Rebase to the llvm-project monorepo #57675

cuviper · 2019-01-16T18:40:31Z

The new git submodule src/llvm-project is a monorepo replacing src/llvm
and src/tools/{clang,lld,lldb}. This also serves as a rebase for these
projects to the new 8.x branch from trunk.

The src/llvm-emscripten fork is unchanged for now.

r? @alexcrichton

rust-highfive · 2019-01-16T18:40:42Z

⚠️ Warning ⚠️

These commits modify submodules.

alexcrichton · 2019-01-16T19:00:00Z

@bors: r+

bors · 2019-01-16T19:00:03Z

📌 Commit 50e655e5cf9a8f9e610106a7e612379b61217e8a has been approved by alexcrichton

nikic · 2019-01-16T20:15:48Z

src/tools/tidy/src/lib.rs

@@ -47,6 +47,7 @@ pub mod libcoretest;
 fn filter_dirs(path: &Path) -> bool {
    let skip = [
        "src/llvm",


Probably can drop this line?

I left the LLVM-related src/tools/* here too. My thought was to keep ignoring those paths in tidy, in case a developer still has those obsolete submodules in place. (For instance, they may be moving between branches and don't want to remove them.)

Centril · 2019-01-17T02:51:04Z

@bors p=3

alexcrichton · 2019-01-17T20:57:19Z

@bors: r+

I've added two more commits from rust-lang/llvm-project#2

bors · 2019-01-17T20:57:19Z

📌 Commit aa9d564d917ee7efd998531ac75062feb2e3caba has been approved by alexcrichton

bors · 2019-01-18T04:34:36Z

⌛ Testing commit aa9d564d917ee7efd998531ac75062feb2e3caba with merge 5862807814fe0437cd5cf017c09e1b8bb66dc657...

bors · 2019-01-18T05:35:43Z

💔 Test failed - status-appveyor

nikic · 2019-01-18T17:11:54Z

Not really clear to me why it fails to compile core from the AppVeyor log, but possibly this is because compiler-rt in https://github.com/rust-lang-nursery/compiler-builtins hasn't been synced with the LLVM version we're updating to here.

cuviper · 2019-01-18T17:31:25Z

Hmm, I wonder if we could get compiler-builtins to just use the monorepo's compiler-rt, so they're always in sync. It would mean the src/llvm-project submodule will always be necessary to checkout, whereas it's currently optional if you're using external LLVM, but that's probably fine for most rustc developers.

Anyway, I do have a machine that I can boot into Windows, so I'll try to reproduce this...

alexcrichton · 2019-01-18T21:53:56Z

I think it's probably also fine to just add an llvm-project in the compiler-builtins repository, we can deal with publishing far less to crates.io and git deps hopefully won't be too too common.

cuviper · 2019-01-18T22:01:42Z

Oh, I forgot that compiler-builtins is now pulled from crates.io -- then yeah, it could control what it publishes from the monorepo.

Anyway, my first build as-is with MSVC cl was fine. I'm trying again with clang-cl as appveyor does.

cuviper · 2019-01-18T23:27:52Z

Well, I can't reproduce the issue with clang-cl either. If anyone else has a ready Windows system to try it, I would appreciate some debugging help.

retep998 · 2019-01-21T19:53:34Z

Did you remember to --enable-profiler when trying to reproduce?

cuviper · 2019-01-21T20:03:05Z

Yes, I just double-checked config.toml to make sure I had --enable-profiler. It doesn't look like the appveyor log reveals the full configure command, but I guess I could try reproducing more from the given configure output.

Another thing I noticed is appveyor's use of Windows Kits\10\include\10.0.14393.0, whereas I have 10.0.16299.0. I don't know if that would make a difference.

nikic · 2019-01-21T20:40:35Z

@cuviper Another thing to check would be whether you have assertions=true and verify-llvm-ir=true, as this is an alt build. Unfortunately the non-alt build was cancelled shortly before the point where one could see whether it also fails. Though I don't really see how either of those would result in a silent failure. (Then again, I would expect nothing to result in a silent failure, but here we are.)

cuviper · 2019-01-21T22:02:39Z

OK, I matched more config settings, and this time got a dialog about a failed assertion:

(If there's a way to get windows/appveyor to show that on the console instead, we should...)

