[len] = 0; + fse->len = len; + hashmap_entry_init(&fse->ent, fsentry_hash(fse)); +} + +/* + * Allocate an fsentry structure on the heap. + */ +static struct fsentry *fsentry_alloc(struct fsentry *list, const char *name, + size_t len) +{ + /* overallocate fsentry and copy the name to the end */ + struct fsentry *fse = xmalloc(sizeof(struct fsentry) + len + 1); + /* init the rest of the structure */ + fsentry_init(fse, list, name, len); + fse->next = NULL; + fse->u.refcnt = 1; + return fse; +} + +/* + * Add a reference to an fsentry. + */ +inline static void fsentry_addref(struct fsentry *fse) +{ + if (fse->list) + fse = fse->list; + + InterlockedIncrement(&(fse->u.refcnt)); +} + +/* + * Release the reference to an fsentry, frees the memory if its the last ref. + */ +static void fsentry_release(struct fsentry *fse) +{ + if (fse->list) + fse = fse->list; + + if (InterlockedDecrement(&(fse->u.refcnt))) + return; + + while (fse) { + struct fsentry *next = fse->next; + free(fse); + fse = next; + } +} + +/* + * Allocate and initialize an fsentry from a WIN32_FIND_DATA structure. + */ +static struct fsentry *fseentry_create_entry(struct fsentry *list, + const WIN32_FIND_DATAW *fdata) +{ + char buf[MAX_PATH * 3]; + int len; + struct fsentry *fse; + len = xwcstoutf(buf, fdata->cFileName, ARRAY_SIZE(buf)); + + fse = fsentry_alloc(list, buf, len); + + fse->st_mode = file_attr_to_st_mode(fdata->dwFileAttributes); + fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG; + fse->u.s.st_size = (((off64_t) (fdata->nFileSizeHigh)) << 32) + | fdata->nFileSizeLow; + filetime_to_timespec(&(fdata->ftLastAccessTime), &(fse->u.s.st_atim)); + filetime_to_timespec(&(fdata->ftLastWriteTime), &(fse->u.s.st_mtim)); + filetime_to_timespec(&(fdata->ftCreationTime), &(fse->u.s.st_ctim)); + + return fse; +} + +/* + * Create an fsentry-based directory listing (similar to opendir / readdir). + * Dir should not contain trailing '/'. Use an empty string for the current + * directory (not "."!). + */ +static struct fsentry *fsentry_create_list(const struct fsentry *dir) +{ + wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ + WIN32_FIND_DATAW fdata; + HANDLE h; + int wlen; + struct fsentry *list, **phead; + DWORD err; + + /* convert name to UTF-16 and check length < MAX_PATH */ + if ((wlen = xutftowcsn(pattern, dir->dirent.d_name, MAX_PATH, + dir->len)) < 0) { + if (errno == ERANGE) + errno = ENAMETOOLONG; + return NULL; + } + + /* append optional '/' and wildcard '*' */ + if (wlen) + pattern[wlen++] = '/'; + pattern[wlen++] = '*'; + pattern[wlen] = 0; + + /* open find handle */ + h = FindFirstFileW(pattern, &fdata); + if (h == INVALID_HANDLE_VALUE) { + err = GetLastError(); + errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err); + return NULL; + } + + /* allocate object to hold directory listing */ + list = fsentry_alloc(NULL, dir->dirent.d_name, dir->len); + + /* walk directory and build linked list of fsentry structures */ + phead = &list->next; + do { + *phead = fseentry_create_entry(list, &fdata); + phead = &(*phead)->next; + } while (FindNextFileW(h, &fdata)); + + /* remember result of last FindNextFile, then close find handle */ + err = GetLastError(); + FindClose(h); + + /* return the list if we've got all the files */ + if (err == ERROR_NO_MORE_FILES) + return list; + + /* otherwise free the list and return error */ + fsentry_release(list); + errno = err_win_to_posix(err); + return NULL; +} + +/* + * Adds a directory listing to the cache. + */ +static void fscache_add(struct fsentry *fse) +{ + if (fse->list) + fse = fse->list; + + for (; fse; fse = fse->next) + hashmap_add(&map, &fse->ent); +} + +/* + * Clears the cache. + */ +static void fscache_clear(void) +{ + hashmap_clear_and_free(&map, struct fsentry, ent); + hashmap_init(&map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0); +} + +/* + * Checks if the cache is enabled for the given path. + */ +static inline int fscache_enabled(const char *path) +{ + return enabled > 0 && !is_absolute_path(path); +} + +/* + * Looks up or creates a cache entry for the specified key. + */ +static struct fsentry *fscache_get(struct fsentry *key) +{ + struct fsentry *fse; + + EnterCriticalSection(&mutex); + /* check if entry is in cache */ + fse = hashmap_get_entry(&map, key, ent, NULL); + if (fse) { + fsentry_addref(fse); + LeaveCriticalSection(&mutex); + return fse; + } + /* if looking for a file, check if directory listing is in cache */ + if (!fse && key->list) { + fse = hashmap_get_entry(&map, key->list, ent, NULL); + if (fse) { + LeaveCriticalSection(&mutex); + /* dir entry without file entry -> file doesn't exist */ + errno = ENOENT; + return NULL; + } + } + + /* create the directory listing (outside mutex!) */ + LeaveCriticalSection(&mutex); + fse = fsentry_create_list(key->list ? key->list : key); + if (!fse) + return NULL; + + EnterCriticalSection(&mutex); + /* add directory listing if it hasn't been added by some other thread */ + if (!hashmap_get_entry(&map, key, ent, NULL)) + fscache_add(fse); + + /* lookup file entry if requested (fse already points to directory) */ + if (key->list) + fse = hashmap_get_entry(&map, key, ent, NULL); + + /* return entry or ENOENT */ + if (fse) + fsentry_addref(fse); + else + errno = ENOENT; + + LeaveCriticalSection(&mutex); + return fse; +} + +/* + * Enables or disables the cache. Note that the cache is read-only, changes to + * the working directory are NOT reflected in the cache while enabled. + */ +int fscache_enable(int enable) +{ + int result; + + if (!initialized) { + /* allow the cache to be disabled entirely */ + if (!core_fscache) + return 0; + + InitializeCriticalSection(&mutex); + hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, 0); + initialized = 1; + } + + result = enable ? InterlockedIncrement(&enabled) + : InterlockedDecrement(&enabled); + + if (enable && result == 1) { + /* redirect opendir and lstat to the fscache implementations */ + opendir = fscache_opendir; + lstat = fscache_lstat; + } else if (!enable && !result) { + /* reset opendir and lstat to the original implementations */ + opendir = dirent_opendir; + lstat = mingw_lstat; + EnterCriticalSection(&mutex); + fscache_clear(); + LeaveCriticalSection(&mutex); + } + return result; +} + +/* + * Lstat replacement, uses the cache if enabled, otherwise redirects to + * mingw_lstat. + */ +int fscache_lstat(const char *filename, struct stat *st) +{ + int dirlen, base, len; +#pragma GCC diagnostic push +#ifdef __clang__ +#pragma GCC diagnostic ignored "-Wflexible-array-extensions" +#endif + struct heap_fsentry key[2]; +#pragma GCC diagnostic pop + struct fsentry *fse; + + if (!fscache_enabled(filename)) + return mingw_lstat(filename, st); + + /* split filename into path + name */ + len = strlen(filename); + if (len && is_dir_sep(filename[len - 1])) + len--; + base = len; + while (base && !is_dir_sep(filename[base - 1])) + base--; + dirlen = base ? base - 1 : 0; + + /* lookup entry for path + name in cache */ + fsentry_init(&key[0].u.ent, NULL, filename, dirlen); + fsentry_init(&key[1].u.ent, &key[0].u.ent, filename + base, len - base); + fse = fscache_get(&key[1].u.ent); + if (!fse) { + errno = ENOENT; + return -1; + } + + /* copy stat data */ + st->st_ino = 0; + st->st_gid = 0; + st->st_uid = 0; + st->st_dev = 0; + st->st_rdev = 0; + st->st_nlink = 1; + st->st_mode = fse->st_mode; + st->st_size = fse->u.s.st_size; + st->st_atim = fse->u.s.st_atim; + st->st_mtim = fse->u.s.st_mtim; + st->st_ctim = fse->u.s.st_ctim; + + /* don't forget to release fsentry */ + fsentry_release(fse); + return 0; +} + +typedef struct fscache_DIR { + struct DIR base_dir; /* extend base struct DIR */ + struct fsentry *pfsentry; + struct dirent *dirent; +} fscache_DIR; + +/* + * Readdir replacement. + */ +static struct dirent *fscache_readdir(DIR *base_dir) +{ + fscache_DIR *dir = (fscache_DIR*) base_dir; + struct fsentry *next = dir->pfsentry->next; + if (!next) + return NULL; + dir->pfsentry = next; + dir->dirent = &next->dirent; + return dir->dirent; +} + +/* + * Closedir replacement. + */ +static int fscache_closedir(DIR *base_dir) +{ + fscache_DIR *dir = (fscache_DIR*) base_dir; + fsentry_release(dir->pfsentry); + free(dir); + return 0; +} + +/* + * Opendir replacement, uses a directory listing from the cache if enabled, + * otherwise calls original dirent implementation. + */ +DIR *fscache_opendir(const char *dirname) +{ + struct heap_fsentry key; + struct fsentry *list; + fscache_DIR *dir; + int len; + + if (!fscache_enabled(dirname)) + return dirent_opendir(dirname); + + /* prepare name (strip trailing '/', replace '.') */ + len = strlen(dirname); + if ((len == 1 && dirname[0] == '.') || + (len && is_dir_sep(dirname[len - 1]))) + len--; + + /* get directory listing from cache */ + fsentry_init(&key.u.ent, NULL, dirname, len); + list = fscache_get(&key.u.ent); + if (!list) + return NULL; + + /* alloc and return DIR structure */ + dir = (fscache_DIR*) xmalloc(sizeof(fscache_DIR)); + dir->base_dir.preaddir = fscache_readdir; + dir->base_dir.pclosedir = fscache_closedir; + dir->pfsentry = list; + return (DIR*) dir; +} diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h new file mode 100644 index 00000000000000..ed518b422d705e --- /dev/null +++ b/compat/win32/fscache.h @@ -0,0 +1,10 @@ +#ifndef FSCACHE_H +#define FSCACHE_H + +int fscache_enable(int enable); +#define enable_fscache(x) fscache_enable(x) + +DIR *fscache_opendir(const char *dir); +int fscache_lstat(const char *file_name, struct stat *buf); + +#endif diff --git a/config.mak.uname b/config.mak.uname index 8d56e75aa8f57e..d46c50e20e4627 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -528,7 +528,7 @@ endif compat/win32/path-utils.o \ compat/win32/pthread.o compat/win32/syslog.o \ compat/win32/trace2_win32_process_info.o \ - compat/win32/dirent.o + compat/win32/dirent.o compat/win32/fscache.o COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\" BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO # invalidcontinue.obj allows Git's source code to close the same file @@ -731,7 +731,7 @@ ifeq ($(uname_S),MINGW) compat/win32/flush.o \ compat/win32/path-utils.o \ compat/win32/pthread.o compat/win32/syslog.o \ - compat/win32/dirent.o + compat/win32/dirent.o compat/win32/fscache.o BASIC_CFLAGS += -DWIN32 EXTLIBS += -lws2_32 GITLIBS += git.res diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index 09e13cee93cd8e..4805206b10a997 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -294,7 +294,8 @@ if(CMAKE_SYSTEM_NAME STREQUAL "Windows") compat/win32/trace2_win32_process_info.c compat/win32/dirent.c compat/nedmalloc/nedmalloc.c - compat/strdup.c) + compat/strdup.c + compat/win32/fscache.c) set(NO_UNIX_SOCKETS 1) elseif(CMAKE_SYSTEM_NAME STREQUAL "Linux") diff --git a/git-compat-util.h b/git-compat-util.h index 77547563b7b082..bdc4de3d3f8c4a 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -191,9 +191,11 @@ static inline int is_xplatform_dir_sep(int c) /* pull in Windows compatibility stuff */ #include "compat/win32/path-utils.h" #include "compat/mingw.h" +#include "compat/win32/fscache.h" #elif defined(_MSC_VER) #include "compat/win32/path-utils.h" #include "compat/msvc.h" +#include "compat/win32/fscache.h" #endif /* used on Mac OS X */ diff --git a/meson.build b/meson.build index 617cf9b7facb8c..327ce7e7387b95 100644 --- a/meson.build +++ b/meson.build @@ -1247,6 +1247,7 @@ elif host_machine.system() == 'windows' 'compat/winansi.c', 'compat/win32/dirent.c', 'compat/win32/flush.c', + 'compat/win32/fscache.c', 'compat/win32/path-utils.c', 'compat/win32/pthread.c', 'compat/win32/syslog.c', From 7fa25b25693e7c43e938a141018f706f23811197 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 24 Jun 2014 13:22:35 +0200 Subject: [PATCH 150/248] fscache: load directories only once If multiple threads access a directory that is not yet in the cache, the directory will be loaded by each thread. Only one of the results is added to the cache, all others are leaked. This wastes performance and memory. On cache miss, add a future object to the cache to indicate that the directory is currently being loaded. Subsequent threads register themselves with the future object and wait. When the first thread has loaded the directory, it replaces the future object with the result and notifies waiting threads. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/win32/fscache.c | 65 ++++++++++++++++++++++++++++++++++++------ 1 file changed, 56 insertions(+), 9 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index dc765ddd57b5bc..ff2479c7387f13 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -27,6 +27,8 @@ struct fsentry { union { /* Reference count of the directory listing. */ volatile long refcnt; + /* Handle to wait on the loading thread. */ + HANDLE hwait; struct { /* More stat members (only used for file entries). */ off64_t st_size; @@ -266,16 +268,43 @@ static inline int fscache_enabled(const char *path) return enabled > 0 && !is_absolute_path(path); } +/* + * Looks up a cache entry, waits if its being loaded by another thread. + * The mutex must be owned by the calling thread. + */ +static struct fsentry *fscache_get_wait(struct fsentry *key) +{ + struct fsentry *fse = hashmap_get_entry(&map, key, ent, NULL); + + /* return if its a 'real' entry (future entries have refcnt == 0) */ + if (!fse || fse->list || fse->u.refcnt) + return fse; + + /* create an event and link our key to the future entry */ + key->u.hwait = CreateEvent(NULL, TRUE, FALSE, NULL); + key->next = fse->next; + fse->next = key; + + /* wait for the loading thread to signal us */ + LeaveCriticalSection(&mutex); + WaitForSingleObject(key->u.hwait, INFINITE); + CloseHandle(key->u.hwait); + EnterCriticalSection(&mutex); + + /* repeat cache lookup */ + return hashmap_get_entry(&map, key, ent, NULL); +} + /* * Looks up or creates a cache entry for the specified key. */ static struct fsentry *fscache_get(struct fsentry *key) { - struct fsentry *fse; + struct fsentry *fse, *future, *waiter; EnterCriticalSection(&mutex); /* check if entry is in cache */ - fse = hashmap_get_entry(&map, key, ent, NULL); + fse = fscache_get_wait(key); if (fse) { fsentry_addref(fse); LeaveCriticalSection(&mutex); @@ -283,7 +312,7 @@ static struct fsentry *fscache_get(struct fsentry *key) } /* if looking for a file, check if directory listing is in cache */ if (!fse && key->list) { - fse = hashmap_get_entry(&map, key->list, ent, NULL); + fse = fscache_get_wait(key->list); if (fse) { LeaveCriticalSection(&mutex); /* dir entry without file entry -> file doesn't exist */ @@ -292,16 +321,34 @@ static struct fsentry *fscache_get(struct fsentry *key) } } + /* add future entry to indicate that we're loading it */ + future = key->list ? key->list : key; + future->next = NULL; + future->u.refcnt = 0; + hashmap_add(&map, &future->ent); + /* create the directory listing (outside mutex!) */ LeaveCriticalSection(&mutex); - fse = fsentry_create_list(key->list ? key->list : key); - if (!fse) + fse = fsentry_create_list(future); + EnterCriticalSection(&mutex); + + /* remove future entry and signal waiting threads */ + hashmap_remove(&map, &future->ent, NULL); + waiter = future->next; + while (waiter) { + HANDLE h = waiter->u.hwait; + waiter = waiter->next; + SetEvent(h); + } + + /* leave on error (errno set by fsentry_create_list) */ + if (!fse) { + LeaveCriticalSection(&mutex); return NULL; + } - EnterCriticalSection(&mutex); - /* add directory listing if it hasn't been added by some other thread */ - if (!hashmap_get_entry(&map, key, ent, NULL)) - fscache_add(fse); + /* add directory listing to the cache */ + fscache_add(fse); /* lookup file entry if requested (fse already points to directory) */ if (key->list) From 271aa38236d95679edf7ff87d693d9be406834fe Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Tue, 24 Jan 2017 15:12:13 -0500 Subject: [PATCH 151/248] fscache: add key for GIT_TRACE_FSCACHE Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index ff2479c7387f13..d67dc918d6b71c 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -4,11 +4,13 @@ #include "fscache.h" #include "../../dir.h" #include "../../abspath.h" +#include "../../trace.h" static int initialized; static volatile long enabled; static struct hashmap map; static CRITICAL_SECTION mutex; +static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); /* * An entry in the file system cache. Used for both entire directory listings @@ -212,6 +214,8 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir) if (h == INVALID_HANDLE_VALUE) { err = GetLastError(); errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err); + trace_printf_key(&trace_fscache, "fscache: error(%d) '%s'\n", + errno, dir->dirent.d_name); return NULL; } @@ -397,6 +401,7 @@ int fscache_enable(int enable) fscache_clear(); LeaveCriticalSection(&mutex); } + trace_printf_key(&trace_fscache, "fscache: enable(%d)\n", enable); return result; } From 2ed1272cceb8e25a80ff8c53135091d3a50ab901 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Tue, 13 Dec 2016 14:05:32 -0500 Subject: [PATCH 152/248] fscache: remember not-found directories Teach FSCACHE to remember "not found" directories. This is a performance optimization. FSCACHE is a performance optimization available for Windows. It intercepts Posix-style lstat() calls into an in-memory directory using FindFirst/FindNext. It improves performance on Windows by catching the first lstat() call in a directory, using FindFirst/ FindNext to read the list of files (and attribute data) for the entire directory into the cache, and short-cut subsequent lstat() calls in the same directory. This gives a major performance boost on Windows. However, it does not remember "not found" directories. When STATUS runs and there are missing directories, the lstat() interception fails to find the parent directory and simply return ENOENT for the file -- it does not remember that the FindFirst on the directory failed. Thus subsequent lstat() calls in the same directory, each re-attempt the FindFirst. This completely defeats any performance gains. This can be seen by doing a sparse-checkout on a large repo and then doing a read-tree to reset the skip-worktree bits and then running status. This change reduced status times for my very large repo by 60%. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 36 ++++++++++++++++++++++++++++++++---- 1 file changed, 32 insertions(+), 4 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index d67dc918d6b71c..7aa3450e7edf47 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -186,7 +186,8 @@ static struct fsentry *fseentry_create_entry(struct fsentry *list, * Dir should not contain trailing '/'. Use an empty string for the current * directory (not "."!). */ -static struct fsentry *fsentry_create_list(const struct fsentry *dir) +static struct fsentry *fsentry_create_list(const struct fsentry *dir, + int *dir_not_found) { wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ WIN32_FIND_DATAW fdata; @@ -195,6 +196,8 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir) struct fsentry *list, **phead; DWORD err; + *dir_not_found = 0; + /* convert name to UTF-16 and check length < MAX_PATH */ if ((wlen = xutftowcsn(pattern, dir->dirent.d_name, MAX_PATH, dir->len)) < 0) { @@ -213,6 +216,7 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir) h = FindFirstFileW(pattern, &fdata); if (h == INVALID_HANDLE_VALUE) { err = GetLastError(); + *dir_not_found = 1; /* or empty directory */ errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err); trace_printf_key(&trace_fscache, "fscache: error(%d) '%s'\n", errno, dir->dirent.d_name); @@ -221,6 +225,8 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir) /* allocate object to hold directory listing */ list = fsentry_alloc(NULL, dir->dirent.d_name, dir->len); + list->st_mode = S_IFDIR; + list->dirent.d_type = DT_DIR; /* walk directory and build linked list of fsentry structures */ phead = &list->next; @@ -305,12 +311,16 @@ static struct fsentry *fscache_get_wait(struct fsentry *key) static struct fsentry *fscache_get(struct fsentry *key) { struct fsentry *fse, *future, *waiter; + int dir_not_found; EnterCriticalSection(&mutex); /* check if entry is in cache */ fse = fscache_get_wait(key); if (fse) { - fsentry_addref(fse); + if (fse->st_mode) + fsentry_addref(fse); + else + fse = NULL; /* non-existing directory */ LeaveCriticalSection(&mutex); return fse; } @@ -319,7 +329,10 @@ static struct fsentry *fscache_get(struct fsentry *key) fse = fscache_get_wait(key->list); if (fse) { LeaveCriticalSection(&mutex); - /* dir entry without file entry -> file doesn't exist */ + /* + * dir entry without file entry, or dir does not + * exist -> file doesn't exist + */ errno = ENOENT; return NULL; } @@ -333,7 +346,7 @@ static struct fsentry *fscache_get(struct fsentry *key) /* create the directory listing (outside mutex!) */ LeaveCriticalSection(&mutex); - fse = fsentry_create_list(future); + fse = fsentry_create_list(future, &dir_not_found); EnterCriticalSection(&mutex); /* remove future entry and signal waiting threads */ @@ -347,6 +360,18 @@ static struct fsentry *fscache_get(struct fsentry *key) /* leave on error (errno set by fsentry_create_list) */ if (!fse) { + if (dir_not_found && key->list) { + /* + * Record that the directory does not exist (or is + * empty, which for all practical matters is the same + * thing as far as fscache is concerned). + */ + fse = fsentry_alloc(key->list->list, + key->list->dirent.d_name, + key->list->len); + fse->st_mode = 0; + hashmap_add(&map, &fse->ent); + } LeaveCriticalSection(&mutex); return NULL; } @@ -358,6 +383,9 @@ static struct fsentry *fscache_get(struct fsentry *key) if (key->list) fse = hashmap_get_entry(&map, key, ent, NULL); + if (fse && !fse->st_mode) + fse = NULL; /* non-existing directory */ + /* return entry or ENOENT */ if (fse) fsentry_addref(fse); From 6a6f95808bc88bf8ed6f53d3e03f54d3ae39ddb1 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 25 Jan 2017 18:39:16 +0100 Subject: [PATCH 153/248] fscache: add a test for the dir-not-found optimization Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t1090-sparse-checkout-scope.sh | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/t/t1090-sparse-checkout-scope.sh b/t/t1090-sparse-checkout-scope.sh index 3a14218b245d4c..529844e2862c74 100755 --- a/t/t1090-sparse-checkout-scope.sh +++ b/t/t1090-sparse-checkout-scope.sh @@ -106,4 +106,24 @@ test_expect_success 'in partial clone, sparse checkout only fetches needed blobs test_cmp expect actual ' +test_expect_success MINGW 'no unnecessary opendir() with fscache' ' + git clone . fscache-test && + ( + cd fscache-test && + git config core.fscache 1 && + echo "/excluded/*" >.git/info/sparse-checkout && + for f in $(test_seq 10) + do + sha1=$(echo $f | git hash-object -w --stdin) && + git update-index --add \ + --cacheinfo 100644,$sha1,excluded/$f || exit 1 + done && + test_tick && + git commit -m excluded && + GIT_TRACE_FSCACHE=1 git status >out 2>err && + grep excluded err >grep.out && + test_line_count = 1 grep.out + ) +' + test_done From d6e1c122344a3a61bb78eb974c5218e3faeb734c Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Tue, 22 Nov 2016 11:26:38 -0500 Subject: [PATCH 154/248] add: use preload-index and fscache for performance Teach "add" to use preload-index and fscache features to improve performance on very large repositories. During an "add", a call is made to run_diff_files() which calls check_remove() for each index-entry. This calls lstat(). On Windows, the fscache code intercepts the lstat() calls and builds a private cache using the FindFirst/FindNext routines, which are much faster. Somewhat independent of this, is the preload-index code which distributes some of the start-up costs across multiple threads. We need to keep the call to read_cache() before parsing the pathspecs (and hence cannot use the pathspecs to limit any preload) because parse_pathspec() is using the index to determine whether a pathspec is, in fact, in a submodule. If we would not read the index first, parse_pathspec() would not error out on a path that is inside a submodule, and t7400-submodule-basic.sh would fail with not ok 47 - do not add files from a submodule We still want the nice preload performance boost, though, so we simply call read_cache_preload(&pathspecs) after parsing the pathspecs. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- builtin/add.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/builtin/add.c b/builtin/add.c index 0235854f8099c4..7f5077e7fa17de 100644 --- a/builtin/add.c +++ b/builtin/add.c @@ -491,6 +491,10 @@ int cmd_add(int argc, die_in_unpopulated_submodule(repo->index, prefix); die_path_inside_submodule(repo->index, &pathspec); + enable_fscache(1); + /* We do not really re-read the index but update the up-to-date flags */ + preload_index(repo->index, &pathspec, 0); + if (add_new_files) { int baselen; @@ -603,5 +607,6 @@ int cmd_add(int argc, free(ps_matched); dir_clear(&dir); clear_pathspec(&pathspec); + enable_fscache(0); return exit_status; } From 1c71ed0b69e650f72ba1d5375b12c7c9429886c1 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Wed, 1 Nov 2017 15:05:44 -0400 Subject: [PATCH 155/248] dir.c: make add_excludes aware of fscache during status Teach read_directory_recursive() and add_excludes() to be aware of optional fscache and avoid trying to open() and fstat() non-existant ".gitignore" files in every directory in the worktree. The current code in add_excludes() calls open() and then fstat() for a ".gitignore" file in each directory present in the worktree. Change that when fscache is enabled to call lstat() first and if present, call open(). This seems backwards because both lstat needs to do more work than fstat. But when fscache is enabled, fscache will already know if the .gitignore file exists and can completely avoid the IO calls. This works because of the lstat diversion to mingw_lstat when fscache is enabled. This reduced status times on a 350K file enlistment of the Windows repo on a NVMe SSD by 0.25 seconds. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> --- compat/win32/fscache.c | 5 +++++ compat/win32/fscache.h | 3 +++ dir.c | 39 ++++++++++++++++++++++++++++++--------- git-compat-util.h | 4 ++++ 4 files changed, 42 insertions(+), 9 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 7aa3450e7edf47..edec8f5813fcf1 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -12,6 +12,11 @@ static struct hashmap map; static CRITICAL_SECTION mutex; static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); +int fscache_is_enabled(void) +{ + return enabled; +} + /* * An entry in the file system cache. Used for both entire directory listings * and file entries. diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index ed518b422d705e..9a21fd5709c5bc 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -4,6 +4,9 @@ int fscache_enable(int enable); #define enable_fscache(x) fscache_enable(x) +int fscache_is_enabled(void); +#define is_fscache_enabled() (fscache_is_enabled()) + DIR *fscache_opendir(const char *dir); int fscache_lstat(const char *file_name, struct stat *buf); diff --git a/dir.c b/dir.c index 7731f7b9d305b7..f20ccc5b2d7c5d 100644 --- a/dir.c +++ b/dir.c @@ -1156,16 +1156,37 @@ static int add_patterns(const char *fname, const char *base, int baselen, size_t size = 0; char *buf; - if (flags & PATTERN_NOFOLLOW) - fd = open_nofollow(fname, O_RDONLY); - else - fd = open(fname, O_RDONLY); - - if (fd < 0 || fstat(fd, &st) < 0) { - if (fd < 0) - warn_on_fopen_errors(fname); + /* + * Since `clang`'s `-Wunreachable-code` mode is clever, it would figure + * out that on non-Windows platforms, this `lstat()` is unreachable. + * We do want to keep the conditional block for the sake of Windows, + * though, so let's use the `NOT_CONSTANT()` trick to suppress that error. + */ + if (NOT_CONSTANT(is_fscache_enabled(fname))) { + if (lstat(fname, &st) < 0) { + fd = -1; + } else { + fd = open(fname, O_RDONLY); + if (fd < 0) + warn_on_fopen_errors(fname); + } + } else { + if (flags & PATTERN_NOFOLLOW) + fd = open_nofollow(fname, O_RDONLY); else - close(fd); + fd = open(fname, O_RDONLY); + + if (fd < 0 || fstat(fd, &st) < 0) { + if (fd < 0) + warn_on_fopen_errors(fname); + else { + close(fd); + fd = -1; + } + } + } + + if (fd < 0) { if (!istate) return -1; r = read_skip_worktree_file_from_index(istate, fname, diff --git a/git-compat-util.h b/git-compat-util.h index bdc4de3d3f8c4a..18e773deaf45c1 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1065,6 +1065,10 @@ static inline int is_missing_file_error(int errno_) #define enable_fscache(x) /* noop */ #endif +#ifndef is_fscache_enabled +#define is_fscache_enabled() (0) +#endif + int cmd_main(int, const char **); /* From 07a12fa9e52c7709672dba5e8a1342b9f85c6531 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Wed, 20 Dec 2017 10:43:41 -0500 Subject: [PATCH 156/248] fscache: make fscache_enabled() public Make fscache_enabled() function public rather than static. Remove unneeded fscache_is_enabled() function. Change is_fscache_enabled() macro to call fscache_enabled(). is_fscache_enabled() now takes a pathname so that the answer is more precise and mean "is fscache enabled for this pathname", since fscache only stores repo-relative paths and not absolute paths, we can avoid attempting lookups for absolute paths. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> --- compat/win32/fscache.c | 7 +------ compat/win32/fscache.h | 4 ++-- git-compat-util.h | 2 +- 3 files changed, 4 insertions(+), 9 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index edec8f5813fcf1..6e44df0a2dc2e7 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -12,11 +12,6 @@ static struct hashmap map; static CRITICAL_SECTION mutex; static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); -int fscache_is_enabled(void) -{ - return enabled; -} - /* * An entry in the file system cache. Used for both entire directory listings * and file entries. @@ -278,7 +273,7 @@ static void fscache_clear(void) /* * Checks if the cache is enabled for the given path. */ -static inline int fscache_enabled(const char *path) +int fscache_enabled(const char *path) { return enabled > 0 && !is_absolute_path(path); } diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index 9a21fd5709c5bc..660ada053b4309 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -4,8 +4,8 @@ int fscache_enable(int enable); #define enable_fscache(x) fscache_enable(x) -int fscache_is_enabled(void); -#define is_fscache_enabled() (fscache_is_enabled()) +int fscache_enabled(const char *path); +#define is_fscache_enabled(path) fscache_enabled(path) DIR *fscache_opendir(const char *dir); int fscache_lstat(const char *file_name, struct stat *buf); diff --git a/git-compat-util.h b/git-compat-util.h index 18e773deaf45c1..6011794646c53a 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1066,7 +1066,7 @@ static inline int is_missing_file_error(int errno_) #endif #ifndef is_fscache_enabled -#define is_fscache_enabled() (0) +#define is_fscache_enabled(path) (0) #endif int cmd_main(int, const char **); From f049733bb11306e30027faa2c744d184d03eac84 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Wed, 20 Dec 2017 11:19:27 -0500 Subject: [PATCH 157/248] dir.c: regression fix for add_excludes with fscache Fix regression described in: https://github.com/git-for-windows/git/issues/1392 which was introduced in: https://github.com/git-for-windows/git/commit/b2353379bba414e6c00dde913497cc9c827366f2 Problem Symptoms ================ When the user has a .gitignore file that is a symlink, the fscache optimization introduced above caused the stat-data from the symlink, rather that of the target file, to be returned. Later when the ignore file was read, the buffer length did not match the stat.st_size field and we called die("cannot use <path> as an exclude file") Optimization Rationale ====================== The above optimization calls lstat() before open() primarily to ask fscache if the file exists. It gets the current stat-data as a side effect essentially for free (since we already have it in memory). If the file does not exist, it does not need to call open(). And since very few directories have .gitignore files, we can greatly reduce time spent in the filesystem. Discussion of Fix ================= The above optimization calls lstat() rather than stat() because the fscache only intercepts lstat() calls. Calls to stat() stay directed to the mingw_stat() completly bypassing fscache. Furthermore, calls to mingw_stat() always call {open, fstat, close} so that symlinks are properly dereferenced, which adds *additional* open/close calls on top of what the original code in dir.c is doing. Since the problem only manifests for symlinks, we add code to overwrite the stat-data when the path is a symlink. This preserves the effect of the performance gains provided by the fscache in the normal case. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> --- dir.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/dir.c b/dir.c index f20ccc5b2d7c5d..6dc4ea2c9f385d 100644 --- a/dir.c +++ b/dir.c @@ -1157,6 +1157,28 @@ static int add_patterns(const char *fname, const char *base, int baselen, char *buf; /* + * A performance optimization for status. + * + * During a status scan, git looks in each directory for a .gitignore + * file before scanning the directory. Since .gitignore files are not + * that common, we can waste a lot of time looking for files that are + * not there. Fortunately, the fscache already knows if the directory + * contains a .gitignore file, since it has already read the directory + * and it already has the stat-data. + * + * If the fscache is enabled, use the fscache-lstat() interlude to see + * if the file exists (in the fscache hash maps) before trying to open() + * it. + * + * This causes problem when the .gitignore file is a symlink, because + * we call lstat() rather than stat() on the symlnk and the resulting + * stat-data is for the symlink itself rather than the target file. + * We CANNOT use stat() here because the fscache DOES NOT install an + * interlude for stat() and mingw_stat() always calls "open-fstat-close" + * on the file and defeats the purpose of the optimization here. Since + * symlinks are even more rare than .gitignore files, we force a fstat() + * after our open() to get stat-data for the target file. + * * Since `clang`'s `-Wunreachable-code` mode is clever, it would figure * out that on non-Windows platforms, this `lstat()` is unreachable. * We do want to keep the conditional block for the sake of Windows, @@ -1169,6 +1191,11 @@ static int add_patterns(const char *fname, const char *base, int baselen, fd = open(fname, O_RDONLY); if (fd < 0) warn_on_fopen_errors(fname); + else if (S_ISLNK(st.st_mode) && fstat(fd, &st) < 0) { + warn_on_fopen_errors(fname); + close(fd); + fd = -1; + } } } else { if (flags & PATTERN_NOFOLLOW) From d1d7ebe12dc193182ea6a94339f6766c1aed37af Mon Sep 17 00:00:00 2001 From: Takuto Ikuta <tikuta@chromium.org> Date: Wed, 22 Nov 2017 20:39:38 +0900 Subject: [PATCH 158/248] fetch-pack.c: enable fscache for stats under .git/objects When I do git fetch, git call file stats under .git/objects for each refs. This takes time when there are many refs. By enabling fscache, git takes file stats by directory traversing and that improved the speed of fetch-pack for repository having large number of refs. In my windows workstation, this improves the time of `git fetch` for chromium repository like below. I took stats 3 times. * With this patch TotalSeconds: 9.9825165 TotalSeconds: 9.1862075 TotalSeconds: 10.1956256 Avg: 9.78811653333333 * Without this patch TotalSeconds: 15.8406702 TotalSeconds: 15.6248053 TotalSeconds: 15.2085938 Avg: 15.5580231 Signed-off-by: Takuto Ikuta <tikuta@chromium.org> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- fetch-pack.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fetch-pack.c b/fetch-pack.c index a9f5e6b5106943..9dd7993544b6f5 100644 --- a/fetch-pack.c +++ b/fetch-pack.c @@ -768,6 +768,7 @@ static void mark_complete_and_common_ref(struct fetch_negotiator *negotiator, save_commit_buffer = 0; trace2_region_enter("fetch-pack", "parse_remote_refs_and_find_cutoff", NULL); + enable_fscache(1); for (ref = *refs; ref; ref = ref->next) { struct commit *commit; @@ -792,6 +793,7 @@ static void mark_complete_and_common_ref(struct fetch_negotiator *negotiator, if (!cutoff || cutoff < commit->date) cutoff = commit->date; } + enable_fscache(0); trace2_region_leave("fetch-pack", "parse_remote_refs_and_find_cutoff", NULL); /* From 3dc226fc9a6a49cc3705d79969edf7c906b8a8ce Mon Sep 17 00:00:00 2001 From: Takuto Ikuta <tikuta@chromium.org> Date: Tue, 30 Jan 2018 22:42:58 +0900 Subject: [PATCH 159/248] checkout.c: enable fscache for checkout again This is retry of #1419. I added flush_fscache macro to flush cached stats after disk writing with tests for regression reported in #1438 and #1442. git checkout checks each file path in sorted order, so cache flushing does not make performance worse unless we have large number of modified files in a directory containing many files. Using chromium repository, I tested `git checkout .` performance when I delete 10 files in different directories. With this patch: TotalSeconds: 4.307272 TotalSeconds: 4.4863595 TotalSeconds: 4.2975562 Avg: 4.36372923333333 Without this patch: TotalSeconds: 20.9705431 TotalSeconds: 22.4867685 TotalSeconds: 18.8968292 Avg: 20.7847136 I confirmed this patch passed all tests in t/ with core_fscache=1. Signed-off-by: Takuto Ikuta <tikuta@chromium.org> --- builtin/checkout.c | 2 ++ compat/win32/fscache.c | 12 ++++++++++++ compat/win32/fscache.h | 3 +++ entry.c | 3 +++ git-compat-util.h | 4 ++++ parallel-checkout.c | 1 + t/t7201-co.sh | 36 ++++++++++++++++++++++++++++++++++++ 7 files changed, 61 insertions(+) diff --git a/builtin/checkout.c b/builtin/checkout.c index f9453473fe2a20..d0689ac6ab82ce 100644 --- a/builtin/checkout.c +++ b/builtin/checkout.c @@ -415,6 +415,7 @@ static int checkout_worktree(const struct checkout_opts *opts, if (pc_workers > 1) init_parallel_checkout(); + enable_fscache(1); for (pos = 0; pos < the_repository->index->cache_nr; pos++) { struct cache_entry *ce = the_repository->index->cache[pos]; if (ce->ce_flags & CE_MATCHED) { @@ -440,6 +441,7 @@ static int checkout_worktree(const struct checkout_opts *opts, errs |= run_parallel_checkout(&state, pc_workers, pc_threshold, NULL, NULL); mem_pool_discard(&ce_mem_pool, should_validate_cache_entries()); + enable_fscache(0); remove_marked_cache_entries(the_repository->index, 1); remove_scheduled_dirs(); errs |= finish_delayed_checkout(&state, opts->show_progress); diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 6e44df0a2dc2e7..b6de459c4d88a6 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -433,6 +433,18 @@ int fscache_enable(int enable) return result; } +/* + * Flush cached stats result when fscache is enabled. + */ +void fscache_flush(void) +{ + if (enabled) { + EnterCriticalSection(&mutex); + fscache_clear(); + LeaveCriticalSection(&mutex); + } +} + /* * Lstat replacement, uses the cache if enabled, otherwise redirects to * mingw_lstat. diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index 660ada053b4309..2f06f8df97dcd0 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -7,6 +7,9 @@ int fscache_enable(int enable); int fscache_enabled(const char *path); #define is_fscache_enabled(path) fscache_enabled(path) +void fscache_flush(void); +#define flush_fscache() fscache_flush() + DIR *fscache_opendir(const char *dir); int fscache_lstat(const char *file_name, struct stat *buf); diff --git a/entry.c b/entry.c index cae02eb50398d7..679d11b1dda8cc 100644 --- a/entry.c +++ b/entry.c @@ -411,6 +411,9 @@ static int write_entry(struct cache_entry *ce, char *path, struct conv_attrs *ca } finish: + /* Flush cached lstat in fscache after writing to disk. */ + flush_fscache(); + if (state->refresh_cache) { if (!fstat_done && lstat(ce->name, &st) < 0) return error_errno("unable to stat just-written file %s", diff --git a/git-compat-util.h b/git-compat-util.h index 6011794646c53a..e01a355e4ebb78 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1069,6 +1069,10 @@ static inline int is_missing_file_error(int errno_) #define is_fscache_enabled(path) (0) #endif +#ifndef flush_fscache +#define flush_fscache() /* noop */ +#endif + int cmd_main(int, const char **); /* diff --git a/parallel-checkout.c b/parallel-checkout.c index fba6aa65a6e852..dc690a3b210f7b 100644 --- a/parallel-checkout.c +++ b/parallel-checkout.c @@ -639,6 +639,7 @@ static void write_items_sequentially(struct checkout *state) { size_t i; + flush_fscache(); for (i = 0; i < parallel_checkout.nr; i++) { struct parallel_checkout_item *pc_item = ¶llel_checkout.items[i]; write_pc_item(pc_item, state); diff --git a/t/t7201-co.sh b/t/t7201-co.sh index 9bcf7c0b40461f..545f388c44a515 100755 --- a/t/t7201-co.sh +++ b/t/t7201-co.sh @@ -35,6 +35,42 @@ fill () { } +test_expect_success MINGW 'fscache flush cache' ' + + git init fscache-test && + cd fscache-test && + git config core.fscache 1 && + echo A > test.txt && + git add test.txt && + git commit -m A && + echo B >> test.txt && + git checkout . && + test -z "$(git status -s)" && + echo A > expect.txt && + test_cmp expect.txt test.txt && + cd .. && + rm -rf fscache-test +' + +test_expect_success MINGW 'fscache flush cache dir' ' + + git init fscache-test && + cd fscache-test && + git config core.fscache 1 && + echo A > test.txt && + git add test.txt && + git commit -m A && + rm test.txt && + mkdir test.txt && + touch test.txt/test.txt && + git checkout . && + test -z "$(git status -s)" && + echo A > expect.txt && + test_cmp expect.txt test.txt && + cd .. && + rm -rf fscache-test +' + test_expect_success setup ' fill x y z >same && fill 1 2 3 4 5 6 7 8 >one && From 8aeff6f30d209ffaad7cc8bcb05e8a28fdb81e4d Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Fri, 7 Sep 2018 11:39:57 -0400 Subject: [PATCH 160/248] Enable the filesystem cache (fscache) in refresh_index(). On file systems that support it, this can dramatically speed up operations like add, commit, describe, rebase, reset, rm that would otherwise have to lstat() every file to "re-match" the stat information in the index to that of the file system. On a synthetic repo with 1M files, "git reset" dropped from 52.02 seconds to 14.42 seconds for a savings of 72%. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- read-cache.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/read-cache.c b/read-cache.c index 06ad74db2286ae..169ca23de86bbc 100644 --- a/read-cache.c +++ b/read-cache.c @@ -1505,6 +1505,7 @@ int refresh_index(struct index_state *istate, unsigned int flags, typechange_fmt = in_porcelain ? "T\t%s\n" : "%s: needs update\n"; added_fmt = in_porcelain ? "A\t%s\n" : "%s: needs update\n"; unmerged_fmt = in_porcelain ? "U\t%s\n" : "%s: needs merge\n"; + enable_fscache(1); /* * Use the multi-threaded preload_index() to refresh most of the * cache entries quickly then in the single threaded loop below, @@ -1599,6 +1600,7 @@ int refresh_index(struct index_state *istate, unsigned int flags, display_progress(progress, istate->cache_nr); stop_progress(&progress); trace_performance_leave("refresh index"); + enable_fscache(0); return has_errors; } From 7c096a447b0ec3ba0bd1f289119c8dfd147e550a Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Tue, 23 Oct 2018 11:42:06 -0400 Subject: [PATCH 161/248] fscache: use FindFirstFileExW to avoid retrieving the short name Use FindFirstFileExW with FindExInfoBasic to avoid forcing NTFS to look up the short name. Also switch to a larger (64K vs 4K) buffer using FIND_FIRST_EX_LARGE_FETCH to minimize round trips to the kernel. In a repo with ~200K files, this drops warm cache status times from 3.19 seconds to 2.67 seconds for a 16% savings. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- compat/win32/fscache.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index b6de459c4d88a6..c6ab9f1a2c7286 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -213,7 +213,8 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir, pattern[wlen] = 0; /* open find handle */ - h = FindFirstFileW(pattern, &fdata); + h = FindFirstFileExW(pattern, FindExInfoBasic, &fdata, FindExSearchNameMatch, + NULL, FIND_FIRST_EX_LARGE_FETCH); if (h == INVALID_HANDLE_VALUE) { err = GetLastError(); *dir_not_found = 1; /* or empty directory */ From 280ec8db2ae45a87f82348f01154fb10d91e6211 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Thu, 4 Oct 2018 18:10:21 -0400 Subject: [PATCH 162/248] fscache: add GIT_TEST_FSCACHE support Add support to fscache to enable running the entire test suite with the fscache enabled. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- compat/win32/fscache.c | 5 +++++ t/README | 3 +++ 2 files changed, 8 insertions(+) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index c6ab9f1a2c7286..13b38104732592 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -5,6 +5,7 @@ #include "../../dir.h" #include "../../abspath.h" #include "../../trace.h" +#include "config.h" static int initialized; static volatile long enabled; @@ -406,7 +407,11 @@ int fscache_enable(int enable) int result; if (!initialized) { + int fscache = git_env_bool("GIT_TEST_FSCACHE", -1); + /* allow the cache to be disabled entirely */ + if (fscache != -1) + core_fscache = fscache; if (!core_fscache) return 0; diff --git a/t/README b/t/README index adbbd9acf4ab27..f19468151410eb 100644 --- a/t/README +++ b/t/README @@ -479,6 +479,9 @@ GIT_TEST_NAME_HASH_VERSION=<int>, when set, causes 'git pack-objects' to assume '--name-hash-version=<n>'. +GIT_TEST_FSCACHE=<boolean> exercises the uncommon fscache code path +which adds a cache below mingw's lstat and dirent implementations. + Naming Tests ------------ From b5888fb8f46da0c3e85c1934559e886bdfdf89f2 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Tue, 25 Sep 2018 16:28:16 -0400 Subject: [PATCH 163/248] fscache: add fscache hit statistics Track fscache hits and misses for lstat and opendir requests. Reporting of statistics is done when the cache is disabled for the last time and freed and is only reported if GIT_TRACE_FSCACHE is set. Sample output is: 11:33:11.836428 compat/win32/fscache.c:433 fscache: lstat 3775, opendir 263, total requests/misses 4052/269 Signed-off-by: Ben Peart <benpeart@microsoft.com> --- compat/win32/fscache.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 13b38104732592..cf8ed5c63573a6 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -11,6 +11,10 @@ static int initialized; static volatile long enabled; static struct hashmap map; static CRITICAL_SECTION mutex; +static unsigned int lstat_requests; +static unsigned int opendir_requests; +static unsigned int fscache_requests; +static unsigned int fscache_misses; static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); /* @@ -270,6 +274,8 @@ static void fscache_clear(void) { hashmap_clear_and_free(&map, struct fsentry, ent); hashmap_init(&map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0); + lstat_requests = opendir_requests = 0; + fscache_misses = fscache_requests = 0; } /* @@ -316,6 +322,7 @@ static struct fsentry *fscache_get(struct fsentry *key) int dir_not_found; EnterCriticalSection(&mutex); + fscache_requests++; /* check if entry is in cache */ fse = fscache_get_wait(key); if (fse) { @@ -379,6 +386,7 @@ static struct fsentry *fscache_get(struct fsentry *key) } /* add directory listing to the cache */ + fscache_misses++; fscache_add(fse); /* lookup file entry if requested (fse already points to directory) */ @@ -416,6 +424,8 @@ int fscache_enable(int enable) return 0; InitializeCriticalSection(&mutex); + lstat_requests = opendir_requests = 0; + fscache_misses = fscache_requests = 0; hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, 0); initialized = 1; } @@ -432,6 +442,10 @@ int fscache_enable(int enable) opendir = dirent_opendir; lstat = mingw_lstat; EnterCriticalSection(&mutex); + trace_printf_key(&trace_fscache, "fscache: lstat %u, opendir %u, " + "total requests/misses %u/%u\n", + lstat_requests, opendir_requests, + fscache_requests, fscache_misses); fscache_clear(); LeaveCriticalSection(&mutex); } @@ -469,6 +483,7 @@ int fscache_lstat(const char *filename, struct stat *st) if (!fscache_enabled(filename)) return mingw_lstat(filename, st); + lstat_requests++; /* split filename into path + name */ len = strlen(filename); if (len && is_dir_sep(filename[len - 1])) @@ -550,6 +565,7 @@ DIR *fscache_opendir(const char *dirname) if (!fscache_enabled(dirname)) return dirent_opendir(dirname); + opendir_requests++; /* prepare name (strip trailing '/', replace '.') */ len = strlen(dirname); if ((len == 1 && dirname[0] == '.') || From 232a0c28273abc4f727e07b8bff8453269f906e1 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Thu, 4 Oct 2018 18:10:21 -0400 Subject: [PATCH 164/248] mem_pool: add GIT_TRACE_MEMPOOL support Add tracing around initializing and discarding mempools. In discard report on the amount of memory unused in the current block to help tune setting the initial_size. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- mem-pool.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/mem-pool.c b/mem-pool.c index 62441dcc71968f..0fab0a5ef26472 100644 --- a/mem-pool.c +++ b/mem-pool.c @@ -7,7 +7,9 @@ #include "git-compat-util.h" #include "mem-pool.h" #include "gettext.h" +#include "trace.h" +static struct trace_key trace_mem_pool = TRACE_KEY_INIT(MEMPOOL); #define BLOCK_GROWTH_SIZE (1024 * 1024 - sizeof(struct mp_block)) /* @@ -65,12 +67,20 @@ void mem_pool_init(struct mem_pool *pool, size_t initial_size) if (initial_size > 0) mem_pool_alloc_block(pool, initial_size, NULL); + + trace_printf_key(&trace_mem_pool, + "mem_pool (%p): init (%"PRIuMAX") initial size\n", + (void *)pool, (uintmax_t)initial_size); } void mem_pool_discard(struct mem_pool *pool, int invalidate_memory) { struct mp_block *block, *block_to_free; + trace_printf_key(&trace_mem_pool, + "mem_pool (%p): discard (%"PRIuMAX") unused\n", + (void *)pool, + (uintmax_t)(pool->mp_block->end - pool->mp_block->next_free)); block = pool->mp_block; while (block) { From 65322b5a460c99d844d9a04fb465ca1d82de9658 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Fri, 2 Nov 2018 11:19:10 -0400 Subject: [PATCH 165/248] fscache: fscache takes an initial size Update enable_fscache() to take an optional initial size parameter which is used to initialize the hashmap so that it can avoid having to rehash as additional entries are added. Add a separate disable_fscache() macro to make the code clearer and easier to read. Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- builtin/add.c | 2 +- builtin/checkout.c | 4 ++-- builtin/commit.c | 4 ++-- compat/win32/fscache.c | 8 ++++++-- compat/win32/fscache.h | 5 +++-- fetch-pack.c | 4 ++-- git-compat-util.h | 4 ++++ preload-index.c | 4 ++-- read-cache.c | 4 ++-- 9 files changed, 24 insertions(+), 15 deletions(-) diff --git a/builtin/add.c b/builtin/add.c index 7f5077e7fa17de..30a5d71d8e7898 100644 --- a/builtin/add.c +++ b/builtin/add.c @@ -491,7 +491,7 @@ int cmd_add(int argc, die_in_unpopulated_submodule(repo->index, prefix); die_path_inside_submodule(repo->index, &pathspec); - enable_fscache(1); + enable_fscache(0); /* We do not really re-read the index but update the up-to-date flags */ preload_index(repo->index, &pathspec, 0); diff --git a/builtin/checkout.c b/builtin/checkout.c index d0689ac6ab82ce..72347addd208b8 100644 --- a/builtin/checkout.c +++ b/builtin/checkout.c @@ -415,7 +415,7 @@ static int checkout_worktree(const struct checkout_opts *opts, if (pc_workers > 1) init_parallel_checkout(); - enable_fscache(1); + enable_fscache(the_repository->index->cache_nr); for (pos = 0; pos < the_repository->index->cache_nr; pos++) { struct cache_entry *ce = the_repository->index->cache[pos]; if (ce->ce_flags & CE_MATCHED) { @@ -441,7 +441,7 @@ static int checkout_worktree(const struct checkout_opts *opts, errs |= run_parallel_checkout(&state, pc_workers, pc_threshold, NULL, NULL); mem_pool_discard(&ce_mem_pool, should_validate_cache_entries()); - enable_fscache(0); + disable_fscache(); remove_marked_cache_entries(the_repository->index, 1); remove_scheduled_dirs(); errs |= finish_delayed_checkout(&state, opts->show_progress); diff --git a/builtin/commit.c b/builtin/commit.c index 34a563bb9e4b8f..64a8bfd14115e0 100644 --- a/builtin/commit.c +++ b/builtin/commit.c @@ -1618,7 +1618,7 @@ struct repository *repo UNUSED) PATHSPEC_PREFER_FULL, prefix, argv); - enable_fscache(1); + enable_fscache(0); if (status_format != STATUS_FORMAT_PORCELAIN && status_format != STATUS_FORMAT_PORCELAIN_V2) progress_flag = REFRESH_PROGRESS; @@ -1659,7 +1659,7 @@ struct repository *repo UNUSED) wt_status_print(&s); wt_status_collect_free_buffers(&s); - enable_fscache(0); + disable_fscache(); return 0; } diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index cf8ed5c63573a6..e9c10908d0e686 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -410,7 +410,7 @@ static struct fsentry *fscache_get(struct fsentry *key) * Enables or disables the cache. Note that the cache is read-only, changes to * the working directory are NOT reflected in the cache while enabled. */ -int fscache_enable(int enable) +int fscache_enable(int enable, size_t initial_size) { int result; @@ -426,7 +426,11 @@ int fscache_enable(int enable) InitializeCriticalSection(&mutex); lstat_requests = opendir_requests = 0; fscache_misses = fscache_requests = 0; - hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, 0); + /* + * avoid having to rehash by leaving room for the parent dirs. + * '4' was determined empirically by testing several repos + */ + hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, initial_size * 4); initialized = 1; } diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index 2f06f8df97dcd0..d49c9381114da6 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -1,8 +1,9 @@ #ifndef FSCACHE_H #define FSCACHE_H -int fscache_enable(int enable); -#define enable_fscache(x) fscache_enable(x) +int fscache_enable(int enable, size_t initial_size); +#define enable_fscache(initial_size) fscache_enable(1, initial_size) +#define disable_fscache() fscache_enable(0, 0) int fscache_enabled(const char *path); #define is_fscache_enabled(path) fscache_enabled(path) diff --git a/fetch-pack.c b/fetch-pack.c index 9dd7993544b6f5..f07ba095b8ef1a 100644 --- a/fetch-pack.c +++ b/fetch-pack.c @@ -768,7 +768,7 @@ static void mark_complete_and_common_ref(struct fetch_negotiator *negotiator, save_commit_buffer = 0; trace2_region_enter("fetch-pack", "parse_remote_refs_and_find_cutoff", NULL); - enable_fscache(1); + enable_fscache(0); for (ref = *refs; ref; ref = ref->next) { struct commit *commit; @@ -793,7 +793,7 @@ static void mark_complete_and_common_ref(struct fetch_negotiator *negotiator, if (!cutoff || cutoff < commit->date) cutoff = commit->date; } - enable_fscache(0); + disable_fscache(); trace2_region_leave("fetch-pack", "parse_remote_refs_and_find_cutoff", NULL); /* diff --git a/git-compat-util.h b/git-compat-util.h index e01a355e4ebb78..bdf888748f148f 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1065,6 +1065,10 @@ static inline int is_missing_file_error(int errno_) #define enable_fscache(x) /* noop */ #endif +#ifndef disable_fscache +#define disable_fscache() /* noop */ +#endif + #ifndef is_fscache_enabled #define is_fscache_enabled(path) (0) #endif diff --git a/preload-index.c b/preload-index.c index 61e8f3a1f6ec84..e466fef15bcd79 100644 --- a/preload-index.c +++ b/preload-index.c @@ -141,7 +141,7 @@ void preload_index(struct index_state *index, pthread_mutex_init(&pd.mutex, NULL); } - enable_fscache(1); + enable_fscache(index->cache_nr); for (i = 0; i < threads; i++) { struct thread_data *p = data+i; int err; @@ -178,7 +178,7 @@ void preload_index(struct index_state *index, trace2_data_intmax("index", NULL, "preload/sum_lstat", t2_sum_lstat); trace2_region_leave("index", "preload", NULL); - enable_fscache(0); + disable_fscache(); } int repo_read_index_preload(struct repository *repo, diff --git a/read-cache.c b/read-cache.c index 169ca23de86bbc..eccf7b606c09b0 100644 --- a/read-cache.c +++ b/read-cache.c @@ -1505,7 +1505,7 @@ int refresh_index(struct index_state *istate, unsigned int flags, typechange_fmt = in_porcelain ? "T\t%s\n" : "%s: needs update\n"; added_fmt = in_porcelain ? "A\t%s\n" : "%s: needs update\n"; unmerged_fmt = in_porcelain ? "U\t%s\n" : "%s: needs merge\n"; - enable_fscache(1); + enable_fscache(0); /* * Use the multi-threaded preload_index() to refresh most of the * cache entries quickly then in the single threaded loop below, @@ -1600,7 +1600,7 @@ int refresh_index(struct index_state *istate, unsigned int flags, display_progress(progress, istate->cache_nr); stop_progress(&progress); trace_performance_leave("refresh index"); - enable_fscache(0); + disable_fscache(); return has_errors; } From b1f344bf305538991d6ea33a6afbc12f1f12fdea Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Thu, 4 Oct 2018 15:38:08 -0400 Subject: [PATCH 166/248] fscache: update fscache to be thread specific instead of global The threading model for fscache has been to have a single, global cache. This puts requirements on it to be thread safe so that callers like preload-index can call it from multiple threads. This was implemented with a single mutex and completion events which introduces contention between the calling threads. Simplify the threading model by making fscache thread specific. This allows us to remove the global mutex and synchronization events entirely and instead associate a fscache with every thread that requests one. This works well with the current multi-threading which divides the cache entries into blocks with a separate thread processing each block. At the end of each worker thread, if there is a fscache on the primary thread, merge the cached results from the worker into the primary thread cache. This enables us to reuse the cache later especially when scanning for untracked files. In testing, this reduced the time spent in preload_index() by about 25% and also reduced the CPU utilization significantly. On a repo with ~200K files, it reduced overall status times by ~12%. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- compat/win32/fscache.c | 294 +++++++++++++++++++++++++---------------- compat/win32/fscache.h | 22 ++- git-compat-util.h | 12 ++ preload-index.c | 8 +- 4 files changed, 215 insertions(+), 121 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index e9c10908d0e686..f27a7e45e365f4 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -7,14 +7,24 @@ #include "../../trace.h" #include "config.h" -static int initialized; -static volatile long enabled; -static struct hashmap map; +static volatile long initialized; +static DWORD dwTlsIndex; static CRITICAL_SECTION mutex; -static unsigned int lstat_requests; -static unsigned int opendir_requests; -static unsigned int fscache_requests; -static unsigned int fscache_misses; + +/* + * Store one fscache per thread to avoid thread contention and locking. + * This is ok because multi-threaded access is 1) uncommon and 2) always + * splitting up the cache entries across multiple threads so there isn't + * any overlap between threads anyway. + */ +struct fscache { + volatile long enabled; + struct hashmap map; + unsigned int lstat_requests; + unsigned int opendir_requests; + unsigned int fscache_requests; + unsigned int fscache_misses; +}; static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); /* @@ -34,8 +44,6 @@ struct fsentry { union { /* Reference count of the directory listing. */ volatile long refcnt; - /* Handle to wait on the loading thread. */ - HANDLE hwait; struct { /* More stat members (only used for file entries). */ off64_t st_size; @@ -258,86 +266,63 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir, /* * Adds a directory listing to the cache. */ -static void fscache_add(struct fsentry *fse) +static void fscache_add(struct fscache *cache, struct fsentry *fse) { if (fse->list) fse = fse->list; for (; fse; fse = fse->next) - hashmap_add(&map, &fse->ent); + hashmap_add(&cache->map, &fse->ent); } /* * Clears the cache. */ -static void fscache_clear(void) +static void fscache_clear(struct fscache *cache) { - hashmap_clear_and_free(&map, struct fsentry, ent); - hashmap_init(&map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0); - lstat_requests = opendir_requests = 0; - fscache_misses = fscache_requests = 0; + hashmap_clear_and_free(&cache->map, struct fsentry, ent); + hashmap_init(&cache->map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0); + cache->lstat_requests = cache->opendir_requests = 0; + cache->fscache_misses = cache->fscache_requests = 0; } /* * Checks if the cache is enabled for the given path. */ -int fscache_enabled(const char *path) +static int do_fscache_enabled(struct fscache *cache, const char *path) { - return enabled > 0 && !is_absolute_path(path); + return cache->enabled > 0 && !is_absolute_path(path); } -/* - * Looks up a cache entry, waits if its being loaded by another thread. - * The mutex must be owned by the calling thread. - */ -static struct fsentry *fscache_get_wait(struct fsentry *key) +int fscache_enabled(const char *path) { - struct fsentry *fse = hashmap_get_entry(&map, key, ent, NULL); - - /* return if its a 'real' entry (future entries have refcnt == 0) */ - if (!fse || fse->list || fse->u.refcnt) - return fse; - - /* create an event and link our key to the future entry */ - key->u.hwait = CreateEvent(NULL, TRUE, FALSE, NULL); - key->next = fse->next; - fse->next = key; - - /* wait for the loading thread to signal us */ - LeaveCriticalSection(&mutex); - WaitForSingleObject(key->u.hwait, INFINITE); - CloseHandle(key->u.hwait); - EnterCriticalSection(&mutex); + struct fscache *cache = fscache_getcache(); - /* repeat cache lookup */ - return hashmap_get_entry(&map, key, ent, NULL); + return cache ? do_fscache_enabled(cache, path) : 0; } /* * Looks up or creates a cache entry for the specified key. */ -static struct fsentry *fscache_get(struct fsentry *key) +static struct fsentry *fscache_get(struct fscache *cache, struct fsentry *key) { - struct fsentry *fse, *future, *waiter; + struct fsentry *fse; int dir_not_found; - EnterCriticalSection(&mutex); - fscache_requests++; + cache->fscache_requests++; /* check if entry is in cache */ - fse = fscache_get_wait(key); + fse = hashmap_get_entry(&cache->map, key, ent, NULL); if (fse) { if (fse->st_mode) fsentry_addref(fse); else fse = NULL; /* non-existing directory */ - LeaveCriticalSection(&mutex); return fse; } /* if looking for a file, check if directory listing is in cache */ if (!fse && key->list) { - fse = fscache_get_wait(key->list); + fse = hashmap_get_entry(&cache->map, key->list, ent, NULL); if (fse) { - LeaveCriticalSection(&mutex); /* * dir entry without file entry, or dir does not * exist -> file doesn't exist @@ -347,25 +332,8 @@ static struct fsentry *fscache_get(struct fsentry *key) } } - /* add future entry to indicate that we're loading it */ - future = key->list ? key->list : key; - future->next = NULL; - future->u.refcnt = 0; - hashmap_add(&map, &future->ent); - - /* create the directory listing (outside mutex!) */ - LeaveCriticalSection(&mutex); - fse = fsentry_create_list(future, &dir_not_found); - EnterCriticalSection(&mutex); - - /* remove future entry and signal waiting threads */ - hashmap_remove(&map, &future->ent, NULL); - waiter = future->next; - while (waiter) { - HANDLE h = waiter->u.hwait; - waiter = waiter->next; - SetEvent(h); - } + /* create the directory listing */ + fse = fsentry_create_list(key->list ? key->list : key, &dir_not_found); /* leave on error (errno set by fsentry_create_list) */ if (!fse) { @@ -379,19 +347,18 @@ static struct fsentry *fscache_get(struct fsentry *key) key->list->dirent.d_name, key->list->len); fse->st_mode = 0; - hashmap_add(&map, &fse->ent); + hashmap_add(&cache->map, &fse->ent); } - LeaveCriticalSection(&mutex); return NULL; } /* add directory listing to the cache */ - fscache_misses++; - fscache_add(fse); + cache->fscache_misses++; + fscache_add(cache, fse); /* lookup file entry if requested (fse already points to directory) */ if (key->list) - fse = hashmap_get_entry(&map, key, ent, NULL); + fse = hashmap_get_entry(&cache->map, key, ent, NULL); if (fse && !fse->st_mode) fse = NULL; /* non-existing directory */ @@ -402,59 +369,104 @@ static struct fsentry *fscache_get(struct fsentry *key) else errno = ENOENT; - LeaveCriticalSection(&mutex); return fse; } /* - * Enables or disables the cache. Note that the cache is read-only, changes to + * Enables the cache. Note that the cache is read-only, changes to * the working directory are NOT reflected in the cache while enabled. */ -int fscache_enable(int enable, size_t initial_size) +int fscache_enable(size_t initial_size) { - int result; + int fscache; + struct fscache *cache; + int result = 0; + + /* allow the cache to be disabled entirely */ + fscache = git_env_bool("GIT_TEST_FSCACHE", -1); + if (fscache != -1) + core_fscache = fscache; + if (!core_fscache) + return 0; + /* + * refcount the global fscache initialization so that the + * opendir and lstat function pointers are redirected if + * any threads are using the fscache. + */ if (!initialized) { - int fscache = git_env_bool("GIT_TEST_FSCACHE", -1); - - /* allow the cache to be disabled entirely */ - if (fscache != -1) - core_fscache = fscache; - if (!core_fscache) - return 0; - InitializeCriticalSection(&mutex); - lstat_requests = opendir_requests = 0; - fscache_misses = fscache_requests = 0; + if (!dwTlsIndex) { + dwTlsIndex = TlsAlloc(); + if (dwTlsIndex == TLS_OUT_OF_INDEXES) { + LeaveCriticalSection(&mutex); + return 0; + } + } + + /* redirect opendir and lstat to the fscache implementations */ + opendir = fscache_opendir; + lstat = fscache_lstat; + } + InterlockedIncrement(&initialized); + + /* refcount the thread specific initialization */ + cache = fscache_getcache(); + if (cache) { + InterlockedIncrement(&cache->enabled); + } else { + cache = (struct fscache *)xcalloc(1, sizeof(*cache)); + cache->enabled = 1; /* * avoid having to rehash by leaving room for the parent dirs. * '4' was determined empirically by testing several repos */ - hashmap_init(&map, (hashmap_cmp_fn) fsentry_cmp, NULL, initial_size * 4); - initialized = 1; + hashmap_init(&cache->map, (hashmap_cmp_fn)fsentry_cmp, NULL, initial_size * 4); + if (!TlsSetValue(dwTlsIndex, cache)) + BUG("TlsSetValue error"); } - result = enable ? InterlockedIncrement(&enabled) - : InterlockedDecrement(&enabled); + trace_printf_key(&trace_fscache, "fscache: enable\n"); + return result; +} - if (enable && result == 1) { - /* redirect opendir and lstat to the fscache implementations */ - opendir = fscache_opendir; - lstat = fscache_lstat; - } else if (!enable && !result) { +/* + * Disables the cache. + */ +void fscache_disable(void) +{ + struct fscache *cache; + + if (!core_fscache) + return; + + /* update the thread specific fscache initialization */ + cache = fscache_getcache(); + if (!cache) + BUG("fscache_disable() called on a thread where fscache has not been initialized"); + if (!cache->enabled) + BUG("fscache_disable() called on an fscache that is already disabled"); + InterlockedDecrement(&cache->enabled); + if (!cache->enabled) { + TlsSetValue(dwTlsIndex, NULL); + trace_printf_key(&trace_fscache, "fscache_disable: lstat %u, opendir %u, " + "total requests/misses %u/%u\n", + cache->lstat_requests, cache->opendir_requests, + cache->fscache_requests, cache->fscache_misses); + fscache_clear(cache); + free(cache); + } + + /* update the global fscache initialization */ + InterlockedDecrement(&initialized); + if (!initialized) { /* reset opendir and lstat to the original implementations */ opendir = dirent_opendir; lstat = mingw_lstat; - EnterCriticalSection(&mutex); - trace_printf_key(&trace_fscache, "fscache: lstat %u, opendir %u, " - "total requests/misses %u/%u\n", - lstat_requests, opendir_requests, - fscache_requests, fscache_misses); - fscache_clear(); - LeaveCriticalSection(&mutex); } - trace_printf_key(&trace_fscache, "fscache: enable(%d)\n", enable); - return result; + + trace_printf_key(&trace_fscache, "fscache: disable\n"); + return; } /* @@ -462,10 +474,10 @@ int fscache_enable(int enable, size_t initial_size) */ void fscache_flush(void) { - if (enabled) { - EnterCriticalSection(&mutex); - fscache_clear(); - LeaveCriticalSection(&mutex); + struct fscache *cache = fscache_getcache(); + + if (cache && cache->enabled) { + fscache_clear(cache); } } @@ -483,11 +495,12 @@ int fscache_lstat(const char *filename, struct stat *st) struct heap_fsentry key[2]; #pragma GCC diagnostic pop struct fsentry *fse; + struct fscache *cache = fscache_getcache(); - if (!fscache_enabled(filename)) + if (!cache || !do_fscache_enabled(cache, filename)) return mingw_lstat(filename, st); - lstat_requests++; + cache->lstat_requests++; /* split filename into path + name */ len = strlen(filename); if (len && is_dir_sep(filename[len - 1])) @@ -500,7 +513,7 @@ int fscache_lstat(const char *filename, struct stat *st) /* lookup entry for path + name in cache */ fsentry_init(&key[0].u.ent, NULL, filename, dirlen); fsentry_init(&key[1].u.ent, &key[0].u.ent, filename + base, len - base); - fse = fscache_get(&key[1].u.ent); + fse = fscache_get(cache, &key[1].u.ent); if (!fse) { errno = ENOENT; return -1; @@ -565,11 +578,12 @@ DIR *fscache_opendir(const char *dirname) struct fsentry *list; fscache_DIR *dir; int len; + struct fscache *cache = fscache_getcache(); - if (!fscache_enabled(dirname)) + if (!cache || !do_fscache_enabled(cache, dirname)) return dirent_opendir(dirname); - opendir_requests++; + cache->opendir_requests++; /* prepare name (strip trailing '/', replace '.') */ len = strlen(dirname); if ((len == 1 && dirname[0] == '.') || @@ -578,7 +592,7 @@ DIR *fscache_opendir(const char *dirname) /* get directory listing from cache */ fsentry_init(&key.u.ent, NULL, dirname, len); - list = fscache_get(&key.u.ent); + list = fscache_get(cache, &key.u.ent); if (!list) return NULL; @@ -589,3 +603,53 @@ DIR *fscache_opendir(const char *dirname) dir->pfsentry = list; return (DIR*) dir; } + +struct fscache *fscache_getcache(void) +{ + return (struct fscache *)TlsGetValue(dwTlsIndex); +} + +void fscache_merge(struct fscache *dest) +{ + struct hashmap_iter iter; + struct hashmap_entry *e; + struct fscache *cache = fscache_getcache(); + + /* + * Only do the merge if fscache was enabled and we have a dest + * cache to merge into. + */ + if (!dest) { + fscache_enable(0); + return; + } + if (!cache) + BUG("fscache_merge() called on a thread where fscache has not been initialized"); + + TlsSetValue(dwTlsIndex, NULL); + trace_printf_key(&trace_fscache, "fscache_merge: lstat %u, opendir %u, " + "total requests/misses %u/%u\n", + cache->lstat_requests, cache->opendir_requests, + cache->fscache_requests, cache->fscache_misses); + + /* + * This is only safe because the primary thread we're merging into + * isn't being used so the critical section only needs to prevent + * the the child threads from stomping on each other. + */ + EnterCriticalSection(&mutex); + + hashmap_iter_init(&cache->map, &iter); + while ((e = hashmap_iter_next(&iter))) + hashmap_add(&dest->map, e); + + dest->lstat_requests += cache->lstat_requests; + dest->opendir_requests += cache->opendir_requests; + dest->fscache_requests += cache->fscache_requests; + dest->fscache_misses += cache->fscache_misses; + LeaveCriticalSection(&mutex); + + free(cache); + + InterlockedDecrement(&initialized); +} diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index d49c9381114da6..2eb8bf3f5cfee8 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -1,9 +1,16 @@ #ifndef FSCACHE_H #define FSCACHE_H -int fscache_enable(int enable, size_t initial_size); -#define enable_fscache(initial_size) fscache_enable(1, initial_size) -#define disable_fscache() fscache_enable(0, 0) +/* + * The fscache is thread specific. enable_fscache() must be called + * for each thread where caching is desired. + */ + +int fscache_enable(size_t initial_size); +#define enable_fscache(initial_size) fscache_enable(initial_size) + +void fscache_disable(void); +#define disable_fscache() fscache_disable() int fscache_enabled(const char *path); #define is_fscache_enabled(path) fscache_enabled(path) @@ -14,4 +21,13 @@ void fscache_flush(void); DIR *fscache_opendir(const char *dir); int fscache_lstat(const char *file_name, struct stat *buf); +/* opaque fscache structure */ +struct fscache; + +struct fscache *fscache_getcache(void); +#define getcache_fscache() fscache_getcache() + +void fscache_merge(struct fscache *dest); +#define merge_fscache(dest) fscache_merge(dest) + #endif diff --git a/git-compat-util.h b/git-compat-util.h index bdf888748f148f..92a4148845e1a5 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -1061,6 +1061,10 @@ static inline int is_missing_file_error(int errno_) * data or even file content without the need to synchronize with the file * system. */ + + /* opaque fscache structure */ +struct fscache; + #ifndef enable_fscache #define enable_fscache(x) /* noop */ #endif @@ -1077,6 +1081,14 @@ static inline int is_missing_file_error(int errno_) #define flush_fscache() /* noop */ #endif +#ifndef getcache_fscache +#define getcache_fscache() (NULL) /* noop */ +#endif + +#ifndef merge_fscache +#define merge_fscache(dest) /* noop */ +#endif + int cmd_main(int, const char **); /* diff --git a/preload-index.c b/preload-index.c index e466fef15bcd79..ac0310008754a3 100644 --- a/preload-index.c +++ b/preload-index.c @@ -20,6 +20,8 @@ #include "trace2.h" #include "config.h" +static struct fscache *fscache; + /* * Mostly randomly chosen maximum thread counts: we * cap the parallelism to 20 threads, and we want @@ -57,6 +59,7 @@ static void *preload_thread(void *_data) nr = index->cache_nr - p->offset; last_nr = nr; + enable_fscache(nr); do { struct cache_entry *ce = *cep++; struct stat st; @@ -100,6 +103,7 @@ static void *preload_thread(void *_data) pthread_mutex_unlock(&pd->mutex); } cache_def_clear(&cache); + merge_fscache(fscache); return NULL; } @@ -118,6 +122,7 @@ void preload_index(struct index_state *index, if (!HAVE_THREADS || !core_preload_index) return; + fscache = getcache_fscache(); threads = index->cache_nr / THREAD_COST; if ((index->cache_nr > 1) && (threads < 2) && git_env_bool("GIT_TEST_PRELOAD_INDEX", 0)) threads = 2; @@ -141,7 +146,6 @@ void preload_index(struct index_state *index, pthread_mutex_init(&pd.mutex, NULL); } - enable_fscache(index->cache_nr); for (i = 0; i < threads; i++) { struct thread_data *p = data+i; int err; @@ -177,8 +181,6 @@ void preload_index(struct index_state *index, trace2_data_intmax("index", NULL, "preload/sum_lstat", t2_sum_lstat); trace2_region_leave("index", "preload", NULL); - - disable_fscache(); } int repo_read_index_preload(struct repository *repo, From b8237147b696b49d90246f278e8dfd9620efb3b7 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Fri, 2 Nov 2018 11:19:10 -0400 Subject: [PATCH 167/248] fscache: teach fscache to use mempool Now that the fscache is single threaded, take advantage of the mem_pool as the allocator to significantly reduce the cost of allocations and frees. With the reduced cost of free, in future patches, we can start freeing the fscache at the end of commands instead of just leaking it. Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 45 ++++++++++++++++++++++-------------------- 1 file changed, 24 insertions(+), 21 deletions(-) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index f27a7e45e365f4..2d967bd62f129f 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -6,6 +6,7 @@ #include "../../abspath.h" #include "../../trace.h" #include "config.h" +#include "../../mem-pool.h" static volatile long initialized; static DWORD dwTlsIndex; @@ -20,6 +21,7 @@ static CRITICAL_SECTION mutex; struct fscache { volatile long enabled; struct hashmap map; + struct mem_pool mem_pool; unsigned int lstat_requests; unsigned int opendir_requests; unsigned int fscache_requests; @@ -129,11 +131,12 @@ static void fsentry_init(struct fsentry *fse, struct fsentry *list, /* * Allocate an fsentry structure on the heap. */ -static struct fsentry *fsentry_alloc(struct fsentry *list, const char *name, +static struct fsentry *fsentry_alloc(struct fscache *cache, struct fsentry *list, const char *name, size_t len) { /* overallocate fsentry and copy the name to the end */ - struct fsentry *fse = xmalloc(sizeof(struct fsentry) + len + 1); + struct fsentry *fse = + mem_pool_alloc(&cache->mem_pool, sizeof(*fse) + len + 1); /* init the rest of the structure */ fsentry_init(fse, list, name, len); fse->next = NULL; @@ -153,27 +156,21 @@ inline static void fsentry_addref(struct fsentry *fse) } /* - * Release the reference to an fsentry, frees the memory if its the last ref. + * Release the reference to an fsentry. */ static void fsentry_release(struct fsentry *fse) { if (fse->list) fse = fse->list; - if (InterlockedDecrement(&(fse->u.refcnt))) - return; - - while (fse) { - struct fsentry *next = fse->next; - free(fse); - fse = next; - } + InterlockedDecrement(&(fse->u.refcnt)); } /* * Allocate and initialize an fsentry from a WIN32_FIND_DATA structure. */ -static struct fsentry *fseentry_create_entry(struct fsentry *list, +static struct fsentry *fseentry_create_entry(struct fscache *cache, + struct fsentry *list, const WIN32_FIND_DATAW *fdata) { char buf[MAX_PATH * 3]; @@ -181,7 +178,7 @@ static struct fsentry *fseentry_create_entry(struct fsentry *list, struct fsentry *fse; len = xwcstoutf(buf, fdata->cFileName, ARRAY_SIZE(buf)); - fse = fsentry_alloc(list, buf, len); + fse = fsentry_alloc(cache, list, buf, len); fse->st_mode = file_attr_to_st_mode(fdata->dwFileAttributes); fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG; @@ -199,7 +196,7 @@ static struct fsentry *fseentry_create_entry(struct fsentry *list, * Dir should not contain trailing '/'. Use an empty string for the current * directory (not "."!). */ -static struct fsentry *fsentry_create_list(const struct fsentry *dir, +static struct fsentry *fsentry_create_list(struct fscache *cache, const struct fsentry *dir, int *dir_not_found) { wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ @@ -238,14 +235,14 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir, } /* allocate object to hold directory listing */ - list = fsentry_alloc(NULL, dir->dirent.d_name, dir->len); + list = fsentry_alloc(cache, NULL, dir->dirent.d_name, dir->len); list->st_mode = S_IFDIR; list->dirent.d_type = DT_DIR; /* walk directory and build linked list of fsentry structures */ phead = &list->next; do { - *phead = fseentry_create_entry(list, &fdata); + *phead = fseentry_create_entry(cache, list, &fdata); phead = &(*phead)->next; } while (FindNextFileW(h, &fdata)); @@ -257,7 +254,7 @@ static struct fsentry *fsentry_create_list(const struct fsentry *dir, if (err == ERROR_NO_MORE_FILES) return list; - /* otherwise free the list and return error */ + /* otherwise release the list and return error */ fsentry_release(list); errno = err_win_to_posix(err); return NULL; @@ -280,7 +277,9 @@ static void fscache_add(struct fscache *cache, struct fsentry *fse) */ static void fscache_clear(struct fscache *cache) { - hashmap_clear_and_free(&cache->map, struct fsentry, ent); + mem_pool_discard(&cache->mem_pool, 0); + mem_pool_init(&cache->mem_pool, 0); + hashmap_clear(&cache->map); hashmap_init(&cache->map, (hashmap_cmp_fn)fsentry_cmp, NULL, 0); cache->lstat_requests = cache->opendir_requests = 0; cache->fscache_misses = cache->fscache_requests = 0; @@ -333,7 +332,7 @@ static struct fsentry *fscache_get(struct fscache *cache, struct fsentry *key) } /* create the directory listing */ - fse = fsentry_create_list(key->list ? key->list : key, &dir_not_found); + fse = fsentry_create_list(cache, key->list ? key->list : key, &dir_not_found); /* leave on error (errno set by fsentry_create_list) */ if (!fse) { @@ -343,7 +342,7 @@ static struct fsentry *fscache_get(struct fscache *cache, struct fsentry *key) * empty, which for all practical matters is the same * thing as far as fscache is concerned). */ - fse = fsentry_alloc(key->list->list, + fse = fsentry_alloc(cache, key->list->list, key->list->dirent.d_name, key->list->len); fse->st_mode = 0; @@ -422,6 +421,7 @@ int fscache_enable(size_t initial_size) * '4' was determined empirically by testing several repos */ hashmap_init(&cache->map, (hashmap_cmp_fn)fsentry_cmp, NULL, initial_size * 4); + mem_pool_init(&cache->mem_pool, 0); if (!TlsSetValue(dwTlsIndex, cache)) BUG("TlsSetValue error"); } @@ -453,7 +453,8 @@ void fscache_disable(void) "total requests/misses %u/%u\n", cache->lstat_requests, cache->opendir_requests, cache->fscache_requests, cache->fscache_misses); - fscache_clear(cache); + mem_pool_discard(&cache->mem_pool, 0); + hashmap_clear(&cache->map); free(cache); } @@ -643,6 +644,8 @@ void fscache_merge(struct fscache *dest) while ((e = hashmap_iter_next(&iter))) hashmap_add(&dest->map, e); + mem_pool_combine(&dest->mem_pool, &cache->mem_pool); + dest->lstat_requests += cache->lstat_requests; dest->opendir_requests += cache->opendir_requests; dest->fscache_requests += cache->fscache_requests; From 679965624b36908d15832bd2ca746997ff470b09 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Fri, 16 Nov 2018 10:59:18 -0500 Subject: [PATCH 168/248] fscache: make fscache_enable() thread safe The recent change to make fscache thread specific relied on fscache_enable() being called first from the primary thread before being called in parallel from worker threads. Make that more robust and protect it with a critical section to avoid any issues. Helped-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Ben Peart <benpeart@microsoft.com> --- compat/mingw.c | 4 ++++ compat/win32/fscache.c | 23 +++++++++++++---------- compat/win32/fscache.h | 2 ++ 3 files changed, 19 insertions(+), 10 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 436be66f6c5fc0..8d8029d9000403 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -14,6 +14,7 @@ #include "symlinks.h" #include "trace2.h" #include "win32.h" +#include "win32/fscache.h" #include "win32/lazyload.h" #include "wrapper.h" #include "write-or-die.h" @@ -3735,6 +3736,9 @@ int wmain(int argc, const wchar_t **wargv) /* initialize critical section for waitpid pinfo_t list */ InitializeCriticalSection(&pinfo_cs); + /* initialize critical section for fscache */ + InitializeCriticalSection(&fscache_cs); + /* set up default file mode and file modes for stdin/out/err */ _fmode = _O_BINARY; _setmode(_fileno(stdin), _O_BINARY); diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 2d967bd62f129f..7234318520b8e8 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -10,7 +10,7 @@ static volatile long initialized; static DWORD dwTlsIndex; -static CRITICAL_SECTION mutex; +CRITICAL_SECTION fscache_cs; /* * Store one fscache per thread to avoid thread contention and locking. @@ -393,12 +393,12 @@ int fscache_enable(size_t initial_size) * opendir and lstat function pointers are redirected if * any threads are using the fscache. */ + EnterCriticalSection(&fscache_cs); if (!initialized) { - InitializeCriticalSection(&mutex); if (!dwTlsIndex) { dwTlsIndex = TlsAlloc(); if (dwTlsIndex == TLS_OUT_OF_INDEXES) { - LeaveCriticalSection(&mutex); + LeaveCriticalSection(&fscache_cs); return 0; } } @@ -407,12 +407,13 @@ int fscache_enable(size_t initial_size) opendir = fscache_opendir; lstat = fscache_lstat; } - InterlockedIncrement(&initialized); + initialized++; + LeaveCriticalSection(&fscache_cs); /* refcount the thread specific initialization */ cache = fscache_getcache(); if (cache) { - InterlockedIncrement(&cache->enabled); + cache->enabled++; } else { cache = (struct fscache *)xcalloc(1, sizeof(*cache)); cache->enabled = 1; @@ -446,7 +447,7 @@ void fscache_disable(void) BUG("fscache_disable() called on a thread where fscache has not been initialized"); if (!cache->enabled) BUG("fscache_disable() called on an fscache that is already disabled"); - InterlockedDecrement(&cache->enabled); + cache->enabled--; if (!cache->enabled) { TlsSetValue(dwTlsIndex, NULL); trace_printf_key(&trace_fscache, "fscache_disable: lstat %u, opendir %u, " @@ -459,12 +460,14 @@ void fscache_disable(void) } /* update the global fscache initialization */ - InterlockedDecrement(&initialized); + EnterCriticalSection(&fscache_cs); + initialized--; if (!initialized) { /* reset opendir and lstat to the original implementations */ opendir = dirent_opendir; lstat = mingw_lstat; } + LeaveCriticalSection(&fscache_cs); trace_printf_key(&trace_fscache, "fscache: disable\n"); return; @@ -638,7 +641,7 @@ void fscache_merge(struct fscache *dest) * isn't being used so the critical section only needs to prevent * the the child threads from stomping on each other. */ - EnterCriticalSection(&mutex); + EnterCriticalSection(&fscache_cs); hashmap_iter_init(&cache->map, &iter); while ((e = hashmap_iter_next(&iter))) @@ -650,9 +653,9 @@ void fscache_merge(struct fscache *dest) dest->opendir_requests += cache->opendir_requests; dest->fscache_requests += cache->fscache_requests; dest->fscache_misses += cache->fscache_misses; - LeaveCriticalSection(&mutex); + initialized--; + LeaveCriticalSection(&fscache_cs); free(cache); - InterlockedDecrement(&initialized); } diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index 2eb8bf3f5cfee8..042b247a542554 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -6,6 +6,8 @@ * for each thread where caching is desired. */ +extern CRITICAL_SECTION fscache_cs; + int fscache_enable(size_t initial_size); #define enable_fscache(initial_size) fscache_enable(initial_size) From 48a0f62f825c8b56b3b89eac25e49a5a949ff071 Mon Sep 17 00:00:00 2001 From: Derrick Stolee <dstolee@microsoft.com> Date: Wed, 12 Jun 2019 00:58:49 +0000 Subject: [PATCH 169/248] unpack-trees: enable fscache for sparse-checkout When updating the skip-worktree bits in the index to align with new values in a sparse-checkout file, Git scans the entire working directory with lstat() calls. In a sparse-checkout, many of these lstat() calls are for paths that do not exist. Enable the fscache feature during this scan. Since enable_fscache() calls nest, the disable_fscache() method decrements a counter and would only clear the cache if that counter reaches zero. In a local test of a repo with ~2.2 million paths, updating the index with git read-tree -m -u HEAD with a sparse-checkout file containing only /.gitattributes improved from 2-3 minutes to ~6 seconds. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- unpack-trees.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/unpack-trees.c b/unpack-trees.c index f38c761ab987a6..450dbdf7c1bd6c 100644 --- a/unpack-trees.c +++ b/unpack-trees.c @@ -1823,7 +1823,9 @@ static void mark_new_skip_worktree(struct pattern_list *pl, * 2. Widen worktree according to sparse-checkout file. * Matched entries will have skip_wt_flag cleared (i.e. "in") */ + enable_fscache(istate->cache_nr); clear_ce_flags(istate, select_flag, skip_wt_flag, pl, show_progress); + disable_fscache(); } static void populate_from_existing_patterns(struct unpack_trees_options *o, From 3e0848c863091bb8dc50d6482aec33f14bca9e99 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 11 Dec 2018 12:59:29 +0100 Subject: [PATCH 170/248] fscache: remember the reparse tag for each entry We will use this in the next commit to implement an FSCache-aware version of is_mount_point(). Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index ca16b35302a3ce..c05b931455945c 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -46,6 +46,7 @@ static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); struct fsentry { struct hashmap_entry ent; mode_t st_mode; + ULONG reparse_tag; /* Pointer to the directory listing, or NULL for the listing itself. */ struct fsentry *list; /* Pointer to the next file entry of the list. */ @@ -202,6 +203,10 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, fse = fsentry_alloc(cache, list, buf, len); + fse->reparse_tag = + fdata->FileAttributes & FILE_ATTRIBUTE_REPARSE_POINT ? + fdata->EaSize : 0; + fse->st_mode = file_attr_to_st_mode(fdata->FileAttributes); fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG; fse->u.s.st_size = fdata->EndOfFile.LowPart | From 97a16c9111c4c275e4948940012b14814567aac4 Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Thu, 1 Nov 2018 11:40:51 -0400 Subject: [PATCH 171/248] status: disable and free fscache at the end of the status command At the end of the status command, disable and free the fscache so that we don't leak the memory and so that we can dump the fscache statistics. Signed-off-by: Ben Peart <benpeart@microsoft.com> --- builtin/commit.c | 1 + 1 file changed, 1 insertion(+) diff --git a/builtin/commit.c b/builtin/commit.c index 39c595276fdc1a..34a563bb9e4b8f 100644 --- a/builtin/commit.c +++ b/builtin/commit.c @@ -1659,6 +1659,7 @@ struct repository *repo UNUSED) wt_status_print(&s); wt_status_collect_free_buffers(&s); + enable_fscache(0); return 0; } From 8ee062d022657e764783a902aec63446e3672dba Mon Sep 17 00:00:00 2001 From: Ben Peart <benpeart@microsoft.com> Date: Thu, 15 Nov 2018 14:15:40 -0500 Subject: [PATCH 172/248] fscache: teach fscache to use NtQueryDirectoryFile Using FindFirstFileExW() requires the OS to allocate a 64K buffer for each directory and then free it when we call FindClose(). Update fscache to call the underlying kernel API NtQueryDirectoryFile so that we can do the buffer management ourselves. That allows us to allocate a single buffer for the lifetime of the cache and reuse it for each directory. This change improves performance of 'git status' by 18% in a repo with ~200K files and 30k folders. Documentation for NtQueryDirectoryFile can be found at: https://docs.microsoft.com/en-us/windows-hardware/drivers/ddi/content/ntifs/nf-ntifs-ntquerydirectoryfile https://docs.microsoft.com/en-us/windows/desktop/FileIO/file-attribute-constants https://docs.microsoft.com/en-us/windows/desktop/fileio/reparse-point-tags To determine if the specified directory is a symbolic link, inspect the FileAttributes member to see if the FILE_ATTRIBUTE_REPARSE_POINT flag is set. If so, EaSize will contain the reparse tag (this is a so far undocumented feature, but confirmed by the NTFS developers). To determine if the reparse point is a symbolic link (and not some other form of reparse point), test whether the tag value equals the value IO_REPARSE_TAG_SYMLINK. The NtQueryDirectoryFile() call works best (and on Windows 8.1 and earlier, it works *only*) with buffer sizes up to 64kB. Which is 32k wide characters, so let's use that as our buffer size. Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/win32/fscache.c | 123 ++++++++++++++++++++++++++++---------- compat/win32/ntifs.h | 131 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 224 insertions(+), 30 deletions(-) create mode 100644 compat/win32/ntifs.h diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 7234318520b8e8..ca16b35302a3ce 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -7,6 +7,7 @@ #include "../../trace.h" #include "config.h" #include "../../mem-pool.h" +#include "ntifs.h" static volatile long initialized; static DWORD dwTlsIndex; @@ -26,6 +27,13 @@ struct fscache { unsigned int opendir_requests; unsigned int fscache_requests; unsigned int fscache_misses; + /* + * 32k wide characters translates to 64kB, which is the maximum that + * Windows 8.1 and earlier can handle. On network drives, not only + * the client's Windows version matters, but also the server's, + * therefore we need to keep this to 64kB. + */ + WCHAR buffer[32 * 1024]; }; static struct trace_key trace_fscache = TRACE_KEY_INIT(FSCACHE); @@ -166,27 +174,44 @@ static void fsentry_release(struct fsentry *fse) InterlockedDecrement(&(fse->u.refcnt)); } +static int xwcstoutfn(char *utf, int utflen, const wchar_t *wcs, int wcslen) +{ + if (!wcs || !utf || utflen < 1) { + errno = EINVAL; + return -1; + } + utflen = WideCharToMultiByte(CP_UTF8, 0, wcs, wcslen, utf, utflen, NULL, NULL); + if (utflen) + return utflen; + errno = ERANGE; + return -1; +} + /* - * Allocate and initialize an fsentry from a WIN32_FIND_DATA structure. + * Allocate and initialize an fsentry from a FILE_FULL_DIR_INFORMATION structure. */ static struct fsentry *fseentry_create_entry(struct fscache *cache, struct fsentry *list, - const WIN32_FIND_DATAW *fdata) + PFILE_FULL_DIR_INFORMATION fdata) { char buf[MAX_PATH * 3]; int len; struct fsentry *fse; - len = xwcstoutf(buf, fdata->cFileName, ARRAY_SIZE(buf)); + + len = xwcstoutfn(buf, ARRAY_SIZE(buf), fdata->FileName, fdata->FileNameLength / sizeof(wchar_t)); fse = fsentry_alloc(cache, list, buf, len); - fse->st_mode = file_attr_to_st_mode(fdata->dwFileAttributes); + fse->st_mode = file_attr_to_st_mode(fdata->FileAttributes); fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG; - fse->u.s.st_size = (((off64_t) (fdata->nFileSizeHigh)) << 32) - | fdata->nFileSizeLow; - filetime_to_timespec(&(fdata->ftLastAccessTime), &(fse->u.s.st_atim)); - filetime_to_timespec(&(fdata->ftLastWriteTime), &(fse->u.s.st_mtim)); - filetime_to_timespec(&(fdata->ftCreationTime), &(fse->u.s.st_ctim)); + fse->u.s.st_size = fdata->EndOfFile.LowPart | + (((off_t)fdata->EndOfFile.HighPart) << 32); + filetime_to_timespec((FILETIME *)&(fdata->LastAccessTime), + &(fse->u.s.st_atim)); + filetime_to_timespec((FILETIME *)&(fdata->LastWriteTime), + &(fse->u.s.st_mtim)); + filetime_to_timespec((FILETIME *)&(fdata->CreationTime), + &(fse->u.s.st_ctim)); return fse; } @@ -199,8 +224,10 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, static struct fsentry *fsentry_create_list(struct fscache *cache, const struct fsentry *dir, int *dir_not_found) { - wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ - WIN32_FIND_DATAW fdata; + wchar_t pattern[MAX_PATH]; + NTSTATUS status; + IO_STATUS_BLOCK iosb; + PFILE_FULL_DIR_INFORMATION di; HANDLE h; int wlen; struct fsentry *list, **phead; @@ -216,15 +243,18 @@ static struct fsentry *fsentry_create_list(struct fscache *cache, const struct f return NULL; } - /* append optional '/' and wildcard '*' */ - if (wlen) - pattern[wlen++] = '/'; - pattern[wlen++] = '*'; - pattern[wlen] = 0; + /* handle CWD */ + if (!wlen) { + wlen = GetCurrentDirectoryW(ARRAY_SIZE(pattern), pattern); + if (!wlen || wlen >= (ssize_t)ARRAY_SIZE(pattern)) { + errno = wlen ? ENAMETOOLONG : err_win_to_posix(GetLastError()); + return NULL; + } + } - /* open find handle */ - h = FindFirstFileExW(pattern, FindExInfoBasic, &fdata, FindExSearchNameMatch, - NULL, FIND_FIRST_EX_LARGE_FETCH); + h = CreateFileW(pattern, FILE_LIST_DIRECTORY, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, + NULL, OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); if (h == INVALID_HANDLE_VALUE) { err = GetLastError(); *dir_not_found = 1; /* or empty directory */ @@ -241,22 +271,55 @@ static struct fsentry *fsentry_create_list(struct fscache *cache, const struct f /* walk directory and build linked list of fsentry structures */ phead = &list->next; - do { - *phead = fseentry_create_entry(cache, list, &fdata); + status = NtQueryDirectoryFile(h, NULL, 0, 0, &iosb, cache->buffer, + sizeof(cache->buffer), FileFullDirectoryInformation, FALSE, NULL, FALSE); + if (!NT_SUCCESS(status)) { + /* + * NtQueryDirectoryFile returns STATUS_INVALID_PARAMETER when + * asked to enumerate an invalid directory (ie it is a file + * instead of a directory). Verify that is the actual cause + * of the error. + */ + if (status == (NTSTATUS)STATUS_INVALID_PARAMETER) { + DWORD attributes = GetFileAttributesW(pattern); + if (!(attributes & FILE_ATTRIBUTE_DIRECTORY)) + status = ERROR_DIRECTORY; + } + goto Error; + } + di = (PFILE_FULL_DIR_INFORMATION)(cache->buffer); + for (;;) { + + *phead = fseentry_create_entry(cache, list, di); phead = &(*phead)->next; - } while (FindNextFileW(h, &fdata)); - /* remember result of last FindNextFile, then close find handle */ - err = GetLastError(); - FindClose(h); + /* If there is no offset in the entry, the buffer has been exhausted. */ + if (di->NextEntryOffset == 0) { + status = NtQueryDirectoryFile(h, NULL, 0, 0, &iosb, cache->buffer, + sizeof(cache->buffer), FileFullDirectoryInformation, FALSE, NULL, FALSE); + if (!NT_SUCCESS(status)) { + if (status == STATUS_NO_MORE_FILES) + break; + goto Error; + } + + di = (PFILE_FULL_DIR_INFORMATION)(cache->buffer); + continue; + } + + /* Advance to the next entry. */ + di = (PFILE_FULL_DIR_INFORMATION)(((PUCHAR)di) + di->NextEntryOffset); + } - /* return the list if we've got all the files */ - if (err == ERROR_NO_MORE_FILES) - return list; + CloseHandle(h); + return list; - /* otherwise release the list and return error */ +Error: + trace_printf_key(&trace_fscache, + "fscache: status(%ld) unable to query directory " + "contents '%s'\n", status, dir->dirent.d_name); + CloseHandle(h); fsentry_release(list); - errno = err_win_to_posix(err); return NULL; } diff --git a/compat/win32/ntifs.h b/compat/win32/ntifs.h new file mode 100644 index 00000000000000..64ed792c52f352 --- /dev/null +++ b/compat/win32/ntifs.h @@ -0,0 +1,131 @@ +#ifndef _NTIFS_ +#define _NTIFS_ + +/* + * Copy necessary structures and definitions out of the Windows DDK + * to enable calling NtQueryDirectoryFile() + */ + +typedef _Return_type_success_(return >= 0) LONG NTSTATUS; +#define NT_SUCCESS(Status) (((NTSTATUS)(Status)) >= 0) + +#if !defined(_NTSECAPI_) && !defined(_WINTERNL_) && \ + !defined(__UNICODE_STRING_DEFINED) +#define __UNICODE_STRING_DEFINED +typedef struct _UNICODE_STRING { + USHORT Length; + USHORT MaximumLength; + PWSTR Buffer; +} UNICODE_STRING; +typedef UNICODE_STRING *PUNICODE_STRING; +typedef const UNICODE_STRING *PCUNICODE_STRING; +#endif /* !_NTSECAPI_ && !_WINTERNL_ && !__UNICODE_STRING_DEFINED */ + +typedef enum _FILE_INFORMATION_CLASS { + FileDirectoryInformation = 1, + FileFullDirectoryInformation, + FileBothDirectoryInformation, + FileBasicInformation, + FileStandardInformation, + FileInternalInformation, + FileEaInformation, + FileAccessInformation, + FileNameInformation, + FileRenameInformation, + FileLinkInformation, + FileNamesInformation, + FileDispositionInformation, + FilePositionInformation, + FileFullEaInformation, + FileModeInformation, + FileAlignmentInformation, + FileAllInformation, + FileAllocationInformation, + FileEndOfFileInformation, + FileAlternateNameInformation, + FileStreamInformation, + FilePipeInformation, + FilePipeLocalInformation, + FilePipeRemoteInformation, + FileMailslotQueryInformation, + FileMailslotSetInformation, + FileCompressionInformation, + FileObjectIdInformation, + FileCompletionInformation, + FileMoveClusterInformation, + FileQuotaInformation, + FileReparsePointInformation, + FileNetworkOpenInformation, + FileAttributeTagInformation, + FileTrackingInformation, + FileIdBothDirectoryInformation, + FileIdFullDirectoryInformation, + FileValidDataLengthInformation, + FileShortNameInformation, + FileIoCompletionNotificationInformation, + FileIoStatusBlockRangeInformation, + FileIoPriorityHintInformation, + FileSfioReserveInformation, + FileSfioVolumeInformation, + FileHardLinkInformation, + FileProcessIdsUsingFileInformation, + FileNormalizedNameInformation, + FileNetworkPhysicalNameInformation, + FileIdGlobalTxDirectoryInformation, + FileIsRemoteDeviceInformation, + FileAttributeCacheInformation, + FileNumaNodeInformation, + FileStandardLinkInformation, + FileRemoteProtocolInformation, + FileMaximumInformation +} FILE_INFORMATION_CLASS, *PFILE_INFORMATION_CLASS; + +typedef struct _FILE_FULL_DIR_INFORMATION { + ULONG NextEntryOffset; + ULONG FileIndex; + LARGE_INTEGER CreationTime; + LARGE_INTEGER LastAccessTime; + LARGE_INTEGER LastWriteTime; + LARGE_INTEGER ChangeTime; + LARGE_INTEGER EndOfFile; + LARGE_INTEGER AllocationSize; + ULONG FileAttributes; + ULONG FileNameLength; + ULONG EaSize; + WCHAR FileName[1]; +} FILE_FULL_DIR_INFORMATION, *PFILE_FULL_DIR_INFORMATION; + +typedef struct _IO_STATUS_BLOCK { + union { + NTSTATUS Status; + PVOID Pointer; + } u; + ULONG_PTR Information; +} IO_STATUS_BLOCK, *PIO_STATUS_BLOCK; + +typedef VOID +(NTAPI *PIO_APC_ROUTINE)( + IN PVOID ApcContext, + IN PIO_STATUS_BLOCK IoStatusBlock, + IN ULONG Reserved); + +NTSYSCALLAPI +NTSTATUS +NTAPI +NtQueryDirectoryFile( + _In_ HANDLE FileHandle, + _In_opt_ HANDLE Event, + _In_opt_ PIO_APC_ROUTINE ApcRoutine, + _In_opt_ PVOID ApcContext, + _Out_ PIO_STATUS_BLOCK IoStatusBlock, + _Out_writes_bytes_(Length) PVOID FileInformation, + _In_ ULONG Length, + _In_ FILE_INFORMATION_CLASS FileInformationClass, + _In_ BOOLEAN ReturnSingleEntry, + _In_opt_ PUNICODE_STRING FileName, + _In_ BOOLEAN RestartScan +); + +#define STATUS_NO_MORE_FILES ((NTSTATUS)0x80000006L) + +#endif From 246a4e370158276ad0ef58b8c4c7a0ffc16c1212 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 11 Dec 2018 12:17:49 +0100 Subject: [PATCH 173/248] fscache: implement an FSCache-aware is_mount_point() When FSCache is active, we can cache the reparse tag and use it directly to determine whether a path refers to an NTFS junction, without any additional, costly I/O. Note: this change only makes a difference with the next commit, which will make use of the FSCache in `git clean` (contingent on `core.fscache` set, of course). Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 2 ++ compat/mingw.h | 3 ++- compat/win32/fscache.c | 40 ++++++++++++++++++++++++++++++++++++++++ compat/win32/fscache.h | 1 + 4 files changed, 45 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8d8029d9000403..d72819df33583c 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2994,6 +2994,8 @@ pid_t waitpid(pid_t pid, int *status, int options) return -1; } +int (*win32_is_mount_point)(struct strbuf *path) = mingw_is_mount_point; + int mingw_is_mount_point(struct strbuf *path) { WIN32_FIND_DATAW findbuf = { 0 }; diff --git a/compat/mingw.h b/compat/mingw.h index 65df57d2a786e4..96677cbb86716d 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -40,7 +40,8 @@ static inline void convert_slashes(char *path) } struct strbuf; int mingw_is_mount_point(struct strbuf *path); -#define is_mount_point mingw_is_mount_point +extern int (*win32_is_mount_point)(struct strbuf *path); +#define is_mount_point win32_is_mount_point #define CAN_UNLINK_MOUNT_POINTS 1 #define PATH_SEP ';' char *mingw_query_user_email(void); diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index c05b931455945c..75dd33dc66bea0 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -474,6 +474,7 @@ int fscache_enable(size_t initial_size) /* redirect opendir and lstat to the fscache implementations */ opendir = fscache_opendir; lstat = fscache_lstat; + win32_is_mount_point = fscache_is_mount_point; } initialized++; LeaveCriticalSection(&fscache_cs); @@ -534,6 +535,7 @@ void fscache_disable(void) /* reset opendir and lstat to the original implementations */ opendir = dirent_opendir; lstat = mingw_lstat; + win32_is_mount_point = mingw_is_mount_point; } LeaveCriticalSection(&fscache_cs); @@ -609,6 +611,44 @@ int fscache_lstat(const char *filename, struct stat *st) return 0; } +/* + * is_mount_point() replacement, uses cache if enabled, otherwise falls + * back to mingw_is_mount_point(). + */ +int fscache_is_mount_point(struct strbuf *path) +{ + int dirlen, base, len; +#pragma GCC diagnostic push +#ifdef __clang__ +#pragma GCC diagnostic ignored "-Wflexible-array-extensions" +#endif + struct heap_fsentry key[2]; +#pragma GCC diagnostic pop + struct fsentry *fse; + struct fscache *cache = fscache_getcache(); + + if (!cache || !do_fscache_enabled(cache, path->buf)) + return mingw_is_mount_point(path); + + cache->lstat_requests++; + /* split path into path + name */ + len = path->len; + if (len && is_dir_sep(path->buf[len - 1])) + len--; + base = len; + while (base && !is_dir_sep(path->buf[base - 1])) + base--; + dirlen = base ? base - 1 : 0; + + /* lookup entry for path + name in cache */ + fsentry_init(&key[0].u.ent, NULL, path->buf, dirlen); + fsentry_init(&key[1].u.ent, &key[0].u.ent, path->buf + base, len - base); + fse = fscache_get(cache, &key[1].u.ent); + if (!fse) + return mingw_is_mount_point(path); + return fse->reparse_tag == IO_REPARSE_TAG_MOUNT_POINT; +} + typedef struct fscache_DIR { struct DIR base_dir; /* extend base struct DIR */ struct fsentry *pfsentry; diff --git a/compat/win32/fscache.h b/compat/win32/fscache.h index 042b247a542554..386c770a85d321 100644 --- a/compat/win32/fscache.h +++ b/compat/win32/fscache.h @@ -22,6 +22,7 @@ void fscache_flush(void); DIR *fscache_opendir(const char *dir); int fscache_lstat(const char *file_name, struct stat *buf); +int fscache_is_mount_point(struct strbuf *path); /* opaque fscache structure */ struct fscache; From 529d28c9876205727a88fcc4435ef42b66b487e8 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 11 Dec 2018 12:17:49 +0100 Subject: [PATCH 174/248] clean: make use of FSCache The `git clean` command needs to enumerate plenty of files and directories, and can therefore benefit from the FSCache. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- builtin/clean.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/builtin/clean.c b/builtin/clean.c index b447155ab2cf2a..a7a523a6ea7f30 100644 --- a/builtin/clean.c +++ b/builtin/clean.c @@ -1044,6 +1044,7 @@ int cmd_clean(int argc, if (repo_read_index(the_repository) < 0) die(_("index file corrupt")); + enable_fscache(the_repository->index->cache_nr); pl = add_pattern_list(&dir, EXC_CMDL, "--exclude option"); for (i = 0; i < exclude_list.nr; i++) @@ -1118,6 +1119,7 @@ int cmd_clean(int argc, } } + disable_fscache(); strbuf_release(&abs_path); strbuf_release(&buf); string_list_clear(&del_list, 0); From e683b9ab9a6243a64f1c839308676048690c7594 Mon Sep 17 00:00:00 2001 From: Doug Kelly <dougk.ff7@gmail.com> Date: Wed, 8 Jan 2014 20:28:15 -0600 Subject: [PATCH 175/248] pack-objects (mingw): demonstrate a segmentation fault with large deltas There is a problem in the way 9ac3f0e5b3e4 (pack-objects: fix performance issues on packing large deltas, 2018-07-22) initializes that mutex in the `packing_data` struct. The problem manifests in a segmentation fault on Windows, when a mutex (AKA critical section) is accessed without being initialized. (With pthreads, you apparently do not really have to initialize them?) This was reported in https://github.com/git-for-windows/git/issues/1839. Signed-off-by: Doug Kelly <dougk.ff7@gmail.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/meson.build | 1 + t/t7429-submodule-long-path.sh | 106 +++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+) create mode 100755 t/t7429-submodule-long-path.sh diff --git a/t/meson.build b/t/meson.build index 25d9e33697adff..7d9d96fd7c3c9c 100644 --- a/t/meson.build +++ b/t/meson.build @@ -876,6 +876,7 @@ integration_tests = [ 't7422-submodule-output.sh', 't7423-submodule-symlinks.sh', 't7424-submodule-mixed-ref-formats.sh', + 't7429-submodule-long-path.sh', 't7450-bad-git-dotfiles.sh', 't7500-commit-template-squash-signoff.sh', 't7501-commit-basic-functionality.sh', diff --git a/t/t7429-submodule-long-path.sh b/t/t7429-submodule-long-path.sh new file mode 100755 index 00000000000000..f692cedbff7ff8 --- /dev/null +++ b/t/t7429-submodule-long-path.sh @@ -0,0 +1,106 @@ +#!/bin/sh +# +# Copyright (c) 2013 Doug Kelly +# + +test_description='Test submodules with a path near PATH_MAX + +This test verifies that "git submodule" initialization, update and clones work, including with recursive submodules and paths approaching PATH_MAX (260 characters on Windows) +' + +TEST_NO_CREATE_REPO=1 +. ./test-lib.sh + +longpath="" +for (( i=0; i<4; i++ )); do + longpath="0123456789abcdefghijklmnopqrstuvwxyz$longpath" +done +# Pick a substring maximum of 90 characters +# This should be good, since we'll add on a lot for temp directories +longpath=${longpath:0:90}; export longpath + +test_expect_failure 'submodule with a long path' ' + git config --global protocol.file.allow always && + GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME= \ + git -c init.defaultBranch=long init --bare remote && + test_create_repo bundle1 && + ( + cd bundle1 && + test_commit "shoot" && + git rev-parse --verify HEAD >../expect + ) && + mkdir home && + ( + cd home && + git clone ../remote test && + cd test && + git checkout -B long && + git submodule add ../bundle1 $longpath && + test_commit "sogood" && + ( + cd $longpath && + git rev-parse --verify HEAD >actual && + test_cmp ../../../expect actual + ) && + git push origin long + ) && + mkdir home2 && + ( + cd home2 && + git clone ../remote test && + cd test && + git checkout long && + git submodule update --init && + ( + cd $longpath && + git rev-parse --verify HEAD >actual && + test_cmp ../../../expect actual + ) + ) +' + +test_expect_failure 'recursive submodule with a long path' ' + GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME= \ + git -c init.defaultBranch=long init --bare super && + test_create_repo child && + ( + cd child && + test_commit "shoot" && + git rev-parse --verify HEAD >../expect + ) && + test_create_repo parent && + ( + cd parent && + git submodule add ../child $longpath && + test_commit "aim" + ) && + mkdir home3 && + ( + cd home3 && + git clone ../super test && + cd test && + git checkout -B long && + git submodule add ../parent foo && + git submodule update --init --recursive && + test_commit "sogood" && + ( + cd foo/$longpath && + git rev-parse --verify HEAD >actual && + test_cmp ../../../../expect actual + ) && + git push origin long + ) && + mkdir home4 && + ( + cd home4 && + git clone ../super test --recursive && + ( + cd test/foo/$longpath && + git rev-parse --verify HEAD >actual && + test_cmp ../../../../expect actual + ) + ) +' +unset longpath + +test_done From 4703d0e07abd645bd3570617bb0bc39621972c46 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 28 Jul 2015 21:07:41 +0200 Subject: [PATCH 176/248] mingw: support long paths Windows paths are typically limited to MAX_PATH = 260 characters, even though the underlying NTFS file system supports paths up to 32,767 chars. This limitation is also evident in Windows Explorer, cmd.exe and many other applications (including IDEs). Particularly annoying is that most Windows APIs return bogus error codes if a relative path only barely exceeds MAX_PATH in conjunction with the current directory, e.g. ERROR_PATH_NOT_FOUND / ENOENT instead of the infinitely more helpful ERROR_FILENAME_EXCED_RANGE / ENAMETOOLONG. Many Windows wide char APIs support longer than MAX_PATH paths through the file namespace prefix ('\\?\' or '\\?\UNC\') followed by an absolute path. Notable exceptions include functions dealing with executables and the current directory (CreateProcess, LoadLibrary, Get/SetCurrentDirectory) as well as the entire shell API (ShellExecute, SHGetSpecialFolderPath...). Introduce a handle_long_path function to check the length of a specified path properly (and fail with ENAMETOOLONG), and to optionally expand long paths using the '\\?\' file namespace prefix. Short paths will not be modified, so we don't need to worry about device names (NUL, CON, AUX). Contrary to MSDN docs, the GetFullPathNameW function doesn't seem to be limited to MAX_PATH (at least not on Win7), so we can use it to do the heavy lifting of the conversion (translate '/' to '\', eliminate '.' and '..', and make an absolute path). Add long path error checking to xutftowcs_path for APIs with hard MAX_PATH limit. Add a new MAX_LONG_PATH constant and xutftowcs_long_path function for APIs that support long paths. While improved error checking is always active, long paths support must be explicitly enabled via 'core.longpaths' option. This is to prevent end users to shoot themselves in the foot by checking out files that Windows Explorer, cmd/bash or their favorite IDE cannot handle. Test suite: Test the case is when the full pathname length of a dir is close to 260 (MAX_PATH). Bug report and an original reproducer by Andrey Rogozhnikov: https://github.com/msysgit/git/pull/122#issuecomment-43604199 [jes: adjusted test number to avoid conflicts, added support for chdir(), etc] Thanks-to: Martin W. Kirst <maki@bitkings.de> Thanks-to: Doug Kelly <dougk.ff7@gmail.com> Original-test-by: Andrey Rogozhnikov <rogozhnikov.andrey@gmail.com> Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- Documentation/config/core.adoc | 7 ++ compat/mingw.c | 174 ++++++++++++++++++++++++++------- compat/mingw.h | 75 +++++++++++++- compat/win32/dirent.c | 17 ++-- compat/win32/fscache.c | 16 ++- t/meson.build | 1 + t/t2031-checkout-long-paths.sh | 102 +++++++++++++++++++ t/t7429-submodule-long-path.sh | 24 +++-- 8 files changed, 349 insertions(+), 67 deletions(-) create mode 100755 t/t2031-checkout-long-paths.sh diff --git a/Documentation/config/core.adoc b/Documentation/config/core.adoc index 32a5872888c5eb..1010bc85842911 100644 --- a/Documentation/config/core.adoc +++ b/Documentation/config/core.adoc @@ -696,6 +696,13 @@ core.fscache:: Git for Windows uses this to bulk-read and cache lstat data of entire directories (instead of doing lstat file by file). +core.longpaths:: + Enable long path (> 260) support for builtin commands in Git for + Windows. This is disabled by default, as long paths are not supported + by Windows Explorer, cmd.exe and the Git for Windows tool chain + (msys, bash, tcl, perl...). Only enable this if you know what you're + doing and are prepared to live with a few quirks. + core.unsetenvvars:: Windows-only: comma-separated list of environment variables' names that need to be unset before spawning any other process. diff --git a/compat/mingw.c b/compat/mingw.c index d72819df33583c..c9e8e6e1bd3698 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -251,6 +251,27 @@ static enum hide_dotfiles_type hide_dotfiles = HIDE_DOTFILES_DOTGITONLY; static char *unset_environment_variables; int core_fscache; +int are_long_paths_enabled(void) +{ + /* default to `false` during initialization */ + static const int fallback = 0; + + static int enabled = -1; + + if (enabled < 0) { + /* avoid infinite recursion */ + if (!the_repository) + return fallback; + + if (the_repository->config && + the_repository->config->hash_initialized && + repo_config_get_bool(the_repository, "core.longpaths", &enabled) < 0) + enabled = 0; + } + + return enabled < 0 ? fallback : enabled; +} + int mingw_core_config(const char *var, const char *value, const struct config_context *ctx UNUSED, void *cb UNUSED) @@ -307,8 +328,8 @@ static wchar_t *normalize_ntpath(wchar_t *wbuf) int mingw_unlink(const char *pathname, int handle_in_use_error) { int ret, tries = 0; - wchar_t wpathname[MAX_PATH]; - if (xutftowcs_path(wpathname, pathname) < 0) + wchar_t wpathname[MAX_LONG_PATH]; + if (xutftowcs_long_path(wpathname, pathname) < 0) return -1; if (DeleteFileW(wpathname)) @@ -343,7 +364,7 @@ static int is_dir_empty(const wchar_t *wpath) { WIN32_FIND_DATAW findbuf; HANDLE handle; - wchar_t wbuf[MAX_PATH + 2]; + wchar_t wbuf[MAX_LONG_PATH + 2]; wcscpy(wbuf, wpath); wcscat(wbuf, L"\\*"); handle = FindFirstFileW(wbuf, &findbuf); @@ -364,7 +385,7 @@ static int is_dir_empty(const wchar_t *wpath) int mingw_rmdir(const char *pathname) { int ret, tries = 0; - wchar_t wpathname[MAX_PATH]; + wchar_t wpathname[MAX_LONG_PATH]; struct stat st; /* @@ -386,7 +407,7 @@ int mingw_rmdir(const char *pathname) return -1; } - if (xutftowcs_path(wpathname, pathname) < 0) + if (xutftowcs_long_path(wpathname, pathname) < 0) return -1; while ((ret = _wrmdir(wpathname)) == -1 && tries < ARRAY_SIZE(delay)) { @@ -465,15 +486,18 @@ static int set_hidden_flag(const wchar_t *path, int set) int mingw_mkdir(const char *path, int mode UNUSED) { int ret; - wchar_t wpath[MAX_PATH]; + wchar_t wpath[MAX_LONG_PATH]; if (!is_valid_win32_path(path, 0)) { errno = EINVAL; return -1; } - if (xutftowcs_path(wpath, path) < 0) + /* CreateDirectoryW path limit is 248 (MAX_PATH - 8.3 file name) */ + if (xutftowcs_path_ex(wpath, path, MAX_LONG_PATH, -1, 248, + are_long_paths_enabled()) < 0) return -1; + ret = _wmkdir(wpath); if (!ret && needs_hiding(path)) return set_hidden_flag(wpath, 1); @@ -637,7 +661,7 @@ int mingw_open (const char *filename, int oflags, ...) va_list args; unsigned mode; int fd, create = (oflags & (O_CREAT | O_EXCL)) == (O_CREAT | O_EXCL); - wchar_t wfilename[MAX_PATH]; + wchar_t wfilename[MAX_LONG_PATH]; open_fn_t open_fn; DECLARE_PROC_ADDR(ntdll.dll, NTSTATUS, NTAPI, RtlGetLastNtStatus, void); @@ -669,7 +693,7 @@ int mingw_open (const char *filename, int oflags, ...) if (filename && !strcmp(filename, "/dev/null")) wcscpy(wfilename, L"nul"); - else if (xutftowcs_path(wfilename, filename) < 0) + else if (xutftowcs_long_path(wfilename, filename) < 0) return -1; fd = open_fn(wfilename, oflags, mode); @@ -742,14 +766,14 @@ FILE *mingw_fopen (const char *filename, const char *otype) { int hide = needs_hiding(filename); FILE *file; - wchar_t wfilename[MAX_PATH], wotype[4]; + wchar_t wfilename[MAX_LONG_PATH], wotype[4]; if (filename && !strcmp(filename, "/dev/null")) wcscpy(wfilename, L"nul"); else if (!is_valid_win32_path(filename, 1)) { int create = otype && strchr(otype, 'w'); errno = create ? EINVAL : ENOENT; return NULL; - } else if (xutftowcs_path(wfilename, filename) < 0) + } else if (xutftowcs_long_path(wfilename, filename) < 0) return NULL; if (xutftowcs(wotype, otype, ARRAY_SIZE(wotype)) < 0) @@ -771,14 +795,14 @@ FILE *mingw_freopen (const char *filename, const char *otype, FILE *stream) { int hide = needs_hiding(filename); FILE *file; - wchar_t wfilename[MAX_PATH], wotype[4]; + wchar_t wfilename[MAX_LONG_PATH], wotype[4]; if (filename && !strcmp(filename, "/dev/null")) wcscpy(wfilename, L"nul"); else if (!is_valid_win32_path(filename, 1)) { int create = otype && strchr(otype, 'w'); errno = create ? EINVAL : ENOENT; return NULL; - } else if (xutftowcs_path(wfilename, filename) < 0) + } else if (xutftowcs_long_path(wfilename, filename) < 0) return NULL; if (xutftowcs(wotype, otype, ARRAY_SIZE(wotype)) < 0) @@ -828,7 +852,7 @@ ssize_t mingw_write(int fd, const void *buf, size_t len) HANDLE h = (HANDLE) _get_osfhandle(fd); if (GetFileType(h) != FILE_TYPE_PIPE) { if (orig == EINVAL) { - wchar_t path[MAX_PATH]; + wchar_t path[MAX_LONG_PATH]; DWORD ret = GetFinalPathNameByHandleW(h, path, ARRAY_SIZE(path), 0); UINT drive_type = ret > 0 && ret < ARRAY_SIZE(path) ? @@ -865,27 +889,33 @@ ssize_t mingw_write(int fd, const void *buf, size_t len) int mingw_access(const char *filename, int mode) { - wchar_t wfilename[MAX_PATH]; + wchar_t wfilename[MAX_LONG_PATH]; if (!strcmp("nul", filename) || !strcmp("/dev/null", filename)) return 0; - if (xutftowcs_path(wfilename, filename) < 0) + if (xutftowcs_long_path(wfilename, filename) < 0) return -1; /* X_OK is not supported by the MSVCRT version */ return _waccess(wfilename, mode & ~X_OK); } +/* cached length of current directory for handle_long_path */ +static int current_directory_len = 0; + int mingw_chdir(const char *dirname) { - wchar_t wdirname[MAX_PATH]; - if (xutftowcs_path(wdirname, dirname) < 0) + int result; + wchar_t wdirname[MAX_LONG_PATH]; + if (xutftowcs_long_path(wdirname, dirname) < 0) return -1; - return _wchdir(wdirname); + result = _wchdir(wdirname); + current_directory_len = GetCurrentDirectoryW(0, NULL); + return result; } int mingw_chmod(const char *filename, int mode) { - wchar_t wfilename[MAX_PATH]; - if (xutftowcs_path(wfilename, filename) < 0) + wchar_t wfilename[MAX_LONG_PATH]; + if (xutftowcs_long_path(wfilename, filename) < 0) return -1; return _wchmod(wfilename, mode); } @@ -933,8 +963,8 @@ static int has_valid_directory_prefix(wchar_t *wfilename) static int do_lstat(int follow, const char *file_name, struct stat *buf) { WIN32_FILE_ATTRIBUTE_DATA fdata; - wchar_t wfilename[MAX_PATH]; - if (xutftowcs_path(wfilename, file_name) < 0) + wchar_t wfilename[MAX_LONG_PATH]; + if (xutftowcs_long_path(wfilename, file_name) < 0) return -1; if (GetFileAttributesExW(wfilename, GetFileExInfoStandard, &fdata)) { @@ -1105,10 +1135,10 @@ int mingw_utime (const char *file_name, const struct utimbuf *times) FILETIME mft, aft; int rc; DWORD attrs; - wchar_t wfilename[MAX_PATH]; + wchar_t wfilename[MAX_LONG_PATH]; HANDLE osfilehandle; - if (xutftowcs_path(wfilename, file_name) < 0) + if (xutftowcs_long_path(wfilename, file_name) < 0) return -1; /* must have write permission */ @@ -1191,6 +1221,7 @@ char *mingw_mktemp(char *template) wchar_t wtemplate[MAX_PATH]; int offset = 0; + /* we need to return the path, thus no long paths here! */ if (xutftowcs_path(wtemplate, template) < 0) return NULL; @@ -1832,6 +1863,10 @@ static pid_t mingw_spawnve_fd(const char *cmd, const char **argv, char **deltaen if (*argv && !strcmp(cmd, *argv)) wcmd[0] = L'\0'; + /* + * Paths to executables and to the current directory do not support + * long paths, therefore we cannot use xutftowcs_long_path() here. + */ else if (xutftowcs_path(wcmd, cmd) < 0) return -1; if (dir && xutftowcs_path(wdir, dir) < 0) @@ -2521,12 +2556,12 @@ int mingw_rename(const char *pold, const char *pnew) static int supports_file_rename_info_ex = 1; DWORD attrs, gle; int tries = 0; - wchar_t wpold[MAX_PATH], wpnew[MAX_PATH]; + wchar_t wpold[MAX_LONG_PATH], wpnew[MAX_LONG_PATH]; int wpnew_len; - if (xutftowcs_path(wpold, pold) < 0) + if (xutftowcs_long_path(wpold, pold) < 0) return -1; - wpnew_len = xutftowcs_path(wpnew, pnew); + wpnew_len = xutftowcs_long_path(wpnew, pnew); if (wpnew_len < 0) return -1; @@ -2565,9 +2600,9 @@ int mingw_rename(const char *pold, const char *pnew) * flex array so that the structure has to be allocated on * the heap. As we declare this structure ourselves though * we can avoid the allocation and define FileName to have - * MAX_PATH bytes. + * MAX_LONG_PATH bytes. */ - WCHAR FileName[MAX_PATH]; + WCHAR FileName[MAX_LONG_PATH]; } rename_info = { 0 }; HANDLE old_handle = INVALID_HANDLE_VALUE; BOOL success; @@ -2930,9 +2965,9 @@ int mingw_raise(int sig) int link(const char *oldpath, const char *newpath) { - wchar_t woldpath[MAX_PATH], wnewpath[MAX_PATH]; - if (xutftowcs_path(woldpath, oldpath) < 0 || - xutftowcs_path(wnewpath, newpath) < 0) + wchar_t woldpath[MAX_LONG_PATH], wnewpath[MAX_LONG_PATH]; + if (xutftowcs_long_path(woldpath, oldpath) < 0 || + xutftowcs_long_path(wnewpath, newpath) < 0) return -1; if (!CreateHardLinkW(wnewpath, woldpath, NULL)) { @@ -3000,8 +3035,8 @@ int mingw_is_mount_point(struct strbuf *path) { WIN32_FIND_DATAW findbuf = { 0 }; HANDLE handle; - wchar_t wfilename[MAX_PATH]; - int wlen = xutftowcs_path(wfilename, path->buf); + wchar_t wfilename[MAX_LONG_PATH]; + int wlen = xutftowcs_long_path(wfilename, path->buf); if (wlen < 0) die(_("could not get long path for '%s'"), path->buf); @@ -3153,9 +3188,9 @@ static size_t append_system_bin_dirs(char *path, size_t size) static int is_system32_path(const char *path) { - WCHAR system32[MAX_PATH], wpath[MAX_PATH]; + WCHAR system32[MAX_LONG_PATH], wpath[MAX_LONG_PATH]; - if (xutftowcs_path(wpath, path) < 0 || + if (xutftowcs_long_path(wpath, path) < 0 || !GetSystemDirectoryW(system32, ARRAY_SIZE(system32)) || _wcsicmp(system32, wpath)) return 0; @@ -3588,6 +3623,68 @@ int is_valid_win32_path(const char *path, int allow_literal_nul) } } +int handle_long_path(wchar_t *path, int len, int max_path, int expand) +{ + int result; + wchar_t buf[MAX_LONG_PATH]; + + /* + * we don't need special handling if path is relative to the current + * directory, and current directory + path don't exceed the desired + * max_path limit. This should cover > 99 % of cases with minimal + * performance impact (git almost always uses relative paths). + */ + if ((len < 2 || (!is_dir_sep(path[0]) && path[1] != ':')) && + (current_directory_len + len < max_path)) + return len; + + /* + * handle everything else: + * - absolute paths: "C:\dir\file" + * - absolute UNC paths: "\\server\share\dir\file" + * - absolute paths on current drive: "\dir\file" + * - relative paths on other drive: "X:file" + * - prefixed paths: "\\?\...", "\\.\..." + */ + + /* convert to absolute path using GetFullPathNameW */ + result = GetFullPathNameW(path, MAX_LONG_PATH, buf, NULL); + if (!result) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + + /* + * return absolute path if it fits within max_path (even if + * "cwd + path" doesn't due to '..' components) + */ + if (result < max_path) { + wcscpy(path, buf); + return result; + } + + /* error out if we shouldn't expand the path or buf is too small */ + if (!expand || result >= MAX_LONG_PATH - 6) { + errno = ENAMETOOLONG; + return -1; + } + + /* prefix full path with "\\?\" or "\\?\UNC\" */ + if (buf[0] == '\\') { + /* ...unless already prefixed */ + if (buf[1] == '\\' && (buf[2] == '?' || buf[2] == '.')) + return len; + + wcscpy(path, L"\\\\?\\UNC\\"); + wcscpy(path + 8, buf + 2); + return result + 6; + } else { + wcscpy(path, L"\\\\?\\"); + wcscpy(path + 4, buf); + return result + 4; + } +} + #if !defined(_MSC_VER) /* * Disable MSVCRT command line wildcard expansion (__getmainargs called from @@ -3750,6 +3847,9 @@ int wmain(int argc, const wchar_t **wargv) /* initialize Unicode console */ winansi_init(); + /* init length of current directory for handle_long_path */ + current_directory_len = GetCurrentDirectoryW(0, NULL); + /* invoke the real main() using our utf8 version of argv. */ exit_status = main(argc, argv); diff --git a/compat/mingw.h b/compat/mingw.h index 96677cbb86716d..ad1166b775322a 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -1,6 +1,7 @@ #include "mingw-posix.h" extern int core_fscache; +int are_long_paths_enabled(void); struct config_context; int mingw_core_config(const char *var, const char *value, @@ -78,6 +79,42 @@ int is_path_owned_by_current_sid(const char *path, struct strbuf *report); int is_valid_win32_path(const char *path, int allow_literal_nul); #define is_valid_path(path) is_valid_win32_path(path, 0) +/** + * Max length of long paths (exceeding MAX_PATH). The actual maximum supported + * by NTFS is 32,767 (* sizeof(wchar_t)), but we choose an arbitrary smaller + * value to limit required stack memory. + */ +#define MAX_LONG_PATH 4096 + +/** + * Handles paths that would exceed the MAX_PATH limit of Windows Unicode APIs. + * + * With expand == false, the function checks for over-long paths and fails + * with ENAMETOOLONG. The path parameter is not modified, except if cwd + path + * exceeds max_path, but the resulting absolute path doesn't (e.g. due to + * eliminating '..' components). The path parameter must point to a buffer + * of max_path wide characters. + * + * With expand == true, an over-long path is automatically converted in place + * to an absolute path prefixed with '\\?\', and the new length is returned. + * The path parameter must point to a buffer of MAX_LONG_PATH wide characters. + * + * Parameters: + * path: path to check and / or convert + * len: size of path on input (number of wide chars without \0) + * max_path: max short path length to check (usually MAX_PATH = 260, but just + * 248 for CreateDirectoryW) + * expand: false to only check the length, true to expand the path to a + * '\\?\'-prefixed absolute path + * + * Return: + * length of the resulting path, or -1 on failure + * + * Errors: + * ENAMETOOLONG if path is too long + */ +int handle_long_path(wchar_t *path, int len, int max_path, int expand); + /** * Converts UTF-8 encoded string to UTF-16LE. * @@ -136,18 +173,46 @@ static inline int xutftowcs(wchar_t *wcs, const char *utf, size_t wcslen) } /** - * Simplified file system specific variant of xutftowcsn, assumes output - * buffer size is MAX_PATH wide chars and input string is \0-terminated, - * fails with ENAMETOOLONG if input string is too long. + * Simplified file system specific wrapper of xutftowcsn and handle_long_path. + * Converts ERANGE to ENAMETOOLONG. If expand is true, wcs must be at least + * MAX_LONG_PATH wide chars (see handle_long_path). */ -static inline int xutftowcs_path(wchar_t *wcs, const char *utf) +static inline int xutftowcs_path_ex(wchar_t *wcs, const char *utf, + size_t wcslen, int utflen, int max_path, int expand) { - int result = xutftowcsn(wcs, utf, MAX_PATH, -1); + int result = xutftowcsn(wcs, utf, wcslen, utflen); if (result < 0 && errno == ERANGE) errno = ENAMETOOLONG; + if (result >= 0) + result = handle_long_path(wcs, result, max_path, expand); return result; } +/** + * Simplified file system specific variant of xutftowcsn, assumes output + * buffer size is MAX_PATH wide chars and input string is \0-terminated, + * fails with ENAMETOOLONG if input string is too long. Typically used for + * Windows APIs that don't support long paths, e.g. SetCurrentDirectory, + * LoadLibrary, CreateProcess... + */ +static inline int xutftowcs_path(wchar_t *wcs, const char *utf) +{ + return xutftowcs_path_ex(wcs, utf, MAX_PATH, -1, MAX_PATH, 0); +} + +/** + * Simplified file system specific variant of xutftowcsn for Windows APIs + * that support long paths via '\\?\'-prefix, assumes output buffer size is + * MAX_LONG_PATH wide chars, fails with ENAMETOOLONG if input string is too + * long. The 'core.longpaths' git-config option controls whether the path + * is only checked or expanded to a long path. + */ +static inline int xutftowcs_long_path(wchar_t *wcs, const char *utf) +{ + return xutftowcs_path_ex(wcs, utf, MAX_LONG_PATH, -1, MAX_PATH, + are_long_paths_enabled()); +} + /** * Converts UTF-16LE encoded string to UTF-8. * diff --git a/compat/win32/dirent.c b/compat/win32/dirent.c index 139d2ba3c4da34..c9fe2454efc01c 100644 --- a/compat/win32/dirent.c +++ b/compat/win32/dirent.c @@ -65,19 +65,24 @@ static int dirent_closedir(dirent_DIR *dir) DIR *dirent_opendir(const char *name) { - wchar_t pattern[MAX_PATH + 2]; /* + 2 for '/' '*' */ + wchar_t pattern[MAX_LONG_PATH + 2]; /* + 2 for "\*" */ WIN32_FIND_DATAW fdata; HANDLE h; int len; dirent_DIR *dir; - /* convert name to UTF-16 and check length < MAX_PATH */ - if ((len = xutftowcs_path(pattern, name)) < 0) + /* convert name to UTF-16 and check length */ + if ((len = xutftowcs_path_ex(pattern, name, MAX_LONG_PATH, -1, + MAX_PATH - 2, + are_long_paths_enabled())) < 0) return NULL; - /* append optional '/' and wildcard '*' */ + /* + * append optional '\' and wildcard '*'. Note: we need to use '\' as + * Windows doesn't translate '/' to '\' for "\\?\"-prefixed paths. + */ if (len && !is_dir_sep(pattern[len - 1])) - pattern[len++] = '/'; + pattern[len++] = '\\'; pattern[len++] = '*'; pattern[len] = 0; @@ -90,7 +95,7 @@ DIR *dirent_opendir(const char *name) } /* initialize DIR structure and copy first dir entry */ - dir = xmalloc(sizeof(dirent_DIR) + MAX_PATH); + dir = xmalloc(sizeof(dirent_DIR) + MAX_LONG_PATH); dir->base_dir.preaddir = (struct dirent *(*)(DIR *dir)) dirent_readdir; dir->base_dir.pclosedir = (int (*)(DIR *dir)) dirent_closedir; dir->dd_handle = h; diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 75dd33dc66bea0..dbf640ca790fde 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -84,7 +84,7 @@ struct fsentry { struct heap_fsentry { union { struct fsentry ent; - char dummy[sizeof(struct fsentry) + MAX_PATH]; + char dummy[sizeof(struct fsentry) + MAX_LONG_PATH]; } u; }; #pragma GCC diagnostic pop @@ -128,7 +128,7 @@ static void fsentry_init(struct fsentry *fse, struct fsentry *list, const char *name, size_t len) { fse->list = list; - if (len > MAX_PATH) + if (len > MAX_LONG_PATH) BUG("Trying to allocate fsentry for long path '%.*s'", (int)len, name); memcpy(fse->dirent.d_name, name, len); @@ -229,7 +229,7 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, static struct fsentry *fsentry_create_list(struct fscache *cache, const struct fsentry *dir, int *dir_not_found) { - wchar_t pattern[MAX_PATH]; + wchar_t pattern[MAX_LONG_PATH]; NTSTATUS status; IO_STATUS_BLOCK iosb; PFILE_FULL_DIR_INFORMATION di; @@ -240,13 +240,11 @@ static struct fsentry *fsentry_create_list(struct fscache *cache, const struct f *dir_not_found = 0; - /* convert name to UTF-16 and check length < MAX_PATH */ - if ((wlen = xutftowcsn(pattern, dir->dirent.d_name, MAX_PATH, - dir->len)) < 0) { - if (errno == ERANGE) - errno = ENAMETOOLONG; + /* convert name to UTF-16 and check length */ + if ((wlen = xutftowcs_path_ex(pattern, dir->dirent.d_name, + MAX_LONG_PATH, dir->len, MAX_PATH - 2, + are_long_paths_enabled())) < 0) return NULL; - } /* handle CWD */ if (!wlen) { diff --git a/t/meson.build b/t/meson.build index 7d9d96fd7c3c9c..dc167eadb7041a 100644 --- a/t/meson.build +++ b/t/meson.build @@ -260,6 +260,7 @@ integration_tests = [ 't2026-checkout-pathspec-file.sh', 't2027-checkout-track.sh', 't2030-unresolve-info.sh', + 't2031-checkout-long-paths.sh', 't2050-git-dir-relative.sh', 't2060-switch.sh', 't2070-restore.sh', diff --git a/t/t2031-checkout-long-paths.sh b/t/t2031-checkout-long-paths.sh new file mode 100755 index 00000000000000..f30f8920ca689c --- /dev/null +++ b/t/t2031-checkout-long-paths.sh @@ -0,0 +1,102 @@ +#!/bin/sh + +test_description='checkout long paths on Windows + +Ensures that Git for Windows can deal with long paths (>260) enabled via core.longpaths' + +. ./test-lib.sh + +if test_have_prereq !MINGW +then + skip_all='skipping MINGW specific long paths test' + test_done +fi + +test_expect_success setup ' + p=longpathxx && # -> 10 + p=$p$p$p$p$p && # -> 50 + p=$p$p$p$p$p && # -> 250 + + path=${p}/longtestfile && # -> 263 (MAX_PATH = 260) + + blob=$(echo foobar | git hash-object -w --stdin) && + + printf "100644 %s 0\t%s\n" "$blob" "$path" | + git update-index --add --index-info && + git commit -m initial -q +' + +test_expect_success 'checkout of long paths without core.longpaths fails' ' + git config core.longpaths false && + test_must_fail git checkout -f 2>error && + grep -q "Filename too long" error && + test ! -d longpa* +' + +test_expect_success 'checkout of long paths with core.longpaths works' ' + git config core.longpaths true && + git checkout -f && + test_path_is_file longpa*/longtestfile +' + +test_expect_success 'update of long paths' ' + echo frotz >>$(ls longpa*/longtestfile) && + echo $path > expect && + git ls-files -m > actual && + test_cmp expect actual && + git add $path && + git commit -m second && + git grep "frotz" HEAD -- $path +' + +test_expect_success cleanup ' + # bash cannot delete the trash dir if it contains a long path + # lets help cleaning up (unless in debug mode) + if test -z "$debug" + then + rm -rf longpa~1 + fi +' + +# check that the template used in the test won't be too long: +abspath="$(pwd)"/testdir +test ${#abspath} -gt 230 || +test_set_prereq SHORTABSPATH + +test_expect_success SHORTABSPATH 'clean up path close to MAX_PATH' ' + p=/123456789abcdef/123456789abcdef/123456789abcdef/123456789abc/ef && + p=y$p$p$p$p && + subdir="x$(echo "$p" | tail -c $((253 - ${#abspath})) - )" && + # Now, $abspath/$subdir has exactly 254 characters, and is inside CWD + p2="$abspath/$subdir" && + test 254 = ${#p2} && + + # Be careful to overcome path limitations of the MSys tools and split + # the $subdir into two parts. ($subdir2 has to contain 16 chars and a + # slash somewhere following; that is why we asked for abspath <= 230 and + # why we placed a slash near the end of the $subdir template.) + subdir2=${subdir#????????????????*/} && + subdir1=testdir/${subdir%/$subdir2} && + mkdir -p "$subdir1" && + i=0 && + # The most important case is when absolute path is 258 characters long, + # and that will be when i == 4. + while test $i -le 7 + do + mkdir -p $subdir2 && + touch $subdir2/one-file && + mv ${subdir2%%/*} "$subdir1/" && + subdir2=z${subdir2} && + i=$(($i+1)) || + exit 1 + done && + + # now check that git is able to clear the tree: + (cd testdir && + git init && + git config core.longpaths yes && + git clean -fdx) && + test ! -d "$subdir1" +' + +test_done diff --git a/t/t7429-submodule-long-path.sh b/t/t7429-submodule-long-path.sh index f692cedbff7ff8..458519eafd6f03 100755 --- a/t/t7429-submodule-long-path.sh +++ b/t/t7429-submodule-long-path.sh @@ -11,15 +11,20 @@ This test verifies that "git submodule" initialization, update and clones work, TEST_NO_CREATE_REPO=1 . ./test-lib.sh -longpath="" -for (( i=0; i<4; i++ )); do - longpath="0123456789abcdefghijklmnopqrstuvwxyz$longpath" -done -# Pick a substring maximum of 90 characters -# This should be good, since we'll add on a lot for temp directories -longpath=${longpath:0:90}; export longpath +# cloning a submodule calls is_git_directory("$path/../.git/modules/$path"), +# which effectively limits the maximum length to PATH_MAX / 2 minus some +# overhead; start with 3 * 36 = 108 chars (test 2 fails if >= 110) +longpath36=0123456789abcdefghijklmnopqrstuvwxyz +longpath180=$longpath36$longpath36$longpath36$longpath36$longpath36 -test_expect_failure 'submodule with a long path' ' +# the git database must fit within PATH_MAX, which limits the submodule name +# to PATH_MAX - len(pwd) - ~90 (= len("/objects//") + 40-byte sha1 + some +# overhead from the test case) +pwd=$(pwd) +pwdlen=$(echo "$pwd" | wc -c) +longpath=$(echo $longpath180 | cut -c 1-$((170-$pwdlen))) + +test_expect_success 'submodule with a long path' ' git config --global protocol.file.allow always && GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME= \ git -c init.defaultBranch=long init --bare remote && @@ -59,7 +64,7 @@ test_expect_failure 'submodule with a long path' ' ) ' -test_expect_failure 'recursive submodule with a long path' ' +test_expect_success 'recursive submodule with a long path' ' GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME= \ git -c init.defaultBranch=long init --bare super && test_create_repo child && @@ -101,6 +106,5 @@ test_expect_failure 'recursive submodule with a long path' ' ) ) ' -unset longpath test_done From 65479771f67155d3046d7ce005015f73d26411b5 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 5 Jul 2014 00:00:36 +0200 Subject: [PATCH 177/248] Win32: fix 'lstat("dir/")' with long paths Use a suffciently large buffer to strip the trailing slash. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index c9e8e6e1bd3698..132125d35eede8 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1035,7 +1035,7 @@ static int do_lstat(int follow, const char *file_name, struct stat *buf) static int do_stat_internal(int follow, const char *file_name, struct stat *buf) { size_t namelen; - char alt_name[PATH_MAX]; + char alt_name[MAX_LONG_PATH]; if (!do_lstat(follow, file_name, buf)) return 0; @@ -1051,7 +1051,7 @@ static int do_stat_internal(int follow, const char *file_name, struct stat *buf) return -1; while (namelen && file_name[namelen-1] == '/') --namelen; - if (!namelen || namelen >= PATH_MAX) + if (!namelen || namelen >= MAX_LONG_PATH) return -1; memcpy(alt_name, file_name, namelen); From bc34cd2d00ab2f66e39fe7f9fc12d34414b36e55 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 6 Sep 2023 09:14:47 +0200 Subject: [PATCH 178/248] win32(long path support): leave drive-less absolute paths intact When trying to ensure that long paths are handled correctly, we first normalize absolute paths as we encounter them. However, if the path is a so-called "drive-less" absolute path, i.e. if it is relative to the current drive but _does_ start with a directory separator, we would want the normalized path to be such a drive-less absolute path, too. Let's do that, being careful to still include the drive prefix when we need to go through the `\\?\` dance (because there, the drive prefix is absolutely required). This fixes https://github.com/git-for-windows/git/issues/4586. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 12 +++++++++++- t/t2031-checkout-long-paths.sh | 9 +++++++++ 2 files changed, 20 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 132125d35eede8..b3b52719762f4c 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -300,6 +300,11 @@ int mingw_core_config(const char *var, const char *value, return 0; } +static inline int is_wdir_sep(wchar_t wchar) +{ + return wchar == L'/' || wchar == L'\\'; +} + /* Normalizes NT paths as returned by some low-level APIs. */ static wchar_t *normalize_ntpath(wchar_t *wbuf) { @@ -3659,7 +3664,12 @@ int handle_long_path(wchar_t *path, int len, int max_path, int expand) * "cwd + path" doesn't due to '..' components) */ if (result < max_path) { - wcscpy(path, buf); + /* Be careful not to add a drive prefix if there was none */ + if (is_wdir_sep(path[0]) && + !is_wdir_sep(buf[0]) && buf[1] == L':' && is_wdir_sep(buf[2])) + wcscpy(path, buf + 2); + else + wcscpy(path, buf); return result; } diff --git a/t/t2031-checkout-long-paths.sh b/t/t2031-checkout-long-paths.sh index f30f8920ca689c..15416a1d6ee8c7 100755 --- a/t/t2031-checkout-long-paths.sh +++ b/t/t2031-checkout-long-paths.sh @@ -99,4 +99,13 @@ test_expect_success SHORTABSPATH 'clean up path close to MAX_PATH' ' test ! -d "$subdir1" ' +test_expect_success SYMLINKS_WINDOWS 'leave drive-less, short paths intact' ' + printf "/Program Files" >symlink-target && + symlink_target_oid="$(git hash-object -w --stdin <symlink-target)" && + git update-index --add --cacheinfo 120000,$symlink_target_oid,PF && + git -c core.symlinks=true checkout -- PF && + cmd //c dir >actual && + grep "<SYMLINKD\\?> *PF *\\[\\\\Program Files\\]" actual +' + test_done From 341494fece0591a045464a28ffa29bdd71104a56 Mon Sep 17 00:00:00 2001 From: Jeff Hostetler <jeffhost@microsoft.com> Date: Fri, 25 Mar 2022 16:56:04 -0400 Subject: [PATCH 179/248] compat/fsmonitor/fsm-*-win32: support long paths Update wchar_t buffers to use MAX_LONG_PATH instead of MAX_PATH and call xutftowcs_long_path() in the Win32 backend source files. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> --- compat/fsmonitor/fsm-health-win32.c | 6 +++--- compat/fsmonitor/fsm-listen-win32.c | 18 +++++++++--------- compat/fsmonitor/fsm-path-utils-win32.c | 8 ++++---- 3 files changed, 16 insertions(+), 16 deletions(-) diff --git a/compat/fsmonitor/fsm-health-win32.c b/compat/fsmonitor/fsm-health-win32.c index 2aa8c219acee4d..4b53360d194105 100644 --- a/compat/fsmonitor/fsm-health-win32.c +++ b/compat/fsmonitor/fsm-health-win32.c @@ -34,7 +34,7 @@ struct fsm_health_data struct wt_moved { - wchar_t wpath[MAX_PATH + 1]; + wchar_t wpath[MAX_LONG_PATH + 1]; BY_HANDLE_FILE_INFORMATION bhfi; } wt_moved; }; @@ -143,8 +143,8 @@ static int has_worktree_moved(struct fsmonitor_daemon_state *state, return 0; case CTX_INIT: - if (xutftowcs_path(data->wt_moved.wpath, - state->path_worktree_watch.buf) < 0) { + if (xutftowcs_long_path(data->wt_moved.wpath, + state->path_worktree_watch.buf) < 0) { error(_("could not convert to wide characters: '%s'"), state->path_worktree_watch.buf); return -1; diff --git a/compat/fsmonitor/fsm-listen-win32.c b/compat/fsmonitor/fsm-listen-win32.c index 9a6efc9bea340b..afcc172750af10 100644 --- a/compat/fsmonitor/fsm-listen-win32.c +++ b/compat/fsmonitor/fsm-listen-win32.c @@ -28,7 +28,7 @@ struct one_watch DWORD count; struct strbuf path; - wchar_t wpath_longname[MAX_PATH + 1]; + wchar_t wpath_longname[MAX_LONG_PATH + 1]; DWORD wpath_longname_len; HANDLE hDir; @@ -131,8 +131,8 @@ static int normalize_path_in_utf8(wchar_t *wpath, DWORD wpath_len, */ static void check_for_shortnames(struct one_watch *watch) { - wchar_t buf_in[MAX_PATH + 1]; - wchar_t buf_out[MAX_PATH + 1]; + wchar_t buf_in[MAX_LONG_PATH + 1]; + wchar_t buf_out[MAX_LONG_PATH + 1]; wchar_t *last; wchar_t *p; @@ -197,8 +197,8 @@ static enum get_relative_result get_relative_longname( const wchar_t *wpath, DWORD wpath_len, wchar_t *wpath_longname, size_t bufsize_wpath_longname) { - wchar_t buf_in[2 * MAX_PATH + 1]; - wchar_t buf_out[MAX_PATH + 1]; + wchar_t buf_in[2 * MAX_LONG_PATH + 1]; + wchar_t buf_out[MAX_LONG_PATH + 1]; DWORD root_len; DWORD out_len; @@ -298,10 +298,10 @@ static struct one_watch *create_watch(const char *path) FILE_SHARE_WRITE | FILE_SHARE_READ | FILE_SHARE_DELETE; HANDLE hDir; DWORD len_longname; - wchar_t wpath[MAX_PATH + 1]; - wchar_t wpath_longname[MAX_PATH + 1]; + wchar_t wpath[MAX_LONG_PATH + 1]; + wchar_t wpath_longname[MAX_LONG_PATH + 1]; - if (xutftowcs_path(wpath, path) < 0) { + if (xutftowcs_long_path(wpath, path) < 0) { error(_("could not convert to wide characters: '%s'"), path); return NULL; } @@ -545,7 +545,7 @@ static int process_worktree_events(struct fsmonitor_daemon_state *state) struct string_list cookie_list = STRING_LIST_INIT_DUP; struct fsmonitor_batch *batch = NULL; const char *p = watch->buffer; - wchar_t wpath_longname[MAX_PATH + 1]; + wchar_t wpath_longname[MAX_LONG_PATH + 1]; /* * If the kernel gets more events than will fit in the kernel diff --git a/compat/fsmonitor/fsm-path-utils-win32.c b/compat/fsmonitor/fsm-path-utils-win32.c index f4f9cc1f336720..c6eb065bde48b4 100644 --- a/compat/fsmonitor/fsm-path-utils-win32.c +++ b/compat/fsmonitor/fsm-path-utils-win32.c @@ -69,8 +69,8 @@ static int check_remote_protocol(wchar_t *wpath) */ int fsmonitor__get_fs_info(const char *path, struct fs_info *fs_info) { - wchar_t wpath[MAX_PATH]; - wchar_t wfullpath[MAX_PATH]; + wchar_t wpath[MAX_LONG_PATH]; + wchar_t wfullpath[MAX_LONG_PATH]; size_t wlen; UINT driveType; @@ -78,7 +78,7 @@ int fsmonitor__get_fs_info(const char *path, struct fs_info *fs_info) * Do everything in wide chars because the drive letter might be * a multi-byte sequence. See win32_has_dos_drive_prefix(). */ - if (xutftowcs_path(wpath, path) < 0) { + if (xutftowcs_long_path(wpath, path) < 0) { return -1; } @@ -97,7 +97,7 @@ int fsmonitor__get_fs_info(const char *path, struct fs_info *fs_info) * slashes to backslashes. This is essential to get GetDriveTypeW() * correctly handle some UNC "\\server\share\..." paths. */ - if (!GetFullPathNameW(wpath, MAX_PATH, wfullpath, NULL)) { + if (!GetFullPathNameW(wpath, MAX_LONG_PATH, wfullpath, NULL)) { return -1; } From c3eb596ef6c967b0aba6c091a18449fede68c0b2 Mon Sep 17 00:00:00 2001 From: Ben Boeckel <mathstuf@gmail.com> Date: Fri, 22 Apr 2022 09:06:23 -0400 Subject: [PATCH 180/248] clean: suggest using `core.longPaths` if paths are too long to remove On Windows, git repositories may have extra files which need cleaned (e.g., a build directory) that may be arbitrarily deep. Suggest using `core.longPaths` if such situations are encountered. Fixes: #2715 Signed-off-by: Ben Boeckel <mathstuf@gmail.com> --- Documentation/config/advice.adoc | 3 +++ advice.c | 1 + advice.h | 1 + builtin/clean.c | 13 +++++++++++++ 4 files changed, 18 insertions(+) diff --git a/Documentation/config/advice.adoc b/Documentation/config/advice.adoc index 257db58918179a..0b3199f4660886 100644 --- a/Documentation/config/advice.adoc +++ b/Documentation/config/advice.adoc @@ -64,6 +64,9 @@ all advice messages. set their identity configuration. mergeConflict:: Shown when various commands stop because of conflicts. + nameTooLong:: + Advice shown if a filepath operation is attempted where the + path was too long. nestedTag:: Shown when a user attempts to recursively tag a tag object. pushAlreadyExists:: diff --git a/advice.c b/advice.c index e5f0ff844917ec..d2f174310abc21 100644 --- a/advice.c +++ b/advice.c @@ -61,6 +61,7 @@ static struct { [ADVICE_IGNORED_HOOK] = { "ignoredHook" }, [ADVICE_IMPLICIT_IDENTITY] = { "implicitIdentity" }, [ADVICE_MERGE_CONFLICT] = { "mergeConflict" }, + [ADVICE_NAME_TOO_LONG] = { "nameTooLong" }, [ADVICE_NESTED_TAG] = { "nestedTag" }, [ADVICE_OBJECT_NAME_WARNING] = { "objectNameWarning" }, [ADVICE_PUSH_ALREADY_EXISTS] = { "pushAlreadyExists" }, diff --git a/advice.h b/advice.h index 727dcecf4a3ee0..c6692dc48775aa 100644 --- a/advice.h +++ b/advice.h @@ -28,6 +28,7 @@ enum advice_type { ADVICE_IGNORED_HOOK, ADVICE_IMPLICIT_IDENTITY, ADVICE_MERGE_CONFLICT, + ADVICE_NAME_TOO_LONG, ADVICE_NESTED_TAG, ADVICE_OBJECT_NAME_WARNING, ADVICE_PUSH_ALREADY_EXISTS, diff --git a/builtin/clean.c b/builtin/clean.c index a7a523a6ea7f30..002303306d9e51 100644 --- a/builtin/clean.c +++ b/builtin/clean.c @@ -26,6 +26,7 @@ #include "pathspec.h" #include "help.h" #include "prompt.h" +#include "advice.h" static int require_force = -1; /* unset */ static int interactive; @@ -221,6 +222,9 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, quote_path(path->buf, prefix, "ed, 0); errno = saved_errno; warning_errno(_(msg_warn_remove_failed), quoted.buf); + if (saved_errno == ENAMETOOLONG) { + advise_if_enabled(ADVICE_NAME_TOO_LONG, _("Setting `core.longPaths` may allow the deletion to succeed.")); + } *dir_gone = 0; } ret = res; @@ -256,6 +260,9 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, quote_path(path->buf, prefix, "ed, 0); errno = saved_errno; warning_errno(_(msg_warn_remove_failed), quoted.buf); + if (saved_errno == ENAMETOOLONG) { + advise_if_enabled(ADVICE_NAME_TOO_LONG, _("Setting `core.longPaths` may allow the deletion to succeed.")); + } *dir_gone = 0; ret = 1; } @@ -299,6 +306,9 @@ static int remove_dirs(struct strbuf *path, const char *prefix, int force_flag, quote_path(path->buf, prefix, "ed, 0); errno = saved_errno; warning_errno(_(msg_warn_remove_failed), quoted.buf); + if (saved_errno == ENAMETOOLONG) { + advise_if_enabled(ADVICE_NAME_TOO_LONG, _("Setting `core.longPaths` may allow the deletion to succeed.")); + } *dir_gone = 0; ret = 1; } @@ -1111,6 +1121,9 @@ int cmd_clean(int argc, qname = quote_path(item->string, NULL, &buf, 0); errno = saved_errno; warning_errno(_(msg_warn_remove_failed), qname); + if (saved_errno == ENAMETOOLONG) { + advise_if_enabled(ADVICE_NAME_TOO_LONG, _("Setting `core.longPaths` may allow the deletion to succeed.")); + } errors++; } else if (!quiet) { qname = quote_path(item->string, NULL, &buf, 0); From 1f4d3a6c94fd1d382606f48dcd1b7adf39e299eb Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Mon, 11 May 2015 19:54:23 +0200 Subject: [PATCH 181/248] strbuf_readlink: don't call readlink twice if hint is the exact link size strbuf_readlink() calls readlink() twice if the hint argument specifies the exact size of the link target (e.g. by passing stat.st_size as returned by lstat()). This is necessary because 'readlink(..., hint) == hint' could mean that the buffer was too small. Use hint + 1 as buffer size to prevent this. Signed-off-by: Karsten Blees <blees@dcon.de> --- strbuf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/strbuf.c b/strbuf.c index 6c3851a7f84d72..44a8f6a554ee43 100644 --- a/strbuf.c +++ b/strbuf.c @@ -578,12 +578,12 @@ int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint) while (hint < STRBUF_MAXLINK) { ssize_t len; - strbuf_grow(sb, hint); - len = readlink(path, sb->buf, hint); + strbuf_grow(sb, hint + 1); + len = readlink(path, sb->buf, hint + 1); if (len < 0) { if (errno != ERANGE) break; - } else if (len < hint) { + } else if (len <= hint) { strbuf_setlen(sb, len); return 0; } From 1b3ab9a12764ca181d20d69dac07f7a412925f39 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Mon, 11 May 2015 22:15:40 +0200 Subject: [PATCH 182/248] strbuf_readlink: support link targets that exceed PATH_MAX strbuf_readlink() refuses to read link targets that exceed PATH_MAX (even if a sufficient size was specified by the caller). As some platforms support longer paths, remove this restriction (similar to strbuf_getcwd()). Signed-off-by: Karsten Blees <blees@dcon.de> --- strbuf.