sys-devel/binutils-apple: fix build on macOS 13, including ARM64. #30781
Conversation
Pull Request assignmentSubmitter: @biergaizi sys-devel/binutils-apple: @gentoo/prefix Linked bugsNo bugs to link found. If your pull request references any of the Gentoo bug reports, please add appropriate GLEP 66 tags to the commit message and request reassignment. If you do not receive any reply to this pull request, please open or link a bug to attract the attention of maintainers. In order to force reassignment and/or bug reference scan, please append Docs: Code of Conduct ● Copyright policy (expl.) ● Devmanual ● GitHub PRs ● Proxy-maint guide |
gentoo-bot
added
assigned
biergaizi
force-pushed
the
macos13-binutils-apple
branch
2 times, most recently
from
fc8be0e
to
cbf3f84
Compare
This commits introduces three patches to fix build on macOS 13
(Intel), and also on ARM64 (Apple Silicon).
1. On macOS 13, Gentoo Prefix fails to compile cctools due to
missing Blocks support in GCC, creating the following error:
/Users/ec2-user/gentoo/MacOSX.sdk/usr/include/objc/runtime.h:373:29: error: expected ')' before '^' token
373 | void (^ _Nonnull block)(Class _Nonnull aClass, BOOL * _Nonnull stop)
| ^
| )
Fortunately the header objc-runtime.h is not strictly necessary.
As a workaround, stop including objc-runtime.h in print_objc.c
and print_bitcode.c.
2. In ar(1), strcmp() checks are used to determine the value
of argument argv[1], even when no argument is given. In the
past, they were possibly harmless out-of-bound reads and
comparison with garbage, without consequences.
However, running it on macOS 13 w/ Apple Silicon immediately
crashes it with Segmentation Fault, because argv[1] is now
NULL and generates EXC_BAD_ACCESS in strcmp().
$ ./bin/ar
Segmentation fault: 11
This commit checks whether argc is equal or greater than 2
before doing strcmp(). No revbump is needed since the package
couldn't be built on ARM64 before.
3. Fabian Groffen's downstream patch "fix compilation on arm64",
which adds missing CPU register definitions.
Closes: https://bugs.gentoo.org/905131
Signed-off-by: Yifeng Li <tomli@tomli.me>
biergaizi
force-pushed
the
macos13-binutils-apple
branch
from
cbf3f84
to
f17f786
Compare
Pull request CI reportReport generated at: 2023-04-27 20:54 UTC There are existing issues already. Please look into the report to make sure none of them affect the packages in question: |
| run_ranlib = 1; | ||
|
|
||
| if (argc < 3) { | ||
| - if(strcmp(argv[1], "--version") == 0){ |
There was a problem hiding this comment.
This changes what it compiles to, so please revbump for this even if it didn't crash before.
|
How should this package be revbumped properly? Its "upstream" is in fact a downstream repo that Fabian Groffen maintained specially for use in Gentoo, so its "upstream" version number is already "gentoo-8.2.1-r101" that contains "r101". Should I merge these patches into Groffen's repo instead? |
Pull request CI reportReport generated at: 2023-04-27 21:04 UTC There are existing issues already. Please look into the report to make sure none of them affect the packages in question: |
|
Which revision number should I use for this ebuild? r102? r101-r1? I'm afraid that r102 has a risk of clashing with a future git tag in Fabian Groffen's repo, which is currently r101. |
|
I'm going to let grobian answer because I don't want to complicate things if he has a different idea in mind. Personally, in this case, I'd say we should swap the "upstream" (fork/grobian repo) versioning to _pN, and use -rN for Gentoo revisions. |
|
hmmm, I once gave it a spin and then got to the conclusion that we were missing a lot of arm64 stuff, but you're saying it actually works with these reasonably small changes? I noticed Ian has a newer version that I meant to look at to see if we can upgrade to it. In any case, I maintain indeed a fork for our usage, so let's patch it there, and then spin a new release. |
I'm mot sure about the functionality of every single tool, possibly there are still problems, but at least it can be built from source now, Portage can even link programs with it, so I'd call it progress. |
|
I just opened another Pull Request to submit the ar(1) patch. After that change, binutils-apple, let's make a new release. |
|
I've pushed these fixes as binutils-apple-8.2.1-r102 |
This commits introduces three patches to fix build on macOS 13 (Intel), and also on ARM64 (Apple Silicon).
Fortunately the header objc-runtime.h is not strictly necessary. As a workaround, stop including objc-runtime.h in print_objc.c and print_bitcode.c.
However, running it on macOS 13 w/ Apple Silicon immediately crashes it with Segmentation Fault, because argv[1] is now NULL and generates EXC_BAD_ACCESS in strcmp().
This commit checks whether argc is equal or greater than 2 before doing strcmp(). No revbump is needed since the package couldn't be built on ARM64 before.
Closes: 905131