That assertion is:

class MDNode : public Metadata {
//...
  const MDOperand &getOperand(unsigned I) const {
    assert(I < NumOperands && "Out of range");
    return op_begin()[I];
  }

I'll try to get that in a debugger...

glaubitz · 2019-01-22T01:18:12Z

I have already tried building Rust against a newer LLVM version but that fails for me due to LLVM commit eaa73537bb48767af73b2aa8b03eef00d1f6f7c8:

cargo:warning=../rustllvm/RustWrapper.cpp: In function ‘LLVMOpaqueMetadata* LLVMRustDIBuilderCreateFunction(LLVMRustDIBuilderRef, LLVMMetadataRef, const char*, const char*, LLVMMetadataRef, unsigned int, LLVMMetadataRef, bool, bool, unsigned int, LLVMRustDIFlags, bool, LLVMValueRef, LLVMMetadataRef, LLVMMetadataRef)’:
cargo:warning=../rustllvm/RustWrapper.cpp:587:38: error: no matching function for call to ‘llvm::DIBuilder::createFunction(llvm::DIScope*, const char*&, const char*&, llvm::DIFile*, unsigned int&, llvm::DISubroutineType*, bool&, bool&, unsigned int&, llvm::DINode::DIFlags, bool&, llvm::DITemplateParameterArray&, llvm::DISubprogram*)’
cargo:warning=       unwrapDIPtr<DISubprogram>(Decl));
cargo:warning=                                      ^
cargo:warning=In file included from ../rustllvm/rustllvm.h:53,
cargo:warning=                 from ../rustllvm/RustWrapper.cpp:1:
cargo:warning=/local_scratch/glaubitz/M680x0-llvm/include/llvm/IR/DIBuilder.h:663:5: note: candidate: ‘llvm::DISubprogram* llvm::DIBuilder::createFunction(llvm::DIScope*, llvm::StringRef, llvm::StringRef, llvm::DIFile*, unsigned int, llvm::DISubroutineType*, unsigned int, llvm::DINode::DIFlags, llvm::DISubprogram::DISPFlags, llvm::DITemplateParameterArray, llvm::DISubprogram*, llvm::DITypeArray)’
cargo:warning=     createFunction(DIScope *Scope, StringRef Name, StringRef LinkageName,
cargo:warning=     ^~~~~~~~~~~~~~
cargo:warning=/local_scratch/glaubitz/M680x0-llvm/include/llvm/IR/DIBuilder.h:663:5: note:   candidate expects 12 arguments, 13 provided
cargo:warning=../rustllvm/RustWrapper.cpp: In function ‘void LLVMRustUnpackOptimizationDiagnostic(LLVMDiagnosticInfoRef, RustStringRef, LLVMOpaqueValue**, unsigned int*, unsigned int*, RustStringRef, RustStringRef)’:
cargo:warning=../rustllvm/RustWrapper.cpp:923:23: error: ‘class llvm::DiagnosticLocation’ has no member named ‘getFilename’; did you mean ‘getLine’?
cargo:warning=     FilenameOS << loc.getFilename();
cargo:warning=                       ^~~~~~~~~~~
cargo:warning=                       getLine

Are there already patches to update rustllvm for the new LLVM version?

Edit: Ah, it's actually part of this PR ;). Will give it a try.

mati865 · 2019-01-25T15:32:40Z

Cargo.lock

@@ -1813,7 +1813,7 @@ name = "rand_chacha"
 version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "rand_core 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "rand_core 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",


Why would upgrading dlmalloc downgrade rand_core?

Both of those rand_core versions are already in the dependency tree, and rand_chacha requires rand_core >= 0.2, < 0.4. I guess it's just a quirk of how cargo resolved this when making the other change. But rand_core 0.2.2 just re-exports from 0.3 anyway, so it doesn't really matter.

The new git submodule src/llvm-project is a monorepo replacing src/llvm and src/tools/{clang,lld,lldb}. This also serves as a rebase for these projects to the new 8.x branch from trunk. The src/llvm-emscripten fork is unchanged for now.

Remove usage of an old and removed wasm intrinsic

cuviper · 2019-01-26T00:20:07Z

I rebased to avoid Cargo.lock conflicts.

@bors r=alexcrichton

bors · 2019-01-26T00:20:08Z

📌 Commit 059ed4f has been approved by alexcrichton

bors · 2019-01-26T02:10:22Z

⌛ Testing commit 059ed4f with merge 9df043b...

@alexcrichton

Rebase to the llvm-project monorepo The new git submodule src/llvm-project is a monorepo replacing src/llvm and src/tools/{clang,lld,lldb}. This also serves as a rebase for these projects to the new 8.x branch from trunk. The src/llvm-emscripten fork is unchanged for now. r? @alexcrichton

bors · 2019-01-26T04:47:50Z

☀️ Test successful - checks-travis, status-appveyor
Approved by: alexcrichton
Pushing 9df043b to master...

alexcrichton · 2019-01-26T05:05:04Z

🎊

nikic · 2019-02-01T08:02:33Z

Perf results are in: https://perf.rust-lang.org/compare.html?start=37d51aa8f3bca674a50eb7c6204deed6fb4dff80&end=9df043b543bb9bc3e50bc243811c58d52a3aefea&stat=instructions:u

Looks like mostly minor improvements for simple crates / incremental. Only one regression in kekkac-opt that might be worth taking a glance at.

