[PATCH] D154324: [C++20] [Modules] [ODRHash] Use CanonicalType for base classes

Hahnfeld added a comment. In D154324#4522551 , @alexfh wrote: > BTW, if in a.h I change > > typename std::enable_if<::p::P::value>::type> > > to > > typename std::enable_if::value>::type> > > Compilation succeeds. For the fun of it, could you test ht

[PATCH] D156425: [clang-repl] Remove redundant tests

Hahnfeld created this revision. Hahnfeld added a reviewer: v.g.vassilev. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. They only need to be tested once in `execute.cpp` and `fail.cpp`. Repository:

[PATCH] D156425: [clang-repl] Remove redundant tests

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGac6e9e69bac7: [clang-repl] Remove redundant tests (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https:/

[PATCH] D157838: [clang-repl] Disambiguate declarations with private typedefs

Hahnfeld created this revision. Hahnfeld added reviewers: v.g.vassilev, rsmith. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Member functions and static variable definitions may use typedefs that are

[PATCH] D157480: [clang-repl] Disambiguate global namespace identifiers

This revision was automatically updated to reflect the committed changes. Closed by commit rGba475a4a3440: [clang-repl] Disambiguate global namespace identifiers (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D157480/new/ https

[PATCH] D156891: [CodeGen] Remove Constant arguments from linkage functions, NFCI.

Hahnfeld added a comment. ping - this one is the only prerequisite left for three of my patches that were accepted yesterday... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156891/new/ https://reviews.llvm.org/D156891 ___

[PATCH] D156891: [CodeGen] Remove Constant arguments from linkage functions, NFCI.

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG2f3fe3ed97bc: [CodeGen] Remove Constant arguments from linkage functions, NFCI. (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANG

[PATCH] D157750: Properly handle -fsplit-machine-functions for fatbinary compilation

Hahnfeld added a comment. I dsabled two tests without `{arm,aarch64}-registered-target` in rGeeac4321c517ee8afc30ebe62c5b1778efc1173d ; two post-commit comments inline Comment at: llvm/test/CodeGen/Generic/

[PATCH] D156897: [CodeGen] Clean up access to EmittedDeferredDecls, NFCI.

This revision was automatically updated to reflect the committed changes. Closed by commit rGb719e410781c: [CodeGen] Clean up access to EmittedDeferredDecls, NFCI. (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156897/new/ htt

[PATCH] D157379: [CodeGen] Restrict addEmittedDeferredDecl to incremental extensions

This revision was automatically updated to reflect the committed changes. Closed by commit rGd43a3d634696: [CodeGen] Restrict addEmittedDeferredDecl to incremental extensions (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D15737

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGf8dadefd4afc: [CodeGen] Keep track of eagerly emitted globals (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTI

[PATCH] D157750: Properly handle -fsplit-machine-functions for fatbinary compilation

Hahnfeld added inline comments. Comment at: clang/test/Driver/fsplit-machine-functions-with-cuda-nvptx.c:16 +// causes a warning. +// RUN: %clang --target=x86_64-unknown-linux-gnu -nogpulib -nogpuinc \ +// RUN: --cuda-gpu-arch=sm_70 -x cuda -fsplit-machine-functions -S %s 2

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

Hahnfeld updated this revision to Diff 551401. Hahnfeld added a comment. Disable RTTI to (hopefully) avoid problems on Windows. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156537/new/ https://reviews.llvm.org/D156537 Files: clang/lib/CodeGen/CodeGenModule.cpp clang/test/Interpret

[PATCH] D158252: Fix regression of D157680

Hahnfeld accepted this revision. Hahnfeld added a comment. This revision is now accepted and ready to land. LG. IMHO such fixes to the tests can be committed without review :) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158252/new/ https://review

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

Hahnfeld added a comment. I reverted the change yesterday because the test wasn't passing on Windows, for example https://lab.llvm.org/buildbot/#/builders/216/builds/25769/steps/7/logs/FAIL__Clang__inline-virtual_cpp. The problem is that the JIT cannot find `??_7type_info@@6B@` related to RTTI.

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGc861d32d7c27: [CodeGen] Keep track of eagerly emitted globals (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTI

[PATCH] D157838: [clang-repl] Disambiguate declarations with private typedefs

Hahnfeld added inline comments. Comment at: clang/lib/Parse/ParseTentative.cpp:94 + // involve namespaces and friend declarations. + if (NextToken().is(tok::identifier)) +return true; v.g.vassilev wrote: > I am wondering what is the

[PATCH] D158413: [Lex] Introduce Preprocessor::LexAll()

Hahnfeld created this revision. Hahnfeld added reviewers: hokein, aaron.ballman. Herald added subscribers: kbarton, nemanjai. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This new method repeatedly c

