[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-02-21 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee abandoned this revision. bc-lee added a comment. Closed in favor of https://reviews.llvm.org/D118862 and https://reviews.llvm.org/D118875 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/D117924 __

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-02-03 Thread Thorsten via Phabricator via cfe-commits
tschuett added a comment. https://reviews.llvm.org/D118862 https://reviews.llvm.org/D118875 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/D117924 ___ cfe-commits mail

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-02-02 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee updated this revision to Diff 405461. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/D117924 Files: clang/lib/Driver/ToolChains/Darwin.cpp clang/test/Driver/darwin-sanitizer-ld.c compiler-rt/cmake/bui

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-02-02 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee added a comment. Some changes in this patch were applied as https://reviews.llvm.org/D118759, so I'm re-uploading this patch to get rid of merge conflicts. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-28 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee added a comment. For the Chromium side issue is https://crbug.com/1259122. Also I opened LLVM issue https://github.com/llvm/llvm-project/issues/53477. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/D11792

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-28 Thread Nico Weber via Phabricator via cfe-commits
thakis added a comment. In D117924#3274425 , @bc-lee wrote: > It may not be appropriate to add other runtime libraries specifically for Mac > Catalyst. However, currently `lld` does not allow linking with dynamic > libraries with different types of buil

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-26 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee added a comment. It may not be appropriate to add other runtime libraries specifically for Mac Catalyst. However, currently `lld` does not allow linking with dynamic libraries with different types of build_version_command, whereas `ld64` does allow for Mac Catalyst. Considering compatibi

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-26 Thread Dan Liew via Phabricator via cfe-commits
delcypher added a comment. @bc-lee Thanks for the patch. While I get what you're trying to do I have doubts about being able to accept the patch in its current form. Apple's ASan catalyst doesn't work by building a separate dylib, instead it builds the macosx dylib in a different way to make it

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-22 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee updated this revision to Diff 402201. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D117924/new/ https://reviews.llvm.org/D117924 Files: clang/lib/Driver/ToolChains/Darwin.cpp clang/test/Driver/darwin-sanitizer-ld.c compiler-rt/cmake/bui

[PATCH] D117924: [compiler_rt] Add a seperate runtime for Mac Catalyst

2022-01-21 Thread Byoungchan Lee via Phabricator via cfe-commits
bc-lee created this revision. Herald added a subscriber: mgorny. bc-lee requested review of this revision. Herald added projects: clang, Sanitizers. Herald added subscribers: Sanitizers, cfe-commits. Mac Catalyst has different platform fields in build_version_command, so it's natural to create a s