Fix rust-lldb wrapper scripts. Currently the `rust-lldb` wrapper provided by Rust project is broken. The error messages it produces on launch are as follows: ``` warning: ignoring unknown option: --one-line-before-file=command script import "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/etc/lldb_rust_formatters.py" warning: ignoring unknown option: --one-line-before-file=type summary add --no-value --python-function lldb_rust_formatters.print_val -x ".*" --category Rust warning: ignoring unknown option: --one-line-before-file=type category enable Rust (lldb) target create "target/debug/nagare" Traceback (most recent call last): File "<string>", line 1, in <module> File "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/x86_64-apple-darwin/lib/python2.7/site-packages/lldb/__init__.py", line 1481, in <module> class SBAddress(object): File "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/x86_64-apple-darwin/lib/python2.7/site-packages/lldb/__init__.py", line 1647, in SBAddress __swig_getmethods__["module"] = GetModule NameError: name '__swig_getmethods__' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined ...etc. ``` The errors stem from two regressions: one caused by an LLVM upgrade and one caused by unintended upgrade to SWIG 4.0 (SWIG is a wrapper generator that is used to generate Python bindings for LLVM and LLDB.) (Edit: found the exact dates) The SWIG breakage happened because of a Homebrew version upgrade on `nightly-2019-05-01-x86_64-apple-darwin` and the LLVM breakage happened on `nightly-2019-01-27-x86_64-apple-darwin` (likely to have been caused by rust-lang#57675 ). The fix is to update the LLVM parameter syntax and to "downgrade" to SWIG 3.0.x. SWIG 3.0.x is not going to be supported by Homebrew forever, but should be good for now, until LLDB upgrades to support SWIG 4.0.0. Here's some more info about Homebrew support: Homebrew/homebrew-core#39929 & Homebrew/homebrew-core#40882 I'm going to send a bug & fix to LLDB about SWIG 4.0.0 to get the situation fixed in the future. It would be good to also backport this to beta, since it's such a small change, and will fix an obvious regression.

Fix rust-lldb wrapper scripts. Currently the `rust-lldb` wrapper provided by Rust project is broken. The error messages it produces on launch are as follows: ``` warning: ignoring unknown option: --one-line-before-file=command script import "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/etc/lldb_rust_formatters.py" warning: ignoring unknown option: --one-line-before-file=type summary add --no-value --python-function lldb_rust_formatters.print_val -x ".*" --category Rust warning: ignoring unknown option: --one-line-before-file=type category enable Rust (lldb) target create "target/debug/nagare" Traceback (most recent call last): File "<string>", line 1, in <module> File "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/x86_64-apple-darwin/lib/python2.7/site-packages/lldb/__init__.py", line 1481, in <module> class SBAddress(object): File "/Users/kon/.rustup/toolchains/nightly-2019-05-02-x86_64-apple-darwin/lib/rustlib/x86_64-apple-darwin/lib/python2.7/site-packages/lldb/__init__.py", line 1647, in SBAddress __swig_getmethods__["module"] = GetModule NameError: name '__swig_getmethods__' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined Traceback (most recent call last): File "<string>", line 1, in <module> NameError: name 'run_one_line' is not defined ...etc. ``` The errors stem from two regressions: one caused by an LLVM upgrade and one caused by unintended upgrade to SWIG 4.0 (SWIG is a wrapper generator that is used to generate Python bindings for LLVM and LLDB.) (Edit: found the exact dates) The SWIG breakage happened because of a Homebrew version upgrade on `nightly-2019-05-01-x86_64-apple-darwin` and the LLVM breakage happened on `nightly-2019-01-27-x86_64-apple-darwin` (likely to have been caused by #57675 ). The fix is to update the LLVM parameter syntax and to "downgrade" to SWIG 3.0.x. SWIG 3.0.x is not going to be supported by Homebrew forever, but should be good for now, until LLDB upgrades to support SWIG 4.0.0. Here's some more info about Homebrew support: Homebrew/homebrew-core#39929 & Homebrew/homebrew-core#40882 I'm going to send a bug & fix to LLDB about SWIG 4.0.0 to get the situation fixed in the future. It would be good to also backport this to beta, since it's such a small change, and will fix an obvious regression.

rust-highfive assigned alexcrichton Jan 16, 2019

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jan 16, 2019

cuviper mentioned this pull request Jan 16, 2019

[DO NOT MERGE] Apply Rust patches on release/8.x rust-lang/llvm-project#1

Closed

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jan 16, 2019

nikic reviewed Jan 16, 2019

View reviewed changes

alexcrichton mentioned this pull request Jan 17, 2019

[WIP] Encode WebAssembly specific locations in DBG_VALUEs and DW_AT_frame_base rust-lang/llvm-project#2

Merged

bors added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Jan 18, 2019

kennytm added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jan 20, 2019

bors added the S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. label Jan 25, 2019

mati865 reviewed Jan 25, 2019

View reviewed changes

cuviper and others added 8 commits January 25, 2019 15:39

Rebase to the llvm-project monorepo

df0466d

The new git submodule src/llvm-project is a monorepo replacing src/llvm and src/tools/{clang,lld,lldb}. This also serves as a rebase for these projects to the new 8.x branch from trunk. The src/llvm-emscripten fork is unchanged for now.