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/strbuf.c b/strbuf.c index 44a8f6a554ee43..fa4e30f112bb09 100644 --- a/strbuf.c +++ b/strbuf.c @@ -566,8 +566,6 @@ ssize_t strbuf_write(struct strbuf *sb, FILE *f) return sb->len ? fwrite(sb->buf, 1, sb->len, f) : 0; } -#define STRBUF_MAXLINK (2*PATH_MAX) - int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint) { size_t oldalloc = sb->alloc; @@ -575,7 +573,7 @@ int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint) if (hint < 32) hint = 32; - while (hint < STRBUF_MAXLINK) { + for (;;) { ssize_t len; strbuf_grow(sb, hint + 1); From e0ea4c258afe212343eacec5f260b7cd914cb9e8 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Mon, 11 May 2015 19:58:14 +0200 Subject: [PATCH 183/248] lockfile.c: use is_dir_sep() instead of hardcoded '/' checks Signed-off-by: Karsten Blees <blees@dcon.de> --- lockfile.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lockfile.c b/lockfile.c index 1d5ed016828746..67082a9caaeb18 100644 --- a/lockfile.c +++ b/lockfile.c @@ -19,14 +19,14 @@ static void trim_last_path_component(struct strbuf *path) int i = path->len; /* back up past trailing slashes, if any */ - while (i && path->buf[i - 1] == '/') + while (i && is_dir_sep(path->buf[i - 1])) i--; /* * then go backwards until a slash, or the beginning of the * string */ - while (i && path->buf[i - 1] != '/') + while (i && !is_dir_sep(path->buf[i - 1])) i--; strbuf_setlen(path, i); From 42912d4db7c0b0cedd707c100ec064e40e408297 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 12 May 2015 11:09:01 +0200 Subject: [PATCH 184/248] Win32: don't call GetFileAttributes twice in mingw_lstat() GetFileAttributes cannot handle paths with trailing dir separator. The current [l]stat implementation calls GetFileAttributes twice if the path has trailing slashes (first with the original path passed to [l]stat, and and a second time with a path copy with trailing '/' removed). With Unicode conversion, we get the length of the path for free and also have a (wide char) buffer that can be modified. Remove trailing directory separators before calling the Win32 API. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 48 ++++++++++++------------------------------------ 1 file changed, 12 insertions(+), 36 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index b3b52719762f4c..b734b1d1f17938 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -969,8 +969,17 @@ static int do_lstat(int follow, const char *file_name, struct stat *buf) { WIN32_FILE_ATTRIBUTE_DATA fdata; wchar_t wfilename[MAX_LONG_PATH]; - if (xutftowcs_long_path(wfilename, file_name) < 0) + int wlen = xutftowcs_long_path(wfilename, file_name); + if (wlen < 0) + return -1; + + /* strip trailing '/', or GetFileAttributes will fail */ + while (wlen && is_dir_sep(wfilename[wlen - 1])) + wfilename[--wlen] = 0; + if (!wlen) { + errno = ENOENT; return -1; + } if (GetFileAttributesExW(wfilename, GetFileExInfoStandard, &fdata)) { buf->st_ino = 0; @@ -1031,39 +1040,6 @@ static int do_lstat(int follow, const char *file_name, struct stat *buf) return -1; } -/* We provide our own lstat/fstat functions, since the provided - * lstat/fstat functions are so slow. These stat functions are - * tailored for Git's usage (read: fast), and are not meant to be - * complete. Note that Git stat()s are redirected to mingw_lstat() - * too, since Windows doesn't really handle symlinks that well. - */ -static int do_stat_internal(int follow, const char *file_name, struct stat *buf) -{ - size_t namelen; - char alt_name[MAX_LONG_PATH]; - - if (!do_lstat(follow, file_name, buf)) - return 0; - - /* if file_name ended in a '/', Windows returned ENOENT; - * try again without trailing slashes - */ - if (errno != ENOENT) - return -1; - - namelen = strlen(file_name); - if (namelen && file_name[namelen-1] != '/') - return -1; - while (namelen && file_name[namelen-1] == '/') - --namelen; - if (!namelen || namelen >= MAX_LONG_PATH) - return -1; - - memcpy(alt_name, file_name, namelen); - alt_name[namelen] = 0; - return do_lstat(follow, alt_name, buf); -} - int (*lstat)(const char *file_name, struct stat *buf) = mingw_lstat; static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) @@ -1091,11 +1067,11 @@ static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) int mingw_lstat(const char *file_name, struct stat *buf) { - return do_stat_internal(0, file_name, buf); + return do_lstat(0, file_name, buf); } int mingw_stat(const char *file_name, struct stat *buf) { - return do_stat_internal(1, file_name, buf); + return do_lstat(1, file_name, buf); } int mingw_fstat(int fd, struct stat *buf) From 1f99439675260c276dbef81aa59aef90ea751cfb Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 16 May 2015 01:18:14 +0200 Subject: [PATCH 185/248] Win32: implement stat() with symlink support With respect to symlinks, the current stat() implementation is almost the same as lstat(): except for the file type (st_mode & S_IFMT), it returns information about the link rather than the target. Implement stat by opening the file with as little permissions as possible and calling GetFileInformationByHandle on it. This way, all link resoltion is handled by the Windows file system layer. If symlinks are disabled, use lstat() as before, but fail with ELOOP if a symlink would have to be resolved. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index b734b1d1f17938..349ce89ae6dfe7 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1069,9 +1069,26 @@ int mingw_lstat(const char *file_name, struct stat *buf) { return do_lstat(0, file_name, buf); } + int mingw_stat(const char *file_name, struct stat *buf) { - return do_lstat(1, file_name, buf); + wchar_t wfile_name[MAX_LONG_PATH]; + HANDLE hnd; + int result; + + /* open the file and let Windows resolve the links */ + if (xutftowcs_long_path(wfile_name, file_name) < 0) + return -1; + hnd = CreateFileW(wfile_name, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); + if (hnd == INVALID_HANDLE_VALUE) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + result = get_file_info_by_handle(hnd, buf); + CloseHandle(hnd); + return result; } int mingw_fstat(int fd, struct stat *buf) From 5f4a6a47cea9ee68c03bb1e078c36550642689a8 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 12 May 2015 00:58:39 +0200 Subject: [PATCH 186/248] Win32: remove separate do_lstat() function With the new mingw_stat() implementation, do_lstat() is only called from mingw_lstat() (with follow == 0). Remove the extra function and the old mingw_stat()-specific (follow == 1) logic. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 22 ++-------------------- 1 file changed, 2 insertions(+), 20 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 349ce89ae6dfe7..65a0859fd80eda 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -958,14 +958,7 @@ static int has_valid_directory_prefix(wchar_t *wfilename) return 1; } -/* We keep the do_lstat code in a separate function to avoid recursion. - * When a path ends with a slash, the stat will fail with ENOENT. In - * this case, we strip the trailing slashes and stat again. - * - * If follow is true then act like stat() and report on the link - * target. Otherwise report on the link itself. - */ -static int do_lstat(int follow, const char *file_name, struct stat *buf) +int mingw_lstat(const char *file_name, struct stat *buf) { WIN32_FILE_ATTRIBUTE_DATA fdata; wchar_t wfilename[MAX_LONG_PATH]; @@ -999,13 +992,7 @@ static int do_lstat(int follow, const char *file_name, struct stat *buf) if (handle != INVALID_HANDLE_VALUE) { if ((findbuf.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) && (findbuf.dwReserved0 == IO_REPARSE_TAG_SYMLINK)) { - if (follow) { - char buffer[MAXIMUM_REPARSE_DATA_BUFFER_SIZE]; - buf->st_size = readlink(file_name, buffer, MAXIMUM_REPARSE_DATA_BUFFER_SIZE); - } else { - buf->st_mode = S_IFLNK; - } - buf->st_mode |= S_IREAD; + buf->st_mode = S_IFLNK | S_IREAD; if (!(findbuf.dwFileAttributes & FILE_ATTRIBUTE_READONLY)) buf->st_mode |= S_IWRITE; } @@ -1065,11 +1052,6 @@ static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) return 0; } -int mingw_lstat(const char *file_name, struct stat *buf) -{ - return do_lstat(0, file_name, buf); -} - int mingw_stat(const char *file_name, struct stat *buf) { wchar_t wfile_name[MAX_LONG_PATH]; From f84ce7f5f4de030e6f35379bdc36303615824f29 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 00:17:56 +0200 Subject: [PATCH 187/248] Win32: let mingw_lstat() error early upon problems with reparse points When obtaining lstat information for reparse points, we need to call FindFirstFile() in addition to GetFileInformationEx() to obtain the type of the reparse point (symlink, mount point etc.). However, currently there is no error handling whatsoever if FindFirstFile() fails. Call FindFirstFile() before modifying the stat *buf output parameter and error out if the call fails. Note: The FindFirstFile() return value includes all the data that we get from GetFileAttributesEx(), so we could replace GetFileAttributesEx() with FindFirstFile(). We don't do that because GetFileAttributesEx() is about twice as fast for single files. I.e. we only pay the extra cost of calling FindFirstFile() in the rare case that we encounter a reparse point. Note: The indentation of the remaining reparse point code will be fixed in the next patch. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 65a0859fd80eda..abd1000d9158f0 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -961,6 +961,7 @@ static int has_valid_directory_prefix(wchar_t *wfilename) int mingw_lstat(const char *file_name, struct stat *buf) { WIN32_FILE_ATTRIBUTE_DATA fdata; + WIN32_FIND_DATAW findbuf = { 0 }; wchar_t wfilename[MAX_LONG_PATH]; int wlen = xutftowcs_long_path(wfilename, file_name); if (wlen < 0) @@ -975,6 +976,13 @@ int mingw_lstat(const char *file_name, struct stat *buf) } if (GetFileAttributesExW(wfilename, GetFileExInfoStandard, &fdata)) { + /* for reparse points, use FindFirstFile to get the reparse tag */ + if (fdata.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) { + HANDLE handle = FindFirstFileW(wfilename, &findbuf); + if (handle == INVALID_HANDLE_VALUE) + goto error; + FindClose(handle); + } buf->st_ino = 0; buf->st_gid = 0; buf->st_uid = 0; @@ -987,20 +995,16 @@ int mingw_lstat(const char *file_name, struct stat *buf) filetime_to_timespec(&(fdata.ftLastWriteTime), &(buf->st_mtim)); filetime_to_timespec(&(fdata.ftCreationTime), &(buf->st_ctim)); if (fdata.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) { - WIN32_FIND_DATAW findbuf; - HANDLE handle = FindFirstFileW(wfilename, &findbuf); - if (handle != INVALID_HANDLE_VALUE) { if ((findbuf.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) && (findbuf.dwReserved0 == IO_REPARSE_TAG_SYMLINK)) { buf->st_mode = S_IFLNK | S_IREAD; if (!(findbuf.dwFileAttributes & FILE_ATTRIBUTE_READONLY)) buf->st_mode |= S_IWRITE; } - FindClose(handle); - } } return 0; } +error: switch (GetLastError()) { case ERROR_ACCESS_DENIED: case ERROR_SHARING_VIOLATION: From 6772d603559e609fa0518b2abe935a676b485123 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Thu, 19 Mar 2015 16:33:44 +0100 Subject: [PATCH 188/248] mingw: Support `git_terminal_prompt` with more terminals The `git_terminal_prompt()` function expects the terminal window to be attached to a Win32 Console. However, this is not the case with terminal windows other than `cmd.exe`'s, e.g. with MSys2's own `mintty`. Non-cmd terminals such as `mintty` still have to have a Win32 Console to be proper console programs, but have to hide the Win32 Console to be able to provide more flexibility (such as being resizeable not only vertically but also horizontally). By writing to that Win32 Console, `git_terminal_prompt()` manages only to send the prompt to nowhere and to wait for input from a Console to which the user has no access. This commit introduces a function specifically to support `mintty` -- or other terminals that are compatible with MSys2's `/dev/tty` emulation. We use the `TERM` environment variable as an indicator for that: if the value starts with "xterm" (such as `mintty`'s "xterm_256color"), we prefer to let `xterm_prompt()` handle the user interaction. The most prominent user of `git_terminal_prompt()` is certainly `git-remote-https.exe`. It is an interesting use case because both `stdin` and `stdout` are redirected when Git calls said executable, yet it still wants to access the terminal. When running inside a `mintty`, the terminal is not accessible to the `git-remote-https.exe` program, though, because it is a MinGW program and the `mintty` terminal is not backed by a Win32 console. To solve that problem, we simply call out to the shell -- which is an *MSys2* program and can therefore access `/dev/tty`. Helped-by: nalla <nalla@hamal.uberspace.de> Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/terminal.c | 54 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 54 insertions(+) diff --git a/compat/terminal.c b/compat/terminal.c index 584f27bf7e1078..cdcde283644e41 100644 --- a/compat/terminal.c +++ b/compat/terminal.c @@ -418,6 +418,54 @@ static int getchar_with_timeout(int timeout) return getchar(); } +static char *shell_prompt(const char *prompt, int echo) +{ + const char *read_input[] = { + /* Note: call 'bash' explicitly, as 'read -s' is bash-specific */ + "bash", "-c", echo ? + "cat >/dev/tty && read -r line </dev/tty && echo \"$line\"" : + "cat >/dev/tty && read -r -s line </dev/tty && echo \"$line\" && echo >/dev/tty", + NULL + }; + struct child_process child = CHILD_PROCESS_INIT; + static struct strbuf buffer = STRBUF_INIT; + int prompt_len = strlen(prompt), len = -1, code; + + strvec_pushv(&child.args, read_input); + child.in = -1; + child.out = -1; + + if (start_command(&child)) + return NULL; + + if (write_in_full(child.in, prompt, prompt_len) != prompt_len) { + error("could not write to prompt script"); + close(child.in); + goto ret; + } + close(child.in); + + strbuf_reset(&buffer); + len = strbuf_read(&buffer, child.out, 1024); + if (len < 0) { + error("could not read from prompt script"); + goto ret; + } + + strbuf_strip_suffix(&buffer, "\n"); + strbuf_strip_suffix(&buffer, "\r"); + +ret: + close(child.out); + code = finish_command(&child); + if (code) { + error("failed to execute prompt script (exit code %d)", code); + return NULL; + } + + return len < 0 ? NULL : buffer.buf; +} + #endif #ifndef FORCE_TEXT @@ -429,6 +477,12 @@ char *git_terminal_prompt(const char *prompt, int echo) static struct strbuf buf = STRBUF_INIT; int r; FILE *input_fh, *output_fh; +#ifdef GIT_WINDOWS_NATIVE + const char *term = getenv("TERM"); + + if (term && starts_with(term, "xterm")) + return shell_prompt(prompt, echo); +#endif input_fh = fopen(INPUT_PATH, "r" FORCE_TEXT); if (!input_fh) From e5e4ea964f2972b11b096e79151f59d9cebc65a5 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 10 Jan 2017 23:21:56 +0100 Subject: [PATCH 189/248] mingw: teach fscache and dirent about symlinks Move S_IFLNK detection to file_attr_to_st_mode() and reuse it in fscache. Implement DT_LNK detection in dirent.c and the fscache readdir version. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 13 +++---------- compat/win32.h | 6 ++++-- compat/win32/dirent.c | 5 ++++- compat/win32/fscache.c | 11 +++++++---- 4 files changed, 18 insertions(+), 17 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index abd1000d9158f0..511de58092fd88 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -987,21 +987,14 @@ int mingw_lstat(const char *file_name, struct stat *buf) buf->st_gid = 0; buf->st_uid = 0; buf->st_nlink = 1; - buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes); + buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, + findbuf.dwReserved0); buf->st_size = fdata.nFileSizeLow | (((off_t)fdata.nFileSizeHigh)<<32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ filetime_to_timespec(&(fdata.ftLastAccessTime), &(buf->st_atim)); filetime_to_timespec(&(fdata.ftLastWriteTime), &(buf->st_mtim)); filetime_to_timespec(&(fdata.ftCreationTime), &(buf->st_ctim)); - if (fdata.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) { - if ((findbuf.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) && - (findbuf.dwReserved0 == IO_REPARSE_TAG_SYMLINK)) { - buf->st_mode = S_IFLNK | S_IREAD; - if (!(findbuf.dwFileAttributes & FILE_ATTRIBUTE_READONLY)) - buf->st_mode |= S_IWRITE; - } - } return 0; } error: @@ -1046,7 +1039,7 @@ static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) buf->st_gid = 0; buf->st_uid = 0; buf->st_nlink = 1; - buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes); + buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, 0); buf->st_size = fdata.nFileSizeLow | (((off_t)fdata.nFileSizeHigh)<<32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ diff --git a/compat/win32.h b/compat/win32.h index a97e880757b6f1..671bcc81f93351 100644 --- a/compat/win32.h +++ b/compat/win32.h @@ -6,10 +6,12 @@ #include <windows.h> #endif -static inline int file_attr_to_st_mode (DWORD attr) +static inline int file_attr_to_st_mode (DWORD attr, DWORD tag) { int fMode = S_IREAD; - if (attr & FILE_ATTRIBUTE_DIRECTORY) + if ((attr & FILE_ATTRIBUTE_REPARSE_POINT) && tag == IO_REPARSE_TAG_SYMLINK) + fMode |= S_IFLNK; + else if (attr & FILE_ATTRIBUTE_DIRECTORY) fMode |= S_IFDIR; else fMode |= S_IFREG; diff --git a/compat/win32/dirent.c b/compat/win32/dirent.c index c9fe2454efc01c..87063101f57202 100644 --- a/compat/win32/dirent.c +++ b/compat/win32/dirent.c @@ -18,7 +18,10 @@ static inline void finddata2dirent(struct dirent *ent, WIN32_FIND_DATAW *fdata) xwcstoutf(ent->d_name, fdata->cFileName, MAX_PATH * 3); /* Set file type, based on WIN32_FIND_DATA */ - if (fdata->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) + if ((fdata->dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) + && fdata->dwReserved0 == IO_REPARSE_TAG_SYMLINK) + ent->d_type = DT_LNK; + else if (fdata->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) ent->d_type = DT_DIR; else ent->d_type = DT_REG; diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index dbf640ca790fde..41fae636c12a41 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -207,10 +207,13 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, fdata->FileAttributes & FILE_ATTRIBUTE_REPARSE_POINT ? fdata->EaSize : 0; - fse->st_mode = file_attr_to_st_mode(fdata->FileAttributes); - fse->dirent.d_type = S_ISDIR(fse->st_mode) ? DT_DIR : DT_REG; - fse->u.s.st_size = fdata->EndOfFile.LowPart | - (((off_t)fdata->EndOfFile.HighPart) << 32); + fse->st_mode = file_attr_to_st_mode(fdata->FileAttributes, + fdata->EaSize); + fse->dirent.d_type = S_ISREG(fse->st_mode) ? DT_REG : + S_ISDIR(fse->st_mode) ? DT_DIR : DT_LNK; + fse->u.s.st_size = S_ISLNK(fse->st_mode) ? MAX_LONG_PATH : + fdata->EndOfFile.LowPart | + (((off_t)fdata->EndOfFile.HighPart) << 32); filetime_to_timespec((FILETIME *)&(fdata->LastAccessTime), &(fse->u.s.st_atim)); filetime_to_timespec((FILETIME *)&(fdata->LastWriteTime), From 4a7fae8509dd94353d50ea66f0d700d729286ccb Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 9 May 2015 02:11:48 +0200 Subject: [PATCH 190/248] compat/terminal.c: only use the Windows console if bash 'read -r' fails MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Accessing the Windows console through the special CONIN$ / CONOUT$ devices doesn't work properly for non-ASCII usernames an passwords. It also doesn't work for terminal emulators that hide the native console window (such as mintty), and 'TERM=xterm*' is not necessarily a reliable indicator for such terminals. The new shell_prompt() function, on the other hand, works fine for both MSys1 and MSys2, in native console windows as well as mintty, and properly supports Unicode. It just needs bash on the path (for 'read -s', which is bash-specific). On Windows, try to use the shell to read from the terminal. If that fails with ENOENT (i.e. bash was not found), use CONIN/OUT as fallback. Note: To test this, create a UTF-8 credential file with non-ASCII chars, e.g. in git-bash: 'echo url=http://täst.com > cred.txt'. Then in git-cmd, 'git credential fill <cred.txt' works (shell version), while calling git without the git-wrapper (i.e. 'mingw64\bin\git credential fill <cred.txt') mangles non-ASCII chars in both console output and input. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/terminal.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/compat/terminal.c b/compat/terminal.c index cdcde283644e41..a89c5cd9ccf604 100644 --- a/compat/terminal.c +++ b/compat/terminal.c @@ -434,6 +434,7 @@ static char *shell_prompt(const char *prompt, int echo) strvec_pushv(&child.args, read_input); child.in = -1; child.out = -1; + child.silent_exec_failure = 1; if (start_command(&child)) return NULL; @@ -477,11 +478,14 @@ char *git_terminal_prompt(const char *prompt, int echo) static struct strbuf buf = STRBUF_INIT; int r; FILE *input_fh, *output_fh; + #ifdef GIT_WINDOWS_NATIVE - const char *term = getenv("TERM"); - if (term && starts_with(term, "xterm")) - return shell_prompt(prompt, echo); + /* try shell_prompt first, fall back to CONIN/OUT if bash is missing */ + char *result = shell_prompt(prompt, echo); + if (result || errno != ENOENT) + return result; + #endif input_fh = fopen(INPUT_PATH, "r" FORCE_TEXT); From 51b009e87ece3ab7774412acf47393204ddcd1e4 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 16 May 2015 01:11:37 +0200 Subject: [PATCH 191/248] Win32: lstat(): return adequate stat.st_size for symlinks Git typically doesn't trust the stat.st_size member of symlinks (e.g. see strbuf_readlink()). However, some functions take shortcuts if st_size is 0 (e.g. diff_populate_filespec()). In mingw_lstat() and fscache_lstat(), make sure to return an adequate size. The extra overhead of opening and reading the reparse point to calculate the exact size is not necessary, as git doesn't rely on the value anyway. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 511de58092fd88..4b14fd2e0a3103 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -989,8 +989,8 @@ int mingw_lstat(const char *file_name, struct stat *buf) buf->st_nlink = 1; buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, findbuf.dwReserved0); - buf->st_size = fdata.nFileSizeLow | - (((off_t)fdata.nFileSizeHigh)<<32); + buf->st_size = S_ISLNK(buf->st_mode) ? MAX_LONG_PATH : + fdata.nFileSizeLow | (((off_t) fdata.nFileSizeHigh) << 32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ filetime_to_timespec(&(fdata.ftLastAccessTime), &(buf->st_atim)); filetime_to_timespec(&(fdata.ftLastWriteTime), &(buf->st_mtim)); From 9f9ce12788b83a59246240eebda7348094d02b74 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 23 Feb 2018 02:50:03 +0100 Subject: [PATCH 192/248] mingw (git_terminal_prompt): do fall back to CONIN$/CONOUT$ method To support Git Bash running in a MinTTY, we use a dirty trick to access the MSYS2 pseudo terminal: we execute a Bash snippet that accesses /dev/tty. The idea was to fall back to writing to/reading from CONOUT$/CONIN$ if that Bash call failed because Bash was not found. However, we should fall back even in other error conditions, because we have not successfully read the user input. Let's make it so. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/terminal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/compat/terminal.c b/compat/terminal.c index a89c5cd9ccf604..882b027e41e52b 100644 --- a/compat/terminal.c +++ b/compat/terminal.c @@ -483,7 +483,7 @@ char *git_terminal_prompt(const char *prompt, int echo) /* try shell_prompt first, fall back to CONIN/OUT if bash is missing */ char *result = shell_prompt(prompt, echo); - if (result || errno != ENOENT) + if (result) return result; #endif From d6e6e29907362802be23619a1b2f3398c0e303c9 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 19 May 2015 21:48:55 +0200 Subject: [PATCH 193/248] Win32: factor out retry logic The retry pattern is duplicated in three places. It also seems to be too hard to use: mingw_unlink() and mingw_rmdir() duplicate the code to retry, and both of them do so incompletely. They also do not restore errno if the user answers 'no'. Introduce a retry_ask_yes_no() helper function that handles retry with small delay, asking the user, and restoring errno. mingw_unlink: include _wchmod in the retry loop (which may fail if the file is locked exclusively). mingw_rmdir: include special error handling in the retry loop. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 104 ++++++++++++++++++++++--------------------------- 1 file changed, 46 insertions(+), 58 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 4b14fd2e0a3103..1586932c632704 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -30,8 +30,6 @@ #define HCAST(type, handle) ((type)(intptr_t)handle) -static const int delay[] = { 0, 1, 10, 20, 40 }; - void open_in_gdb(void) { static struct child_process cp = CHILD_PROCESS_INIT; @@ -207,15 +205,12 @@ static int read_yes_no_answer(void) return -1; } -static int ask_yes_no_if_possible(const char *format, ...) +static int ask_yes_no_if_possible(const char *format, va_list args) { char question[4096]; const char *retry_hook; - va_list args; - va_start(args, format); vsnprintf(question, sizeof(question), format, args); - va_end(args); retry_hook = mingw_getenv("GIT_ASK_YESNO"); if (retry_hook) { @@ -240,6 +235,31 @@ static int ask_yes_no_if_possible(const char *format, ...) } } +static int retry_ask_yes_no(int *tries, const char *format, ...) +{ + static const int delay[] = { 0, 1, 10, 20, 40 }; + va_list args; + int result, saved_errno = errno; + + if ((*tries) < ARRAY_SIZE(delay)) { + /* + * We assume that some other process had the file open at the wrong + * moment and retry. In order to give the other process a higher + * chance to complete its operation, we give up our time slice now. + * If we have to retry again, we do sleep a bit. + */ + Sleep(delay[*tries]); + (*tries)++; + return 1; + } + + va_start(args, format); + result = ask_yes_no_if_possible(format, args); + va_end(args); + errno = saved_errno; + return result; +} + /* Windows only */ enum hide_dotfiles_type { HIDE_DOTFILES_FALSE = 0, @@ -332,7 +352,7 @@ static wchar_t *normalize_ntpath(wchar_t *wbuf) int mingw_unlink(const char *pathname, int handle_in_use_error) { - int ret, tries = 0; + int tries = 0; wchar_t wpathname[MAX_LONG_PATH]; if (xutftowcs_long_path(wpathname, pathname) < 0) return -1; @@ -340,29 +360,19 @@ int mingw_unlink(const char *pathname, int handle_in_use_error) if (DeleteFileW(wpathname)) return 0; - /* read-only files cannot be removed */ - _wchmod(wpathname, 0666); - while ((ret = _wunlink(wpathname)) == -1 && tries < ARRAY_SIZE(delay)) { + do { + /* read-only files cannot be removed */ + _wchmod(wpathname, 0666); + if (!_wunlink(wpathname)) + return 0; if (!is_file_in_use_error(GetLastError())) break; if (!handle_in_use_error) - return ret; + return -1; - /* - * We assume that some other process had the source or - * destination file open at the wrong moment and retry. - * In order to give the other process a higher chance to - * complete its operation, we give up our time slice now. - * If we have to retry again, we do sleep a bit. - */ - Sleep(delay[tries]); - tries++; - } - while (ret == -1 && is_file_in_use_error(GetLastError()) && - ask_yes_no_if_possible("Unlink of file '%s' failed. " - "Should I try again?", pathname)) - ret = _wunlink(wpathname); - return ret; + } while (retry_ask_yes_no(&tries, "Unlink of file '%s' failed. " + "Should I try again?", pathname)); + return -1; } static int is_dir_empty(const wchar_t *wpath) @@ -389,7 +399,7 @@ static int is_dir_empty(const wchar_t *wpath) int mingw_rmdir(const char *pathname) { - int ret, tries = 0; + int tries = 0; wchar_t wpathname[MAX_LONG_PATH]; struct stat st; @@ -415,7 +425,11 @@ int mingw_rmdir(const char *pathname) if (xutftowcs_long_path(wpathname, pathname) < 0) return -1; - while ((ret = _wrmdir(wpathname)) == -1 && tries < ARRAY_SIZE(delay)) { + do { + if (!_wrmdir(wpathname)) { + invalidate_lstat_cache(); + return 0; + } if (!is_file_in_use_error(GetLastError())) errno = err_win_to_posix(GetLastError()); if (errno != EACCES) @@ -424,23 +438,9 @@ int mingw_rmdir(const char *pathname) errno = ENOTEMPTY; break; } - /* - * We assume that some other process had the source or - * destination file open at the wrong moment and retry. - * In order to give the other process a higher chance to - * complete its operation, we give up our time slice now. - * If we have to retry again, we do sleep a bit. - */ - Sleep(delay[tries]); - tries++; - } - while (ret == -1 && errno == EACCES && is_file_in_use_error(GetLastError()) && - ask_yes_no_if_possible("Deletion of directory '%s' failed. " - "Should I try again?", pathname)) - ret = _wrmdir(wpathname); - if (!ret) - invalidate_lstat_cache(); - return ret; + } while (retry_ask_yes_no(&tries, "Deletion of directory '%s' failed. " + "Should I try again?", pathname)); + return -1; } static inline int needs_hiding(const char *path) @@ -2652,20 +2652,8 @@ int mingw_rename(const char *pold, const char *pnew) SetFileAttributesW(wpnew, attrs); } } - if (tries < ARRAY_SIZE(delay) && gle == ERROR_ACCESS_DENIED) { - /* - * We assume that some other process had the source or - * destination file open at the wrong moment and retry. - * In order to give the other process a higher chance to - * complete its operation, we give up our time slice now. - * If we have to retry again, we do sleep a bit. - */ - Sleep(delay[tries]); - tries++; - goto repeat; - } if (gle == ERROR_ACCESS_DENIED && - ask_yes_no_if_possible("Rename from '%s' to '%s' failed. " + retry_ask_yes_no(&tries, "Rename from '%s' to '%s' failed. " "Should I try again?", pold, pnew)) goto repeat; From 94655c73defd484ed8add4d60e68a3181f877679 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:55:05 +0200 Subject: [PATCH 194/248] Win32: change default of 'core.symlinks' to false Symlinks on Windows don't work the same way as on Unix systems. E.g. there are different types of symlinks for directories and files, creating symlinks requires administrative privileges etc. By default, disable symlink support on Windows. I.e. users explicitly have to enable it with 'git config [--system|--global] core.symlinks true'. The test suite ignores system / global config files. Allow testing *with* symlink support by checking if native symlinks are enabled in MSys2 (via 'MSYS=winsymlinks:nativestrict'). Reminder: This would need to be changed if / when we find a way to run the test suite in a non-MSys-based shell (e.g. dash). Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 1586932c632704..a29732974a2c2d 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -3261,6 +3261,15 @@ static void setup_windows_environment(void) if (!getenv("LC_ALL") && !getenv("LC_CTYPE") && !getenv("LANG")) setenv("LC_CTYPE", "C.UTF-8", 1); + + /* + * Change 'core.symlinks' default to false, unless native symlinks are + * enabled in MSys2 (via 'MSYS=winsymlinks:nativestrict'). Thus we can + * run the test suite (which doesn't obey config files) with or without + * symlink support. + */ + if (!(tmp = getenv("MSYS")) || !strstr(tmp, "winsymlinks:nativestrict")) + has_symlinks = 0; } static void get_current_user_sid(PSID *sid, HANDLE *linked_token) From 79c8553e4d7b51d983a30831db2c926217eb7d8a Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sat, 16 May 2015 00:32:03 +0200 Subject: [PATCH 195/248] Win32: add symlink-specific error codes Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index a29732974a2c2d..097583feafbf6d 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -104,6 +104,7 @@ int err_win_to_posix(DWORD winerr) case ERROR_INVALID_PARAMETER: error = EINVAL; break; case ERROR_INVALID_PASSWORD: error = EPERM; break; case ERROR_INVALID_PRIMARY_GROUP: error = EINVAL; break; + case ERROR_INVALID_REPARSE_DATA: error = EINVAL; break; case ERROR_INVALID_SIGNAL_NUMBER: error = EINVAL; break; case ERROR_INVALID_TARGET_HANDLE: error = EIO; break; case ERROR_INVALID_WORKSTATION: error = EACCES; break; @@ -118,6 +119,7 @@ int err_win_to_posix(DWORD winerr) case ERROR_NEGATIVE_SEEK: error = ESPIPE; break; case ERROR_NOACCESS: error = EFAULT; break; case ERROR_NONE_MAPPED: error = EINVAL; break; + case ERROR_NOT_A_REPARSE_POINT: error = EINVAL; break; case ERROR_NOT_ENOUGH_MEMORY: error = ENOMEM; break; case ERROR_NOT_READY: error = EAGAIN; break; case ERROR_NOT_SAME_DEVICE: error = EXDEV; break; @@ -138,6 +140,9 @@ int err_win_to_posix(DWORD winerr) case ERROR_PIPE_NOT_CONNECTED: error = EPIPE; break; case ERROR_PRIVILEGE_NOT_HELD: error = EACCES; break; case ERROR_READ_FAULT: error = EIO; break; + case ERROR_REPARSE_ATTRIBUTE_CONFLICT: error = EINVAL; break; + case ERROR_REPARSE_TAG_INVALID: error = EINVAL; break; + case ERROR_REPARSE_TAG_MISMATCH: error = EINVAL; break; case ERROR_SEEK: error = EIO; break; case ERROR_SEEK_ON_DEVICE: error = ESPIPE; break; case ERROR_SHARING_BUFFER_EXCEEDED: error = ENFILE; break; From 98c4afe6ace7ba5c783390b802c3107ec404aefa Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:06:10 +0200 Subject: [PATCH 196/248] Win32: mingw_unlink: support symlinks to directories _wunlink() / DeleteFileW() refuses to delete symlinks to directories. If _wunlink() fails with ERROR_ACCESS_DENIED, try _wrmdir() as well. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 097583feafbf6d..c3a529d96746ce 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -372,9 +372,16 @@ int mingw_unlink(const char *pathname, int handle_in_use_error) return 0; if (!is_file_in_use_error(GetLastError())) break; + /* + * _wunlink() / DeleteFileW() for directory symlinks fails with + * ERROR_ACCESS_DENIED (EACCES), so try _wrmdir() as well. This is the + * same error we get if a file is in use (already checked above). + */ + if (!_wrmdir(wpathname)) + return 0; + if (!handle_in_use_error) return -1; - } while (retry_ask_yes_no(&tries, "Unlink of file '%s' failed. " "Should I try again?", pathname)); return -1; From 457dc39f010f86e34dbdc6af2f1b0efeea022655 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Tue, 19 May 2015 22:42:48 +0200 Subject: [PATCH 197/248] Win32: mingw_rename: support renaming symlinks MSVCRT's _wrename() cannot rename symlinks over existing files: it returns success without doing anything. Newer MSVCR*.dll versions probably do not have this problem: according to CRT sources, they just call MoveFileEx() with the MOVEFILE_COPY_ALLOWED flag. Get rid of _wrename() and call MoveFileEx() with proper error handling. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw.c | 38 ++++++++++++++++---------------------- 1 file changed, 16 insertions(+), 22 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index c3a529d96746ce..8dc015119a4cde 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2543,7 +2543,7 @@ int mingw_accept(int sockfd1, struct sockaddr *sa, socklen_t *sz) int mingw_rename(const char *pold, const char *pnew) { static int supports_file_rename_info_ex = 1; - DWORD attrs, gle; + DWORD attrs = INVALID_FILE_ATTRIBUTES, gle; int tries = 0; wchar_t wpold[MAX_LONG_PATH], wpnew[MAX_LONG_PATH]; int wpnew_len; @@ -2554,15 +2554,6 @@ int mingw_rename(const char *pold, const char *pnew) if (wpnew_len < 0) return -1; - /* - * Try native rename() first to get errno right. - * It is based on MoveFile(), which cannot overwrite existing files. - */ - if (!_wrename(wpold, wpnew)) - return 0; - if (errno != EEXIST) - return -1; - repeat: if (supports_file_rename_info_ex) { /* @@ -2638,13 +2629,22 @@ int mingw_rename(const char *pold, const char *pnew) * to retry. */ } else { - if (MoveFileExW(wpold, wpnew, MOVEFILE_REPLACE_EXISTING)) + if (MoveFileExW(wpold, wpnew, + MOVEFILE_REPLACE_EXISTING | MOVEFILE_COPY_ALLOWED)) return 0; gle = GetLastError(); } - /* TODO: translate more errors */ - if (gle == ERROR_ACCESS_DENIED && + /* revert file attributes on failure */ + if (attrs != INVALID_FILE_ATTRIBUTES) + SetFileAttributesW(wpnew, attrs); + + if (!is_file_in_use_error(gle)) { + errno = err_win_to_posix(gle); + return -1; + } + + if (attrs == INVALID_FILE_ATTRIBUTES && (attrs = GetFileAttributesW(wpnew)) != INVALID_FILE_ATTRIBUTES) { if (attrs & FILE_ATTRIBUTE_DIRECTORY) { DWORD attrsold = GetFileAttributesW(wpold); @@ -2656,16 +2656,10 @@ int mingw_rename(const char *pold, const char *pnew) return -1; } if ((attrs & FILE_ATTRIBUTE_READONLY) && - SetFileAttributesW(wpnew, attrs & ~FILE_ATTRIBUTE_READONLY)) { - if (MoveFileExW(wpold, wpnew, MOVEFILE_REPLACE_EXISTING)) - return 0; - gle = GetLastError(); - /* revert file attributes on failure */ - SetFileAttributesW(wpnew, attrs); - } + SetFileAttributesW(wpnew, attrs & ~FILE_ATTRIBUTE_READONLY)) + goto repeat; } - if (gle == ERROR_ACCESS_DENIED && - retry_ask_yes_no(&tries, "Rename from '%s' to '%s' failed. " + if (retry_ask_yes_no(&tries, "Rename from '%s' to '%s' failed. " "Should I try again?", pold, pnew)) goto repeat; From ade12d381386e24ac62d56df75bdfac0fbb3110c Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:17:31 +0200 Subject: [PATCH 198/248] Win32: mingw_chdir: change to symlink-resolved directory If symlinks are enabled, resolve all symlinks when changing directories, as required by POSIX. Note: Git's real_path() function bases its link resolution algorithm on this property of chdir(). Unfortunately, the current directory on Windows is limited to only MAX_PATH (260) characters. Therefore using symlinks and long paths in combination may be problematic. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 8dc015119a4cde..dd2a5db6a3b626 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -924,7 +924,24 @@ int mingw_chdir(const char *dirname) wchar_t wdirname[MAX_LONG_PATH]; if (xutftowcs_long_path(wdirname, dirname) < 0) return -1; - result = _wchdir(wdirname); + + if (has_symlinks) { + HANDLE hnd = CreateFileW(wdirname, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); + if (hnd == INVALID_HANDLE_VALUE) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + if (!GetFinalPathNameByHandleW(hnd, wdirname, ARRAY_SIZE(wdirname), 0)) { + errno = err_win_to_posix(GetLastError()); + CloseHandle(hnd); + return -1; + } + CloseHandle(hnd); + } + + result = _wchdir(normalize_ntpath(wdirname)); current_directory_len = GetCurrentDirectoryW(0, NULL); return result; } From 5ae7d2950a3304e683b3b85d7aeb92c16c138975 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:24:41 +0200 Subject: [PATCH 199/248] Win32: implement readlink() Implement readlink() by reading NTFS reparse points. Works for symlinks and directory junctions. If symlinks are disabled, fail with ENOSYS. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw-posix.h | 3 +- compat/mingw.c | 98 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+), 2 deletions(-) diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index 626e33280a1664..5d8853ac69aab8 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -121,8 +121,6 @@ struct utsname { * trivial stubs */ -static inline int readlink(const char *path UNUSED, char *buf UNUSED, size_t bufsiz UNUSED) -{ errno = ENOSYS; return -1; } static inline int symlink(const char *oldpath UNUSED, const char *newpath UNUSED) { errno = ENOSYS; return -1; } static inline int fchmod(int fildes UNUSED, mode_t mode UNUSED) @@ -197,6 +195,7 @@ int setitimer(int type, struct itimerval *in, struct itimerval *out); int sigaction(int sig, struct sigaction *in, struct sigaction *out); int link(const char *oldpath, const char *newpath); int uname(struct utsname *buf); +int readlink(const char *path, char *buf, size_t bufsiz); /* * replacements of existing functions diff --git a/compat/mingw.c b/compat/mingw.c index dd2a5db6a3b626..f6a0eb21724b2e 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -24,6 +24,7 @@ #define SECURITY_WIN32 #include <sspi.h> #include <wchar.h> +#include <winioctl.h> #include <winternl.h> #define STATUS_DELETE_PENDING ((NTSTATUS) 0xC0000056) @@ -2965,6 +2966,103 @@ int link(const char *oldpath, const char *newpath) return 0; } +#ifndef _WINNT_H +/* + * The REPARSE_DATA_BUFFER structure is defined in the Windows DDK (in + * ntifs.h) and in MSYS1's winnt.h (which defines _WINNT_H). So define + * it ourselves if we are on MSYS2 (whose winnt.h defines _WINNT_). + */ +typedef struct _REPARSE_DATA_BUFFER { + DWORD ReparseTag; + WORD ReparseDataLength; + WORD Reserved; +#ifndef _MSC_VER + _ANONYMOUS_UNION +#endif + union { + struct { + WORD SubstituteNameOffset; + WORD SubstituteNameLength; + WORD PrintNameOffset; + WORD PrintNameLength; + ULONG Flags; + WCHAR PathBuffer[1]; + } SymbolicLinkReparseBuffer; + struct { + WORD SubstituteNameOffset; + WORD SubstituteNameLength; + WORD PrintNameOffset; + WORD PrintNameLength; + WCHAR PathBuffer[1]; + } MountPointReparseBuffer; + struct { + BYTE DataBuffer[1]; + } GenericReparseBuffer; + } DUMMYUNIONNAME; +} REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER; +#endif + +int readlink(const char *path, char *buf, size_t bufsiz) +{ + HANDLE handle; + WCHAR wpath[MAX_LONG_PATH], *wbuf; + REPARSE_DATA_BUFFER *b = alloca(MAXIMUM_REPARSE_DATA_BUFFER_SIZE); + DWORD dummy; + char tmpbuf[MAX_LONG_PATH]; + int len; + + if (xutftowcs_long_path(wpath, path) < 0) + return -1; + + /* read reparse point data */ + handle = CreateFileW(wpath, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, + FILE_FLAG_BACKUP_SEMANTICS | FILE_FLAG_OPEN_REPARSE_POINT, NULL); + if (handle == INVALID_HANDLE_VALUE) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + if (!DeviceIoControl(handle, FSCTL_GET_REPARSE_POINT, NULL, 0, b, + MAXIMUM_REPARSE_DATA_BUFFER_SIZE, &dummy, NULL)) { + errno = err_win_to_posix(GetLastError()); + CloseHandle(handle); + return -1; + } + CloseHandle(handle); + + /* get target path for symlinks or mount points (aka 'junctions') */ + switch (b->ReparseTag) { + case IO_REPARSE_TAG_SYMLINK: + wbuf = (WCHAR*) (((char*) b->SymbolicLinkReparseBuffer.PathBuffer) + + b->SymbolicLinkReparseBuffer.SubstituteNameOffset); + *(WCHAR*) (((char*) wbuf) + + b->SymbolicLinkReparseBuffer.SubstituteNameLength) = 0; + break; + case IO_REPARSE_TAG_MOUNT_POINT: + wbuf = (WCHAR*) (((char*) b->MountPointReparseBuffer.PathBuffer) + + b->MountPointReparseBuffer.SubstituteNameOffset); + *(WCHAR*) (((char*) wbuf) + + b->MountPointReparseBuffer.SubstituteNameLength) = 0; + break; + default: + errno = EINVAL; + return -1; + } + + /* + * Adapt to strange readlink() API: Copy up to bufsiz *bytes*, potentially + * cutting off a UTF-8 sequence. Insufficient bufsize is *not* a failure + * condition. There is no conversion function that produces invalid UTF-8, + * so convert to a (hopefully large enough) temporary buffer, then memcpy + * the requested number of bytes (including '\0' for robustness). + */ + if ((len = xwcstoutf(tmpbuf, normalize_ntpath(wbuf), MAX_LONG_PATH)) < 0) + return -1; + memcpy(buf, tmpbuf, min(bufsiz, len + 1)); + return min(bufsiz, len); +} + pid_t waitpid(pid_t pid, int *status, int options) { HANDLE h = OpenProcess(SYNCHRONIZE | PROCESS_QUERY_INFORMATION, From 154ad49325aec710f2246a589145f33cd85c0673 Mon Sep 17 00:00:00 2001 From: Bill Zissimopoulos <billziss@navimatics.com> Date: Thu, 28 May 2020 16:35:57 -0700 Subject: [PATCH 200/248] mingw: lstat: compute correct size for symlinks This commit fixes mingw_lstat by computing the proper size for symlinks according to POSIX. POSIX specifies that upon successful return from lstat: "the value of the st_size member shall be set to the length of the pathname contained in the symbolic link not including any terminating null byte". Prior to this commit the mingw_lstat function returned a fixed size of 4096. This caused problems in git repositories that were accessed by git for Cygwin or git for WSL. For example, doing `git reset --hard` using git for Windows would update the size of symlinks in the index to be 4096; at a later time git for Cygwin or git for WSL would find that symlinks have changed size during `git status`. Vice versa doing `git reset --hard` in git for Cygwin or git for WSL would update the size of symlinks in the index with the correct value, only for git for Windows to find incorrectly at a later time that the size had changed. Signed-off-by: Bill Zissimopoulos <billziss@navimatics.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 65 ++++++++++++++++++++++++++++-------------- compat/win32/fscache.c | 12 ++++++++ 2 files changed, 56 insertions(+), 21 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index f6a0eb21724b2e..130a5392a2f8aa 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -988,10 +988,14 @@ static int has_valid_directory_prefix(wchar_t *wfilename) return 1; } +static int readlink_1(const WCHAR *wpath, BOOL fail_on_unknown_tag, + char *tmpbuf, int *plen, DWORD *ptag); + int mingw_lstat(const char *file_name, struct stat *buf) { WIN32_FILE_ATTRIBUTE_DATA fdata; - WIN32_FIND_DATAW findbuf = { 0 }; + DWORD reparse_tag = 0; + int link_len = 0; wchar_t wfilename[MAX_LONG_PATH]; int wlen = xutftowcs_long_path(wfilename, file_name); if (wlen < 0) @@ -1006,20 +1010,21 @@ int mingw_lstat(const char *file_name, struct stat *buf) } if (GetFileAttributesExW(wfilename, GetFileExInfoStandard, &fdata)) { - /* for reparse points, use FindFirstFile to get the reparse tag */ + /* for reparse points, get the link tag and length */ if (fdata.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) { - HANDLE handle = FindFirstFileW(wfilename, &findbuf); - if (handle == INVALID_HANDLE_VALUE) - goto error; - FindClose(handle); + char tmpbuf[MAX_LONG_PATH]; + + if (readlink_1(wfilename, FALSE, tmpbuf, &link_len, + &reparse_tag) < 0) + return -1; } buf->st_ino = 0; buf->st_gid = 0; buf->st_uid = 0; buf->st_nlink = 1; buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, - findbuf.dwReserved0); - buf->st_size = S_ISLNK(buf->st_mode) ? MAX_LONG_PATH : + reparse_tag); + buf->st_size = S_ISLNK(buf->st_mode) ? link_len : fdata.nFileSizeLow | (((off_t) fdata.nFileSizeHigh) << 32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ filetime_to_timespec(&(fdata.ftLastAccessTime), &(buf->st_atim)); @@ -1027,7 +1032,7 @@ int mingw_lstat(const char *file_name, struct stat *buf) filetime_to_timespec(&(fdata.ftCreationTime), &(buf->st_ctim)); return 0; } -error: + switch (GetLastError()) { case ERROR_ACCESS_DENIED: case ERROR_SHARING_VIOLATION: @@ -3002,17 +3007,13 @@ typedef struct _REPARSE_DATA_BUFFER { } REPARSE_DATA_BUFFER, *PREPARSE_DATA_BUFFER; #endif -int readlink(const char *path, char *buf, size_t bufsiz) +static int readlink_1(const WCHAR *wpath, BOOL fail_on_unknown_tag, + char *tmpbuf, int *plen, DWORD *ptag) { HANDLE handle; - WCHAR wpath[MAX_LONG_PATH], *wbuf; + WCHAR *wbuf; REPARSE_DATA_BUFFER *b = alloca(MAXIMUM_REPARSE_DATA_BUFFER_SIZE); DWORD dummy; - char tmpbuf[MAX_LONG_PATH]; - int len; - - if (xutftowcs_long_path(wpath, path) < 0) - return -1; /* read reparse point data */ handle = CreateFileW(wpath, 0, @@ -3032,7 +3033,7 @@ int readlink(const char *path, char *buf, size_t bufsiz) CloseHandle(handle); /* get target path for symlinks or mount points (aka 'junctions') */ - switch (b->ReparseTag) { + switch ((*ptag = b->ReparseTag)) { case IO_REPARSE_TAG_SYMLINK: wbuf = (WCHAR*) (((char*) b->SymbolicLinkReparseBuffer.PathBuffer) + b->SymbolicLinkReparseBuffer.SubstituteNameOffset); @@ -3046,10 +3047,34 @@ int readlink(const char *path, char *buf, size_t bufsiz) + b->MountPointReparseBuffer.SubstituteNameLength) = 0; break; default: - errno = EINVAL; - return -1; + if (fail_on_unknown_tag) { + errno = EINVAL; + return -1; + } else { + *plen = MAX_LONG_PATH; + return 0; + } } + if ((*plen = + xwcstoutf(tmpbuf, normalize_ntpath(wbuf), MAX_LONG_PATH)) < 0) + return -1; + return 0; +} + +int readlink(const char *path, char *buf, size_t bufsiz) +{ + WCHAR wpath[MAX_LONG_PATH]; + char tmpbuf[MAX_LONG_PATH]; + int len; + DWORD tag; + + if (xutftowcs_long_path(wpath, path) < 0) + return -1; + + if (readlink_1(wpath, TRUE, tmpbuf, &len, &tag) < 0) + return -1; + /* * Adapt to strange readlink() API: Copy up to bufsiz *bytes*, potentially * cutting off a UTF-8 sequence. Insufficient bufsize is *not* a failure @@ -3057,8 +3082,6 @@ int readlink(const char *path, char *buf, size_t bufsiz) * so convert to a (hopefully large enough) temporary buffer, then memcpy * the requested number of bytes (including '\0' for robustness). */ - if ((len = xwcstoutf(tmpbuf, normalize_ntpath(wbuf), MAX_LONG_PATH)) < 0) - return -1; memcpy(buf, tmpbuf, min(bufsiz, len + 1)); return min(bufsiz, len); } diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 41fae636c12a41..0f5e00ae18f949 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -594,6 +594,18 @@ int fscache_lstat(const char *filename, struct stat *st) return -1; } + /* + * Special case symbolic links: FindFirstFile()/FindNextFile() did not + * provide us with the length of the target path. + */ + if (fse->u.s.st_size == MAX_LONG_PATH && S_ISLNK(fse->st_mode)) { + char buf[MAX_LONG_PATH]; + int len = readlink(filename, buf, sizeof(buf) - 1); + + if (len > 0) + fse->u.s.st_size = len; + } + /* copy stat data */ st->st_ino = 0; st->st_gid = 0; From 459c1cc287f91a11e43423b8797ae4205d613e1a Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:32:03 +0200 Subject: [PATCH 201/248] Win32: implement basic symlink() functionality (file symlinks only) Implement symlink() that always creates file symlinks. Fails with ENOSYS if symlinks are disabled or unsupported. Note: CreateSymbolicLinkW() was introduced with symlink support in Windows Vista. For compatibility with Windows XP, we need to load it dynamically and fail gracefully if it isnt's available. Signed-off-by: Karsten Blees <blees@dcon.de> --- compat/mingw-posix.h | 3 +-- compat/mingw.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 2 deletions(-) diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index 5d8853ac69aab8..1f1ea70b048e27 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -121,8 +121,6 @@ struct utsname { * trivial stubs */ -static inline int symlink(const char *oldpath UNUSED, const char *newpath UNUSED) -{ errno = ENOSYS; return -1; } static inline int fchmod(int fildes UNUSED, mode_t mode UNUSED) { errno = ENOSYS; return -1; } #ifndef __MINGW64_VERSION_MAJOR @@ -195,6 +193,7 @@ int setitimer(int type, struct itimerval *in, struct itimerval *out); int sigaction(int sig, struct sigaction *in, struct sigaction *out); int link(const char *oldpath, const char *newpath); int uname(struct utsname *buf); +int symlink(const char *target, const char *link); int readlink(const char *path, char *buf, size_t bufsiz); /* diff --git a/compat/mingw.c b/compat/mingw.c index 130a5392a2f8aa..626ee8bfd3dbc1 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2971,6 +2971,34 @@ int link(const char *oldpath, const char *newpath) return 0; } +int symlink(const char *target, const char *link) +{ + wchar_t wtarget[MAX_LONG_PATH], wlink[MAX_LONG_PATH]; + int len; + + /* fail if symlinks are disabled or API is not supported (WinXP) */ + if (!has_symlinks) { + errno = ENOSYS; + return -1; + } + + if ((len = xutftowcs_long_path(wtarget, target)) < 0 + || xutftowcs_long_path(wlink, link) < 0) + return -1; + + /* convert target dir separators to backslashes */ + while (len--) + if (wtarget[len] == '/') + wtarget[len] = '\\'; + + /* create file symlink */ + if (!CreateSymbolicLinkW(wlink, wtarget, 0)) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + return 0; +} + #ifndef _WINNT_H /* * The REPARSE_DATA_BUFFER structure is defined in the Windows DDK (in From 4f7508555eee2f6a1b03d4fb49ca4d7f81fb5259 Mon Sep 17 00:00:00 2001 From: Karsten Blees <blees@dcon.de> Date: Sun, 24 May 2015 01:48:35 +0200 Subject: [PATCH 202/248] Win32: symlink: add support for symlinks to directories Symlinks on Windows have a flag that indicates whether the target is a file or a directory. Symlinks of wrong type simply don't work. This even affects core Win32 APIs (e.g. DeleteFile() refuses to delete directory symlinks). However, CreateFile() with FILE_FLAG_BACKUP_SEMANTICS doesn't seem to care. Check the target type by first creating a tentative file symlink, opening it, and checking the type of the resulting handle. If it is a directory, recreate the symlink with the directory flag set. It is possible to create symlinks before the target exists (or in case of symlinks to symlinks: before the target type is known). If this happens, create a tentative file symlink and postpone the directory decision: keep a list of phantom symlinks to be processed whenever a new directory is created in mingw_mkdir(). Limitations: This algorithm may fail if a link target changes from file to directory or vice versa, or if the target directory is created in another process. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 159 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 159 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 626ee8bfd3dbc1..7045eafcf94211 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -331,6 +331,126 @@ static inline int is_wdir_sep(wchar_t wchar) return wchar == L'/' || wchar == L'\\'; } +static const wchar_t *make_relative_to(const wchar_t *path, + const wchar_t *relative_to, wchar_t *out, + size_t size) +{ + size_t i = wcslen(relative_to), len; + + /* Is `path` already absolute? */ + if (is_wdir_sep(path[0]) || + (iswalpha(path[0]) && path[1] == L':' && is_wdir_sep(path[2]))) + return path; + + while (i > 0 && !is_wdir_sep(relative_to[i - 1])) + i--; + + /* Is `relative_to` in the current directory? */ + if (!i) + return path; + + len = wcslen(path); + if (i + len + 1 > size) { + error("Could not make '%ls' relative to '%ls' (too large)", + path, relative_to); + return NULL; + } + + memcpy(out, relative_to, i * sizeof(wchar_t)); + wcscpy(out + i, path); + return out; +} + +enum phantom_symlink_result { + PHANTOM_SYMLINK_RETRY, + PHANTOM_SYMLINK_DONE, + PHANTOM_SYMLINK_DIRECTORY +}; + +/* + * Changes a file symlink to a directory symlink if the target exists and is a + * directory. + */ +static enum phantom_symlink_result +process_phantom_symlink(const wchar_t *wtarget, const wchar_t *wlink) +{ + HANDLE hnd; + BY_HANDLE_FILE_INFORMATION fdata; + wchar_t relative[MAX_LONG_PATH]; + const wchar_t *rel; + + /* check that wlink is still a file symlink */ + if ((GetFileAttributesW(wlink) + & (FILE_ATTRIBUTE_REPARSE_POINT | FILE_ATTRIBUTE_DIRECTORY)) + != FILE_ATTRIBUTE_REPARSE_POINT) + return PHANTOM_SYMLINK_DONE; + + /* make it relative, if necessary */ + rel = make_relative_to(wtarget, wlink, relative, ARRAY_SIZE(relative)); + if (!rel) + return PHANTOM_SYMLINK_DONE; + + /* let Windows resolve the link by opening it */ + hnd = CreateFileW(rel, 0, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, + OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); + if (hnd == INVALID_HANDLE_VALUE) { + errno = err_win_to_posix(GetLastError()); + return PHANTOM_SYMLINK_RETRY; + } + + if (!GetFileInformationByHandle(hnd, &fdata)) { + errno = err_win_to_posix(GetLastError()); + CloseHandle(hnd); + return PHANTOM_SYMLINK_RETRY; + } + CloseHandle(hnd); + + /* if target exists and is a file, we're done */ + if (!(fdata.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY)) + return PHANTOM_SYMLINK_DONE; + + /* otherwise recreate the symlink with directory flag */ + if (DeleteFileW(wlink) && CreateSymbolicLinkW(wlink, wtarget, 1)) + return PHANTOM_SYMLINK_DIRECTORY; + + errno = err_win_to_posix(GetLastError()); + return PHANTOM_SYMLINK_RETRY; +} + +/* keep track of newly created symlinks to non-existing targets */ +struct phantom_symlink_info { + struct phantom_symlink_info *next; + wchar_t *wlink; + wchar_t *wtarget; +}; + +static struct phantom_symlink_info *phantom_symlinks = NULL; +static CRITICAL_SECTION phantom_symlinks_cs; + +static void process_phantom_symlinks(void) +{ + struct phantom_symlink_info *current, **psi; + EnterCriticalSection(&phantom_symlinks_cs); + /* process phantom symlinks list */ + psi = &phantom_symlinks; + while ((current = *psi)) { + enum phantom_symlink_result result = process_phantom_symlink( + current->wtarget, current->wlink); + if (result == PHANTOM_SYMLINK_RETRY) { + psi = ¤t->next; + } else { + /* symlink was processed, remove from list */ + *psi = current->next; + free(current); + /* if symlink was a directory, start over */ + if (result == PHANTOM_SYMLINK_DIRECTORY) + psi = &phantom_symlinks; + } + } + LeaveCriticalSection(&phantom_symlinks_cs); +} + /* Normalizes NT paths as returned by some low-level APIs. */ static wchar_t *normalize_ntpath(wchar_t *wbuf) { @@ -517,6 +637,8 @@ int mingw_mkdir(const char *path, int mode UNUSED) return -1; ret = _wmkdir(wpath); + if (!ret) + process_phantom_symlinks(); if (!ret && needs_hiding(path)) return set_hidden_flag(wpath, 1); return ret; @@ -2996,6 +3118,42 @@ int symlink(const char *target, const char *link) errno = err_win_to_posix(GetLastError()); return -1; } + + /* convert to directory symlink if target exists */ + switch (process_phantom_symlink(wtarget, wlink)) { + case PHANTOM_SYMLINK_RETRY: { + /* if target doesn't exist, add to phantom symlinks list */ + wchar_t wfullpath[MAX_LONG_PATH]; + struct phantom_symlink_info *psi; + + /* convert to absolute path to be independent of cwd */ + len = GetFullPathNameW(wlink, MAX_LONG_PATH, wfullpath, NULL); + if (!len || len >= MAX_LONG_PATH) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + + /* over-allocate and fill phantom_symlink_info structure */ + psi = xmalloc(sizeof(struct phantom_symlink_info) + + sizeof(wchar_t) * (len + wcslen(wtarget) + 2)); + psi->wlink = (wchar_t *)(psi + 1); + wcscpy(psi->wlink, wfullpath); + psi->wtarget = psi->wlink + len + 1; + wcscpy(psi->wtarget, wtarget); + + EnterCriticalSection(&phantom_symlinks_cs); + psi->next = phantom_symlinks; + phantom_symlinks = psi; + LeaveCriticalSection(&phantom_symlinks_cs); + break; + } + case PHANTOM_SYMLINK_DIRECTORY: + /* if we created a dir symlink, process other phantom symlinks */ + process_phantom_symlinks(); + break; + default: + break; + } return 0; } @@ -3985,6 +4143,7 @@ int wmain(int argc, const wchar_t **wargv) /* initialize critical section for waitpid pinfo_t list */ InitializeCriticalSection(&pinfo_cs); + InitializeCriticalSection(&phantom_symlinks_cs); /* initialize critical section for fscache */ InitializeCriticalSection(&fscache_cs); From 4ab296f5f398df2c8eadc3c7962a80057f641895 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 30 May 2017 21:50:57 +0200 Subject: [PATCH 203/248] mingw: try to create symlinks without elevated permissions With Windows 10 Build 14972 in Developer Mode, a new flag is supported by CreateSymbolicLink() to create symbolic links even when running outside of an elevated session (which was previously required). This new flag is called SYMBOLIC_LINK_FLAG_ALLOW_UNPRIVILEGED_CREATE and has the numeric value 0x02. Previous Windows 10 versions will not understand that flag and return an ERROR_INVALID_PARAMETER, therefore we have to be careful to try passing that flag only when the build number indicates that it is supported. For more information about the new flag, see this blog post: https://blogs.windows.com/buildingapps/2016/12/02/symlinks-windows-10/ This patch is loosely based on the patch submitted by Samuel D. Leslie as https://github.com/git-for-windows/git/pull/1184. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 7045eafcf94211..e71ab9361ce81b 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -361,6 +361,8 @@ static const wchar_t *make_relative_to(const wchar_t *path, return out; } +static DWORD symlink_file_flags = 0, symlink_directory_flags = 1; + enum phantom_symlink_result { PHANTOM_SYMLINK_RETRY, PHANTOM_SYMLINK_DONE, @@ -411,7 +413,8 @@ process_phantom_symlink(const wchar_t *wtarget, const wchar_t *wlink) return PHANTOM_SYMLINK_DONE; /* otherwise recreate the symlink with directory flag */ - if (DeleteFileW(wlink) && CreateSymbolicLinkW(wlink, wtarget, 1)) + if (DeleteFileW(wlink) && + CreateSymbolicLinkW(wlink, wtarget, symlink_directory_flags)) return PHANTOM_SYMLINK_DIRECTORY; errno = err_win_to_posix(GetLastError()); @@ -3114,7 +3117,7 @@ int symlink(const char *target, const char *link) wtarget[len] = '\\'; /* create file symlink */ - if (!CreateSymbolicLinkW(wlink, wtarget, 0)) { + if (!CreateSymbolicLinkW(wlink, wtarget, symlink_file_flags)) { errno = err_win_to_posix(GetLastError()); return -1; } @@ -4078,6 +4081,24 @@ static void maybe_redirect_std_handles(void) GENERIC_WRITE, FILE_FLAG_NO_BUFFERING); } +static void adjust_symlink_flags(void) +{ + /* + * Starting with Windows 10 Build 14972, symbolic links can be created + * using CreateSymbolicLink() without elevation by passing the flag + * SYMBOLIC_LINK_FLAG_ALLOW_UNPRIVILEGED_CREATE (0x02) as last + * parameter, provided the Developer Mode has been enabled. Some + * earlier Windows versions complain about this flag with an + * ERROR_INVALID_PARAMETER, hence we have to test the build number + * specifically. + */ + if (GetVersion() >= 14972 << 16) { + symlink_file_flags |= 2; + symlink_directory_flags |= 2; + } + +} + #ifdef _MSC_VER #ifdef _DEBUG #include <crtdbg.h> @@ -4113,6 +4134,7 @@ int wmain(int argc, const wchar_t **wargv) #endif maybe_redirect_std_handles(); + adjust_symlink_flags(); fsync_object_files = 1; /* determine size of argv and environ conversion buffer */ From 8ac5c585d015f99c55e821677e325ec71e4b7d76 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 2 Mar 2020 21:54:29 +0100 Subject: [PATCH 204/248] mingw: emulate stat() a little more faithfully When creating directories via `safe_create_leading_directories()`, we might encounter an already-existing directory which is not readable by the current user. To handle that situation, Git's code calls `stat()` to determine whether we're looking at a directory. In such a case, `CreateFile()` will fail, though, no matter what, and consequently `mingw_stat()` will fail, too. But POSIX semantics seem to still allow `stat()` to go forward. So let's call `mingw_lstat()` for the rescue if we fail to get a file handle due to denied permission in `mingw_stat()`, and fill the stat info that way. We need to be careful to not allow this to go forward in case that we're looking at a symbolic link: to resolve the link, we would still have to create a file handle, and we just found out that we cannot. Therefore, `stat()` still needs to fail with `EACCES` in that case. This fixes https://github.com/git-for-windows/git/issues/2531. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index e71ab9361ce81b..0673b4a121ad07 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1222,7 +1222,19 @@ int mingw_stat(const char *file_name, struct stat *buf) FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, NULL, OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL); if (hnd == INVALID_HANDLE_VALUE) { - errno = err_win_to_posix(GetLastError()); + DWORD err = GetLastError(); + + if (err == ERROR_ACCESS_DENIED && + !mingw_lstat(file_name, buf) && + !S_ISLNK(buf->st_mode)) + /* + * POSIX semantics state to still try to fill + * information, even if permission is denied to create + * a file handle. + */ + return 0; + + errno = err_win_to_posix(err); return -1; } result = get_file_info_by_handle(hnd, buf); From 340b04ad076b8e5e2d3ead76de51b67a2abd18f6 Mon Sep 17 00:00:00 2001 From: JiSeop Moon <zcube@zcube.kr> Date: Mon, 23 Apr 2018 22:30:18 +0900 Subject: [PATCH 205/248] mingw: introduce code to detect whether we're inside a Windows container This will come in handy in the next commit. Signed-off-by: JiSeop Moon <zcube@zcube.kr> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 32 ++++++++++++++++++++++++++++++++ compat/mingw.h | 5 +++++ 2 files changed, 37 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 0673b4a121ad07..095741e7bb4e87 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -4238,3 +4238,35 @@ int mingw_have_unix_sockets(void) return ret; } #endif + +/* + * Based on https://stackoverflow.com/questions/43002803 + * + * [HKLM\SYSTEM\CurrentControlSet\Services\cexecsvc] + * "DisplayName"="@%systemroot%\\system32\\cexecsvc.exe,-100" + * "ErrorControl"=dword:00000001 + * "ImagePath"=hex(2):25,00,73,00,79,00,73,00,74,00,65,00,6d,00,72,00,6f,00, + * 6f,00,74,00,25,00,5c,00,73,00,79,00,73,00,74,00,65,00,6d,00,33,00,32,00, + * 5c,00,63,00,65,00,78,00,65,00,63,00,73,00,76,00,63,00,2e,00,65,00,78,00, + * 65,00,00,00 + * "Start"=dword:00000002 + * "Type"=dword:00000010 + * "Description"="@%systemroot%\\system32\\cexecsvc.exe,-101" + * "ObjectName"="LocalSystem" + * "ServiceSidType"=dword:00000001 + */ +int is_inside_windows_container(void) +{ + static int inside_container = -1; /* -1 uninitialized */ + const char *key = "SYSTEM\\CurrentControlSet\\Services\\cexecsvc"; + HKEY handle = NULL; + + if (inside_container != -1) + return inside_container; + + inside_container = ERROR_SUCCESS == + RegOpenKeyExA(HKEY_LOCAL_MACHINE, key, 0, KEY_READ, &handle); + RegCloseKey(handle); + + return inside_container; +} diff --git a/compat/mingw.h b/compat/mingw.h index ad1166b775322a..807ee7b7e2e573 100644 --- a/compat/mingw.h +++ b/compat/mingw.h @@ -288,3 +288,8 @@ int mingw_have_unix_sockets(void); #undef have_unix_sockets #define have_unix_sockets mingw_have_unix_sockets #endif + +/* + * Check current process is inside Windows Container. + */ +int is_inside_windows_container(void); From 1e51c2ed65b4d00cfdfb399778bdbfba92c307b6 Mon Sep 17 00:00:00 2001 From: Bert Belder <bertbelder@gmail.com> Date: Fri, 26 Oct 2018 11:13:45 +0200 Subject: [PATCH 206/248] Win32: symlink: move phantom symlink creation to a separate function Signed-off-by: Bert Belder <bertbelder@gmail.com> --- compat/mingw.c | 91 +++++++++++++++++++++++++++----------------------- 1 file changed, 49 insertions(+), 42 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 0673b4a121ad07..6e8cadc1114931 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -454,6 +454,54 @@ static void process_phantom_symlinks(void) LeaveCriticalSection(&phantom_symlinks_cs); } +static int create_phantom_symlink(wchar_t *wtarget, wchar_t *wlink) +{ + int len; + + /* create file symlink */ + if (!CreateSymbolicLinkW(wlink, wtarget, symlink_file_flags)) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + + /* convert to directory symlink if target exists */ + switch (process_phantom_symlink(wtarget, wlink)) { + case PHANTOM_SYMLINK_RETRY: { + /* if target doesn't exist, add to phantom symlinks list */ + wchar_t wfullpath[MAX_LONG_PATH]; + struct phantom_symlink_info *psi; + + /* convert to absolute path to be independent of cwd */ + len = GetFullPathNameW(wlink, MAX_LONG_PATH, wfullpath, NULL); + if (!len || len >= MAX_LONG_PATH) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + + /* over-allocate and fill phantom_symlink_info structure */ + psi = xmalloc(sizeof(struct phantom_symlink_info) + + sizeof(wchar_t) * (len + wcslen(wtarget) + 2)); + psi->wlink = (wchar_t *)(psi + 1); + wcscpy(psi->wlink, wfullpath); + psi->wtarget = psi->wlink + len + 1; + wcscpy(psi->wtarget, wtarget); + + EnterCriticalSection(&phantom_symlinks_cs); + psi->next = phantom_symlinks; + phantom_symlinks = psi; + LeaveCriticalSection(&phantom_symlinks_cs); + break; + } + case PHANTOM_SYMLINK_DIRECTORY: + /* if we created a dir symlink, process other phantom symlinks */ + process_phantom_symlinks(); + break; + default: + break; + } + return 0; +} + /* Normalizes NT paths as returned by some low-level APIs. */ static wchar_t *normalize_ntpath(wchar_t *wbuf) { @@ -3128,48 +3176,7 @@ int symlink(const char *target, const char *link) if (wtarget[len] == '/') wtarget[len] = '\\'; - /* create file symlink */ - if (!CreateSymbolicLinkW(wlink, wtarget, symlink_file_flags)) { - errno = err_win_to_posix(GetLastError()); - return -1; - } - - /* convert to directory symlink if target exists */ - switch (process_phantom_symlink(wtarget, wlink)) { - case PHANTOM_SYMLINK_RETRY: { - /* if target doesn't exist, add to phantom symlinks list */ - wchar_t wfullpath[MAX_LONG_PATH]; - struct phantom_symlink_info *psi; - - /* convert to absolute path to be independent of cwd */ - len = GetFullPathNameW(wlink, MAX_LONG_PATH, wfullpath, NULL); - if (!len || len >= MAX_LONG_PATH) { - errno = err_win_to_posix(GetLastError()); - return -1; - } - - /* over-allocate and fill phantom_symlink_info structure */ - psi = xmalloc(sizeof(struct phantom_symlink_info) - + sizeof(wchar_t) * (len + wcslen(wtarget) + 2)); - psi->wlink = (wchar_t *)(psi + 1); - wcscpy(psi->wlink, wfullpath); - psi->wtarget = psi->wlink + len + 1; - wcscpy(psi->wtarget, wtarget); - - EnterCriticalSection(&phantom_symlinks_cs); - psi->next = phantom_symlinks; - phantom_symlinks = psi; - LeaveCriticalSection(&phantom_symlinks_cs); - break; - } - case PHANTOM_SYMLINK_DIRECTORY: - /* if we created a dir symlink, process other phantom symlinks */ - process_phantom_symlinks(); - break; - default: - break; - } - return 0; + return create_phantom_symlink(wtarget, wlink); } #ifndef _WINNT_H From 854bdc6488dd00c529866c06122dfef770870555 Mon Sep 17 00:00:00 2001 From: JiSeop Moon <zcube@zcube.kr> Date: Mon, 23 Apr 2018 22:31:42 +0200 Subject: [PATCH 207/248] mingw: when running in a Windows container, try to rename() harder It is a known issue that a rename() can fail with an "Access denied" error at times, when copying followed by deleting the original file works. Let's just fall back to that behavior. Signed-off-by: JiSeop Moon <zcube@zcube.kr> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 095741e7bb4e87..74804c585b0bcc 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2795,6 +2795,13 @@ int mingw_rename(const char *pold, const char *pnew) gle = GetLastError(); } + if (gle == ERROR_ACCESS_DENIED && is_inside_windows_container()) { + /* Fall back to copy to destination & remove source */ + if (CopyFileW(wpold, wpnew, FALSE) && !mingw_unlink(pold)) + return 0; + gle = GetLastError(); + } + /* revert file attributes on failure */ if (attrs != INVALID_FILE_ATTRIBUTES) SetFileAttributesW(wpnew, attrs); From 03f9e726cd4d2c51fc3ee29b583ec6dba03d1750 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 11 Feb 2019 14:19:18 +0100 Subject: [PATCH 208/248] Introduce helper to create symlinks that knows about index_state On Windows, symbolic links actually have a type depending on the target: it can be a file or a directory. In certain circumstances, this poses problems, e.g. when a symbolic link is supposed to point into a submodule that is not checked out, so there is no way for Git to auto-detect the type. To help with that, we will add support over the course of the next commits to specify that symlink type via the Git attributes. This requires an index_state, though, something that Git for Windows' `symlink()` replacement cannot know about because the function signature is defined by the POSIX standard and not ours to change. So let's introduce a helper function to create symbolic links that *does* know about the index_state. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- apply.c | 2 +- builtin/difftool.c | 2 +- compat/mingw-posix.h | 4 +++- compat/mingw.c | 2 +- entry.c | 2 +- git-compat-util.h | 10 ++++++++++ refs/files-backend.c | 2 +- setup.c | 4 ++-- 8 files changed, 20 insertions(+), 8 deletions(-) diff --git a/apply.c b/apply.c index a2ceb3fb40d3b5..bf1260b51d2c2d 100644 --- a/apply.c +++ b/apply.c @@ -4401,7 +4401,7 @@ static int try_create_file(struct apply_state *state, const char *path, /* Although buf:size is counted string, it also is NUL * terminated. */ - return !!symlink(buf, path); + return !!create_symlink(state && state->repo ? state->repo->index : NULL, buf, path); fd = open(path, O_CREAT | O_EXCL | O_WRONLY, (mode & 0100) ? 0777 : 0666); if (fd < 0) diff --git a/builtin/difftool.c b/builtin/difftool.c index e4bc1f831696a8..8d10e2489f088e 100644 --- a/builtin/difftool.c +++ b/builtin/difftool.c @@ -544,7 +544,7 @@ static int run_dir_diff(struct repository *repo, } add_path(&wtdir, wtdir_len, dst_path); if (dt_options->symlinks) { - if (symlink(wtdir.buf, rdir.buf)) { + if (create_symlink(lstate.istate, wtdir.buf, rdir.buf)) { ret = error_errno("could not symlink '%s' to '%s'", wtdir.buf, rdir.buf); goto finish; } diff --git a/compat/mingw-posix.h b/compat/mingw-posix.h index 1f1ea70b048e27..5c168c8a10d811 100644 --- a/compat/mingw-posix.h +++ b/compat/mingw-posix.h @@ -193,8 +193,10 @@ int setitimer(int type, struct itimerval *in, struct itimerval *out); int sigaction(int sig, struct sigaction *in, struct sigaction *out); int link(const char *oldpath, const char *newpath); int uname(struct utsname *buf); -int symlink(const char *target, const char *link); int readlink(const char *path, char *buf, size_t bufsiz); +struct index_state; +int mingw_create_symlink(struct index_state *index, const char *target, const char *link); +#define create_symlink mingw_create_symlink /* * replacements of existing functions diff --git a/compat/mingw.c b/compat/mingw.c index 6e8cadc1114931..36ddca3adeba38 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -3156,7 +3156,7 @@ int link(const char *oldpath, const char *newpath) return 0; } -int symlink(const char *target, const char *link) +int mingw_create_symlink(struct index_state *index UNUSED, const char *target, const char *link) { wchar_t wtarget[MAX_LONG_PATH], wlink[MAX_LONG_PATH]; int len; diff --git a/entry.c b/entry.c index 679d11b1dda8cc..c2dc98f10dbef3 100644 --- a/entry.c +++ b/entry.c @@ -324,7 +324,7 @@ static int write_entry(struct cache_entry *ce, char *path, struct conv_attrs *ca if (!has_symlinks || to_tempfile) goto write_file_entry; - ret = symlink(new_blob, path); + ret = create_symlink(state->istate, new_blob, path); free(new_blob); if (ret) return error_errno("unable to create symlink %s", path); diff --git a/git-compat-util.h b/git-compat-util.h index 92a4148845e1a5..3d1f978d984749 100644 --- a/git-compat-util.h +++ b/git-compat-util.h @@ -396,6 +396,16 @@ static inline int git_has_dir_sep(const char *path) #define is_mount_point is_mount_point_via_stat #endif +#ifndef create_symlink +struct index_state; +static inline int git_create_symlink(struct index_state *index UNUSED, + const char *target, const char *link) +{ + return symlink(target, link); +} +#define create_symlink git_create_symlink +#endif + #ifndef query_user_email #define query_user_email() NULL #endif diff --git a/refs/files-backend.c b/refs/files-backend.c index b2de84fa6932bf..a5a8e21daceac2 100644 --- a/refs/files-backend.c +++ b/refs/files-backend.c @@ -2114,7 +2114,7 @@ static int create_ref_symlink(struct ref_lock *lock, const char *target) char *ref_path = get_locked_file_path(&lock->lk); unlink(ref_path); - ret = symlink(target, ref_path); + ret = create_symlink(NULL, target, ref_path); free(ref_path); if (ret) diff --git a/setup.c b/setup.c index 6000fe3c531858..2baf45f2155e61 100644 --- a/setup.c +++ b/setup.c @@ -2141,7 +2141,7 @@ static void copy_templates_1(struct strbuf *path, struct strbuf *template_path, if (strbuf_readlink(&lnk, template_path->buf, st_template.st_size) < 0) die_errno(_("cannot readlink '%s'"), template_path->buf); - if (symlink(lnk.buf, path->buf)) + if (create_symlink(NULL, lnk.buf, path->buf)) die_errno(_("cannot symlink '%s' '%s'"), lnk.buf, path->buf); strbuf_release(&lnk); @@ -2402,7 +2402,7 @@ static int create_default_files(const char *template_path, repo_git_path_replace(the_repository, &path, "tXXXXXX"); if (!close(xmkstemp(path.buf)) && !unlink(path.buf) && - !symlink("testing", path.buf) && + !create_symlink(NULL, "testing", path.buf) && !lstat(path.buf, &st1) && S_ISLNK(st1.st_mode)) unlink(path.buf); /* good */ From 03cf6a64da28b257169bd55c3c5cc3f1d70ead40 Mon Sep 17 00:00:00 2001 From: JiSeop Moon <zcube@zcube.kr> Date: Mon, 23 Apr 2018 22:35:26 +0200 Subject: [PATCH 209/248] mingw: move the file_attr_to_st_mode() function definition In preparation for making this function a bit more complicated (to allow for special-casing the `ContainerMappedDirectories` in Windows containers, which look like a symbolic link, but are not), let's move it out of the header. Signed-off-by: JiSeop Moon <zcube@zcube.kr> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 14 ++++++++++++++ compat/win32.h | 14 +------------- 2 files changed, 15 insertions(+), 13 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 74804c585b0bcc..7ac2c924bcabea 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -4277,3 +4277,17 @@ int is_inside_windows_container(void) return inside_container; } + +int file_attr_to_st_mode (DWORD attr, DWORD tag) +{ + int fMode = S_IREAD; + if ((attr & FILE_ATTRIBUTE_REPARSE_POINT) && tag == IO_REPARSE_TAG_SYMLINK) + fMode |= S_IFLNK; + else if (attr & FILE_ATTRIBUTE_DIRECTORY) + fMode |= S_IFDIR; + else + fMode |= S_IFREG; + if (!(attr & FILE_ATTRIBUTE_READONLY)) + fMode |= S_IWRITE; + return fMode; +} diff --git a/compat/win32.h b/compat/win32.h index 671bcc81f93351..52169ae19f4371 100644 --- a/compat/win32.h +++ b/compat/win32.h @@ -6,19 +6,7 @@ #include <windows.h> #endif -static inline int file_attr_to_st_mode (DWORD attr, DWORD tag) -{ - int fMode = S_IREAD; - if ((attr & FILE_ATTRIBUTE_REPARSE_POINT) && tag == IO_REPARSE_TAG_SYMLINK) - fMode |= S_IFLNK; - else if (attr & FILE_ATTRIBUTE_DIRECTORY) - fMode |= S_IFDIR; - else - fMode |= S_IFREG; - if (!(attr & FILE_ATTRIBUTE_READONLY)) - fMode |= S_IWRITE; - return fMode; -} +extern int file_attr_to_st_mode (DWORD attr, DWORD tag); static inline int get_file_attr(const char *fname, WIN32_FILE_ATTRIBUTE_DATA *fdata) { From c68b79787f8eb976f2d77f9eab48541b8665ca26 Mon Sep 17 00:00:00 2001 From: Bert Belder <bertbelder@gmail.com> Date: Fri, 26 Oct 2018 11:51:51 +0200 Subject: [PATCH 210/248] mingw: allow to specify the symlink type in .gitattributes On Windows, symbolic links have a type: a "file symlink" must point at a file, and a "directory symlink" must point at a directory. If the type of symlink does not match its target, it doesn't work. Git does not record the type of symlink in the index or in a tree. On checkout it'll guess the type, which only works if the target exists at the time the symlink is created. This may often not be the case, for example when the link points at a directory inside a submodule. By specifying `symlink=file` or `symlink=dir` the user can specify what type of symlink Git should create, so Git doesn't have to rely on unreliable heuristics. Signed-off-by: Bert Belder <bertbelder@gmail.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- Documentation/gitattributes.adoc | 30 ++++++++++++++++ compat/mingw.c | 60 ++++++++++++++++++++++++++++++-- 2 files changed, 88 insertions(+), 2 deletions(-) diff --git a/Documentation/gitattributes.adoc b/Documentation/gitattributes.adoc index f20041a323d174..7794bf0fd98dad 100644 --- a/Documentation/gitattributes.adoc +++ b/Documentation/gitattributes.adoc @@ -403,6 +403,36 @@ sign `$` upon checkout. Any byte sequence that begins with with `$Id$` upon check-in. +`symlink` +^^^^^^^^^ + +On Windows, symbolic links have a type: a "file symlink" must point at +a file, and a "directory symlink" must point at a directory. If the +type of symlink does not match its target, it doesn't work. + +Git does not record the type of symlink in the index or in a tree. On +checkout it'll guess the type, which only works if the target exists +at the time the symlink is created. This may often not be the case, +for example when the link points at a directory inside a submodule. + +The `symlink` attribute allows you to explicitly set the type of symlink +to `file` or `dir`, so Git doesn't have to guess. If you have a set of +symlinks that point at other files, you can do: + +------------------------ +*.gif symlink=file +------------------------ + +To tell Git that a symlink points at a directory, use: + +------------------------ +tools_folder symlink=dir +------------------------ + +The `symlink` attribute is ignored on platforms other than Windows, +since they don't distinguish between different types of symlinks. + + `filter` ^^^^^^^^ diff --git a/compat/mingw.c b/compat/mingw.c index 36ddca3adeba38..7b972c15c1f6a4 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -4,6 +4,7 @@ #include "git-compat-util.h" #include "abspath.h" #include "alloc.h" +#include "attr.h" #include "config.h" #include "dir.h" #include "environment.h" @@ -3156,7 +3157,38 @@ int link(const char *oldpath, const char *newpath) return 0; } -int mingw_create_symlink(struct index_state *index UNUSED, const char *target, const char *link) +enum symlink_type { + SYMLINK_TYPE_UNSPECIFIED = 0, + SYMLINK_TYPE_FILE, + SYMLINK_TYPE_DIRECTORY, +}; + +static enum symlink_type check_symlink_attr(struct index_state *index, const char *link) +{ + static struct attr_check *check; + const char *value; + + if (!index) + return SYMLINK_TYPE_UNSPECIFIED; + + if (!check) + check = attr_check_initl("symlink", NULL); + + git_check_attr(index, link, check); + + value = check->items[0].value; + if (ATTR_UNSET(value)) + return SYMLINK_TYPE_UNSPECIFIED; + if (!strcmp(value, "file")) + return SYMLINK_TYPE_FILE; + if (!strcmp(value, "dir") || !strcmp(value, "directory")) + return SYMLINK_TYPE_DIRECTORY; + + warning(_("ignoring invalid symlink type '%s' for '%s'"), value, link); + return SYMLINK_TYPE_UNSPECIFIED; +} + +int mingw_create_symlink(struct index_state *index, const char *target, const char *link) { wchar_t wtarget[MAX_LONG_PATH], wlink[MAX_LONG_PATH]; int len; @@ -3176,7 +3208,31 @@ int mingw_create_symlink(struct index_state *index UNUSED, const char *target, c if (wtarget[len] == '/') wtarget[len] = '\\'; - return create_phantom_symlink(wtarget, wlink); + switch (check_symlink_attr(index, link)) { + case SYMLINK_TYPE_UNSPECIFIED: + /* Create a phantom symlink: it is initially created as a file + * symlink, but may change to a directory symlink later if/when + * the target exists. */ + return create_phantom_symlink(wtarget, wlink); + case SYMLINK_TYPE_FILE: + if (!CreateSymbolicLinkW(wlink, wtarget, symlink_file_flags)) + break; + return 0; + case SYMLINK_TYPE_DIRECTORY: + if (!CreateSymbolicLinkW(wlink, wtarget, + symlink_directory_flags)) + break; + /* There may be dangling phantom symlinks that point at this + * one, which should now morph into directory symlinks. */ + process_phantom_symlinks(); + return 0; + default: + BUG("unhandled symlink type"); + } + + /* CreateSymbolicLinkW failed. */ + errno = err_win_to_posix(GetLastError()); + return -1; } #ifndef _WINNT_H From ccdf9bfbb99be892955eef9ab67195517c25b1e3 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 4 Jun 2020 23:16:07 +0200 Subject: [PATCH 211/248] mingw: special-case index entries for symlinks with buggy size In https://github.com/git-for-windows/git/pull/2637, we fixed a bug where symbolic links' target path sizes were recorded incorrectly in the index. The downside of this fix was that every user with tracked symbolic links in their checkouts would see them as modified in `git status`, but not in `git diff`, and only a `git add <path>` (or `git add -u`) would "fix" this. Let's do better than that: we can detect that situation and simply pretend that a symbolic link with a known bad size (or a size that just happens to be that bad size, a _very_ unlikely scenario because it would overflow our buffers due to the trailing NUL byte) means that it needs to be re-checked as if we had just checked it out. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- read-cache.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/read-cache.c b/read-cache.c index eccf7b606c09b0..817d6edade55c9 100644 --- a/read-cache.c +++ b/read-cache.c @@ -471,6 +471,17 @@ int ie_modified(struct index_state *istate, * then we know it is. */ if ((changed & DATA_CHANGED) && +#ifdef GIT_WINDOWS_NATIVE + /* + * Work around Git for Windows v2.27.0 fixing a bug where symlinks' + * target path lengths were not read at all, and instead recorded + * as 4096: now, all symlinks would appear as modified. + * + * So let's just special-case symlinks with a target path length + * (i.e. `sd_size`) of 4096 and force them to be re-checked. + */ + (!S_ISLNK(st->st_mode) || ce->ce_stat_data.sd_size != MAX_LONG_PATH) && +#endif (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0)) return changed; From dc350a12bc102c9bc6692e6a296a1fa51f9af097 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 23 Apr 2018 23:20:00 +0200 Subject: [PATCH 212/248] mingw: Windows Docker volumes are *not* symbolic links ... even if they may look like them. As looking up the target of the "symbolic link" (just to see whether it starts with `/ContainerMappedDirectories/`) is pretty expensive, we do it when we can be *really* sure that there is a possibility that this might be the case. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: JiSeop Moon <zcube@zcube.kr> --- compat/mingw.c | 25 +++++++++++++++++++------ compat/win32.h | 2 +- compat/win32/fscache.c | 24 +++++++++++++++++++++++- 3 files changed, 43 insertions(+), 8 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 7ac2c924bcabea..f88f1059db4159 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1148,7 +1148,7 @@ int mingw_lstat(const char *file_name, struct stat *buf) buf->st_uid = 0; buf->st_nlink = 1; buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, - reparse_tag); + reparse_tag, file_name); buf->st_size = S_ISLNK(buf->st_mode) ? link_len : fdata.nFileSizeLow | (((off_t) fdata.nFileSizeHigh) << 32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ @@ -1199,7 +1199,7 @@ static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) buf->st_gid = 0; buf->st_uid = 0; buf->st_nlink = 1; - buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, 0); + buf->st_mode = file_attr_to_st_mode(fdata.dwFileAttributes, 0, NULL); buf->st_size = fdata.nFileSizeLow | (((off_t)fdata.nFileSizeHigh)<<32); buf->st_dev = buf->st_rdev = 0; /* not used by Git */ @@ -4278,12 +4278,25 @@ int is_inside_windows_container(void) return inside_container; } -int file_attr_to_st_mode (DWORD attr, DWORD tag) +int file_attr_to_st_mode (DWORD attr, DWORD tag, const char *path) { int fMode = S_IREAD; - if ((attr & FILE_ATTRIBUTE_REPARSE_POINT) && tag == IO_REPARSE_TAG_SYMLINK) - fMode |= S_IFLNK; - else if (attr & FILE_ATTRIBUTE_DIRECTORY) + if ((attr & FILE_ATTRIBUTE_REPARSE_POINT) && + tag == IO_REPARSE_TAG_SYMLINK) { + int flag = S_IFLNK; + char buf[MAX_LONG_PATH]; + + /* + * Windows containers' mapped volumes are marked as reparse + * points and look like symbolic links, but they are not. + */ + if (path && is_inside_windows_container() && + readlink(path, buf, sizeof(buf)) > 27 && + starts_with(buf, "/ContainerMappedDirectories/")) + flag = S_IFDIR; + + fMode |= flag; + } else if (attr & FILE_ATTRIBUTE_DIRECTORY) fMode |= S_IFDIR; else fMode |= S_IFREG; diff --git a/compat/win32.h b/compat/win32.h index 52169ae19f4371..299f01bdf0f5a4 100644 --- a/compat/win32.h +++ b/compat/win32.h @@ -6,7 +6,7 @@ #include <windows.h> #endif -extern int file_attr_to_st_mode (DWORD attr, DWORD tag); +extern int file_attr_to_st_mode (DWORD attr, DWORD tag, const char *path); static inline int get_file_attr(const char *fname, WIN32_FILE_ATTRIBUTE_DATA *fdata) { diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 0f5e00ae18f949..3f9a70e15df853 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -207,8 +207,30 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, fdata->FileAttributes & FILE_ATTRIBUTE_REPARSE_POINT ? fdata->EaSize : 0; + /* + * On certain Windows versions, host directories mapped into + * Windows Containers ("Volumes", see https://docs.docker.com/storage/volumes/) + * look like symbolic links, but their targets are paths that + * are valid only in kernel mode. + * + * Let's work around this by detecting that situation and + * telling Git that these are *not* symbolic links. + */ + if (fse->reparse_tag == IO_REPARSE_TAG_SYMLINK && + sizeof(buf) > (size_t)(list ? list->len + 1 : 0) + fse->len + 1 && + is_inside_windows_container()) { + size_t off = 0; + if (list) { + memcpy(buf, list->dirent.d_name, list->len); + buf[list->len] = '/'; + off = list->len + 1; + } + memcpy(buf + off, fse->dirent.d_name, fse->len); + buf[off + fse->len] = '\0'; + } + fse->st_mode = file_attr_to_st_mode(fdata->FileAttributes, - fdata->EaSize); + fdata->EaSize, buf); fse->dirent.d_type = S_ISREG(fse->st_mode) ? DT_REG : S_ISDIR(fse->st_mode) ? DT_DIR : DT_LNK; fse->u.s.st_size = S_ISLNK(fse->st_mode) ? MAX_LONG_PATH : From e7ab56af2b1246e7d3700257b20140d3b89af697 Mon Sep 17 00:00:00 2001 From: David Lomas <dl3@pale-eds.co.uk> Date: Fri, 28 Jul 2023 15:20:43 +0100 Subject: [PATCH 213/248] mingw: work around rename() failing on a read-only file At least on _some_ APFS network shares, Git fails to rename the object files because they are marked as read-only, because that has the effect of setting the uchg flag on APFS, which then means the file can't be renamed or deleted. To work around that, when a rename failed, and the read-only flag is set, try to turn it off and on again. This fixes https://github.com/git-for-windows/git/issues/4482 Signed-off-by: David Lomas <dl3@pale-eds.co.uk> Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> --- compat/mingw.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index f88f1059db4159..1b6a12934446c0 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -2703,7 +2703,7 @@ int mingw_accept(int sockfd1, struct sockaddr *sa, socklen_t *sz) int mingw_rename(const char *pold, const char *pnew) { static int supports_file_rename_info_ex = 1; - DWORD attrs = INVALID_FILE_ATTRIBUTES, gle; + DWORD attrs = INVALID_FILE_ATTRIBUTES, gle, attrsold; int tries = 0; wchar_t wpold[MAX_LONG_PATH], wpnew[MAX_LONG_PATH]; int wpnew_len; @@ -2795,11 +2795,24 @@ int mingw_rename(const char *pold, const char *pnew) gle = GetLastError(); } - if (gle == ERROR_ACCESS_DENIED && is_inside_windows_container()) { - /* Fall back to copy to destination & remove source */ - if (CopyFileW(wpold, wpnew, FALSE) && !mingw_unlink(pold)) - return 0; - gle = GetLastError(); + if (gle == ERROR_ACCESS_DENIED) { + if (is_inside_windows_container()) { + /* Fall back to copy to destination & remove source */ + if (CopyFileW(wpold, wpnew, FALSE) && !mingw_unlink(pold, 1)) + return 0; + gle = GetLastError(); + } else if ((attrsold = GetFileAttributesW(wpold)) & FILE_ATTRIBUTE_READONLY) { + /* if file is read-only, change and retry */ + SetFileAttributesW(wpold, attrsold & ~FILE_ATTRIBUTE_READONLY); + if (MoveFileExW(wpold, wpnew, + MOVEFILE_REPLACE_EXISTING | MOVEFILE_COPY_ALLOWED)) { + SetFileAttributesW(wpnew, attrsold); + return 0; + } + gle = GetLastError(); + /* revert attribute change on failure */ + SetFileAttributesW(wpold, attrsold); + } } /* revert file attributes on failure */ From a945010d5e6f61bb75472181b9391ebb39dc8a79 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 20 Jul 2017 22:45:01 +0200 Subject: [PATCH 214/248] mingw: explicitly specify with which cmd to prefix the cmdline The main idea of this patch is that even if we have to look up the absolute path of the script, if only the basename was specified as argv[0], then we should use that basename on the command line, too, not the absolute path. This patch will also help with the upcoming patch where we automatically substitute "sh ..." by "busybox sh ..." if "sh" is not in the PATH but "busybox" is: we will do that by substituting the actual executable, but still keep prepending "sh" to the command line. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/compat/mingw.c b/compat/mingw.c index 7b972c15c1f6a4..b15fec02c90bc9 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1992,8 +1992,8 @@ static int is_msys2_sh(const char *cmd) } static pid_t mingw_spawnve_fd(const char *cmd, const char **argv, char **deltaenv, - const char *dir, - int prepend_cmd, int fhin, int fhout, int fherr) + const char *dir, const char *prepend_cmd, + int fhin, int fhout, int fherr) { STARTUPINFOEXW si; PROCESS_INFORMATION pi; @@ -2073,9 +2073,9 @@ static pid_t mingw_spawnve_fd(const char *cmd, const char **argv, char **deltaen /* concatenate argv, quoting args as we go */ strbuf_init(&args, 0); if (prepend_cmd) { - char *quoted = (char *)quote_arg(cmd); + char *quoted = (char *)quote_arg(prepend_cmd); strbuf_addstr(&args, quoted); - if (quoted != cmd) + if (quoted != prepend_cmd) free(quoted); } for (; *argv; argv++) { @@ -2195,7 +2195,8 @@ static pid_t mingw_spawnve_fd(const char *cmd, const char **argv, char **deltaen return (pid_t)pi.dwProcessId; } -static pid_t mingw_spawnv(const char *cmd, const char **argv, int prepend_cmd) +static pid_t mingw_spawnv(const char *cmd, const char **argv, + const char *prepend_cmd) { return mingw_spawnve_fd(cmd, argv, NULL, NULL, prepend_cmd, 0, 1, 2); } @@ -2223,14 +2224,14 @@ pid_t mingw_spawnvpe(const char *cmd, const char **argv, char **deltaenv, pid = -1; } else { - pid = mingw_spawnve_fd(iprog, argv, deltaenv, dir, 1, + pid = mingw_spawnve_fd(iprog, argv, deltaenv, dir, interpr, fhin, fhout, fherr); free(iprog); } argv[0] = argv0; } else - pid = mingw_spawnve_fd(prog, argv, deltaenv, dir, 0, + pid = mingw_spawnve_fd(prog, argv, deltaenv, dir, NULL, fhin, fhout, fherr); free(prog); } @@ -2255,7 +2256,7 @@ static int try_shell_exec(const char *cmd, char *const *argv) argv2[0] = (char *)cmd; /* full path to the script file */ COPY_ARRAY(&argv2[1], &argv[1], argc); exec_id = trace2_exec(prog, (const char **)argv2); - pid = mingw_spawnv(prog, (const char **)argv2, 1); + pid = mingw_spawnv(prog, (const char **)argv2, interpr); if (pid >= 0) { int status; if (waitpid(pid, &status, 0) < 0) @@ -2279,7 +2280,7 @@ int mingw_execv(const char *cmd, char *const *argv) int exec_id; exec_id = trace2_exec(cmd, (const char **)argv); - pid = mingw_spawnv(cmd, (const char **)argv, 0); + pid = mingw_spawnv(cmd, (const char **)argv, NULL); if (pid < 0) { trace2_exec_result(exec_id, -1); return -1; From bfd9ff944989cb7ca5736c42551ad71e6578f2d6 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 20 Jul 2017 20:41:29 +0200 Subject: [PATCH 215/248] mingw: when path_lookup() failed, try BusyBox BusyBox comes with a ton of applets ("applet" being the identical concept to Git's "builtins"). And similar to Git's builtins, the applets can be called via `busybox <command>`, or the BusyBox executable can be copied/hard-linked to the command name. The similarities do not end here. Just as with Git's builtins, it is problematic that BusyBox' hard-linked applets cannot easily be put into a .zip file: .zip archives have no concept of hard-links and therefore would store identical copies (and also extract identical copies, "inflating" the archive unnecessarily). To counteract that issue, MinGit already ships without hard-linked copies of the builtins, and the plan is to do the same with BusyBox' applets: simply ship busybox.exe as single executable, without hard-linked applets. To accommodate that, Git is being taught by this commit a very special trick, exploiting the fact that it is possible to call an executable with a command-line whose argv[0] is different from the executable's name: when `sh` is to be spawned, and no `sh` is found in the PATH, but busybox.exe is, use that executable (with unchanged argv). Likewise, if any executable to be spawned is not on the PATH, but busybox.exe is found, parse the output of `busybox.exe --help` to find out what applets are included, and if the command matches an included applet name, use busybox.exe to execute it. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++ t/t0014-alias.sh | 2 +- 2 files changed, 64 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index b15fec02c90bc9..c40d5cce0feb9f 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -12,6 +12,7 @@ #include "repository.h" #include "run-command.h" #include "strbuf.h" +#include "string-list.h" #include "symlinks.h" #include "trace2.h" #include "win32.h" @@ -1761,6 +1762,65 @@ static char *lookup_prog(const char *dir, int dirlen, const char *cmd, return NULL; } +static char *path_lookup(const char *cmd, int exe_only); + +static char *is_busybox_applet(const char *cmd) +{ + static struct string_list applets = STRING_LIST_INIT_DUP; + static char *busybox_path; + static int busybox_path_initialized; + + /* Avoid infinite loop */ + if (!strncasecmp(cmd, "busybox", 7) && + (!cmd[7] || !strcasecmp(cmd + 7, ".exe"))) + return NULL; + + if (!busybox_path_initialized) { + busybox_path = path_lookup("busybox.exe", 1); + busybox_path_initialized = 1; + } + + /* Assume that sh is compiled in... */ + if (!busybox_path || !strcasecmp(cmd, "sh")) + return xstrdup_or_null(busybox_path); + + if (!applets.nr) { + struct child_process cp = CHILD_PROCESS_INIT; + struct strbuf buf = STRBUF_INIT; + char *p; + + strvec_pushl(&cp.args, busybox_path, "--help", NULL); + + if (capture_command(&cp, &buf, 2048)) { + string_list_append(&applets, ""); + return NULL; + } + + /* parse output */ + p = strstr(buf.buf, "Currently defined functions:\n"); + if (!p) { + warning("Could not parse output of busybox --help"); + string_list_append(&applets, ""); + return NULL; + } + p = strchrnul(p, '\n'); + for (;;) { + size_t len; + + p += strspn(p, "\n\t ,"); + len = strcspn(p, "\n\t ,"); + if (!len) + break; + p[len] = '\0'; + string_list_insert(&applets, p); + p = p + len + 1; + } + } + + return string_list_has_string(&applets, cmd) ? + xstrdup(busybox_path) : NULL; +} + /* * Determines the absolute path of cmd using the split path in path. * If cmd contains a slash or backslash, no lookup is performed. @@ -1789,6 +1849,9 @@ static char *path_lookup(const char *cmd, int exe_only) path = sep + 1; } + if (!prog && !isexe) + prog = is_busybox_applet(cmd); + return prog; } diff --git a/t/t0014-alias.sh b/t/t0014-alias.sh index 62b4d81db875ca..ee0f0a54b6623f 100755 --- a/t/t0014-alias.sh +++ b/t/t0014-alias.sh @@ -53,7 +53,7 @@ test_expect_success 'looping aliases - deprecated builtins' ' test_expect_success 'run-command formats empty args properly' ' test_must_fail env GIT_TRACE=1 git frotz a "" b " " c 2>actual.raw && - sed -ne "/run_command:/s/.*trace: run_command: //p" actual.raw >actual && + sed -ne "/run_command: git-frotz/s/.*trace: run_command: //p" actual.raw >actual && echo "git-frotz a '\'''\'' b '\'' '\'' c" >expect && test_cmp expect actual ' From 89ddc233746659639fd468635ba315acfd1ceddb Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 20 Jul 2017 22:18:56 +0200 Subject: [PATCH 216/248] test-tool: learn to act as a drop-in replacement for `iconv` It is convenient to assume that everybody who wants to build & test Git has access to a working `iconv` executable (after all, we already pretty much require libiconv). However, that limits esoteric test scenarios such as Git for Windows', where an end user installation has to ship with `iconv` for the sole purpose of being testable. That payload serves no other purpose. So let's just have a test helper (to be able to test Git, the test helpers have to be available, after all) to act as `iconv` replacement. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- Makefile | 1 + t/helper/meson.build | 1 + t/helper/test-iconv.c | 47 +++++++++++++++++++++++++++++++++++++++++++ t/helper/test-tool.c | 1 + t/helper/test-tool.h | 1 + 5 files changed, 51 insertions(+) create mode 100644 t/helper/test-iconv.c diff --git a/Makefile b/Makefile index ae79cfa020f4cb..1571b9af99c43c 100644 --- a/Makefile +++ b/Makefile @@ -815,6 +815,7 @@ TEST_BUILTINS_OBJS += test-hash-speed.o TEST_BUILTINS_OBJS += test-hash.o TEST_BUILTINS_OBJS += test-hashmap.o TEST_BUILTINS_OBJS += test-hexdump.o +TEST_BUILTINS_OBJS += test-iconv.o TEST_BUILTINS_OBJS += test-json-writer.o TEST_BUILTINS_OBJS += test-lazy-init-name-hash.o TEST_BUILTINS_OBJS += test-match-trees.o diff --git a/t/helper/meson.build b/t/helper/meson.build index 675e64c0101b61..cba4a9bf4f1434 100644 --- a/t/helper/meson.build +++ b/t/helper/meson.build @@ -29,6 +29,7 @@ test_tool_sources = [ 'test-hash.c', 'test-hashmap.c', 'test-hexdump.c', + 'test-iconv.c', 'test-json-writer.c', 'test-lazy-init-name-hash.c', 'test-match-trees.c', diff --git a/t/helper/test-iconv.c b/t/helper/test-iconv.c new file mode 100644 index 00000000000000..d3c772fddf990b --- /dev/null +++ b/t/helper/test-iconv.c @@ -0,0 +1,47 @@ +#include "test-tool.h" +#include "git-compat-util.h" +#include "strbuf.h" +#include "gettext.h" +#include "parse-options.h" +#include "utf8.h" + +int cmd__iconv(int argc, const char **argv) +{ + struct strbuf buf = STRBUF_INIT; + char *from = NULL, *to = NULL, *p; + size_t len; + int ret = 0; + const char * const iconv_usage[] = { + N_("test-helper --iconv [<options>]"), + NULL + }; + struct option options[] = { + OPT_STRING('f', "from-code", &from, "encoding", "from"), + OPT_STRING('t', "to-code", &to, "encoding", "to"), + OPT_END() + }; + + argc = parse_options(argc, argv, NULL, options, + iconv_usage, 0); + + if (argc > 1 || !from || !to) + usage_with_options(iconv_usage, options); + + if (!argc) { + if (strbuf_read(&buf, 0, 2048) < 0) + die_errno("Could not read from stdin"); + } else if (strbuf_read_file(&buf, argv[0], 2048) < 0) + die_errno("Could not read from '%s'", argv[0]); + + p = reencode_string_len(buf.buf, buf.len, to, from, &len); + if (!p) + die_errno("Could not reencode"); + if (write(1, p, len) < 0) + ret = !!error_errno("Could not write %"PRIuMAX" bytes", + (uintmax_t)len); + + strbuf_release(&buf); + free(p); + + return ret; +} diff --git a/t/helper/test-tool.c b/t/helper/test-tool.c index a7abc618b3887e..9d1b41c8e39b89 100644 --- a/t/helper/test-tool.c +++ b/t/helper/test-tool.c @@ -39,6 +39,7 @@ static struct test_cmd cmds[] = { { "hashmap", cmd__hashmap }, { "hash-speed", cmd__hash_speed }, { "hexdump", cmd__hexdump }, + { "iconv", cmd__iconv }, { "json-writer", cmd__json_writer }, { "lazy-init-name-hash", cmd__lazy_init_name_hash }, { "match-trees", cmd__match_trees }, diff --git a/t/helper/test-tool.h b/t/helper/test-tool.h index 7f150fa1eb9ad2..e18e5a9ed9de81 100644 --- a/t/helper/test-tool.h +++ b/t/helper/test-tool.h @@ -32,6 +32,7 @@ int cmd__getcwd(int argc, const char **argv); int cmd__hashmap(int argc, const char **argv); int cmd__hash_speed(int argc, const char **argv); int cmd__hexdump(int argc, const char **argv); +int cmd__iconv(int argc, const char **argv); int cmd__json_writer(int argc, const char **argv); int cmd__lazy_init_name_hash(int argc, const char **argv); int cmd__match_trees(int argc, const char **argv); From ee5fc6878f5279b1fc51e11c284b1243d1f2c04c Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 20 Jul 2017 22:25:21 +0200 Subject: [PATCH 217/248] tests(mingw): if `iconv` is unavailable, use `test-helper --iconv` Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/test-lib.sh | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/t/test-lib.sh b/t/test-lib.sh index 621cd31ae1dc51..929eee812d68cc 100644 --- a/t/test-lib.sh +++ b/t/test-lib.sh @@ -1674,6 +1674,12 @@ Darwin) test_set_prereq GREP_STRIPS_CR test_set_prereq WINDOWS GIT_TEST_CMP="GIT_DIR=/dev/null git diff --no-index --ignore-cr-at-eol --" + if ! type iconv >/dev/null 2>&1 + then + iconv () { + test-tool iconv "$@" + } + fi ;; *CYGWIN*) test_set_prereq POSIXPERM From 03da4d6c6993babe33bd4dc63ba51e8ad7708e49 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 11 Oct 2018 23:55:44 +0200 Subject: [PATCH 218/248] gitattributes: mark .png files as binary Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- .gitattributes | 1 + 1 file changed, 1 insertion(+) diff --git a/.gitattributes b/.gitattributes index 32583149c2f927..a727d7ab9bdcb1 100644 --- a/.gitattributes +++ b/.gitattributes @@ -6,6 +6,7 @@ *.pm text eol=lf diff=perl *.py text eol=lf diff=python *.bat text eol=crlf +*.png binary CODE_OF_CONDUCT.md -whitespace /Documentation/**/*.adoc text eol=lf /command-list.txt text eol=lf From 697cbf417b3c0eaf38ccd8bafcfeb7be62558ad2 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Sat, 5 Aug 2017 20:28:37 +0200 Subject: [PATCH 219/248] tests: move test PNGs into t/lib-diff/ We already have a directory where we store files intended for use by multiple test scripts. The same directory is a better home for the test-binary-*.png files than t/. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/{ => lib-diff}/test-binary-1.png | Bin t/{ => lib-diff}/test-binary-2.png | Bin t/t3307-notes-man.sh | 2 +- t/t3903-stash.sh | 2 +- t/t4012-diff-binary.sh | 2 +- t/t4049-diff-stat-count.sh | 2 +- t/t4108-apply-threeway.sh | 12 ++++++------ t/t6403-merge-file.sh | 4 ++-- t/t6407-merge-binary.sh | 2 +- t/t9200-git-cvsexportcommit.sh | 14 +++++++------- 10 files changed, 20 insertions(+), 20 deletions(-) rename t/{ => lib-diff}/test-binary-1.png (100%) rename t/{ => lib-diff}/test-binary-2.png (100%) diff --git a/t/test-binary-1.png b/t/lib-diff/test-binary-1.png similarity index 100% rename from t/test-binary-1.png rename to t/lib-diff/test-binary-1.png diff --git a/t/test-binary-2.png b/t/lib-diff/test-binary-2.png similarity index 100% rename from t/test-binary-2.png rename to t/lib-diff/test-binary-2.png diff --git a/t/t3307-notes-man.sh b/t/t3307-notes-man.sh index 1aa366a410e9a3..7e5c06e6615d7a 100755 --- a/t/t3307-notes-man.sh +++ b/t/t3307-notes-man.sh @@ -26,7 +26,7 @@ test_expect_success 'example 1: notes to add an Acked-by line' ' ' test_expect_success 'example 2: binary notes' ' - cp "$TEST_DIRECTORY"/test-binary-1.png . && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png . && git checkout B && blob=$(git hash-object -w test-binary-1.png) && git notes --ref=logo add -C "$blob" && diff --git a/t/t3903-stash.sh b/t/t3903-stash.sh index 0bb4648e3639b2..6ee349a9bd5967 100755 --- a/t/t3903-stash.sh +++ b/t/t3903-stash.sh @@ -1374,7 +1374,7 @@ test_expect_success 'stash -- <subdir> works with binary files' ' git reset && >subdir/untracked && >subdir/tracked && - cp "$TEST_DIRECTORY"/test-binary-1.png subdir/tracked-binary && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png subdir/tracked-binary && git add subdir/tracked* && git stash -- subdir/ && test_path_is_missing subdir/tracked && diff --git a/t/t4012-diff-binary.sh b/t/t4012-diff-binary.sh index d1d30ac2a9474e..73b1e43779783d 100755 --- a/t/t4012-diff-binary.sh +++ b/t/t4012-diff-binary.sh @@ -19,7 +19,7 @@ test_expect_success 'prepare repository' ' echo AIT >a && echo BIT >b && echo CIT >c && echo DIT >d && git update-index --add a b c d && echo git >a && - cat "$TEST_DIRECTORY"/test-binary-1.png >b && + cat "$TEST_DIRECTORY"/lib-diff/test-binary-1.png >b && echo git >c && cat b b >d ' diff --git a/t/t4049-diff-stat-count.sh b/t/t4049-diff-stat-count.sh index eceb47c8594416..2161a1e8cf5ba6 100755 --- a/t/t4049-diff-stat-count.sh +++ b/t/t4049-diff-stat-count.sh @@ -33,7 +33,7 @@ test_expect_success 'binary changes do not count in lines' ' git reset --hard && echo a >a && echo c >c && - cat "$TEST_DIRECTORY"/test-binary-1.png >d && + cat "$TEST_DIRECTORY"/lib-diff/test-binary-1.png >d && cat >expect <<-\EOF && a | 1 + c | 1 + diff --git a/t/t4108-apply-threeway.sh b/t/t4108-apply-threeway.sh index f30e85659dbb87..7f84edd9653a7d 100755 --- a/t/t4108-apply-threeway.sh +++ b/t/t4108-apply-threeway.sh @@ -272,11 +272,11 @@ test_expect_success 'apply with --3way --cached and conflicts' ' test_expect_success 'apply binary file patch' ' git reset --hard main && - cp "$TEST_DIRECTORY/test-binary-1.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-1.png" bin.png && git add bin.png && git commit -m "add binary file" && - cp "$TEST_DIRECTORY/test-binary-2.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-2.png" bin.png && git diff --binary >bin.diff && git reset --hard && @@ -287,11 +287,11 @@ test_expect_success 'apply binary file patch' ' test_expect_success 'apply binary file patch with 3way' ' git reset --hard main && - cp "$TEST_DIRECTORY/test-binary-1.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-1.png" bin.png && git add bin.png && git commit -m "add binary file" && - cp "$TEST_DIRECTORY/test-binary-2.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-2.png" bin.png && git diff --binary >bin.diff && git reset --hard && @@ -302,11 +302,11 @@ test_expect_success 'apply binary file patch with 3way' ' test_expect_success 'apply full-index patch with 3way' ' git reset --hard main && - cp "$TEST_DIRECTORY/test-binary-1.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-1.png" bin.png && git add bin.png && git commit -m "add binary file" && - cp "$TEST_DIRECTORY/test-binary-2.png" bin.png && + cp "$TEST_DIRECTORY/lib-diff/test-binary-2.png" bin.png && git diff --full-index >bin.diff && git reset --hard && diff --git a/t/t6403-merge-file.sh b/t/t6403-merge-file.sh index 06ab4d7aede081..3e06db0cbc579b 100755 --- a/t/t6403-merge-file.sh +++ b/t/t6403-merge-file.sh @@ -355,12 +355,12 @@ test_expect_success "expected conflict markers" ' test_expect_success 'binary files cannot be merged' ' test_must_fail git merge-file -p \ - orig.txt "$TEST_DIRECTORY"/test-binary-1.png new1.txt 2> merge.err && + orig.txt "$TEST_DIRECTORY"/lib-diff/test-binary-1.png new1.txt 2> merge.err && grep "Cannot merge binary files" merge.err ' test_expect_success 'binary files cannot be merged with --object-id' ' - cp "$TEST_DIRECTORY"/test-binary-1.png . && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png . && git add orig.txt new1.txt test-binary-1.png && test_must_fail git merge-file --object-id \ :orig.txt :test-binary-1.png :new1.txt 2> merge.err && diff --git a/t/t6407-merge-binary.sh b/t/t6407-merge-binary.sh index e8a28717cece32..2547f1d504a2c5 100755 --- a/t/t6407-merge-binary.sh +++ b/t/t6407-merge-binary.sh @@ -9,7 +9,7 @@ export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME test_expect_success setup ' - cat "$TEST_DIRECTORY"/test-binary-1.png >m && + cat "$TEST_DIRECTORY"/lib-diff/test-binary-1.png >m && git add m && git ls-files -s | sed -e "s/ 0 / 1 /" >E1 && test_tick && diff --git a/t/t9200-git-cvsexportcommit.sh b/t/t9200-git-cvsexportcommit.sh index a44eabf0d80fa8..5249a9eb886e0b 100755 --- a/t/t9200-git-cvsexportcommit.sh +++ b/t/t9200-git-cvsexportcommit.sh @@ -54,8 +54,8 @@ test_expect_success 'New file' ' mkdir A B C D E F && echo hello1 >A/newfile1.txt && echo hello2 >B/newfile2.txt && - cp "$TEST_DIRECTORY"/test-binary-1.png C/newfile3.png && - cp "$TEST_DIRECTORY"/test-binary-1.png D/newfile4.png && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png C/newfile3.png && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png D/newfile4.png && git add A/newfile1.txt && git add B/newfile2.txt && git add C/newfile3.png && @@ -80,8 +80,8 @@ test_expect_success 'Remove two files, add two and update two' ' rm -f B/newfile2.txt && rm -f C/newfile3.png && echo Hello5 >E/newfile5.txt && - cp "$TEST_DIRECTORY"/test-binary-2.png D/newfile4.png && - cp "$TEST_DIRECTORY"/test-binary-1.png F/newfile6.png && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-2.png D/newfile4.png && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png F/newfile6.png && git add E/newfile5.txt && git add F/newfile6.png && git commit -a -m "Test: Remove, add and update" && @@ -169,7 +169,7 @@ test_expect_success 'New file with spaces in file name' ' mkdir "G g" && echo ok then >"G g/with spaces.txt" && git add "G g/with spaces.txt" && \ - cp "$TEST_DIRECTORY"/test-binary-1.png "G g/with spaces.png" && \ + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png "G g/with spaces.png" && \ git add "G g/with spaces.png" && git commit -a -m "With spaces" && id=$(git rev-list --max-count=1 HEAD) && @@ -181,7 +181,7 @@ test_expect_success 'New file with spaces in file name' ' test_expect_success 'Update file with spaces in file name' ' echo Ok then >>"G g/with spaces.txt" && - cat "$TEST_DIRECTORY"/test-binary-1.png >>"G g/with spaces.png" && \ + cat "$TEST_DIRECTORY"/lib-diff/test-binary-1.png >>"G g/with spaces.png" && \ git add "G g/with spaces.png" && git commit -a -m "Update with spaces" && id=$(git rev-list --max-count=1 HEAD) && @@ -206,7 +206,7 @@ test_expect_success !MINGW 'File with non-ascii file name' ' mkdir -p Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö && echo Foo >Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö/gårdetsågårdet.txt && git add Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö/gårdetsågårdet.txt && - cp "$TEST_DIRECTORY"/test-binary-1.png Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö/gårdetsågårdet.png && + cp "$TEST_DIRECTORY"/lib-diff/test-binary-1.png Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö/gårdetsågårdet.png && git add Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö/gårdetsågårdet.png && git commit -a -m "Går det så går det" && \ id=$(git rev-list --max-count=1 HEAD) && From b396c21d216e372eee7e23ec643f8c8c595b3156 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 18 Jul 2017 01:15:40 +0200 Subject: [PATCH 220/248] tests: only override sort & find if there are usable ones in /usr/bin/ The idea is to allow running the test suite on MinGit with BusyBox installed in /mingw64/bin/sh.exe. In that case, we will want to exclude sort & find (and other Unix utilities) from being bundled. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- git-sh-setup.sh | 21 ++++++++++++++------- t/test-lib.sh | 21 ++++++++++++++------- 2 files changed, 28 insertions(+), 14 deletions(-) diff --git a/git-sh-setup.sh b/git-sh-setup.sh index 19aef72ec25530..fad4f9df94e143 100644 --- a/git-sh-setup.sh +++ b/git-sh-setup.sh @@ -292,13 +292,20 @@ create_virtual_base() { # Platform specific tweaks to work around some commands case $(uname -s) in *MINGW*) - # Windows has its own (incompatible) sort and find - sort () { - /usr/bin/sort "$@" - } - find () { - /usr/bin/find "$@" - } + if test -x /usr/bin/sort + then + # Windows has its own (incompatible) sort; override + sort () { + /usr/bin/sort "$@" + } + fi + if test -x /usr/bin/find + then + # Windows has its own (incompatible) find; override + find () { + /usr/bin/find "$@" + } + fi # git sees Windows-style pwd pwd () { builtin pwd -W diff --git a/t/test-lib.sh b/t/test-lib.sh index 929eee812d68cc..7501c2a16bb7fb 100644 --- a/t/test-lib.sh +++ b/t/test-lib.sh @@ -1654,13 +1654,20 @@ Darwin) test_set_prereq EXECKEEPSPID ;; *MINGW*) - # Windows has its own (incompatible) sort and find - sort () { - /usr/bin/sort "$@" - } - find () { - /usr/bin/find "$@" - } + if test -x /usr/bin/sort + then + # Windows has its own (incompatible) sort; override + sort () { + /usr/bin/sort "$@" + } + fi + if test -x /usr/bin/find + then + # Windows has its own (incompatible) find; override + find () { + /usr/bin/find "$@" + } + fi # git sees Windows-style pwd pwd () { builtin pwd -W From 49eb59d268a566a2091e01a1d7c7963358257722 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 19 Nov 2018 20:34:13 +0100 Subject: [PATCH 221/248] tests: use the correct path separator with BusyBox BusyBox-w32 is a true Win32 application, i.e. it does not come with a POSIX emulation layer. That also means that it does *not* use the Unix convention of separating the entries in the PATH variable using colons, but semicolons. However, there are also BusyBox ports to Windows which use a POSIX emulation layer such as Cygwin's or MSYS2's runtime, i.e. using colons as PATH separators. As a tell-tale, let's use the presence of semicolons in the PATH variable: on Unix, it is highly unlikely that it contains semicolons, and on Windows (without POSIX emulation), it is virtually guaranteed, as everybody should have both $SYSTEMROOT and $SYSTEMROOT/system32 in their PATH. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/interop/interop-lib.sh | 8 ++++++-- t/lib-proto-disable.sh | 2 +- t/t0021-conversion.sh | 2 +- t/t0060-path-utils.sh | 24 ++++++++++++------------ t/t0061-run-command.sh | 6 +++--- t/t0300-credentials.sh | 2 +- t/t1504-ceiling-dirs.sh | 10 +++++----- t/t2300-cd-to-toplevel.sh | 2 +- t/t3418-rebase-continue.sh | 4 ++-- t/t5615-alternate-env.sh | 4 ++-- t/t5802-connect-helper.sh | 2 +- t/t7006-pager.sh | 4 ++-- t/t7606-merge-custom.sh | 2 +- t/t7811-grep-open.sh | 2 +- t/t9003-help-autocorrect.sh | 2 +- t/t9800-git-p4-basic.sh | 2 +- t/test-lib.sh | 17 +++++++++++++---- 17 files changed, 54 insertions(+), 41 deletions(-) diff --git a/t/interop/interop-lib.sh b/t/interop/interop-lib.sh index 1b5864d2a7f22c..1facc69d97741a 100644 --- a/t/interop/interop-lib.sh +++ b/t/interop/interop-lib.sh @@ -4,6 +4,10 @@ . ../../GIT-BUILD-OPTIONS INTEROP_ROOT=$(pwd) BUILD_ROOT=$INTEROP_ROOT/build +case "$PATH" in +*\;*) PATH_SEP=\; ;; +*) PATH_SEP=: ;; +esac build_version () { if test -z "$1" @@ -57,7 +61,7 @@ wrap_git () { write_script "$1" <<-EOF GIT_EXEC_PATH="$2" export GIT_EXEC_PATH - PATH="$2:\$PATH" + PATH="$2$PATH_SEP\$PATH" export GIT_EXEC_PATH exec git "\$@" EOF @@ -71,7 +75,7 @@ generate_wrappers () { echo >&2 fatal: test tried to run generic git: $* exit 1 EOF - PATH=$(pwd)/.bin:$PATH + PATH=$(pwd)/.bin$PATH_SEP$PATH } VERSION_A=${GIT_TEST_VERSION_A:-$VERSION_A} diff --git a/t/lib-proto-disable.sh b/t/lib-proto-disable.sh index 890622be81642b..9db481e1be15b2 100644 --- a/t/lib-proto-disable.sh +++ b/t/lib-proto-disable.sh @@ -214,7 +214,7 @@ setup_ext_wrapper () { cd "$TRASH_DIRECTORY/remote" && eval "$*" EOF - PATH=$TRASH_DIRECTORY:$PATH && + PATH=$TRASH_DIRECTORY$PATH_SEP$PATH && export TRASH_DIRECTORY ' } diff --git a/t/t0021-conversion.sh b/t/t0021-conversion.sh index f0d50d769e9fc5..0c5975336f2104 100755 --- a/t/t0021-conversion.sh +++ b/t/t0021-conversion.sh @@ -8,7 +8,7 @@ export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME . ./test-lib.sh . "$TEST_DIRECTORY"/lib-terminal.sh -PATH=$PWD:$PATH +PATH=$PWD$PATH_SEP$PATH TEST_ROOT="$(pwd)" write_script <<\EOF "$TEST_ROOT/rot13.sh" diff --git a/t/t0060-path-utils.sh b/t/t0060-path-utils.sh index 3cdc4738644dbc..5abfa202c19dca 100755 --- a/t/t0060-path-utils.sh +++ b/t/t0060-path-utils.sh @@ -147,25 +147,25 @@ ancestor /foo /fo -1 ancestor /foo /foo -1 ancestor /foo /bar -1 ancestor /foo /foo/bar -1 -ancestor /foo /foo:/bar -1 -ancestor /foo /:/foo:/bar 0 -ancestor /foo /foo:/:/bar 0 -ancestor /foo /:/bar:/foo 0 +ancestor /foo "/foo$PATH_SEP/bar" -1 +ancestor /foo "/$PATH_SEP/foo$PATH_SEP/bar" 0 +ancestor /foo "/foo$PATH_SEP/$PATH_SEP/bar" 0 +ancestor /foo "/$PATH_SEP/bar$PATH_SEP/foo" 0 ancestor /foo/bar / 0 ancestor /foo/bar /fo -1 ancestor /foo/bar /foo 4 ancestor /foo/bar /foo/ba -1 -ancestor /foo/bar /:/fo 0 -ancestor /foo/bar /foo:/foo/ba 4 +ancestor /foo/bar "/$PATH_SEP/fo" 0 +ancestor /foo/bar "/foo$PATH_SEP/foo/ba" 4 ancestor /foo/bar /bar -1 ancestor /foo/bar /fo -1 -ancestor /foo/bar /foo:/bar 4 -ancestor /foo/bar /:/foo:/bar 4 -ancestor /foo/bar /foo:/:/bar 4 -ancestor /foo/bar /:/bar:/fo 0 -ancestor /foo/bar /:/bar 0 +ancestor /foo/bar "/foo$PATH_SEP/bar" 4 +ancestor /foo/bar "/$PATH_SEP/foo$PATH_SEP/bar" 4 +ancestor /foo/bar "/foo$PATH_SEP/$PATH_SEP/bar" 4 +ancestor /foo/bar "/$PATH_SEP/bar$PATH_SEP/fo" 0 +ancestor /foo/bar "/$PATH_SEP/bar" 0 ancestor /foo/bar /foo 4 -ancestor /foo/bar /foo:/bar 4 +ancestor /foo/bar "/foo$PATH_SEP/bar" 4 ancestor /foo/bar /bar -1 # Windows-specific: DOS drives, network shares diff --git a/t/t0061-run-command.sh b/t/t0061-run-command.sh index 76d4936a879afd..336a0d322fefa1 100755 --- a/t/t0061-run-command.sh +++ b/t/t0061-run-command.sh @@ -69,7 +69,7 @@ test_expect_success 'run_command does not try to execute a directory' ' cat bin2/greet EOF - PATH=$PWD/bin1:$PWD/bin2:$PATH \ + PATH=$PWD/bin1$PATH_SEP$PWD/bin2$PATH_SEP$PATH \ test-tool run-command run-command greet >actual 2>err && test_cmp bin2/greet actual && test_must_be_empty err @@ -86,7 +86,7 @@ test_expect_success POSIXPERM 'run_command passes over non-executable file' ' cat bin2/greet EOF - PATH=$PWD/bin1:$PWD/bin2:$PATH \ + PATH=$PWD/bin1$PATH_SEP$PWD/bin2$PATH_SEP$PATH \ test-tool run-command run-command greet >actual 2>err && test_cmp bin2/greet actual && test_must_be_empty err @@ -106,7 +106,7 @@ test_expect_success POSIXPERM,SANITY 'unreadable directory in PATH' ' git config alias.nitfol "!echo frotz" && chmod a-rx local-command && ( - PATH=./local-command:$PATH && + PATH=./local-command$PATH_SEP$PATH && git nitfol >actual ) && echo frotz >expect && diff --git a/t/t0300-credentials.sh b/t/t0300-credentials.sh index 07aa834d33e248..e740ce362988a5 100755 --- a/t/t0300-credentials.sh +++ b/t/t0300-credentials.sh @@ -80,7 +80,7 @@ test_expect_success 'setup helper scripts' ' printf "username=\\007latrix Lestrange\\n" EOF - PATH="$PWD:$PATH" + PATH="$PWD$PATH_SEP$PATH" ' test_expect_success 'credential_fill invokes helper' ' diff --git a/t/t1504-ceiling-dirs.sh b/t/t1504-ceiling-dirs.sh index e04420f4368b93..ff9fb804827b59 100755 --- a/t/t1504-ceiling-dirs.sh +++ b/t/t1504-ceiling-dirs.sh @@ -84,9 +84,9 @@ then GIT_CEILING_DIRECTORIES="$TRASH_ROOT/top/" test_fail subdir_ceil_at_top_slash - GIT_CEILING_DIRECTORIES=":$TRASH_ROOT/top" + GIT_CEILING_DIRECTORIES="$PATH_SEP$TRASH_ROOT/top" test_prefix subdir_ceil_at_top_no_resolve "sub/dir/" - GIT_CEILING_DIRECTORIES=":$TRASH_ROOT/top/" + GIT_CEILING_DIRECTORIES="$PATH_SEP$TRASH_ROOT/top/" test_prefix subdir_ceil_at_top_slash_no_resolve "sub/dir/" fi @@ -116,13 +116,13 @@ GIT_CEILING_DIRECTORIES="$TRASH_ROOT/subdi" test_prefix subdir_ceil_at_subdi_slash "sub/dir/" -GIT_CEILING_DIRECTORIES="/foo:$TRASH_ROOT/sub" +GIT_CEILING_DIRECTORIES="/foo$PATH_SEP$TRASH_ROOT/sub" test_fail second_of_two -GIT_CEILING_DIRECTORIES="$TRASH_ROOT/sub:/bar" +GIT_CEILING_DIRECTORIES="$TRASH_ROOT/sub$PATH_SEP/bar" test_fail first_of_two -GIT_CEILING_DIRECTORIES="/foo:$TRASH_ROOT/sub:/bar" +GIT_CEILING_DIRECTORIES="/foo$PATH_SEP$TRASH_ROOT/sub$PATH_SEP/bar" test_fail second_of_three diff --git a/t/t2300-cd-to-toplevel.sh b/t/t2300-cd-to-toplevel.sh index c8de6d8a190220..91f523d5198d8d 100755 --- a/t/t2300-cd-to-toplevel.sh +++ b/t/t2300-cd-to-toplevel.sh @@ -16,7 +16,7 @@ test_cd_to_toplevel () { test_expect_success $3 "$2" ' ( cd '"'$1'"' && - PATH="$EXEC_PATH:$PATH" && + PATH="$EXEC_PATH$PATH_SEP$PATH" && . git-sh-setup && cd_to_toplevel && [ "$(pwd -P)" = "$TOPLEVEL" ] diff --git a/t/t3418-rebase-continue.sh b/t/t3418-rebase-continue.sh index b8a8dd77e74408..5aec42e2b07315 100755 --- a/t/t3418-rebase-continue.sh +++ b/t/t3418-rebase-continue.sh @@ -82,7 +82,7 @@ test_expect_success 'rebase --continue remembers merge strategy and options' ' rm -f actual && ( - PATH=./test-bin:$PATH && + PATH=./test-bin$PATH_SEP$PATH && test_must_fail git rebase -s funny -X"option=arg with space" \ -Xop\"tion\\ -X"new${LF}line " main topic ) && @@ -91,7 +91,7 @@ test_expect_success 'rebase --continue remembers merge strategy and options' ' echo "Resolved" >F2 && git add F2 && ( - PATH=./test-bin:$PATH && + PATH=./test-bin$PATH_SEP$PATH && git rebase --continue ) && test_cmp expect actual diff --git a/t/t5615-alternate-env.sh b/t/t5615-alternate-env.sh index 9d6aa2187f2aaa..1bfeccdeb49958 100755 --- a/t/t5615-alternate-env.sh +++ b/t/t5615-alternate-env.sh @@ -39,7 +39,7 @@ test_expect_success 'access alternate via absolute path' ' ' test_expect_success 'access multiple alternates' ' - check_obj "$PWD/one.git/objects:$PWD/two.git/objects" <<-EOF + check_obj "$PWD/one.git/objects$PATH_SEP$PWD/two.git/objects" <<-EOF $one blob $two blob EOF @@ -75,7 +75,7 @@ test_expect_success 'access alternate via relative path (subdir)' ' quoted='"one.git\057objects"' unquoted='two.git/objects' test_expect_success 'mix of quoted and unquoted alternates' ' - check_obj "$quoted:$unquoted" <<-EOF + check_obj "$quoted$PATH_SEP$unquoted" <<-EOF $one blob $two blob EOF diff --git a/t/t5802-connect-helper.sh b/t/t5802-connect-helper.sh index a7be375bceb8d3..26cbcebf3b2b24 100755 --- a/t/t5802-connect-helper.sh +++ b/t/t5802-connect-helper.sh @@ -86,7 +86,7 @@ test_expect_success 'set up fake git-daemon' ' "$TRASH_DIRECTORY/remote" EOF export TRASH_DIRECTORY && - PATH=$TRASH_DIRECTORY:$PATH + PATH=$TRASH_DIRECTORY$PATH_SEP$PATH ' test_expect_success 'ext command can connect to git daemon (no vhost)' ' diff --git a/t/t7006-pager.sh b/t/t7006-pager.sh index 9717e825f0d7a5..e3aa496a286331 100755 --- a/t/t7006-pager.sh +++ b/t/t7006-pager.sh @@ -54,7 +54,7 @@ test_expect_success !MINGW,TTY 'LESS and LV envvars set by git-sh-setup' ' sane_unset LESS LV && PAGER="env >pager-env.out; wc" && export PAGER && - PATH="$(git --exec-path):$PATH" && + PATH="$(git --exec-path)$PATH_SEP$PATH" && export PATH && test_terminal sh -c ". git-sh-setup && git_pager" ) && @@ -388,7 +388,7 @@ test_default_pager() { EOF chmod +x \$less && ( - PATH=.:\$PATH && + PATH=.$PATH_SEP\$PATH && export PATH && $full_command ) && diff --git a/t/t7606-merge-custom.sh b/t/t7606-merge-custom.sh index 81fb7c474c14c1..8197a1c46bb5b6 100755 --- a/t/t7606-merge-custom.sh +++ b/t/t7606-merge-custom.sh @@ -23,7 +23,7 @@ test_expect_success 'set up custom strategy' ' EOF chmod +x git-merge-theirs && - PATH=.:$PATH && + PATH=.$PATH_SEP$PATH && export PATH ' diff --git a/t/t7811-grep-open.sh b/t/t7811-grep-open.sh index 3160be59fd2e26..1a98d733dceb86 100755 --- a/t/t7811-grep-open.sh +++ b/t/t7811-grep-open.sh @@ -52,7 +52,7 @@ test_expect_success SIMPLEPAGER 'git grep -O' ' EOF echo grep.h >expect.notless && - PATH=.:$PATH git grep -O GREP_PATTERN >out && + PATH=.$PATH_SEP$PATH git grep -O GREP_PATTERN >out && { test_cmp expect.less pager-args || test_cmp expect.notless pager-args diff --git a/t/t9003-help-autocorrect.sh b/t/t9003-help-autocorrect.sh index 8da318d2b543da..c7a03aae697ac0 100755 --- a/t/t9003-help-autocorrect.sh +++ b/t/t9003-help-autocorrect.sh @@ -13,7 +13,7 @@ test_expect_success 'setup' ' echo distimdistim was called EOF - PATH="$PATH:." && + PATH="$PATH$PATH_SEP." && export PATH && git commit --allow-empty -m "a single log entry" && diff --git a/t/t9800-git-p4-basic.sh b/t/t9800-git-p4-basic.sh index 0816763e46639c..b3dbd02961fae3 100755 --- a/t/t9800-git-p4-basic.sh +++ b/t/t9800-git-p4-basic.sh @@ -286,7 +286,7 @@ test_expect_success 'exit when p4 fails to produce marshaled output' ' EOF chmod 755 badp4dir/p4 && ( - PATH="$TRASH_DIRECTORY/badp4dir:$PATH" && + PATH="$TRASH_DIRECTORY/badp4dir$PATH_SEP$PATH" && export PATH && test_expect_code 1 git p4 clone --dest="$git" //depot >errs 2>&1 ) && diff --git a/t/test-lib.sh b/t/test-lib.sh index 7501c2a16bb7fb..ca612cb27de340 100644 --- a/t/test-lib.sh +++ b/t/test-lib.sh @@ -15,6 +15,15 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see https://www.gnu.org/licenses/ . +# On Unix/Linux, the path separator is the colon, on other systems it +# may be different, though. On Windows, for example, it is a semicolon. +# If the PATH variable contains semicolons, it is pretty safe to assume +# that the path separator is a semicolon. +case "$PATH" in +*\;*) PATH_SEP=\; ;; +*) PATH_SEP=: ;; +esac + # Test the binaries we have just built. The tests are kept in # t/ subdirectory and are run in 'trash directory' subdirectory. if test -z "$TEST_DIRECTORY" @@ -1384,7 +1393,7 @@ then done done IFS=$OLDIFS - PATH=$GIT_VALGRIND/bin:$PATH + PATH=$GIT_VALGRIND/bin$PATH_SEP$PATH GIT_EXEC_PATH=$GIT_VALGRIND/bin export GIT_VALGRIND GIT_VALGRIND_MODE="$valgrind" @@ -1396,7 +1405,7 @@ elif test -n "$GIT_TEST_INSTALLED" then GIT_EXEC_PATH=$($GIT_TEST_INSTALLED/git --exec-path) || error "Cannot run git from $GIT_TEST_INSTALLED." - PATH=$GIT_TEST_INSTALLED:$GIT_BUILD_DIR/t/helper:$PATH + PATH=$GIT_TEST_INSTALLED$PATH_SEP$GIT_BUILD_DIR/t/helper$PATH_SEP$PATH GIT_EXEC_PATH=${GIT_TEST_EXEC_PATH:-$GIT_EXEC_PATH} else # normal case, use ../bin-wrappers only unless $with_dashes: if test -n "$no_bin_wrappers" @@ -1412,12 +1421,12 @@ else # normal case, use ../bin-wrappers only unless $with_dashes: fi with_dashes=t fi - PATH="$git_bin_dir:$PATH" + PATH="$git_bin_dir$PATH_SEP$PATH" fi GIT_EXEC_PATH=$GIT_BUILD_DIR if test -n "$with_dashes" then - PATH="$GIT_BUILD_DIR:$GIT_BUILD_DIR/t/helper:$PATH" + PATH="$GIT_BUILD_DIR$PATH_SEP$GIT_BUILD_DIR/t/helper$PATH_SEP$PATH" fi fi GIT_TEMPLATE_DIR="$GIT_TEST_TEMPLATE_DIR" From 74eaa8c43200e4c811b6cd89ecc1187247cdedee Mon Sep 17 00:00:00 2001 From: Bert Belder <bertbelder@gmail.com> Date: Fri, 26 Oct 2018 23:42:09 +0200 Subject: [PATCH 222/248] Win32: symlink: add test for `symlink` attribute To verify that the symlink is resolved correctly, we use the fact that `git.exe` is a native Win32 program, and that `git.exe config -f <path>` therefore uses the native symlink resolution. Signed-off-by: Bert Belder <bertbelder@gmail.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/meson.build | 1 + t/t2040-checkout-symlink-attr.sh | 46 ++++++++++++++++++++++++++++++++ 2 files changed, 47 insertions(+) create mode 100755 t/t2040-checkout-symlink-attr.sh diff --git a/t/meson.build b/t/meson.build index dc167eadb7041a..0f5f3d341628ef 100644 --- a/t/meson.build +++ b/t/meson.build @@ -261,6 +261,7 @@ integration_tests = [ 't2027-checkout-track.sh', 't2030-unresolve-info.sh', 't2031-checkout-long-paths.sh', + 't2040-checkout-symlink-attr.sh', 't2050-git-dir-relative.sh', 't2060-switch.sh', 't2070-restore.sh', diff --git a/t/t2040-checkout-symlink-attr.sh b/t/t2040-checkout-symlink-attr.sh new file mode 100755 index 00000000000000..e00c31d096ce88 --- /dev/null +++ b/t/t2040-checkout-symlink-attr.sh @@ -0,0 +1,46 @@ +#!/bin/sh + +test_description='checkout symlinks with `symlink` attribute on Windows + +Ensures that Git for Windows creates symlinks of the right type, +as specified by the `symlink` attribute in `.gitattributes`.' + +# Tell MSYS to create native symlinks. Without this flag test-lib's +# prerequisite detection for SYMLINKS doesn't detect the right thing. +MSYS=winsymlinks:nativestrict && export MSYS + +. ./test-lib.sh + +if ! test_have_prereq MINGW,SYMLINKS +then + skip_all='skipping $0: MinGW-only test, which requires symlink support.' + test_done +fi + +# Adds a symlink to the index without clobbering the work tree. +cache_symlink () { + sha=$(printf '%s' "$1" | git hash-object --stdin -w) && + git update-index --add --cacheinfo 120000,$sha,"$2" +} + +test_expect_success 'checkout symlinks with attr' ' + cache_symlink file1 file-link && + cache_symlink dir dir-link && + + printf "file-link symlink=file\ndir-link symlink=dir\n" >.gitattributes && + git add .gitattributes && + + git checkout . && + + mkdir dir && + echo "[a]b=c" >file1 && + echo "[x]y=z" >dir/file2 && + + # MSYS2 is very forgiving, it will resolve symlinks even if the + # symlink type is incorrect. To make this test meaningful, try + # them with a native, non-MSYS executable, such as `git config`. + test "$(git config -f file-link a.b)" = "c" && + test "$(git config -f dir-link/file2 x.y)" = "z" +' + +test_done From 5c9b85f2fb12736bf50ea99facea877fbf3ea608 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 30 Jun 2017 00:35:40 +0200 Subject: [PATCH 223/248] mingw: only use Bash-ism `builtin pwd -W` when available Traditionally, Git for Windows' SDK uses Bash as its default shell. However, other Unix shells are available, too. Most notably, the Win32 port of BusyBox comes with `ash` whose `pwd` command already prints Windows paths as Git for Windows wants them, while there is not even a `builtin` command. Therefore, let's be careful not to override `pwd` unless we know that the `builtin` command is available. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- git-sh-setup.sh | 14 ++++++++++---- t/test-lib.sh | 14 ++++++++++---- 2 files changed, 20 insertions(+), 8 deletions(-) diff --git a/git-sh-setup.sh b/git-sh-setup.sh index fad4f9df94e143..c51ad34148ccf3 100644 --- a/git-sh-setup.sh +++ b/git-sh-setup.sh @@ -306,10 +306,16 @@ case $(uname -s) in /usr/bin/find "$@" } fi - # git sees Windows-style pwd - pwd () { - builtin pwd -W - } + # On Windows, Git wants Windows paths. But /usr/bin/pwd spits out + # Unix-style paths. At least in Bash, we have a builtin pwd that + # understands the -W option to force "mixed" paths, i.e. with drive + # prefix but still with forward slashes. Let's use that, if available. + if type builtin >/dev/null 2>&1 + then + pwd () { + builtin pwd -W + } + fi is_absolute_path () { case "$1" in [/\\]* | [A-Za-z]:*) diff --git a/t/test-lib.sh b/t/test-lib.sh index ca612cb27de340..ce39c4527f561d 100644 --- a/t/test-lib.sh +++ b/t/test-lib.sh @@ -1677,10 +1677,16 @@ Darwin) /usr/bin/find "$@" } fi - # git sees Windows-style pwd - pwd () { - builtin pwd -W - } + # On Windows, Git wants Windows paths. But /usr/bin/pwd spits out + # Unix-style paths. At least in Bash, we have a builtin pwd that + # understands the -W option to force "mixed" paths, i.e. with drive + # prefix but still with forward slashes. Let's use that, if available. + if type builtin >/dev/null 2>&1 + then + pwd () { + builtin pwd -W + } + fi # no POSIX permissions # backslashes in pathspec are converted to '/' # exec does not inherit the PID From 51fe8dc1ed51fa9221ac26e7a6d582c4a98536fa Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 30 Jun 2017 22:32:33 +0200 Subject: [PATCH 224/248] tests (mingw): remove Bash-specific pwd option The -W option is only understood by MSYS2 Bash's pwd command. We already make sure to override `pwd` by `builtin pwd -W` for MINGW, so let's not double the effort here. This will also help when switching the shell to another one (such as BusyBox' ash) whose pwd does *not* understand the -W option. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t9902-completion.sh | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/t/t9902-completion.sh b/t/t9902-completion.sh index 6650d33fba69c8..a44efa6b11cb8c 100755 --- a/t/t9902-completion.sh +++ b/t/t9902-completion.sh @@ -139,12 +139,7 @@ invalid_variable_name='${foo.bar}' actual="$TRASH_DIRECTORY/actual" -if test_have_prereq MINGW -then - ROOT="$(pwd -W)" -else - ROOT="$(pwd)" -fi +ROOT="$(pwd)" test_expect_success 'setup for __git_find_repo_path/__gitdir tests' ' mkdir -p subdir/subsubdir && From fc3910ad187cae5dbb34ea37d3e110696959437d Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 19 Jul 2017 17:07:56 +0200 Subject: [PATCH 225/248] test-lib: add BUSYBOX prerequisite When running with BusyBox, we will want to avoid calling executables on the PATH that are implemented in BusyBox itself. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/test-lib.sh | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/t/test-lib.sh b/t/test-lib.sh index ce39c4527f561d..d892e7df880aac 100644 --- a/t/test-lib.sh +++ b/t/test-lib.sh @@ -1871,6 +1871,10 @@ test_lazy_prereq UNZIP ' test $? -ne 127 ' +test_lazy_prereq BUSYBOX ' + case "$($SHELL --help 2>&1)" in *BusyBox*) true;; *) false;; esac +' + run_with_limited_cmdline () { (ulimit -s 128 && "$@") } From d0ae914f07d1146584fb29aec93fa093aa4cc27b Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Sat, 5 Aug 2017 21:36:01 +0200 Subject: [PATCH 226/248] t5003: use binary file from t/lib-diff/ At some stage, t5003-archive-zip wants to add a file that is not ASCII. To that end, it uses /bin/sh. But that file may actually not exist (it is too easy to forget that not all the world is Unix/Linux...)! Besides, we already have perfectly fine binary files intended for use solely by the tests. So let's use one of them instead. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t5003-archive-zip.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/t5003-archive-zip.sh b/t/t5003-archive-zip.sh index 961c6aac256135..2c3d5a13ad027f 100755 --- a/t/t5003-archive-zip.sh +++ b/t/t5003-archive-zip.sh @@ -88,7 +88,7 @@ test_expect_success \ 'mkdir a && echo simple textfile >a/a && mkdir a/bin && - cp /bin/sh a/bin && + cp "$TEST_DIRECTORY/lib-diff/test-binary-1.png" a/bin && printf "text\r" >a/text.cr && printf "text\r\n" >a/text.crlf && printf "text\n" >a/text.lf && From 1fab7822602cbf727afb495a9e8f8bfb856e69c2 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 21 Jul 2017 12:48:33 +0200 Subject: [PATCH 227/248] t5532: workaround for BusyBox on Windows While it may seem super convenient to some old Unix hands to simpy require Perl to be available when running the test suite, this is a major hassle on Windows, where we want to verify that Perl is not, actually, required in a NO_PERL build. As a super ugly workaround, we "install" a script into /usr/bin/perl reading like this: #!/bin/sh # We'd much rather avoid requiring Perl altogether when testing # an installed Git. Oh well, that's why we cannot have nice # things. exec c:/git-sdk-64/usr/bin/perl.exe "$@" The problem with that is that BusyBox assumes that the #! line in a script refers to an executable, not to a script. So when it encounters the line #!/usr/bin/perl in t5532's proxy-get-cmd, it barfs. Let's help this situation by simply executing the Perl script with the "interpreter" specified explicitly. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t5532-fetch-proxy.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/t/t5532-fetch-proxy.sh b/t/t5532-fetch-proxy.sh index 95d0f33b29531c..86fe5d8f752147 100755 --- a/t/t5532-fetch-proxy.sh +++ b/t/t5532-fetch-proxy.sh @@ -32,7 +32,7 @@ test_expect_success 'setup proxy script' ' write_script proxy <<-\EOF echo >&2 "proxying for $*" - cmd=$(./proxy-get-cmd) + cmd=$("$PERL_PATH" ./proxy-get-cmd) echo >&2 "Running $cmd" exec $cmd EOF From aa1ea2587d40340ebf15645d9bf0ddaff2441622 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 21 Jul 2017 13:24:55 +0200 Subject: [PATCH 228/248] t5605: special-case hardlink test for BusyBox-w32 When t5605 tries to verify that files are hardlinked (or that they are not), it uses the `-links` option of the `find` utility. BusyBox' implementation does not support that option, and BusyBox-w32's lstat() does not even report the number of hard links correctly (for performance reasons). So let's just switch to a different method that actually works on Windows. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t5605-clone-local.sh | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/t/t5605-clone-local.sh b/t/t5605-clone-local.sh index 2397f8fa618054..a7444acc5f89e4 100755 --- a/t/t5605-clone-local.sh +++ b/t/t5605-clone-local.sh @@ -11,6 +11,21 @@ repo_is_hardlinked() { test_line_count = 0 output } +if test_have_prereq MINGW,BUSYBOX +then + # BusyBox' `find` does not support `-links`. Besides, BusyBox-w32's + # lstat() does not report hard links, just like Git's mingw_lstat() + # (from where BusyBox-w32 got its initial implementation). + repo_is_hardlinked() { + for f in $(find "$1/objects" -type f) + do + "$SYSTEMROOT"/system32/fsutil.exe \ + hardlink list $f >links && + test_line_count -gt 1 links || return 1 + done + } +fi + test_expect_success 'preparing origin repository' ' : >file && git add . && git commit -m1 && git clone --bare . a.git && From 0185f561bfa2fbb10814dbc3131c7d5dc295cc00 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 5 Jul 2017 15:14:50 +0200 Subject: [PATCH 229/248] t5813: allow for $PWD to be a Windows path Git for Windows uses MSYS2's Bash to run the test suite, which comes with benefits but also at a heavy price: on the plus side, MSYS2's POSIX emulation layer allows us to continue pretending that we are on a Unix system, e.g. use Unix paths instead of Windows ones, yet this is bought at a rather noticeable performance penalty. There *are* some more native ports of Unix shells out there, though, most notably BusyBox-w32's ash. These native ports do not use any POSIX emulation layer (or at most a *very* thin one, choosing to avoid features such as fork() that are expensive to emulate on Windows), and they use native Windows paths (usually with forward slashes instead of backslashes, which is perfectly legal in almost all use cases). And here comes the problem: with a $PWD looking like, say, C:/git-sdk-64/usr/src/git/t/trash directory.t5813-proto-disable-ssh Git's test scripts get quite a bit confused, as their assumptions have been shattered. Not only does this path contain a colon (oh no!), it also does not start with a slash. This is a problem e.g. when constructing a URL as t5813 does it: ssh://remote$PWD. Not only is it impossible to separate the "host" from the path with a $PWD as above, even prefixing $PWD by a slash won't work, as /C:/git-sdk-64/... is not a valid path. As a workaround, detect when $PWD does not start with a slash on Windows, and simply strip the drive prefix, using an obscure feature of Windows paths: if an absolute Windows path starts with a slash, it is implicitly prefixed by the drive prefix of the current directory. As we are talking about the current directory here, anyway, that strategy works. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t5813-proto-disable-ssh.sh | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/t/t5813-proto-disable-ssh.sh b/t/t5813-proto-disable-ssh.sh index 045e2fe6ce376a..c78581dc9f4a1e 100755 --- a/t/t5813-proto-disable-ssh.sh +++ b/t/t5813-proto-disable-ssh.sh @@ -15,8 +15,23 @@ test_expect_success 'setup repository to clone' ' ' test_proto "host:path" ssh "remote:repo.git" -test_proto "ssh://" ssh "ssh://remote$PWD/remote/repo.git" -test_proto "git+ssh://" ssh "git+ssh://remote$PWD/remote/repo.git" + +hostdir="$PWD" +if test_have_prereq MINGW && test "/${PWD#/}" != "$PWD" +then + case "$PWD" in + [A-Za-z]:/*) + hostdir="${PWD#?:}" + ;; + *) + skip_all="Unhandled PWD '$PWD'; skipping rest" + test_done + ;; + esac +fi + +test_proto "ssh://" ssh "ssh://remote$hostdir/remote/repo.git" +test_proto "git+ssh://" ssh "git+ssh://remote$hostdir/remote/repo.git" # Don't even bother setting up a "-remote" directory, as ssh would generally # complain about the bogus option rather than completing our request. Our From 6bf3a28afe10a80c5f5b85ca6c2425b652269dce Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 7 Jul 2017 10:15:36 +0200 Subject: [PATCH 230/248] t9200: skip tests when $PWD contains a colon On Windows, the current working directory is pretty much guaranteed to contain a colon. If we feed that path to CVS, it mistakes it for a separator between host and port, though. This has not been a problem so far because Git for Windows uses MSYS2's Bash using a POSIX emulation layer that also pretends that the current directory is a Unix path (at least as long as we're in a shell script). However, that is rather limiting, as Git for Windows also explores other ports of other Unix shells. One of those is BusyBox-w32's ash, which is a native port (i.e. *not* using any POSIX emulation layer, and certainly not emulating Unix paths). So let's just detect if there is a colon in $PWD and punt in that case. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- t/t9200-git-cvsexportcommit.sh | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/t/t9200-git-cvsexportcommit.sh b/t/t9200-git-cvsexportcommit.sh index 5249a9eb886e0b..026089f6806733 100755 --- a/t/t9200-git-cvsexportcommit.sh +++ b/t/t9200-git-cvsexportcommit.sh @@ -11,6 +11,13 @@ if ! test_have_prereq PERL; then test_done fi +case "$PWD" in +*:*) + skip_all='cvs would get confused by the colon in `pwd`; skipping tests' + test_done + ;; +esac + cvs >/dev/null 2>&1 if test $? -ne 1 then From a7c03a1b8198c335c80be5ea363d7c916fe24af2 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 20 Jul 2017 00:23:26 +0200 Subject: [PATCH 231/248] mingw: add a Makefile target to copy test artifacts The Makefile target `install-mingit-test-artifacts` simply copies stuff and things directly into a MinGit directory, including an init.bat script to set everything up so that the tests can be run in a cmd window. Sadly, Git's test suite still relies on a Perl interpreter even if compiled with NO_PERL=YesPlease. We punt for now, installing a small script into /usr/bin/perl that hands off to an existing Perl of a Git for Windows SDK. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- config.mak.uname | 51 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) diff --git a/config.mak.uname b/config.mak.uname index d46c50e20e4627..6d17f9d3a80346 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -781,6 +781,57 @@ ifeq ($(uname_S),MINGW) ETC_GITCONFIG = ../etc/gitconfig ETC_GITATTRIBUTES = ../etc/gitattributes endif + MINGW_PREFIX := $(subst /,,$(prefix)) + + DESTDIR_WINDOWS = $(shell cygpath -aw '$(DESTDIR_SQ)') + DESTDIR_MIXED = $(shell cygpath -am '$(DESTDIR_SQ)') +install-mingit-test-artifacts: + install -m755 -d '$(DESTDIR_SQ)/usr/bin' + printf '%s\n%s\n' >'$(DESTDIR_SQ)/usr/bin/perl' \ + "#!/mingw64/bin/busybox sh" \ + "exec \"$(shell cygpath -am /usr/bin/perl.exe)\" \"\$$@\"" + + install -m755 -d '$(DESTDIR_SQ)' + printf '%s%s\n%s\n%s\n%s\n%s\n' >'$(DESTDIR_SQ)/init.bat' \ + "PATH=$(DESTDIR_WINDOWS)\\$(MINGW_PREFIX)\\bin;" \ + "C:\\WINDOWS;C:\\WINDOWS\\system32" \ + "@set GIT_TEST_INSTALLED=$(DESTDIR_MIXED)/$(MINGW_PREFIX)/bin" \ + "@`echo "$(DESTDIR_WINDOWS)" | sed 's/:.*/:/'`" \ + "@cd `echo "$(DESTDIR_WINDOWS)" | sed 's/^.://'`\\test-git\\t" \ + "@echo Now, run 'helper\\test-run-command testsuite'" + + install -m755 -d '$(DESTDIR_SQ)/test-git' + sed 's/^\(NO_PERL\|NO_PYTHON\)=.*/\1=YesPlease/' \ + <GIT-BUILD-OPTIONS >'$(DESTDIR_SQ)/test-git/GIT-BUILD-OPTIONS' + + install -m755 -d '$(DESTDIR_SQ)/test-git/t/helper' + install -m755 $(TEST_PROGRAMS) '$(DESTDIR_SQ)/test-git/t/helper' + (cd t && $(TAR) cf - t[0-9][0-9][0-9][0-9] lib-diff) | \ + (cd '$(DESTDIR_SQ)/test-git/t' && $(TAR) xf -) + install -m755 t/t556x_common t/*.sh '$(DESTDIR_SQ)/test-git/t' + + install -m755 -d '$(DESTDIR_SQ)/test-git/templates' + (cd templates && $(TAR) cf - blt) | \ + (cd '$(DESTDIR_SQ)/test-git/templates' && $(TAR) xf -) + + # po/build/locale for t0200 + install -m755 -d '$(DESTDIR_SQ)/test-git/po/build/locale' + (cd po/build/locale && $(TAR) cf - .) | \ + (cd '$(DESTDIR_SQ)/test-git/po/build/locale' && $(TAR) xf -) + + # git-daemon.exe for t5802, git-http-backend.exe for t5560 + install -m755 -d '$(DESTDIR_SQ)/$(MINGW_PREFIX)/bin' + install -m755 git-daemon.exe git-http-backend.exe \ + '$(DESTDIR_SQ)/$(MINGW_PREFIX)/bin' + + # git-upload-archive (dashed) for t5000 + install -m755 -d '$(DESTDIR_SQ)/$(MINGW_PREFIX)/bin' + install -m755 git-upload-archive.exe '$(DESTDIR_SQ)/$(MINGW_PREFIX)/bin' + + # git-difftool--helper for t7800 + install -m755 -d '$(DESTDIR_SQ)/$(MINGW_PREFIX)/libexec/git-core' + install -m755 git-difftool--helper \ + '$(DESTDIR_SQ)/$(MINGW_PREFIX)/libexec/git-core' endif ifeq ($(uname_S),QNX) COMPAT_CFLAGS += -DSA_RESTART=0 From c18c47492a3a498752db51ea1892a83bc5cc3ee9 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Wed, 17 May 2017 17:05:09 +0200 Subject: [PATCH 232/248] mingw: kill child processes in a gentler way The TerminateProcess() function does not actually leave the child processes any chance to perform any cleanup operations. This is bad insofar as Git itself expects its signal handlers to run. A symptom is e.g. a left-behind .lock file that would not be left behind if the same operation was run, say, on Linux. To remedy this situation, we use an obscure trick: we inject a thread into the process that needs to be killed and to let that thread run the ExitProcess() function with the desired exit status. Thanks J Wyman for describing this trick. The advantage is that the ExitProcess() function lets the atexit handlers run. While this is still different from what Git expects (i.e. running a signal handler), in practice Git sets up signal handlers and atexit handlers that call the same code to clean up after itself. In case that the gentle method to terminate the process failed, we still fall back to calling TerminateProcess(), but in that case we now also make sure that processes spawned by the spawned process are terminated; TerminateProcess() does not give the spawned process a chance to do so itself. Please note that this change only affects how Git for Windows tries to terminate processes spawned by Git's own executables. Third-party software that *calls* Git and wants to terminate it *still* need to make sure to imitate this gentle method, otherwise this patch will not have any effect. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 29 +++++-- compat/win32/exit-process.h | 165 ++++++++++++++++++++++++++++++++++++ 2 files changed, 186 insertions(+), 8 deletions(-) create mode 100644 compat/win32/exit-process.h diff --git a/compat/mingw.c b/compat/mingw.c index 0673b4a121ad07..54d8a65d80fe0a 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -14,6 +14,7 @@ #include "symlinks.h" #include "trace2.h" #include "win32.h" +#include "win32/exit-process.h" #include "win32/fscache.h" #include "win32/lazyload.h" #include "wrapper.h" @@ -2259,16 +2260,28 @@ int mingw_execvp(const char *cmd, char *const *argv) int mingw_kill(pid_t pid, int sig) { if (pid > 0 && sig == SIGTERM) { - HANDLE h = OpenProcess(PROCESS_TERMINATE, FALSE, pid); - - if (TerminateProcess(h, -1)) { + HANDLE h = OpenProcess(PROCESS_CREATE_THREAD | + PROCESS_QUERY_INFORMATION | + PROCESS_VM_OPERATION | PROCESS_VM_WRITE | + PROCESS_VM_READ | PROCESS_TERMINATE, + FALSE, pid); + int ret; + + if (h) + ret = exit_process(h, 128 + sig); + else { + h = OpenProcess(PROCESS_TERMINATE, FALSE, pid); + if (!h) { + errno = err_win_to_posix(GetLastError()); + return -1; + } + ret = terminate_process_tree(h, 128 + sig); + } + if (ret) { + errno = err_win_to_posix(GetLastError()); CloseHandle(h); - return 0; } - - errno = err_win_to_posix(GetLastError()); - CloseHandle(h); - return -1; + return ret; } else if (pid > 0 && sig == 0) { HANDLE h = OpenProcess(PROCESS_QUERY_INFORMATION, FALSE, pid); if (h) { diff --git a/compat/win32/exit-process.h b/compat/win32/exit-process.h new file mode 100644 index 00000000000000..d53989884cfb0c --- /dev/null +++ b/compat/win32/exit-process.h @@ -0,0 +1,165 @@ +#ifndef EXIT_PROCESS_H +#define EXIT_PROCESS_H + +/* + * This file contains functions to terminate a Win32 process, as gently as + * possible. + * + * At first, we will attempt to inject a thread that calls ExitProcess(). If + * that fails, we will fall back to terminating the entire process tree. + * + * For simplicity, these functions are marked as file-local. + */ + +#include <tlhelp32.h> + +/* + * Terminates the process corresponding to the process ID and all of its + * directly and indirectly spawned subprocesses. + * + * This way of terminating the processes is not gentle: the processes get + * no chance of cleaning up after themselves (closing file handles, removing + * .lock files, terminating spawned processes (if any), etc). + */ +static int terminate_process_tree(HANDLE main_process, int exit_status) +{ + HANDLE snapshot = CreateToolhelp32Snapshot(TH32CS_SNAPPROCESS, 0); + PROCESSENTRY32 entry; + DWORD pids[16384]; + int max_len = sizeof(pids) / sizeof(*pids), i, len, ret = 0; + pid_t pid = GetProcessId(main_process); + + pids[0] = (DWORD)pid; + len = 1; + + /* + * Even if Process32First()/Process32Next() seem to traverse the + * processes in topological order (i.e. parent processes before + * child processes), there is nothing in the Win32 API documentation + * suggesting that this is guaranteed. + * + * Therefore, run through them at least twice and stop when no more + * process IDs were added to the list. + */ + for (;;) { + int orig_len = len; + + memset(&entry, 0, sizeof(entry)); + entry.dwSize = sizeof(entry); + + if (!Process32First(snapshot, &entry)) + break; + + do { + for (i = len - 1; i >= 0; i--) { + if (pids[i] == entry.th32ProcessID) + break; + if (pids[i] == entry.th32ParentProcessID) + pids[len++] = entry.th32ProcessID; + } + } while (len < max_len && Process32Next(snapshot, &entry)); + + if (orig_len == len || len >= max_len) + break; + } + + for (i = len - 1; i > 0; i--) { + HANDLE process = OpenProcess(PROCESS_TERMINATE, FALSE, pids[i]); + + if (process) { + if (!TerminateProcess(process, exit_status)) + ret = -1; + CloseHandle(process); + } + } + if (!TerminateProcess(main_process, exit_status)) + ret = -1; + CloseHandle(main_process); + + return ret; +} + +/** + * Determine whether a process runs in the same architecture as the current + * one. That test is required before we assume that GetProcAddress() returns + * a valid address *for the target process*. + */ +static inline int process_architecture_matches_current(HANDLE process) +{ + static BOOL current_is_wow = -1; + BOOL is_wow; + + if (current_is_wow == -1 && + !IsWow64Process (GetCurrentProcess(), ¤t_is_wow)) + current_is_wow = -2; + if (current_is_wow == -2) + return 0; /* could not determine current process' WoW-ness */ + if (!IsWow64Process (process, &is_wow)) + return 0; /* cannot determine */ + return is_wow == current_is_wow; +} + +/** + * Inject a thread into the given process that runs ExitProcess(). + * + * Note: as kernel32.dll is loaded before any process, the other process and + * this process will have ExitProcess() at the same address. + * + * This function expects the process handle to have the access rights for + * CreateRemoteThread(): PROCESS_CREATE_THREAD, PROCESS_QUERY_INFORMATION, + * PROCESS_VM_OPERATION, PROCESS_VM_WRITE, and PROCESS_VM_READ. + * + * The idea comes from the Dr Dobb's article "A Safer Alternative to + * TerminateProcess()" by Andrew Tucker (July 1, 1999), + * http://www.drdobbs.com/a-safer-alternative-to-terminateprocess/184416547 + * + * If this method fails, we fall back to running terminate_process_tree(). + */ +static int exit_process(HANDLE process, int exit_code) +{ + DWORD code; + + if (GetExitCodeProcess(process, &code) && code == STILL_ACTIVE) { + static int initialized; + static LPTHREAD_START_ROUTINE exit_process_address; + PVOID arg = (PVOID)(intptr_t)exit_code; + DWORD thread_id; + HANDLE thread = NULL; + + if (!initialized) { + HINSTANCE kernel32 = GetModuleHandleA("kernel32"); + if (!kernel32) + die("BUG: cannot find kernel32"); + exit_process_address = + (LPTHREAD_START_ROUTINE)(void (*)(void)) + GetProcAddress(kernel32, "ExitProcess"); + initialized = 1; + } + if (!exit_process_address || + !process_architecture_matches_current(process)) + return terminate_process_tree(process, exit_code); + + thread = CreateRemoteThread(process, NULL, 0, + exit_process_address, + arg, 0, &thread_id); + if (thread) { + CloseHandle(thread); + /* + * If the process survives for 10 seconds (a completely + * arbitrary value picked from thin air), fall back to + * killing the process tree via TerminateProcess(). + */ + if (WaitForSingleObject(process, 10000) == + WAIT_OBJECT_0) { + CloseHandle(process); + return 0; + } + } + + return terminate_process_tree(process, exit_code); + } + + return 0; +} + +#endif From 561554bf48729f7d5aab5b315232d88852358417 Mon Sep 17 00:00:00 2001 From: xungeng li <xungeng@gmail.com> Date: Wed, 7 Jun 2023 20:26:33 +0800 Subject: [PATCH 233/248] mingw: optionally enable wsl compability file mode bits The Windows Subsystem for Linux (WSL) version 2 allows to use `chmod` on NTFS volumes provided that they are mounted with metadata enabled (see https://devblogs.microsoft.com/commandline/chmod-chown-wsl-improvements/ for details), for example: $ chmod 0755 /mnt/d/test/a.sh In order to facilitate better collaboration between the Windows version of Git and the WSL version of Git, we can make the Windows version of Git also support reading and writing NTFS file modes in a manner compatible with WSL. Since this slightly slows down operations where lots of files are created (such as an initial checkout), this feature is only enabled when `core.WSLCompat` is set to true. Note that you also have to set `core.fileMode=true` in repositories that have been initialized without enabling WSL compatibility. There are several ways to enable metadata loading for NTFS volumes in WSL, one of which is to modify `/etc/wsl.conf` by adding: ``` [automount] enabled = true options = "metadata,umask=027,fmask=117" ``` And reboot WSL. It can also be enabled temporarily by this incantation: $ sudo umount /mnt/c && sudo mount -t drvfs C: /mnt/c -o metadata,uid=1000,gid=1000,umask=22,fmask=111 It's important to note that this modification is compatible with, but does not depend on WSL. The helper functions in this commit can operate independently and functions normally on devices where WSL is not installed or properly configured. Signed-off-by: xungeng li <xungeng@gmail.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- Documentation/config/core.adoc | 6 ++ compat/mingw.c | 13 +++ compat/win32/fscache.c | 16 ++++ compat/win32/wsl.c | 142 ++++++++++++++++++++++++++++ compat/win32/wsl.h | 12 +++ config.mak.uname | 4 +- contrib/buildsystems/CMakeLists.txt | 1 + meson.build | 1 + 8 files changed, 193 insertions(+), 2 deletions(-) create mode 100644 compat/win32/wsl.c create mode 100644 compat/win32/wsl.h diff --git a/Documentation/config/core.adoc b/Documentation/config/core.adoc index 1010bc85842911..64643990d4ff6f 100644 --- a/Documentation/config/core.adoc +++ b/Documentation/config/core.adoc @@ -770,3 +770,9 @@ core.maxTreeDepth:: to allow Git to abort cleanly, and should not generally need to be adjusted. When Git is compiled with MSVC, the default is 512. Otherwise, the default is 2048. + +core.WSLCompat:: + Tells Git whether to enable wsl compatibility mode. + The default value is false. When set to true, Git will set the mode + bits of the file in the way of wsl, so that the executable flag of + files can be set or read correctly. diff --git a/compat/mingw.c b/compat/mingw.c index c40d5cce0feb9f..7ad3af111b8e8c 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -18,6 +18,7 @@ #include "win32.h" #include "win32/fscache.h" #include "win32/lazyload.h" +#include "win32/wsl.h" #include "wrapper.h" #include "write-or-die.h" #include <aclapi.h> @@ -906,6 +907,11 @@ int mingw_open (const char *filename, int oflags, ...) if (fd < 0 && create && GetLastError() == ERROR_ACCESS_DENIED && INIT_PROC_ADDR(RtlGetLastNtStatus) && RtlGetLastNtStatus() == STATUS_DELETE_PENDING) errno = EEXIST; + else if ((oflags & O_CREAT) && fd >= 0 && are_wsl_compatible_mode_bits_enabled()) { + _mode_t wsl_mode = S_IFREG | (mode&0777); + set_wsl_mode_bits_by_handle((HANDLE)_get_osfhandle(fd), wsl_mode); + } + if (fd < 0 && (oflags & O_ACCMODE) != O_RDONLY && errno == EACCES) { DWORD attrs = GetFileAttributesW(wfilename); if (attrs != INVALID_FILE_ATTRIBUTES && (attrs & FILE_ATTRIBUTE_DIRECTORY)) @@ -1205,6 +1211,11 @@ int mingw_lstat(const char *file_name, struct stat *buf) filetime_to_timespec(&(fdata.ftLastAccessTime), &(buf->st_atim)); filetime_to_timespec(&(fdata.ftLastWriteTime), &(buf->st_mtim)); filetime_to_timespec(&(fdata.ftCreationTime), &(buf->st_ctim)); + if (S_ISREG(buf->st_mode) && + are_wsl_compatible_mode_bits_enabled()) { + copy_wsl_mode_bits_from_disk(wfilename, -1, + &buf->st_mode); + } return 0; } @@ -1256,6 +1267,8 @@ static int get_file_info_by_handle(HANDLE hnd, struct stat *buf) filetime_to_timespec(&(fdata.ftLastAccessTime), &(buf->st_atim)); filetime_to_timespec(&(fdata.ftLastWriteTime), &(buf->st_mtim)); filetime_to_timespec(&(fdata.ftCreationTime), &(buf->st_ctim)); + if (are_wsl_compatible_mode_bits_enabled()) + get_wsl_mode_bits_by_handle(hnd, &buf->st_mode); return 0; } diff --git a/compat/win32/fscache.c b/compat/win32/fscache.c index 0f5e00ae18f949..d2e67bd5ac0cd1 100644 --- a/compat/win32/fscache.c +++ b/compat/win32/fscache.c @@ -8,6 +8,7 @@ #include "config.h" #include "../../mem-pool.h" #include "ntifs.h" +#include "wsl.h" static volatile long initialized; static DWORD dwTlsIndex; @@ -220,6 +221,21 @@ static struct fsentry *fseentry_create_entry(struct fscache *cache, &(fse->u.s.st_mtim)); filetime_to_timespec((FILETIME *)&(fdata->CreationTime), &(fse->u.s.st_ctim)); + if (fdata->EaSize > 0 && + sizeof(buf) >= (size_t)(list ? list->len+1 : 0) + fse->len+1 && + are_wsl_compatible_mode_bits_enabled()) { + size_t off = 0; + wchar_t wpath[MAX_LONG_PATH]; + if (list && list->len) { + memcpy(buf, list->dirent.d_name, list->len); + buf[list->len] = '/'; + off = list->len + 1; + } + memcpy(buf + off, fse->dirent.d_name, fse->len); + buf[off + fse->len] = '\0'; + if (xutftowcs_long_path(wpath, buf) >= 0) + copy_wsl_mode_bits_from_disk(wpath, -1, &fse->st_mode); + } return fse; } diff --git a/compat/win32/wsl.c b/compat/win32/wsl.c new file mode 100644 index 00000000000000..ab599770138b4e --- /dev/null +++ b/compat/win32/wsl.c @@ -0,0 +1,142 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "../../git-compat-util.h" +#include "../win32.h" +#include "../../repository.h" +#include "config.h" +#include "ntifs.h" +#include "wsl.h" + +int are_wsl_compatible_mode_bits_enabled(void) +{ + /* default to `false` during initialization */ + static const int fallback = 0; + static int enabled = -1; + + if (enabled < 0) { + /* avoid infinite recursion */ + if (!the_repository) + return fallback; + + if (the_repository->config && + the_repository->config->hash_initialized && + repo_config_get_bool(the_repository, "core.wslcompat", &enabled) < 0) + enabled = 0; + } + + return enabled < 0 ? fallback : enabled; +} + +int copy_wsl_mode_bits_from_disk(const wchar_t *wpath, ssize_t wpathlen, + _mode_t *mode) +{ + int ret = -1; + HANDLE h; + if (wpathlen >= 0) { + /* + * It's caller's duty to make sure wpathlen is reasonable so + * it does not overflow. + */ + wchar_t *fn2 = (wchar_t*)alloca((wpathlen + 1) * sizeof(wchar_t)); + memcpy(fn2, wpath, wpathlen * sizeof(wchar_t)); + fn2[wpathlen] = 0; + wpath = fn2; + } + h = CreateFileW(wpath, FILE_READ_EA | SYNCHRONIZE, + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, + NULL, OPEN_EXISTING, + FILE_FLAG_BACKUP_SEMANTICS | + FILE_FLAG_OPEN_REPARSE_POINT, + NULL); + if (h != INVALID_HANDLE_VALUE) { + ret = get_wsl_mode_bits_by_handle(h, mode); + CloseHandle(h); + } + return ret; +} + +#ifndef LX_FILE_METADATA_HAS_UID +#define LX_FILE_METADATA_HAS_UID 0x1 +#define LX_FILE_METADATA_HAS_GID 0x2 +#define LX_FILE_METADATA_HAS_MODE 0x4 +#define LX_FILE_METADATA_HAS_DEVICE_ID 0x8 +#define LX_FILE_CASE_SENSITIVE_DIR 0x10 +typedef struct _FILE_STAT_LX_INFORMATION { + LARGE_INTEGER FileId; + LARGE_INTEGER CreationTime; + LARGE_INTEGER LastAccessTime; + LARGE_INTEGER LastWriteTime; + LARGE_INTEGER ChangeTime; + LARGE_INTEGER AllocationSize; + LARGE_INTEGER EndOfFile; + uint32_t FileAttributes; + uint32_t ReparseTag; + uint32_t NumberOfLinks; + ACCESS_MASK EffectiveAccess; + uint32_t LxFlags; + uint32_t LxUid; + uint32_t LxGid; + uint32_t LxMode; + uint32_t LxDeviceIdMajor; + uint32_t LxDeviceIdMinor; +} FILE_STAT_LX_INFORMATION, *PFILE_STAT_LX_INFORMATION; +#endif + +/* + * This struct is extended from the original FILE_FULL_EA_INFORMATION of + * Microsoft Windows. + */ +struct wsl_full_ea_info_t { + uint32_t NextEntryOffset; + uint8_t Flags; + uint8_t EaNameLength; + uint16_t EaValueLength; + char EaName[7]; + char EaValue[4]; + char Padding[1]; +}; + +enum { + FileStatLxInformation = 70, +}; +__declspec(dllimport) NTSTATUS WINAPI + NtQueryInformationFile(HANDLE FileHandle, + PIO_STATUS_BLOCK IoStatusBlock, + PVOID FileInformation, ULONG Length, + uint32_t FileInformationClass); +__declspec(dllimport) NTSTATUS WINAPI + NtSetInformationFile(HANDLE FileHandle, PIO_STATUS_BLOCK IoStatusBlock, + PVOID FileInformation, ULONG Length, + uint32_t FileInformationClass); +__declspec(dllimport) NTSTATUS WINAPI + NtSetEaFile(HANDLE FileHandle, PIO_STATUS_BLOCK IoStatusBlock, + PVOID EaBuffer, ULONG EaBufferSize); + +int set_wsl_mode_bits_by_handle(HANDLE h, _mode_t mode) +{ + uint32_t value = mode; + struct wsl_full_ea_info_t ea_info; + IO_STATUS_BLOCK iob; + /* mode should be valid to make WSL happy */ + assert(S_ISREG(mode) || S_ISDIR(mode)); + ea_info.NextEntryOffset = 0; + ea_info.Flags = 0; + ea_info.EaNameLength = 6; + ea_info.EaValueLength = sizeof(value); /* 4 */ + strlcpy(ea_info.EaName, "$LXMOD", sizeof(ea_info.EaName)); + memcpy(ea_info.EaValue, &value, sizeof(value)); + ea_info.Padding[0] = 0; + return NtSetEaFile(h, &iob, &ea_info, sizeof(ea_info)); +} + +int get_wsl_mode_bits_by_handle(HANDLE h, _mode_t *mode) +{ + FILE_STAT_LX_INFORMATION fxi; + IO_STATUS_BLOCK iob; + if (NtQueryInformationFile(h, &iob, &fxi, sizeof(fxi), + FileStatLxInformation) == 0) { + if (fxi.LxFlags & LX_FILE_METADATA_HAS_MODE) + *mode = (_mode_t)fxi.LxMode; + return 0; + } + return -1; +} diff --git a/compat/win32/wsl.h b/compat/win32/wsl.h new file mode 100644 index 00000000000000..1f5ad7e67a4fc2 --- /dev/null +++ b/compat/win32/wsl.h @@ -0,0 +1,12 @@ +#ifndef COMPAT_WIN32_WSL_H +#define COMPAT_WIN32_WSL_H + +int are_wsl_compatible_mode_bits_enabled(void); + +int copy_wsl_mode_bits_from_disk(const wchar_t *wpath, ssize_t wpathlen, + _mode_t *mode); + +int get_wsl_mode_bits_by_handle(HANDLE h, _mode_t *mode); +int set_wsl_mode_bits_by_handle(HANDLE h, _mode_t mode); + +#endif diff --git a/config.mak.uname b/config.mak.uname index 6d17f9d3a80346..b28e40caff6ae9 100644 --- a/config.mak.uname +++ b/config.mak.uname @@ -528,7 +528,7 @@ endif compat/win32/path-utils.o \ compat/win32/pthread.o compat/win32/syslog.o \ compat/win32/trace2_win32_process_info.o \ - compat/win32/dirent.o compat/win32/fscache.o + compat/win32/dirent.o compat/win32/fscache.o compat/win32/wsl.o COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\" BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO # invalidcontinue.obj allows Git's source code to close the same file @@ -731,7 +731,7 @@ ifeq ($(uname_S),MINGW) compat/win32/flush.o \ compat/win32/path-utils.o \ compat/win32/pthread.o compat/win32/syslog.o \ - compat/win32/dirent.o compat/win32/fscache.o + compat/win32/dirent.o compat/win32/fscache.o compat/win32/wsl.o BASIC_CFLAGS += -DWIN32 EXTLIBS += -lws2_32 GITLIBS += git.res diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index 4805206b10a997..47d9a0982f078c 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -293,6 +293,7 @@ if(CMAKE_SYSTEM_NAME STREQUAL "Windows") compat/win32/syslog.c compat/win32/trace2_win32_process_info.c compat/win32/dirent.c + compat/win32/wsl.c compat/nedmalloc/nedmalloc.c compat/strdup.c compat/win32/fscache.c) diff --git a/meson.build b/meson.build index 327ce7e7387b95..a287f5372af694 100644 --- a/meson.build +++ b/meson.build @@ -1251,6 +1251,7 @@ elif host_machine.system() == 'windows' 'compat/win32/path-utils.c', 'compat/win32/pthread.c', 'compat/win32/syslog.c', + 'compat/win32/wsl.c', 'compat/win32mmap.c', 'compat/nedmalloc/nedmalloc.c', ] From 5efb26b8df6b989fdfe3ff7eaac749bf18b37ed6 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 23 Apr 2018 00:24:29 +0200 Subject: [PATCH 234/248] mingw: really handle SIGINT Previously, we did not install any handler for Ctrl+C, but now we really want to because the MSYS2 runtime learned the trick to call the ConsoleCtrlHandler when Ctrl+C was pressed. With this, hitting Ctrl+C while `git log` is running will only terminate the Git process, but not the pager. This finally matches the behavior on Linux and on macOS. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/compat/mingw.c b/compat/mingw.c index 54d8a65d80fe0a..eeb3681b9d052a 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -4121,7 +4121,14 @@ static void adjust_symlink_flags(void) symlink_file_flags |= 2; symlink_directory_flags |= 2; } +} +static BOOL WINAPI handle_ctrl_c(DWORD ctrl_type) +{ + if (ctrl_type != CTRL_C_EVENT) + return FALSE; /* we did not handle this */ + mingw_raise(SIGINT); + return TRUE; /* we did handle this */ } #ifdef _MSC_VER @@ -4158,6 +4165,8 @@ int wmain(int argc, const wchar_t **wargv) #endif #endif + SetConsoleCtrlHandler(handle_ctrl_c, TRUE); + maybe_redirect_std_handles(); adjust_symlink_flags(); fsync_object_files = 1; From 3c0040efb93936aba4e0760a568af2acea583e24 Mon Sep 17 00:00:00 2001 From: "Neeraj K. Singh" <neerajsi@microsoft.com> Date: Wed, 27 Oct 2021 14:22:42 -0700 Subject: [PATCH 235/248] mingw: do not call xutftowcs_path in mingw_mktemp The `xutftowcs_path` function canonicalizes absolute paths using GetFullPathNameW. This canonicalization may change the length of the string (e.g. getting rid of \.\), which breaks callers that pass the template string in a strbuf and expect the length of the string to remain the same. In my particular case, the tmp-objdir code is passing a strbuf to mkdtemp and is breaking since the strbuf.len is no longer synchronized with strlen(strbuf.buf). Signed-off-by: Neeraj K. Singh <neerajsi@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- compat/mingw.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/compat/mingw.c b/compat/mingw.c index 0673b4a121ad07..15cce47a9763f2 100644 --- a/compat/mingw.c +++ b/compat/mingw.c @@ -1371,8 +1371,11 @@ char *mingw_mktemp(char *template) int offset = 0; /* we need to return the path, thus no long paths here! */ - if (xutftowcs_path(wtemplate, template) < 0) + if (xutftowcsn(wtemplate, template, MAX_PATH, -1) < 0) { + if (errno == ERANGE) + errno = ENAMETOOLONG; return NULL; + } if (is_dir_sep(template[0]) && !is_dir_sep(template[1]) && iswalpha(wtemplate[0]) && wtemplate[1] == L':') { From c2d851e7053bc653222735c9d9750eb631dfb236 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 29 Sep 2020 13:50:59 +0200 Subject: [PATCH 236/248] Add a GitHub workflow to monitor component updates MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Rather than using private IFTTT Applets that send mails to this maintainer whenever a new version of a Git for Windows component was released, let's use the power of GitHub workflows to make this process publicly visible. This workflow monitors the Atom/RSS feeds, and opens a ticket whenever a new version was released. Note: Bash sometimes releases multiple patched versions within a few minutes of each other (i.e. 5.1p1 through 5.1p4, 5.0p15 and 5.0p16). The MSYS2 runtime also has a similar system. We can address those patches as a group, so we shouldn't get multiple issues about them. Note further: We're not acting on newlib releases, OpenSSL alphas, Perl release candidates or non-stable Perl releases. There's no need to open issues about them. Co-authored-by: Matthias Aßhauer <mha1993@live.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- .github/workflows/monitor-components.yml | 94 ++++++++++++++++++++++++ 1 file changed, 94 insertions(+) create mode 100644 .github/workflows/monitor-components.yml diff --git a/.github/workflows/monitor-components.yml b/.github/workflows/monitor-components.yml new file mode 100644 index 00000000000000..a074403c64f42d --- /dev/null +++ b/.github/workflows/monitor-components.yml @@ -0,0 +1,94 @@ +name: Monitor component updates + +# Git for Windows is a slightly modified subset of MSYS2. Some of its +# components are maintained by Git for Windows, others by MSYS2. To help +# keeping the former up to date, this workflow monitors the Atom/RSS feeds +# and opens new tickets for each new component version. + +on: + schedule: + - cron: "23 8,11,14,17 * * *" + workflow_dispatch: + +env: + CHARACTER_LIMIT: 5000 + MAX_AGE: 7d + +jobs: + job: + # Only run this in Git for Windows' fork + if: github.event.repository.owner.login == 'git-for-windows' + runs-on: ubuntu-latest + permissions: + issues: write + strategy: + matrix: + component: + - label: git + feed: https://github.com/git/git/tags.atom + - label: git-lfs + feed: https://github.com/git-lfs/git-lfs/tags.atom + - label: git-credential-manager + feed: https://github.com/git-ecosystem/git-credential-manager/tags.atom + - label: tig + feed: https://github.com/jonas/tig/tags.atom + - label: cygwin + feed: https://github.com/cygwin/cygwin/releases.atom + title-pattern: ^(?!.*newlib) + - label: msys2-runtime-package + feed: https://github.com/msys2/MSYS2-packages/commits/master/msys2-runtime.atom + - label: msys2-runtime + feed: https://github.com/msys2/msys2-runtime/commits/HEAD.atom + aggregate: true + - label: openssh + feed: https://github.com/openssh/openssh-portable/tags.atom + - label: libfido2 + feed: https://github.com/Yubico/libfido2/tags.atom + - label: libcbor + feed: https://github.com/PJK/libcbor/tags.atom + - label: openssl + feed: https://github.com/openssl/openssl/tags.atom + title-pattern: ^(?!.*alpha) + - label: gnutls + feed: https://gnutls.org/news.atom + - label: heimdal + feed: https://github.com/heimdal/heimdal/tags.atom + - label: git-sizer + feed: https://github.com/github/git-sizer/tags.atom + - label: gitflow + feed: https://github.com/petervanderdoes/gitflow-avh/tags.atom + - label: curl + feed: https://github.com/curl/curl/tags.atom + title-pattern: ^(?!rc-) + - label: mintty + feed: https://github.com/mintty/mintty/releases.atom + - label: 7-zip + feed: https://sourceforge.net/projects/sevenzip/rss?path=/7-Zip + aggregate: true + - label: bash + feed: https://git.savannah.gnu.org/cgit/bash.git/atom/?h=master + aggregate: true + - label: perl + feed: https://github.com/Perl/perl5/tags.atom + title-pattern: ^(?!.*(5\.[0-9]+[13579]|RC)) + - label: pcre2 + feed: https://github.com/PCRE2Project/pcre2/tags.atom + - label: mingw-w64-llvm + feed: https://github.com/msys2/MINGW-packages/commits/master/mingw-w64-llvm.atom + - label: innosetup + feed: https://github.com/jrsoftware/issrc/tags.atom + - label: mimalloc + feed: https://github.com/microsoft/mimalloc/tags.atom + title-pattern: ^(?!v1\.) + fail-fast: false + steps: + - uses: git-for-windows/rss-to-issues@v0 + with: + feed: ${{matrix.component.feed}} + prefix: "[New ${{matrix.component.label}} version]" + labels: component-update + github-token: ${{ secrets.GITHUB_TOKEN }} + character-limit: ${{ env.CHARACTER_LIMIT }} + max-age: ${{ env.MAX_AGE }} + aggregate: ${{matrix.component.aggregate}} + title-pattern: ${{matrix.component.title-pattern}} From 37810912ccf51af3de920979a1e6aa56c619e777 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Thu, 25 Nov 2021 11:26:41 +0100 Subject: [PATCH 237/248] Partially un-revert "editor: save and reset terminal after calling EDITOR" In e3f7e01b50be (Revert "editor: save and reset terminal after calling EDITOR", 2021-11-22), we reverted the commit wholesale where the terminal state would be saved and restored before/after calling an editor. The reverted commit was intended to fix a problem with Windows Terminal where simply calling `vi` would cause problems afterwards. To fix the problem addressed by the revert, but _still_ keep the problem with Windows Terminal fixed, let's revert the revert, with a twist: we restrict the save/restore _specifically_ to the case where `vi` (or `vim`) is called, and do not do the same for any other editor. This should still catch the majority of the cases, and will bridge the time until the original patch is re-done in a way that addresses all concerns. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- editor.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/editor.c b/editor.c index fd174e6a034f1c..f6d960c6f30782 100644 --- a/editor.c +++ b/editor.c @@ -13,6 +13,7 @@ #include "strvec.h" #include "run-command.h" #include "sigchain.h" +#include "compat/terminal.h" #ifndef DEFAULT_EDITOR #define DEFAULT_EDITOR "vi" @@ -64,6 +65,7 @@ static int launch_specified_editor(const char *editor, const char *path, return error("Terminal is dumb, but EDITOR unset"); if (strcmp(editor, ":")) { + int save_and_restore_term = !strcmp(editor, "vi") || !strcmp(editor, "vim"); struct strbuf realpath = STRBUF_INIT; struct child_process p = CHILD_PROCESS_INIT; int ret, sig; @@ -92,7 +94,11 @@ static int launch_specified_editor(const char *editor, const char *path, strvec_pushv(&p.env, (const char **)env); p.use_shell = 1; p.trace2_child_class = "editor"; + if (save_and_restore_term) + save_and_restore_term = !save_term(1); if (start_command(&p) < 0) { + if (save_and_restore_term) + restore_term(); strbuf_release(&realpath); return error("unable to start editor '%s'", editor); } @@ -100,6 +106,8 @@ static int launch_specified_editor(const char *editor, const char *path, sigchain_push(SIGINT, SIG_IGN); sigchain_push(SIGQUIT, SIG_IGN); ret = finish_command(&p); + if (save_and_restore_term) + restore_term(); strbuf_release(&realpath); sig = ret - 128; sigchain_pop(SIGINT); From b846014792d69f396429baa31f9132863a981522 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 10 Dec 2019 21:41:57 +0100 Subject: [PATCH 238/248] reset: reinstate support for the deprecated --stdin option The `--stdin` option was a well-established paradigm in other commands, therefore we implemented it in `git reset` for use by Visual Studio. Unfortunately, upstream Git decided that it is time to introduce `--pathspec-from-file` instead. To keep backwards-compatibility for some grace period, we therefore reinstate the `--stdin` option on top of the `--pathspec-from-file` option, but mark it firmly as deprecated. Helped-by: Victoria Dye <vdye@github.com> Helped-by: Matthew John Cheetham <mjcheetham@outlook.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- Documentation/git-reset.adoc | 11 +++++++++++ builtin/reset.c | 16 ++++++++++++++++ t/meson.build | 1 + t/t7108-reset-stdin.sh | 32 ++++++++++++++++++++++++++++++++ 4 files changed, 60 insertions(+) create mode 100755 t/t7108-reset-stdin.sh diff --git a/Documentation/git-reset.adoc b/Documentation/git-reset.adoc index 50e8a0ba6f6612..4bbf1e306b40ce 100644 --- a/Documentation/git-reset.adoc +++ b/Documentation/git-reset.adoc @@ -12,6 +12,7 @@ git reset [-q] [<tree-ish>] [--] <pathspec>... git reset [-q] [--pathspec-from-file=<file> [--pathspec-file-nul]] [<tree-ish>] git reset (--patch | -p) [<tree-ish>] [--] [<pathspec>...] git reset [--soft | --mixed [-N] | --hard | --merge | --keep] [-q] [<commit>] +DEPRECATED: git reset [-q] [--stdin [-z]] [<tree-ish>] DESCRIPTION ----------- @@ -135,6 +136,16 @@ include::diff-context-options.adoc[] + For more details, see the 'pathspec' entry in linkgit:gitglossary[7]. +--stdin:: + DEPRECATED (use `--pathspec-from-file=-` instead): Instead of taking + list of paths from the command line, read list of paths from the + standard input. Paths are separated by LF (i.e. one path per line) by + default. + +-z:: + DEPRECATED (use `--pathspec-file-nul` instead): Only meaningful with + `--stdin`; paths are separated with NUL character instead of LF. + EXAMPLES -------- diff --git a/builtin/reset.c b/builtin/reset.c index ed35802af15c94..54244b6e32ea5a 100644 --- a/builtin/reset.c +++ b/builtin/reset.c @@ -38,6 +38,8 @@ #include "trace2.h" #include "dir.h" #include "add-interactive.h" +#include "strbuf.h" +#include "quote.h" #define REFRESH_INDEX_DELAY_WARNING_IN_MS (2 * 1000) @@ -46,6 +48,7 @@ static const char * const git_reset_usage[] = { N_("git reset [-q] [<tree-ish>] [--] <pathspec>..."), N_("git reset [-q] [--pathspec-from-file [--pathspec-file-nul]] [<tree-ish>]"), N_("git reset --patch [<tree-ish>] [--] [<pathspec>...]"), + N_("DEPRECATED: git reset [-q] [--stdin [-z]] [<tree-ish>]"), NULL }; @@ -347,6 +350,7 @@ int cmd_reset(int argc, struct pathspec pathspec; int intent_to_add = 0; struct add_p_opt add_p_opt = ADD_P_OPT_INIT; + int nul_term_line = 0, read_from_stdin = 0; const struct option options[] = { OPT__QUIET(&quiet, N_("be quiet, only report errors")), OPT_BOOL(0, "no-refresh", &no_refresh, @@ -377,6 +381,10 @@ int cmd_reset(int argc, N_("record only the fact that removed paths will be added later")), OPT_PATHSPEC_FROM_FILE(&pathspec_from_file), OPT_PATHSPEC_FILE_NUL(&pathspec_file_nul), + OPT_BOOL('z', NULL, &nul_term_line, + N_("DEPRECATED (use --pathspec-file-nul instead): paths are separated with NUL character")), + OPT_BOOL(0, "stdin", &read_from_stdin, + N_("DEPRECATED (use --pathspec-from-file=- instead): read paths from <stdin>")), OPT_END() }; @@ -386,6 +394,14 @@ int cmd_reset(int argc, PARSE_OPT_KEEP_DASHDASH); parse_args(&pathspec, argv, prefix, patch_mode, &rev); + if (read_from_stdin) { + warning(_("--stdin is deprecated, please use --pathspec-from-file=- instead")); + free(pathspec_from_file); + pathspec_from_file = xstrdup("-"); + if (nul_term_line) + pathspec_file_nul = 1; + } + if (pathspec_from_file) { if (patch_mode) die(_("options '%s' and '%s' cannot be used together"), "--pathspec-from-file", "--patch"); diff --git a/t/meson.build b/t/meson.build index 25d9e33697adff..cb25d951e67135 100644 --- a/t/meson.build +++ b/t/meson.build @@ -848,6 +848,7 @@ integration_tests = [ 't7105-reset-patch.sh', 't7106-reset-unborn-branch.sh', 't7107-reset-pathspec-file.sh', + 't7108-reset-stdin.sh', 't7110-reset-merge.sh', 't7111-reset-table.sh', 't7112-reset-submodule.sh', diff --git a/t/t7108-reset-stdin.sh b/t/t7108-reset-stdin.sh new file mode 100755 index 00000000000000..b7cbcbf869296c --- /dev/null +++ b/t/t7108-reset-stdin.sh @@ -0,0 +1,32 @@ +#!/bin/sh + +test_description='reset --stdin' + +. ./test-lib.sh + +test_expect_success 'reset --stdin' ' + test_commit hello && + git rm hello.t && + test -z "$(git ls-files hello.t)" && + echo hello.t | git reset --stdin && + test hello.t = "$(git ls-files hello.t)" +' + +test_expect_success 'reset --stdin -z' ' + test_commit world && + git rm hello.t world.t && + test -z "$(git ls-files hello.t world.t)" && + printf world.tQworld.tQhello.tQ | q_to_nul | git reset --stdin -z && + printf "hello.t\nworld.t\n" >expect && + git ls-files >actual && + test_cmp expect actual +' + +test_expect_success '--stdin requires --mixed' ' + echo hello.t >list && + test_must_fail git reset --soft --stdin <list && + test_must_fail git reset --hard --stdin <list && + git reset --mixed --stdin <list +' + +test_done From bc05c5190a896a111b61a4acbad74950a602f36b Mon Sep 17 00:00:00 2001 From: Victoria Dye <vdye@github.com> Date: Mon, 4 Apr 2022 15:38:58 -0700 Subject: [PATCH 239/248] fsmonitor: reintroduce core.useBuiltinFSMonitor Reintroduce the 'core.useBuiltinFSMonitor' config setting (originally added in 0a756b2a25 (fsmonitor: config settings are repository-specific, 2021-03-05)) after its removal from the upstream version of FSMonitor. Upstream, the 'core.useBuiltinFSMonitor' setting was rendered obsolete by "overloading" the 'core.fsmonitor' setting to take a boolean value. However, several applications (e.g., 'scalar') utilize the original config setting, so it should be preserved for a deprecation period before complete removal: * if 'core.fsmonitor' is a boolean, the user is correctly using the new config syntax; do not use 'core.useBuiltinFSMonitor'. * if 'core.fsmonitor' is unspecified, use 'core.useBuiltinFSMonitor'. * if 'core.fsmonitor' is a path, override and use the builtin FSMonitor if 'core.useBuiltinFSMonitor' is 'true'; otherwise, use the FSMonitor hook indicated by the path. Additionally, for this deprecation period, advise users to switch to using 'core.fsmonitor' to specify their use of the builtin FSMonitor. Signed-off-by: Victoria Dye <vdye@github.com> --- Documentation/config/advice.adoc | 4 ++++ advice.c | 1 + advice.h | 1 + fsmonitor-settings.c | 34 ++++++++++++++++++++++++++++++-- 4 files changed, 38 insertions(+), 2 deletions(-) diff --git a/Documentation/config/advice.adoc b/Documentation/config/advice.adoc index 257db58918179a..f156f638dcd5ee 100644 --- a/Documentation/config/advice.adoc +++ b/Documentation/config/advice.adoc @@ -166,4 +166,8 @@ all advice messages. Shown when the user tries to create a worktree from an invalid reference, to tell the user how to create a new unborn branch instead. + + useCoreFSMonitorConfig:: + Advice shown if the deprecated 'core.useBuiltinFSMonitor' config + setting is in use. -- diff --git a/advice.c b/advice.c index e5f0ff844917ec..e6a8f62185e042 100644 --- a/advice.c +++ b/advice.c @@ -89,6 +89,7 @@ static struct { [ADVICE_SUBMODULE_MERGE_CONFLICT] = { "submoduleMergeConflict" }, [ADVICE_SUGGEST_DETACHING_HEAD] = { "suggestDetachingHead" }, [ADVICE_UPDATE_SPARSE_PATH] = { "updateSparsePath" }, + [ADVICE_USE_CORE_FSMONITOR_CONFIG] = { "useCoreFSMonitorConfig" }, [ADVICE_WAITING_FOR_EDITOR] = { "waitingForEditor" }, [ADVICE_WORKTREE_ADD_ORPHAN] = { "worktreeAddOrphan" }, }; diff --git a/advice.h b/advice.h index 727dcecf4a3ee0..209a4308cb56b2 100644 --- a/advice.h +++ b/advice.h @@ -56,6 +56,7 @@ enum advice_type { ADVICE_SUBMODULE_MERGE_CONFLICT, ADVICE_SUGGEST_DETACHING_HEAD, ADVICE_UPDATE_SPARSE_PATH, + ADVICE_USE_CORE_FSMONITOR_CONFIG, ADVICE_WAITING_FOR_EDITOR, ADVICE_WORKTREE_ADD_ORPHAN, }; diff --git a/fsmonitor-settings.c b/fsmonitor-settings.c index a6587a8972b184..b4c29f44a27827 100644 --- a/fsmonitor-settings.c +++ b/fsmonitor-settings.c @@ -5,6 +5,7 @@ #include "fsmonitor-ipc.h" #include "fsmonitor-settings.h" #include "fsmonitor-path-utils.h" +#include "advice.h" /* * We keep this structure definition private and have getters @@ -100,6 +101,31 @@ static struct fsmonitor_settings *alloc_settings(void) return s; } +static int check_deprecated_builtin_config(struct repository *r) +{ + int core_use_builtin_fsmonitor = 0; + + /* + * If 'core.useBuiltinFSMonitor' is set, print a deprecation warning + * suggesting the use of 'core.fsmonitor' instead. If the config is + * set to true, set the appropriate mode and return 1 indicating that + * the check resulted the config being set by this (deprecated) setting. + */ + if(!repo_config_get_bool(r, "core.useBuiltinFSMonitor", &core_use_builtin_fsmonitor) && + core_use_builtin_fsmonitor) { + if (!git_env_bool("GIT_SUPPRESS_USEBUILTINFSMONITOR_ADVICE", 0)) { + advise_if_enabled(ADVICE_USE_CORE_FSMONITOR_CONFIG, + _("core.useBuiltinFSMonitor=true is deprecated;" + "please set core.fsmonitor=true instead")); + setenv("GIT_SUPPRESS_USEBUILTINFSMONITOR_ADVICE", "1", 1); + } + fsm_settings__set_ipc(r); + return 1; + } + + return 0; +} + static void lookup_fsmonitor_settings(struct repository *r) { const char *const_str; @@ -126,12 +152,16 @@ static void lookup_fsmonitor_settings(struct repository *r) return; case 1: /* config value was unset */ + if (check_deprecated_builtin_config(r)) + return; + const_str = getenv("GIT_TEST_FSMONITOR"); break; case -1: /* config value set to an arbitrary string */ - if (repo_config_get_pathname(r, "core.fsmonitor", &to_free)) - return; /* should not happen */ + if (check_deprecated_builtin_config(r) || + repo_config_get_pathname(r, "core.fsmonitor", &to_free)) + return; const_str = to_free; break; From 1559b38714e28cdcb8a2918eae5f483ba323e9f6 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Mon, 13 Feb 2023 13:31:35 +0100 Subject: [PATCH 240/248] Describe Git for Windows' architecture [no ci] The Git for Windows project has grown quite complex over the years, certainly much more complex than during the first years where the `msysgit.git` repository was abusing Git for package management purposes and the `git/git` fork was called `4msysgit.git`. Let's describe the status quo in a thorough way. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- ARCHITECTURE.md | 116 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 116 insertions(+) create mode 100644 ARCHITECTURE.md diff --git a/ARCHITECTURE.md b/ARCHITECTURE.md new file mode 100644 index 00000000000000..7de4f99bf71ec4 --- /dev/null +++ b/ARCHITECTURE.md @@ -0,0 +1,116 @@ +# Architecture of Git for Windows + +Git for Windows is a complex project. + +## What _is_ Git for Windows? + +### A fork of `git/git` + +First and foremost, it is a friendly fork of [`git/git`](https://github.com/git/git), aiming to improve Git's Windows support. The [`git-for-windows/git`](https://github.com/git-for-windows/git) repository contains dozens of topics on top of `git/git`, some awaiting to be "upstreamed" (i.e. to be contributed to `git/git`), some still being stabilized, and a few topics are specific to the Git for Windows project and are not intended to be integrated into `git/git` at all. + +### Enhancing and maintaining Git's support for Windows + +On the source code side, Git's Windows support is made a bit more tricky than strictly necessary by the fact that Git does not have any platform abstraction layer (unlike other version control systems, such as Subversion). It relies on the presence of POSIX features such as the `hstrerror()` function, and on platforms lacking that functionality, Git provides shims. That leads to some challenges e.g. with the `stat()` function which is very slow on Windows because it has to collect much more metadata than what e.g. the very quick `GetFileAttributesExW()` Win32 API function provides, even when Git calls `stat()` merely to test for the presence of a file (for which all that gathered metadata is totally irrelevant). + +### Providing more than just source code + +In contrast to the Git project, Git for Windows not only publishes tagged source code versions, but full builds of Git. In fact, Git for Windows' primary purpose, as far as most users are concerned, is to provide a convenient installer that end-users can run to have Git on their computer, without ever having to check out `git-for-windows/git` let alone build it. In essence, Git for Windows has to maintain a separate project altogether in addition to the fork of `git/git`, just to build these release artifacts: [`git-for-windows/build-extra`](https://github.com/git-for-windows/build-extra). This repository also contains the definition for a couple of other release artifacts published by Git for Windows, e.g. the "portable" edition of Git for Windows which is a self-extracting 7-Zip archive that does not need to be installed. + +### A software distribution, really + +Another aspect that contributes to the complexity of Git for Windows is that it is not just building `git.exe` and distributes that. Due to its heritage within the Linux project, Git takes certain things for granted, such as the presence of a Unix shell, or for that matter, a package management system from which dependencies can be fetched and updated independently of Git itself. Things that are distinctly not present in most Windows setups. To accommodate for that, Git for Windows originally relied on the MSys project, a minimal fork of Cygwin providing a Unix shell ("Bash"), a Perl interpreter and similar Unix-like tools, and on the MINGW project, a project to build libraries and executables using a GNU C Compiler that relies only on Win32 API functions. As of Git for Windows v2.x, the project has switched away from [MSys](https://sourceforge.net/projects/mingw/files/MSYS/)/[MinGW](https://osdn.net/projects/mingw/) (due to less-than-active maintenance) to [the MSYS2 project](https://msys2.org). That switch brought along the benefit of a robust package management system based on [Pacman](https://archlinux.org/pacman/) (hailing from Arch Linux). To support Windows users, who are in general unfamiliar with Linux-like package management and the need to update installed packages frequently, Git for Windows bundles a subset of its own fork of MSYS2. To put things in perspective: Git for Windows bundles files from ~170 packages, one of which contains Git, and another one contains Git's help files. In that respect, Git for Windows acts like a distribution more than like a mere single software application. + +Most of MSYS2's packages that are bundled in Git for Windows are consumed directly from MSYS2. Others need forks that are maintained by Git for Windows project, to support Git for Windows better. These forks live in the [`git-for-windows/MSYS2-packages`](https://github.com/git-for-windows/MSYS2-packages) and [`git-for-windows/MINGW-packages`](https://github.com/git-for-windows/MINGW-packages) repositories. There are several reasons justifying these forks. For example, the Git for Windows' flavor of the MSYS2 runtime behaves like Git's test suite expects it while MSYS2's flavor does not. Another example: The Bash executable bundled in Git for Windows is code-signed with the same certificate as `git.exe` to help anti-malware programs get out of the users' way. That is why Git for Windows maintains its own `bash` Pacman package. And since MSYS2 dropped 32-bit support already, Git for Windows has to update the 32-bit Pacman packages itself, which is done in the git-for-windows/MSYS2-packages repository. (Side note: the 32-bit issue is a bit more complicated, actually: MSYS2 _still_ builds _MINGW_ packages targeting i686 processors, but no longer any _MSYS_ packages for said processor architecture, and Git for Windows does not keep all of the 32-bit MSYS packages up to date but instead judiciously decides which packages are vital enough as far as Git is concerned to justify the maintenance cost.) + +### Supporting third-party applications that use Git's functionality + +Since the infrastructure required by Git is non-trivial the installer (or for that matter, the Portable Git) is not exactly light-weight: As of January 2023, both artifacts are over fifty megabytes. This is a problem for third-party applications wishing to bundle a version of Git for Windows, which is often advisable given that applications may depend on features that have been introduced only in recent Git versions and therefore relying on an installed Git for Windows could break things. To help with that, the Git for Windows project also provides MinGit as a release artifact, a zip file that is much smaller than the full installer and that contains only the parts of Git for Windows relevant for third-party applications. It lacks Git GUI, for example, as well as the terminal program MinTTY, or for that matter, the documentation. + +### Supporting `git/git`'s GitHub workflows + +The Git for Windows project is also responsible for keeping the Windows part of `git/git`'s automated builds up and running. On Windows, there is no canonical and easy way to get a build environment necessary to build Git and run its test suite, therefore this is a non-trivial task that comes with its own maintenance cost. Git for Windows provides two GitHub Actions to help with that: [`git-for-windows/setup-git-for-windows-sdk`](https://github.com/git-for-windows/setup-git-for-windows-sdk) to set up a tiny subset of Git for Windows' full SDK (which would require about 500MB to be cloned, as opposed to the ~75MB of that subset) and [`git-for-windows/get-azure-pipelines-artifact`](https://github.com/git-for-windows/get-azure-pipelines-artifact) e.g. to download some regularly pre-built artifacts (for example, when `git/git`'s automated tests ran on an Ubuntu version that did not provide an up to date [Coccinelle](https://coccinelle.gitlabpages.inria.fr/website/) package, this GitHub Action was used to download a pre-built version of that Debian package). + +## Maintaining Git for Windows' components + +Git for Windows uses a combination of [a GitHub App called GitForWindowsHelper](https://github.com/git-for-windows/gfw-helper-github-app) (to listen for so-called [slash commands](https://github.com/git-for-windows/gfw-helper-github-app#slash-commands)) combined with workflows in [the `git-for-windows-automation` repository](https://github.com/git-for-windows/git-for-windows-automation/) (for computationally heavy tasks) to support Git for Windows' repetitive tasks. + +This heavy automation serves two purposes: + +1. Document the knowledge about "how things are done" in the Git for Windows project. +2. Make Git for Windows' maintenance less tedious by off-loading as many tasks onto machines as possible. + +One neat trick of some `git-for-windows-automation` workflows is that they "mirror back" check runs to the targeted PRs in another repository. This essentially allows versioning the source code independently of the workflow definition. + +Here is a diagram showing how the bits and pieces fit together. + +```mermaid +graph LR + A[`monitor-components`] --> |opens| B + B{issues labeled<br />`component-update`} --> |/open pr| C + C((GitForWindowsHelper)) --> |triggers| D + D[`open-pr`] --> |opens| E + E{PR in</br>MINGW-packages<br />MSYS2-packages<br />build-extra} --> |closes| B + E --> |/deploy| F + F((GitForWindowsHelper)) --> |triggers| G + G[`build-and-deploy`] --> |deploys to| H + H{Pacman repository} + C --> |backed by| I + F --> |backed by| I + I[[Azure Function]] + D --> |running in| J + G --> | running in| J + J[[git-for-windows-automation]] + K[[git-sdk-32<br />git-sdk-64<br />git-sdk-arm64]] --> |syncing from| H + B --> |/add release note| L + L[`add-release-note`] +``` + +For the curious mind, here are [detailed instructions how the Azure Function backing the GitForWindowsHelper GitHub App was set up](https://github.com/git-for-windows/gfw-helper-github-app#how-this-github-app-was-set-up). + +### The `monitor-components` workflow + +When new versions of components that Git for Windows builds become available, new Pacman packages have to be built. To this end, [the `monitor-components` workflow](https://github.com/git-for-windows/git/blob/main/.github/workflows/monitor-components.yml) monitors a couple of RSS feeds and opens new tickets labeled `component-update` for such new versions. + +### Opening Pull Requests to update Git for Windows' components + +After determining that such a ticket indeed indicates the need for a new Pacman package build, a Git for Windows maintainer issues the `/open pr` command via an issue comment ([example](https://github.com/git-for-windows/git/issues/4281#issuecomment-1426859787)), which gets picked up by the GitForWindowsHelper GitHub App, which in turn triggers [the `open-pr` workflow](https://github.com/git-for-windows/git-for-windows-automation/blob/main/.github/workflows/open-pr.yml) in the `git-for-windows-automation` repository. + +### Deploying the Pacman packages + +This will open a Pull Request in one of Git for Windows' repositories, and once the PR build passes, a Git for Windows maintainer issues the `/deploy` command ([example](https://github.com/git-for-windows/MINGW-packages/pull/69#issuecomment-1427591890)), which gets picked up by the GitForWindowsHelper GitHub App, which triggers [the `build-and-deploy` workflow](https://github.com/git-for-windows/git-for-windows-automation/blob/main/.github/workflows/build-and-deploy.yml). + +### Adding release notes + +Finally, once the packages have been built and deployed to the Pacman repository (which is hosted in Azure Blob Storage), a Git for Windows maintainer will merge the PR(s), which in turn will close the ticket, and the maintainer then issues an `/add release note` command ([example](https://github.com/git-for-windows/MINGW-packages/pull/69#issuecomment-1427782230)), which again gets picked up by the GitForWindowsHelper GitHub App that triggers [the `add-release-note` workflow](https://github.com/git-for-windows/build-extra/blob/main/.github/workflows/add-release-note.yml) that creates and pushes a new commit to the `ReleaseNotes.md` file in `build-extra` ([example](https://github.com/git-for-windows/build-extra/commit/b39c148ff8dc0e987afdb677d17c46a8e99fd0ef)). + +## Releasing official Git for Windows versions + +A relatively infrequent part of Git for Windows' maintainers' duties, if the most rewarding part, is the task of releasing new versions of Git for Windows. + +Most commonly, this is done in response to the "upstream" Git project releasing a new version. When that happens, a Git for Windows maintainer runs [the helper script](https://github.com/git-for-windows/build-extra/blob/main/shears.sh) to perform a "merging rebase" (i.e. a rebase that starts with a fake-merge of the previous tip commit, to maintain both a clean set of commits as well as a [fast-forwarding](https://git-scm.com/docs/git-merge#Documentation/git-merge.txt---ff-only) commit history). + +Once that is done, the maintainer will open a Pull Request to benefit from the automated builds and tests ([example](https://github.com/git-for-windows/git/pull/4160)) as well as from reviews of the [`range-diff`](https://git-scm.com/docs/git-range-diff) relative to the current `main` branch. + +Once everything looks good, the maintainer will issue the `/git-artifacts` command ([example](https://github.com/git-for-windows/git/pull/4160#issuecomment-1346801735)). This will trigger an automated workflow that builds all of the release artifacts: installers, Portable Git, MinGit, `.tar.xz` archive and a NuGet package. Apart from the NuGet package, two sets of artifacts are built: targeting 32-bit ("x86") and 64-bit ("amd64"). + +Once these artifacts are built, the maintainer will download the installer and run [the "pre-flight checklist"](https://github.com/git-for-windows/build-extra/blob/main/installer/checklist.txt). + +If everything looks good, a `/release` command will be issued, which triggers yet another workflow that will download the just-built-and-verified release artifacts, publish them as a new GitHub release, publish the NuGet packages, deploy the Pacman packages to the Pacman repository, send out an announcement mail, and update the respective repositories including [Git for Windows' website](https://gitforwindows.org/). + +As mentioned [before](#architecture-of-git-for-windows), the `/git-artifacts` and `/release` commands are picked up by the GitForWindowsHelper GitHub App which subsequently triggers the respective workflows in the `git-for-windows-automation` repository. Here is a diagram: + +```mermaid +graph LR + A{Pull Request<br />updating to<br />new Git version} --> |/git-artifacts| B + B((GitForWindowsHelper)) --> |triggers| C + C[`tag-git`] --> |upon successful build<br />triggers| D + D((GitForWindowsHelper)) --> |triggers| E + E[`git-artifacts`] + E --> |maintainer verifies artifacts| E + A --> |upon verified `git-artifacts`<br />/release| F + F[`release-git`] + C --> |running in| J + E --> | running in| J + F --> | running in| J + J[[git-for-windows-automation]] +``` \ No newline at end of file From 8f60a54e70c14e991b56c7bacfc7716fadf9c0a4 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 11 Oct 2019 13:22:24 +0200 Subject: [PATCH 241/248] Modify the Code of Conduct for Git for Windows The Git project followed Git for Windows' lead and added their Code of Conduct, based on the Contributor Covenant v1.4, later updated to v2.0. We adapt it slightly to Git for Windows. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- CODE_OF_CONDUCT.md | 58 +++++++++++++++++++++------------------------- 1 file changed, 26 insertions(+), 32 deletions(-) diff --git a/CODE_OF_CONDUCT.md b/CODE_OF_CONDUCT.md index e58917c50a96dc..4daef7e3ce9196 100644 --- a/CODE_OF_CONDUCT.md +++ b/CODE_OF_CONDUCT.md @@ -1,9 +1,9 @@ -# Git Code of Conduct +# Git for Windows Code of Conduct This code of conduct outlines our expectations for participants within -the Git community, as well as steps for reporting unacceptable behavior. -We are committed to providing a welcoming and inspiring community for -all and expect our code of conduct to be honored. Anyone who violates +the **Git for Windows** community, as well as steps for reporting unacceptable +behavior. We are committed to providing a welcoming and inspiring community +for all and expect our code of conduct to be honored. Anyone who violates this code of conduct may be banned from the community. ## Our Pledge @@ -12,8 +12,8 @@ We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, -nationality, personal appearance, race, religion, or sexual identity -and orientation. +nationality, personal appearance, race, caste, color, religion, or sexual +identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community. @@ -28,17 +28,17 @@ community include: * Giving and gracefully accepting constructive feedback * Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience -* Focusing on what is best not just for us as individuals, but for the - overall community +* Focusing on what is best not just for us as individuals, but for the overall + community Examples of unacceptable behavior include: -* The use of sexualized language or imagery, and sexual attention or - advances of any kind +* The use of sexualized language or imagery, and sexual attention or advances of + any kind * Trolling, insulting or derogatory comments, and personal or political attacks * Public or private harassment -* Publishing others' private information, such as a physical or email - address, without their explicit permission +* Publishing others' private information, such as a physical or email address, + without their explicit permission * Other conduct which could reasonably be considered inappropriate in a professional setting @@ -58,20 +58,14 @@ decisions when appropriate. This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. -Examples of representing our community include using an official e-mail address, +Examples of representing our community include using an official email address, posting via an official social media account, or acting as an appointed representative at an online or offline event. ## Enforcement Instances of abusive, harassing, or otherwise unacceptable behavior may be -reported to the community leaders responsible for enforcement at -git@sfconservancy.org, or individually: - - - Ævar Arnfjörð Bjarmason <avarab@gmail.com> - - Christian Couder <christian.couder@gmail.com> - - Junio C Hamano <gitster@pobox.com> - - Taylor Blau <me@ttaylorr.com> +reported by contacting the Git for Windows maintainer. All complaints will be reviewed and investigated promptly and fairly. @@ -94,15 +88,15 @@ behavior was inappropriate. A public apology may be requested. ### 2. Warning -**Community Impact**: A violation through a single incident or series -of actions. +**Community Impact**: A violation through a single incident or series of +actions. **Consequence**: A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels -like social media. Violating these terms may lead to a temporary or -permanent ban. +like social media. Violating these terms may lead to a temporary or permanent +ban. ### 3. Temporary Ban @@ -118,27 +112,27 @@ Violating these terms may lead to a permanent ban. ### 4. Permanent Ban **Community Impact**: Demonstrating a pattern of violation of community -standards, including sustained inappropriate behavior, harassment of an +standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. -**Consequence**: A permanent ban from any sort of public interaction within -the community. +**Consequence**: A permanent ban from any sort of public interaction within the +community. ## Attribution This Code of Conduct is adapted from the [Contributor Covenant][homepage], -version 2.0, available at -[https://www.contributor-covenant.org/version/2/0/code_of_conduct.html][v2.0]. +version 2.1, available at +[https://www.contributor-covenant.org/version/2/1/code_of_conduct.html][v2.1]. Community Impact Guidelines were inspired by [Mozilla's code of conduct enforcement ladder][Mozilla CoC]. For answers to common questions about this code of conduct, see the FAQ at -[https://www.contributor-covenant.org/faq][FAQ]. Translations are available -at [https://www.contributor-covenant.org/translations][translations]. +[https://www.contributor-covenant.org/faq][FAQ]. Translations are available at +[https://www.contributor-covenant.org/translations][translations]. [homepage]: https://www.contributor-covenant.org -[v2.0]: https://www.contributor-covenant.org/version/2/0/code_of_conduct.html +[v2.1]: https://www.contributor-covenant.org/version/2/1/code_of_conduct.html [Mozilla CoC]: https://github.com/mozilla/diversity [FAQ]: https://www.contributor-covenant.org/faq [translations]: https://www.contributor-covenant.org/translations From 68aafe1d2868b2ab41f2605ce5fc2f72dbd42e09 Mon Sep 17 00:00:00 2001 From: Derrick Stolee <dstolee@microsoft.com> Date: Thu, 1 Mar 2018 12:10:14 -0500 Subject: [PATCH 242/248] CONTRIBUTING.md: add guide for first-time contributors Getting started contributing to Git can be difficult on a Windows machine. CONTRIBUTING.md contains a guide to getting started, including detailed steps for setting up build tools, running tests, and submitting patches to upstream. [includes an example by Pratik Karki how to submit v2, v3, v4, etc.] Signed-off-by: Derrick Stolee <dstolee@microsoft.com> --- CONTRIBUTING.md | 417 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 417 insertions(+) create mode 100644 CONTRIBUTING.md diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md new file mode 100644 index 00000000000000..48ff9029374df3 --- /dev/null +++ b/CONTRIBUTING.md @@ -0,0 +1,417 @@ +How to Contribute to Git for Windows +==================================== + +Git was originally designed for Unix systems and still today, all the build tools for the Git +codebase assume you have standard Unix tools available in your path. If you have an open-source +mindset and want to start contributing to Git, but primarily use a Windows machine, then you may +have trouble getting started. This guide is for you. + +Get the Source +-------------- + +Clone the [GitForWindows repository on GitHub](https://github.com/git-for-windows/git). +It is helpful to create your own fork for storing your development branches. + +Windows uses different line endings than Unix systems. See +[this GitHub article on working with line endings](https://help.github.com/articles/dealing-with-line-endings/#refreshing-a-repository-after-changing-line-endings) +if you have trouble with line endings. + +Build the Source +---------------- + +First, download and install the latest [Git for Windows SDK (64-bit)](https://github.com/git-for-windows/build-extra/releases/latest). +When complete, you can run the Git SDK, which creates a new Git Bash terminal window with +the additional development commands, such as `make`. + + As of time of writing, the SDK uses a different credential manager, so you may still want to use normal Git + Bash for interacting with your remotes. Alternatively, use SSH rather than HTTPS and + avoid credential manager problems. + +You should now be ready to type `make` from the root of your `git` source directory. +Here are some helpful variations: + +* `make -j[N] DEVELOPER=1`: Compile new sources using up to N concurrent processes. + The `DEVELOPER` flag turns on all warnings; code failing these warnings will not be + accepted upstream ("upstream" = "the core Git project"). +* `make clean`: Delete all compiled files. + +When running `make`, you can use `-j$(nproc)` to automatically use the number of processors +on your machine as the number of concurrent build processes. + +You can go deeper on the Windows-specific build process by reading the +[technical overview](https://gitforwindows.org/technical-overview) or the +[guide to compiling Git with Visual Studio](https://gitforwindows.org/compiling-git-with-visual-studio). + +## Building `git` on Windows with Visual Studio + +The typical approach to building `git` is to use the standard `Makefile` with GCC, as +above. Developers working in a Windows environment may want to instead build with the +[Microsoft Visual C++ compiler and libraries toolset (MSVC)](https://blogs.msdn.microsoft.com/vcblog/2017/03/07/msvc-the-best-choice-for-windows/). +There are a few benefits to using MSVC over GCC during your development, including creating +symbols for debugging and [performance tracing](https://github.com/Microsoft/perfview#perfview-overview). + +There are two ways to build Git for Windows using MSVC. Each have their own merits. + +### Using SDK Command Line + +Use one of the following commands from the SDK Bash window to build Git for Windows: + +``` + make MSVC=1 -j12 + make MSVC=1 DEBUG=1 -j12 +``` + +The first form produces release-mode binaries; the second produces debug-mode binaries. +Both forms produce PDB files and can be debugged. However, the first is best for perf +tracing and the second is best for single-stepping. + +You can then open Visual Studio and select File -> Open -> Project/Solution and select +the compiled `git.exe` file. This creates a basic solution and you can use the debugging +and performance tracing tools in Visual Studio to monitor a Git process. Use the Debug +Properties page to set the working directory and command line arguments. + +Be sure to clean up before switching back to GCC (or to switch between debug and +release MSVC builds): + +``` + make MSVC=1 -j12 clean + make MSVC=1 DEBUG=1 -j12 clean +``` + +### Using the IDE + +If you prefer working in Visual Studio with a solution full of projects, then you can use +CMake, either by letting Visual Studio configure it automatically (simply open Git's +top-level directory via `File>Open>Folder...`) or by (downloading and) running +[CMake](https://cmake.org) manually. + +What to Change? +--------------- + +Many new contributors ask: What should I start working on? + +One way to win big with the open-source community is to look at the +[issues page](https://github.com/git-for-windows/git/issues) and see if there are any issues that +you can fix quickly, or if anything catches your eye. + +You can also look at [the unofficial Chromium issues page](https://crbug.com/git) for +multi-platform issues. You can look at recent user questions on +[the Git mailing list](https://public-inbox.org/git). + +Or you can "scratch your own itch", i.e. address an issue you have with Git. The team at Microsoft where the Git for Windows maintainer works, for example, is focused almost entirely on [improving performance](https://blogs.msdn.microsoft.com/devops/2018/01/11/microsofts-performance-contributions-to-git-in-2017/). +We approach our work by finding something that is slow and try to speed it up. We start our +investigation by reliably reproducing the slow behavior, then running that example using +the MSVC build and tracing the results in PerfView. + +You could also think of something you wish Git could do, and make it do that thing! The +only concern I would have with this approach is whether or not that feature is something +the community also wants. If this excites you though, go for it! Don't be afraid to +[get involved in the mailing list](http://vger.kernel.org/vger-lists.html#git) early for +feedback on the idea. + +Test Your Changes +----------------- + +After you make your changes, it is important that you test your changes. Manual testing is +important, but checking and extending the existing test suite is even more important. You +want to run the functional tests to see if you broke something else during your change, and +you want to extend the functional tests to be sure no one breaks your feature in the future. + +### Functional Tests + +Navigate to the `t/` directory and type `make` to run all tests or use `prove` as +[described on this Git for Windows page](https://gitforwindows.org/building-git): + +``` +prove -j12 --state=failed,save ./t[0-9]*.sh +``` + +You can also run each test directly by running the corresponding shell script with a name +like `tNNNN-descriptor.sh`. + +If you are adding new functionality, you may need to create unit tests by creating +helper commands that test a very limited action. These commands are stored in `t/helpers`. +When adding a helper, be sure to add a line to `t/Makefile` and to the `.gitignore` for the +binary file you add. The Git community prefers functional tests using the full `git` +executable, so try to exercise your new code using `git` commands before creating a test +helper. + +To find out why a test failed, repeat the test with the `-x -v -d -i` options and then +navigate to the appropriate "trash" directory to see the data shape that was used for the +test failed step. + +Read [`t/README`](t/README) for more details. + +### Performance Tests + +If you are working on improving performance, you will need to be acquainted with the +performance tests in `t/perf`. There are not too many performance tests yet, but adding one +as your first commit in a patch series helps to communicate the boost your change provides. + +To check the change in performance across multiple versions of `git`, you can use the +`t/perf/run` script. For example, to compare the performance of `git rev-list` across the +`core/master` and `core/next` branches compared to a `topic` branch, you can run + +``` +cd t/perf +./run core/master core/next topic -- p0001-rev-list.sh +``` + +You can also set certain environment variables to help test the performance on different +repositories or with more repetitions. The full list is available in +[the `t/perf/README` file](t/perf/README), +but here are a few important ones: + +``` +GIT_PERF_REPO=/path/to/repo +GIT_PERF_LARGE_REPO=/path/to/large/repo +GIT_PERF_REPEAT_COUNT=10 +``` + +When running the performance tests on Linux, you may see a message "Can't locate JSON.pm in +@INC" and that means you need to run `sudo cpanm install JSON` to get the JSON perl package. + +For running performance tests, it can be helpful to set up a few repositories with strange +data shapes, such as: + +**Many objects:** Clone repos such as [Kotlin](https://github.com/jetbrains/kotlin), [Linux](https://github.com/torvalds/linux), or [Android](https://source.android.com/setup/downloading). + +**Many pack-files:** You can split a fresh clone into multiple pack-files of size at most +16MB by running `git repack -adfF --max-pack-size=16m`. See the +[`git repack` documentation](https://git-scm.com/docs/git-repack) for more information. +You can count the number of pack-files using `ls .git/objects/pack/*.pack | wc -l`. + +**Many loose objects:** If you already split your repository into multiple pack-files, then +you can pick one to split into loose objects using `cat .git/objects/pack/[id].pack | git unpack-objects`; +delete the `[id].pack` and `[id].idx` files after this. You can count the number of loose +bjects using `ls .git/objects/??/* | wc -l`. + +**Deep history:** Usually large repositories also have deep histories, but you can use the +[test-many-commits-1m repo](https://github.com/cirosantilli/test-many-commits-1m/) to +target deep histories without the overhead of many objects. One issue with this repository: +there are no merge commits, so you will need to use a different repository to test a "wide" +commit history. + +**Large Index:** You can generate a large index and repo by using the scripts in +`t/perf/repos`. There are two scripts. `many-files.sh` which will generate a repo with +same tree and blobs but different paths. Using `many-files.sh -d 5 -w 10 -f 9` will create +a repo with ~1 million entries in the index. `inflate-repo.sh` will use an existing repo +and copy the current work tree until it is a specified size. + +Test Your Changes on Linux +-------------------------- + +It can be important to work directly on the [core Git codebase](https://github.com/git/git), +such as a recent commit into the `master` or `next` branch that has not been incorporated +into Git for Windows. Also, it can help to run functional and performance tests on your +code in Linux before submitting patches to the mailing list, which focuses on many platforms. +The differences between Windows and Linux are usually enough to catch most cross-platform +issues. + +### Using the Windows Subsystem for Linux + +The [Windows Subsystem for Linux (WSL)](https://docs.microsoft.com/en-us/windows/wsl/install-win10) +allows you to [install Ubuntu Linux as an app](https://www.microsoft.com/en-us/store/p/ubuntu/9nblggh4msv6) +that can run Linux executables on top of the Windows kernel. Internally, +Linux syscalls are interpreted by the WSL, everything else is plain Ubuntu. + +First, open WSL (either type "Bash" in Cortana, or execute "bash.exe" in a CMD window). +Then install the prerequisites, and `git` for the initial clone: + +``` +sudo apt-get update +sudo apt-get install git gcc make libssl-dev libcurl4-openssl-dev \ + libexpat-dev tcl tk gettext git-email zlib1g-dev +``` + +Then, clone and build: + +``` +git clone https://github.com/git-for-windows/git +cd git +git remote add -f upstream https://github.com/git/git +make +``` + +Be sure to clone into `/home/[user]/` and not into any folder under `/mnt/?/` or your build +will fail due to colons in file names. + +### Using a Linux Virtual Machine with Hyper-V + +If you prefer, you can use a virtual machine (VM) to run Linux and test your changes in the +full environment. The test suite runs a lot faster on Linux than on Windows or with the WSL. +You can connect to the VM using an SSH terminal like +[PuTTY](https://www.chiark.greenend.org.uk/~sgtatham/putty/). + +The following instructions are for using Hyper-V, which is available in some versions of Windows. +There are many virtual machine alternatives available, if you do not have such a version installed. + +* [Download an Ubuntu Server ISO](https://www.ubuntu.com/download/server). +* Open [Hyper-V Manager](https://docs.microsoft.com/en-us/virtualization/hyper-v-on-windows/quick-start/enable-hyper-v). +* [Set up a virtual switch](https://docs.microsoft.com/en-us/virtualization/hyper-v-on-windows/quick-start/connect-to-network) + so your VM can reach the network. +* Select "Quick Create", name your machine, select the ISO as installation source, and un-check + "This virtual machine will run Windows." +* Go through the Ubuntu install process, being sure to select to install OpenSSH Server. +* When install is complete, log in and check the SSH server status with `sudo service ssh status`. + * If the service is not found, install with `sudo apt-get install openssh-server`. + * If the service is not running, then use `sudo service ssh start`. +* Use `shutdown -h now` to shutdown the VM, go to the Hyper-V settings for the VM, expand Network Adapter + to select "Advanced Features", and set the MAC address to be static (this can save your VM from losing + network if shut down incorrectly). +* Provide as many cores to your VM as you can (for parallel builds). +* Restart your VM, but do not connect. +* Use `ssh` in Git Bash, download [PuTTY](https://www.chiark.greenend.org.uk/~sgtatham/putty/), or use your favorite SSH client to connect to the VM through SSH. + +In order to build and use `git`, you will need the following libraries via `apt-get`: + +``` +sudo apt-get update +sudo apt-get install git gcc make libssl-dev libcurl4-openssl-dev \ + libexpat-dev tcl tk gettext git-email zlib1g-dev +``` + +To get your code from your Windows machine to the Linux VM, it is easiest to push the branch to your fork of Git and clone your fork in the Linux VM. + +Don't forget to set your `git` config with your preferred name, email, and editor. + +Polish Your Commits +------------------- + +Before submitting your patch, be sure to read the [coding guidelines](https://github.com/git/git/blob/master/Documentation/CodingGuidelines) +and check your code to match as best you can. This can be a lot of effort, but it saves +time during review to avoid style issues. + +The other possibly major difference between the mailing list submissions and GitHub PR workflows +is that each commit will be reviewed independently. Even if you are submitting a +patch series with multiple commits, each commit must stand on it's own and be reviewable +by itself. Make sure the commit message clearly explain the why of the commit not the how. +Describe what is wrong with the current code and how your changes have made the code better. + +When preparing your patch, it is important to put yourself in the shoes of the Git community. +Accepting a patch requires more justification than approving a pull request from someone on +your team. The community has a stable product and is responsible for keeping it stable. If +you introduce a bug, then they cannot count on you being around to fix it. When you decided +to start work on a new feature, they were not part of the design discussion and may not +even believe the feature is worth introducing. + +Questions to answer in your patch message (and commit messages) may include: +* Why is this patch necessary? +* How does the current behavior cause pain for users? +* What kinds of repositories are necessary for noticing a difference? +* What design options did you consider before writing this version? Do you have links to + code for those alternate designs? +* Is this a performance fix? Provide clear performance numbers for various well-known repos. + +Here are some other tips that we use when cleaning up our commits: + +* Commit messages should be wrapped at 76 columns per line (or less; 72 is also a + common choice). +* Make sure the commits are signed off using `git commit (-s|--signoff)`. See + [SubmittingPatches](https://github.com/git/git/blob/v2.8.1/Documentation/SubmittingPatches#L234-L286) + for more details about what this sign-off means. +* Check for whitespace errors using `git diff --check [base]...HEAD` or `git log --check`. +* Run `git rebase --whitespace=fix` to correct upstream issues with whitespace. +* Become familiar with interactive rebase (`git rebase -i`) because you will be reordering, + squashing, and editing commits as your patch or series of patches is reviewed. +* Make sure any shell scripts that you add have the executable bit set on them. This is + usually for test files that you add in the `/t` directory. You can use + `git add --chmod=+x [file]` to update it. You can test whether a file is marked as executable + using `git ls-files --stage \*.sh`; the first number is 100755 for executable files. +* Your commit titles should match the "area: change description" format. Rules of thumb: + * Choose "<area>: " prefix appropriately. + * Keep the description short and to the point. + * The word that follows the "<area>: " prefix is not capitalized. + * Do not include a full-stop at the end of the title. + * Read a few commit messages -- using `git log origin/master`, for instance -- to + become acquainted with the preferred commit message style. +* Build source using `make DEVELOPER=1` for extra-strict compiler warnings. + +Submit Your Patch +----------------- + +Git for Windows [accepts pull requests on GitHub](https://github.com/git-for-windows/git/pulls), but +these are reserved for Windows-specific improvements. For core Git, submissions are accepted on +[the Git mailing list](https://public-inbox.org/git). + +### Configure Git to Send Emails + +There are a bunch of options for configuring the `git send-email` command. These options can +be found in the documentation for +[`git config`](https://git-scm.com/docs/git-config) and +[`git send-email`](https://git-scm.com/docs/git-send-email). + +``` +git config --global sendemail.smtpserver <smtp server> +git config --global sendemail.smtpserverport 587 +git config --global sendemail.smtpencryption tls +git config --global sendemail.smtpuser <email address> +``` + +To avoid storing your password in the config file, store it in the Git credential manager: + +``` +$ git credential fill +protocol=smtp +host=<stmp server> +username=<email address> +password=password +``` + +Before submitting a patch, read the [Git documentation on submitting patches](https://github.com/git/git/blob/master/Documentation/SubmittingPatches). + +To construct a patch set, use the `git format-patch` command. There are three important options: + +* `--cover-letter`: If specified, create a `[v#-]0000-cover-letter.patch` file that can be + edited to describe the patch as a whole. If you previously added a branch description using + `git branch --edit-description`, you will end up with a 0/N mail with that description and + a nice overall diffstat. +* `--in-reply-to=[Message-ID]`: This will mark your cover letter as replying to the given + message (which should correspond to your previous iteration). To determine the correct Message-ID, + find the message you are replying to on [public-inbox.org/git](https://public-inbox.org/git) and take + the ID from between the angle brackets. + +* `--subject-prefix=[prefix]`: This defaults to [PATCH]. For subsequent iterations, you will want to + override it like `--subject-prefix="[PATCH v2]"`. You can also use the `-v` option to have it + automatically generate the version number in the patches. + +If you have multiple commits and use the `--cover-letter` option be sure to open the +`0000-cover-letter.patch` file to update the subject and add some details about the overall purpose +of the patch series. + +### Examples + +To generate a single commit patch file: +``` +git format-patch -s -o [dir] -1 +``` +To generate four patch files from the last three commits with a cover letter: +``` +git format-patch --cover-letter -s -o [dir] HEAD~4 +``` +To generate version 3 with four patch files from the last four commits with a cover letter: +``` +git format-patch --cover-letter -s -o [dir] -v 3 HEAD~4 +``` + +### Submit the Patch + +Run [`git send-email`](https://git-scm.com/docs/git-send-email), starting with a test email: + +``` +git send-email --to=yourself@address.com [dir with patches]/*.patch +``` + +After checking the receipt of your test email, you can send to the list and to any +potentially interested reviewers. + +``` +git send-email --to=git@vger.kernel.org --cc=<email1> --cc=<email2> [dir with patches]/*.patch +``` + +To submit a nth version patch (say version 3): + +``` +git send-email --to=git@vger.kernel.org --cc=<email1> --cc=<email2> \ + --in-reply-to=<the message id of cover letter of patch v2> [dir with patches]/*.patch +``` From 0c84a1de2c894c76026e013e5f54047421f69ce2 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 10 Jan 2014 16:16:03 -0600 Subject: [PATCH 243/248] README.md: Add a Windows-specific preamble MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Includes touch-ups by 마누엘, Philip Oakley and 孙卓识. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- README.md | 78 +++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 76 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index d87bca1b8c3ebf..026d5d85caef09 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,77 @@ -[![Build status](https://github.com/git/git/workflows/CI/badge.svg)](https://github.com/git/git/actions?query=branch%3Amaster+event%3Apush) +Git for Windows +=============== + +[![Contributor Covenant](https://img.shields.io/badge/Contributor%20Covenant-2.1-4baaaa.svg)](CODE_OF_CONDUCT.md) +[![Open in Visual Studio Code](https://img.shields.io/static/v1?logo=visualstudiocode&label=&message=Open%20in%20Visual%20Studio%20Code&labelColor=2c2c32&color=007acc&logoColor=007acc)](https://open.vscode.dev/git-for-windows/git) +[![Build status](https://github.com/git-for-windows/git/workflows/CI/badge.svg)](https://github.com/git-for-windows/git/actions?query=branch%3Amain+event%3Apush) +[![Join the chat at https://gitter.im/git-for-windows/git](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/git-for-windows/git?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) + +This is [Git for Windows](http://git-for-windows.github.io/), the Windows port +of [Git](http://git-scm.com/). + +The Git for Windows project is run using a [governance +model](http://git-for-windows.github.io/governance-model.html). If you +encounter problems, you can report them as [GitHub +issues](https://github.com/git-for-windows/git/issues), discuss them in Git +for Windows' [Discussions](https://github.com/git-for-windows/git/discussions) +or on the [Git mailing list](mailto:git@vger.kernel.org), and [contribute bug +fixes](https://gitforwindows.org/how-to-participate). + +To build Git for Windows, please either install [Git for Windows' +SDK](https://gitforwindows.org/#download-sdk), start its `git-bash.exe`, `cd` +to your Git worktree and run `make`, or open the Git worktree as a folder in +Visual Studio. + +To verify that your build works, use one of the following methods: + +- If you want to test the built executables within Git for Windows' SDK, + prepend `<worktree>/bin-wrappers` to the `PATH`. +- Alternatively, run `make install` in the Git worktree. +- If you need to test this in a full installer, run `sdk build + git-and-installer`. +- You can also "install" Git into an existing portable Git via `make install + DESTDIR=<dir>` where `<dir>` refers to the top-level directory of the + portable Git. In this instance, you will want to prepend that portable Git's + `/cmd` directory to the `PATH`, or test by running that portable Git's + `git-bash.exe` or `git-cmd.exe`. +- If you built using a recent Visual Studio, you can use the menu item + `Build>Install git` (you will want to click on `Project>CMake Settings for + Git` first, then click on `Edit JSON` and then point `installRoot` to the + `mingw64` directory of an already-unpacked portable Git). + + As in the previous bullet point, you will then prepend `/cmd` to the `PATH` + or run using the portable Git's `git-bash.exe` or `git-cmd.exe`. +- If you want to run the built executables in-place, but in a CMD instead of + inside a Bash, you can run a snippet like this in the `git-bash.exe` window + where Git was built (ensure that the `EOF` line has no leading spaces), and + then paste into the CMD window what was put in the clipboard: + + ```sh + clip.exe <<EOF + set GIT_EXEC_PATH=$(cygpath -aw .) + set PATH=$(cygpath -awp ".:contrib/scalar:/mingw64/bin:/usr/bin:$PATH") + set GIT_TEMPLATE_DIR=$(cygpath -aw templates/blt) + set GITPERLLIB=$(cygpath -aw perl/build/lib) + EOF + ``` +- If you want to run the built executables in-place, but outside of Git for + Windows' SDK, and without an option to set/override any environment + variables (e.g. in Visual Studio's debugger), you can call the Git executable + by its absolute path and use the `--exec-path` option, like so: + + ```cmd + C:\git-sdk-64\usr\src\git\git.exe --exec-path=C:\git-sdk-64\usr\src\git help + ``` + + Note: for this to work, you have to hard-link (or copy) the `.dll` files from + the `/mingw64/bin` directory to the Git worktree, or add the `/mingw64/bin` + directory to the `PATH` somehow or other. + +To make sure that you are testing the correct binary, call `./git.exe version` +in the Git worktree, and then call `git version` in a directory/window where +you want to test Git, and verify that they refer to the same version (you may +even want to pass the command-line option `--build-options` to look at the +exact commit from which the Git version was built). Git - fast, scalable, distributed revision control system ========================================================= @@ -29,7 +102,7 @@ CVS users may also want to read [Documentation/gitcvs-migration.adoc][] (`man gitcvs-migration` or `git help cvs-migration` if git is installed). -The user discussion and development of Git take place on the Git +The user discussion and development of core Git take place on the Git mailing list -- everyone is welcome to post bug reports, feature requests, comments and patches to git@vger.kernel.org (read [Documentation/SubmittingPatches][] for instructions on patch submission @@ -43,6 +116,7 @@ To subscribe to the list, send an email to <git+subscribe@vger.kernel.org> (see https://subspace.kernel.org/subscribing.html for details). The mailing list archives are available at <https://lore.kernel.org/git/>, <https://marc.info/?l=git> and other archival sites. +The core git mailing list is plain text (no HTML!). Issues which are security relevant should be disclosed privately to the Git Security mailing list <git-security@googlegroups.com>. From 2078f8964426a229726862f7bb3406c9f092d2cb Mon Sep 17 00:00:00 2001 From: Brendan Forster <brendan@github.com> Date: Thu, 18 Feb 2016 21:29:50 +1100 Subject: [PATCH 244/248] Add an issue template With improvements by Clive Chan, Adric Norris, Ben Bodenmiller and Philip Oakley. Helped-by: Clive Chan <cc@clive.io> Helped-by: Adric Norris <landstander668@gmail.com> Helped-by: Ben Bodenmiller <bbodenmiller@hotmail.com> Helped-by: Philip Oakley <philipoakley@iee.org> Signed-off-by: Brendan Forster <brendan@github.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- .github/ISSUE_TEMPLATE/bug-report.yml | 105 ++++++++++++++++++++++++++ .github/ISSUE_TEMPLATE/config.yml | 1 + 2 files changed, 106 insertions(+) create mode 100644 .github/ISSUE_TEMPLATE/bug-report.yml create mode 100644 .github/ISSUE_TEMPLATE/config.yml diff --git a/.github/ISSUE_TEMPLATE/bug-report.yml b/.github/ISSUE_TEMPLATE/bug-report.yml new file mode 100644 index 00000000000000..b49593339932b2 --- /dev/null +++ b/.github/ISSUE_TEMPLATE/bug-report.yml @@ -0,0 +1,105 @@ +name: Bug report +description: Use this template to report bugs. +body: + - type: checkboxes + id: search + attributes: + label: Existing issues matching what you're seeing + description: Please search for [open](https://github.com/git-for-windows/git/issues?q=is%3Aopen) or [closed](https://github.com/git-for-windows/git/issues?q=is%3Aclosed) issue matching what you're seeing before submitting a new issue. + options: + - label: I was not able to find an open or closed issue matching what I'm seeing + - type: textarea + id: git-for-windows-version + attributes: + label: Git for Windows version + description: Which version of Git for Windows are you using? + placeholder: Please insert the output of `git --version --build-options` here + render: shell + validations: + required: true + - type: dropdown + id: windows-version + attributes: + label: Windows version + description: Which version of Windows are you running? + options: + - Windows 8.1 + - Windows 10 + - Windows 11 + - Other + default: 2 + validations: + required: true + - type: dropdown + id: windows-arch + attributes: + label: Windows CPU architecture + description: What CPU Archtitecture does your Windows target? + options: + - i686 (32-bit) + - x86_64 (64-bit) + - ARM64 + default: 1 + validations: + required: true + - type: textarea + id: windows-version-cmd + attributes: + label: Additional Windows version information + description: This provides us with further information about your Windows such as the build number + placeholder: Please insert the output of `cmd.exe /c ver` here + render: shell + - type: textarea + id: options + attributes: + label: Options set during installation + description: What options did you set as part of the installation? Or did you choose the defaults? + placeholder: | + One of the following: + > type "C:\Program Files\Git\etc\install-options.txt" + > type "C:\Program Files (x86)\Git\etc\install-options.txt" + > type "%USERPROFILE%\AppData\Local\Programs\Git\etc\install-options.txt" + > type "$env:USERPROFILE\AppData\Local\Programs\Git\etc\install-options.txt" + $ cat /etc/install-options.txt + render: shell + validations: + required: true + - type: textarea + id: other-things + attributes: + label: Other interesting things + description: Any other interesting things about your environment that might be related to the issue you're seeing? + - type: input + id: terminal + attributes: + label: Terminal/shell + description: Which terminal/shell are you running Git from? e.g Bash/CMD/PowerShell/other + validations: + required: true + - type: textarea + id: commands + attributes: + label: Commands that trigger the issue + description: What commands did you run to trigger this issue? If you can provide a [Minimal, Complete, and Verifiable example](http://stackoverflow.com/help/mcve) this will help us understand the issue. + render: shell + validations: + required: true + - type: textarea + id: expected-behaviour + attributes: + label: Expected behaviour + description: What did you expect to occur after running these commands? + validations: + required: true + - type: textarea + id: actual-behaviour + attributes: + label: Actual behaviour + description: What actually happened instead? + validations: + required: true + - type: textarea + id: repository + attributes: + label: Repository + description: If the problem was occurring with a specific repository, can you provide the URL to that repository to help us with testing? \ No newline at end of file diff --git a/.github/ISSUE_TEMPLATE/config.yml b/.github/ISSUE_TEMPLATE/config.yml new file mode 100644 index 00000000000000..ec4bb386bcf8a4 --- /dev/null +++ b/.github/ISSUE_TEMPLATE/config.yml @@ -0,0 +1 @@ +blank_issues_enabled: false \ No newline at end of file From b10a6f4cc37d51d24f16827fa0f8dbf0cea1fe22 Mon Sep 17 00:00:00 2001 From: Philip Oakley <philipoakley@iee.org> Date: Fri, 22 Dec 2017 17:15:50 +0000 Subject: [PATCH 245/248] Modify the GitHub Pull Request template (to reflect Git for Windows) Git for Windows accepts pull requests; Core Git does not. Therefore we need to adjust the template (because it only matches core Git's project management style, not ours). Also: direct Git for Windows enhancements to their contributions page, space out the text for easy reading, and clarify that the mailing list is plain text, not HTML. Signed-off-by: Philip Oakley <philipoakley@iee.org> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- .github/PULL_REQUEST_TEMPLATE.md | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/.github/PULL_REQUEST_TEMPLATE.md b/.github/PULL_REQUEST_TEMPLATE.md index 37654cdfd7abcf..7baf31f2c471ec 100644 --- a/.github/PULL_REQUEST_TEMPLATE.md +++ b/.github/PULL_REQUEST_TEMPLATE.md @@ -1,7 +1,19 @@ -Thanks for taking the time to contribute to Git! Please be advised that the -Git community does not use github.com for their contributions. Instead, we use -a mailing list (git@vger.kernel.org) for code submissions, code reviews, and -bug reports. Nevertheless, you can use GitGitGadget (https://gitgitgadget.github.io/) +Thanks for taking the time to contribute to Git! + +Those seeking to contribute to the Git for Windows fork should see +http://gitforwindows.org/#contribute on how to contribute Windows specific +enhancements. + +If your contribution is for the core Git functions and documentation +please be aware that the Git community does not use the github.com issues +or pull request mechanism for their contributions. + +Instead, we use the Git mailing list (git@vger.kernel.org) for code and +documentation submissions, code reviews, and bug reports. The +mailing list is plain text only (anything with HTML is sent directly +to the spam folder). + +Nevertheless, you can use GitGitGadget (https://gitgitgadget.github.io/) to conveniently send your Pull Requests commits to our mailing list. For a single-commit pull request, please *leave the pull request description From e93b2bc5142556a2862109e7f775fe7d5d9fa607 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Tue, 6 Feb 2024 18:45:35 +0100 Subject: [PATCH 246/248] dependabot: help keeping GitHub Actions versions up to date See https://docs.github.com/en/code-security/dependabot/working-with-dependabot/keeping-your-actions-up-to-date-with-dependabot#enabling-dependabot-version-updates-for-actions for details. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- .github/dependabot.yml | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 .github/dependabot.yml diff --git a/.github/dependabot.yml b/.github/dependabot.yml new file mode 100644 index 00000000000000..22d5376407abf1 --- /dev/null +++ b/.github/dependabot.yml @@ -0,0 +1,13 @@ +# To get started with Dependabot version updates, you'll need to specify which +# package ecosystems to update and where the package manifests are located. +# Please see the documentation for all configuration options: +# https://docs.github.com/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file +# especially +# https://docs.github.com/en/code-security/dependabot/working-with-dependabot/keeping-your-actions-up-to-date-with-dependabot#enabling-dependabot-version-updates-for-actions + +version: 2 +updates: + - package-ecosystem: "github-actions" # See documentation for possible values + directory: "/" # Location of package manifests + schedule: + interval: "weekly" From 736564b630d92b2ce9a394e3a289334df5ae084d Mon Sep 17 00:00:00 2001 From: Johannes Schindelin <johannes.schindelin@gmx.de> Date: Fri, 23 Aug 2019 14:14:42 +0200 Subject: [PATCH 247/248] SECURITY.md: document Git for Windows' policies This is the recommended way on GitHub to describe policies revolving around security issues and about supported versions. Helped-by: Sven Strickroth <email@cs-ware.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- SECURITY.md | 56 +++++++++++++++++++++++++++++++++-------------------- 1 file changed, 35 insertions(+), 21 deletions(-) diff --git a/SECURITY.md b/SECURITY.md index c720c2ae7f9580..42b6d458bfd557 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -28,24 +28,38 @@ Examples for details to include: ## Supported Versions -There are no official "Long Term Support" versions in Git. -Instead, the maintenance track (i.e. the versions based on the -most recently published feature release, also known as ".0" -version) sees occasional updates with bug fixes. - -Fixes to vulnerabilities are made for the maintenance track for -the latest feature release and merged up to the in-development -branches. The Git project makes no formal guarantee for any -older maintenance tracks to receive updates. In practice, -though, critical vulnerability fixes are applied not only to the -most recent track, but to at least a couple more maintenance -tracks. - -This is typically done by making the fix on the oldest and still -relevant maintenance track, and merging it upwards to newer and -newer maintenance tracks. - -For example, v2.24.1 was released to address a couple of -[CVEs](https://cve.mitre.org/), and at the same time v2.14.6, -v2.15.4, v2.16.6, v2.17.3, v2.18.2, v2.19.3, v2.20.2, v2.21.1, -v2.22.2 and v2.23.1 were released. +Git for Windows is a "friendly fork" of [Git](https://git-scm.com/), i.e. changes in Git for Windows are frequently contributed back, and Git for Windows' release cycle closely following Git's. + +While Git maintains several release trains (when v2.19.1 was released, there were updates to v2.14.x-v2.18.x, too, for example), Git for Windows follows only the latest Git release. For example, there is no Git for Windows release corresponding to Git v2.16.5 (which was released after v2.19.0). + +One exception is [MinGit for Windows](https://gitforwindows.org/mingit) (a minimal subset of Git for Windows, intended for bundling with third-party applications that do not need any interactive commands nor support for `git svn`): critical security fixes are backported to the v2.11.x, v2.14.x, v2.19.x, v2.21.x and v2.23.x release trains. + +## Version number scheme + +The Git for Windows versions reflect the Git version on which they are based. For example, Git for Windows v2.21.0 is based on Git v2.21.0. + +As Git for Windows bundles more than just Git (such as Bash, OpenSSL, OpenSSH, GNU Privacy Guard), sometimes there are interim releases without corresponding Git releases. In these cases, Git for Windows appends a number in parentheses, starting with the number 2, then 3, etc. For example, both Git for Windows v2.17.1 and v2.17.1(2) were based on Git v2.17.1, but the latter included updates for Git Credential Manager and Git LFS, fixing critical regressions. + +## Tag naming scheme + +Every Git for Windows version is tagged using a name that starts with the Git version on which it is based, with the suffix `.windows.<patchlevel>` appended. For example, Git for Windows v2.17.1' source code is tagged as [`v2.17.1.windows.1`](https://github.com/git-for-windows/git/releases/tag/v2.17.1.windows.1) (the patch level is always at least 1, given that Git for Windows always has patches on top of Git). Likewise, Git for Windows v2.17.1(2)' source code is tagged as [`v2.17.1.windows.2`](https://github.com/git-for-windows/git/releases/tag/v2.17.1.windows.2). + +## Release Candidate (rc) versions + +As a friendly fork of Git (the "upstream" project), Git for Windows is closely corelated to that project. + +Consequently, Git for Windows publishes versions based on Git's release candidates (for upcoming "`.0`" versions, see [Git's release schedule](https://tinyurl.com/gitCal)). These versions end in `-rc<n>`, starting with `-rc0` for a very early preview of what is to come, and as with regular versions, Git for Windows tries to follow Git's releases as quickly as possible. + +Note: there is currently a bug in the "Check daily for updates" code, where it mistakes the final version as a downgrade from release candidates. Example: if you installed Git for Windows v2.23.0-rc3 and enabled the auto-updater, it would ask you whether you want to "downgrade" to v2.23.0 when that version was available. + +[All releases](https://github.com/git-for-windows/git/releases/), including release candidates, are listed via a link at the footer of the [Git for Windows](https://gitforwindows.org/) home page. + +## Snapshot versions ('nightly builds') + +Git for Windows also provides snapshots (these are not releases) of the current development as per git-for-Windows/git's `master` branch at the [Snapshots](https://gitforwindows.org/git-snapshots/) page. This link is also listed in the footer of the [Git for Windows](https://gitforwindows.org/) home page. + +Note: even if those builds are not exactly "nightly", they are sometimes referred to as "nightly builds" to keep with other projects' nomenclature. + +## Following upstream's developments + +The [gitforwindows/git repository](https://github.com/git-for-windows/git) also provides the `shears/*` branches. The `shears/*` branches reflect Git for Windows' patches, rebased onto the upstream integration branches, [updated (mostly) via automated CI builds](https://dev.azure.com/git-for-windows/git/_build?definitionId=25). From bd2ecbae58213046a468256b95fc4864de25bdf5 Mon Sep 17 00:00:00 2001 From: Lidong Yan <yldhome2d2@gmail.com> Date: Fri, 17 Oct 2025 10:38:36 +0800 Subject: [PATCH 248/248] diff: stop output garbled message in dry run mode In dry run mode, diff_flush_patch() should not produce any output. However, in commit b55e6d36eb (diff: ensure consistent diff behavior with ignore options, 2025-08-08), only the output during the comparison of two file contents was suppressed. For file deletions or mode changes, diff_flush_patch() still produces output. In run_extern_diff(), set quiet to true if in dry run mode. In emit_diff_symbol_from_struct(), directly return if in dry run mode. Signed-off-by: Lidong Yan <yldhome2d2@gmail.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> --- diff.c | 5 ++++- t/t4013-diff-various.sh | 21 +++++++++++++++++++++ 2 files changed, 25 insertions(+), 1 deletion(-) diff --git a/diff.c b/diff.c index 7b5601de2f3c80..eeeca39c30dfb7 100644 --- a/diff.c +++ b/diff.c @@ -1357,6 +1357,9 @@ static void emit_diff_symbol_from_struct(struct diff_options *o, int len = eds->len; unsigned flags = eds->flags; + if (o->dry_run) + return; + switch (s) { case DIFF_SYMBOL_NO_LF_EOF: context = diff_get_color_opt(o, DIFF_CONTEXT); @@ -4426,7 +4429,7 @@ static void run_external_diff(const struct external_diff *pgm, { struct child_process cmd = CHILD_PROCESS_INIT; struct diff_queue_struct *q = &diff_queued_diff; - int quiet = !(o->output_format & DIFF_FORMAT_PATCH); + int quiet = !(o->output_format & DIFF_FORMAT_PATCH) || o->dry_run; int rc; /* diff --git a/t/t4013-diff-various.sh b/t/t4013-diff-various.sh index cfeec239e0946c..3a47b2f7cbae73 100755 --- a/t/t4013-diff-various.sh +++ b/t/t4013-diff-various.sh @@ -661,6 +661,27 @@ test_expect_success 'diff -I<regex>: ignore matching file' ' test_grep ! "file1" actual ' +test_expect_success 'diff -I<regex>: ignore all content changes' ' + test_when_finished "git rm -f file1 file2" && + : >file1 && + git add file1 && + : >file2 && + git add file2 && + + rm -f file1 file2 && + mkdir file2 && + test_diff_no_content_changes () { + git diff $1 --ignore-blank-lines -I".*" >actual && + test_line_count = 2 actual && + test_grep "file1" actual && + test_grep "file2" actual && + test_grep ! "diff --git" actual + } && + test_diff_no_content_changes "--raw" && + test_diff_no_content_changes "--name-only" && + test_diff_no_content_changes "--name-status" +' + # check_prefix <patch> <src> <dst> # check only lines with paths to avoid dependency on exact oid/contents check_prefix () {