Add two more wasm-related LLVM commits

75d0bce

Set the DICompileUnit emissionKind

cff0750

Update LLVM for a CodeView fix

8b88324

librustc_llvm: default to libc++ for darwin

400dec0

[rust-lldb] Adapt to changes in LLDB APIs

a920ed9

Set LLDB_NO_DEBUGSERVER=ON

15cf179

Update dlmalloc to 0.1.2

059ed4f

Remove usage of an old and removed wasm intrinsic

cuviper force-pushed the llvm-monorepo branch from fe93b3d to 059ed4f Compare January 26, 2019 00:19

bors merged commit 059ed4f into rust-lang:master Jan 26, 2019

nikic mentioned this pull request Jan 26, 2019

Enable RISC-V atomic compare and swap #57925

Merged

This was referenced Jan 27, 2019

Field ordering causes extra memcpy #56356

Closed

Weird copy happening during initialization #56191

Closed

andersk mentioned this pull request Feb 12, 2019

Small SIMD test fails with --release but passes without #50154

Closed

cuviper deleted the llvm-monorepo branch March 27, 2019 22:19

cuviper mentioned this pull request Apr 22, 2019

linking failed: undefined reference to STATIC #60184

Closed

froydnj mentioned this pull request Apr 29, 2019

Rust 1.34 generates significantly less debug information for libstd functions vs. Rust 1.33 #60020

Closed

golddranks mentioned this pull request Jun 14, 2019

Fix rust-lldb wrapper scripts. #61827

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Rebase to the llvm-project monorepo #57675

Rebase to the llvm-project monorepo #57675

cuviper commented Jan 16, 2019

rust-highfive commented Jan 16, 2019

alexcrichton commented Jan 16, 2019

bors commented Jan 16, 2019

nikic Jan 16, 2019

cuviper Jan 16, 2019

Centril commented Jan 17, 2019

alexcrichton commented Jan 17, 2019

bors commented Jan 17, 2019

bors commented Jan 18, 2019

bors commented Jan 18, 2019

nikic commented Jan 18, 2019

cuviper commented Jan 18, 2019

alexcrichton commented Jan 18, 2019

cuviper commented Jan 18, 2019

cuviper commented Jan 18, 2019 •

edited

retep998 commented Jan 21, 2019

cuviper commented Jan 21, 2019

nikic commented Jan 21, 2019

cuviper commented Jan 21, 2019

glaubitz commented Jan 22, 2019 •

edited

mati865 Jan 25, 2019

cuviper Jan 25, 2019

cuviper commented Jan 26, 2019

bors commented Jan 26, 2019

bors commented Jan 26, 2019

bors commented Jan 26, 2019

alexcrichton commented Jan 26, 2019

nikic commented Feb 1, 2019

Rebase to the llvm-project monorepo #57675

Rebase to the llvm-project monorepo #57675

Conversation

cuviper commented Jan 16, 2019

rust-highfive commented Jan 16, 2019

alexcrichton commented Jan 16, 2019

bors commented Jan 16, 2019

nikic Jan 16, 2019

Choose a reason for hiding this comment

cuviper Jan 16, 2019

Choose a reason for hiding this comment

Centril commented Jan 17, 2019

alexcrichton commented Jan 17, 2019

bors commented Jan 17, 2019

bors commented Jan 18, 2019

bors commented Jan 18, 2019

nikic commented Jan 18, 2019

cuviper commented Jan 18, 2019

alexcrichton commented Jan 18, 2019

cuviper commented Jan 18, 2019

cuviper commented Jan 18, 2019 • edited

retep998 commented Jan 21, 2019

cuviper commented Jan 21, 2019

nikic commented Jan 21, 2019

cuviper commented Jan 21, 2019

glaubitz commented Jan 22, 2019 • edited

mati865 Jan 25, 2019

Choose a reason for hiding this comment

cuviper Jan 25, 2019

Choose a reason for hiding this comment

cuviper commented Jan 26, 2019

bors commented Jan 26, 2019

bors commented Jan 26, 2019

bors commented Jan 26, 2019

alexcrichton commented Jan 26, 2019

nikic commented Feb 1, 2019

cuviper commented Jan 18, 2019 •

edited

glaubitz commented Jan 22, 2019 •

edited