[PATCH] D158414: [LexerTest] Use LexAll() in StringifyArgs

Hahnfeld created this revision. Hahnfeld added a reviewer: erichkeane. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Rely on `Preprocessor::LexAll()` and just replace `tok::comma` by `tok::eof` and p

[PATCH] D158415: [Lex] Handle repl_input_end in Preprocessor::LexAll()

Hahnfeld created this revision. Hahnfeld added reviewers: aaron.ballman, v.g.vassilev. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This fixes many unit tests when trying to enable `IncrementalExtens

[PATCH] D157838: [clang-repl] Disambiguate declarations with private typedefs

This revision was automatically updated to reflect the committed changes. Closed by commit rG6c274ba4108b: [clang-repl] Disambiguate declarations with private typedefs (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D157838/new/

[PATCH] D158414: [LexerTest] Use LexTokensUntilEOF() in StringifyArgs

Hahnfeld added a comment. ping! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158414/new/ https://reviews.llvm.org/D158414 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

Hahnfeld created this revision. Hahnfeld added reviewers: v.g.vassilev, rjmccall, aaron.ballman. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. An inline virtual function must be emitted, but we need t

[PATCH] D154328: [AST] Add API to iterate already loaded specializations

Hahnfeld added a comment. Right, it's not ideal to land APIs without usage in Clang itself... The context is, once again, ROOT and the Cling interpreter (where we are already using this API via downstream patches): When unloading a "transaction" (now `PartialTranslationUnit` in `clang-repl`) we

[PATCH] D156660: [CodeGen] Assert that EmittedDeferredDecls is empty

Hahnfeld created this revision. Hahnfeld added reviewers: v.g.vassilev, junaire, rjmccall. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Its contents are transferred into `DeferredDecls` in `Release()

[PATCH] D156660: [CodeGen] Assert that EmittedDeferredDecls is empty

This revision was automatically updated to reflect the committed changes. Closed by commit rG5ea647dea635: [CodeGen] Assert that EmittedDeferredDecls is empty (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156660/new/ https://

[PATCH] D156806: [Modules] Add test for merging of template member parent

Hahnfeld created this revision. Hahnfeld added reviewers: rsmith, ChuanqiXu. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This is a reduced test case originally meant to be addressed by https://revi

[PATCH] D137787: [CodeGen] Relax assertion on generating destructor call

Hahnfeld added a comment. The problem is back, but has been fixed upstream by https://reviews.llvm.org/rG61c7a9140becb19c5b1bc644e54452c6f782f5d5. I managed to reduce a test case triggering the assert touched in this revision, see https://reviews.llvm.org/D156806 Repository: rG LLVM Github

[PATCH] D156806: [Modules] Add test for merging of template member parent

This revision was automatically updated to reflect the committed changes. Closed by commit rG99b54743106b: [Modules] Add test for merging of template member parent (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156806/new/ htt

[PATCH] D156363: [Driver] -###: exit with code 1 if hasErrorOccurred

Hahnfeld added a comment. Hi, after this change I have three failing tests: Failed Tests (3): Clang :: Driver/lto.cu Clang :: Driver/openmp-offload-gpu.c Clang :: Driver/openmp-offload-jit.c I have CUDA installed on my system, not sure if a bot is testing this configuration... Cou

[PATCH] D156891: [CodeGen] Remove Constant arguments from linkage functions, NFCI.

Hahnfeld created this revision. Hahnfeld added reviewers: rjmccall, efriedma, asl. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This was unused since commit rGdd2362a8ba

[PATCH] D156897: [CodeGen] Clean up access to EmittedDeferredDecls, NFCI.

Hahnfeld created this revision. Hahnfeld added reviewers: rjmccall, v.g.vassilev, junaire. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. `GlobalDecl`s should only be added to `EmittedDeferredDecl` if

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

Hahnfeld added a comment. @v.g.vassilev do we have a means to detect this case? In D156897 , I'm refactoring all accesses to `EmittedDeferredDecl` to go via `addEmittedDeferredDecl`, so we should just add an early return when not emitting for a REPL. Reposito

[PATCH] D156363: [Driver] -###: exit with code 1 if hasErrorOccurred

Hahnfeld added a comment. `test/Driver/openmp-offload-jit.c` was still failing for me with CUDA installed, so I pushed rGcb3136b0d3596f5c3984e4fb49359e2a1a28a547 to add an explicit `--cuda-path` and make it return exit code 0

[PATCH] D157379: [CodeGen] Restrict addEmittedDeferredDecl to incremental extensions

Hahnfeld created this revision. Hahnfeld added reviewers: rjmccall, v.g.vassilev, junaire. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Reemission is only needed in incremental mode. With this early

[PATCH] D156537: [CodeGen] Keep track of eagerly emitted globals

Hahnfeld added a comment. In D156537#4554413 , @v.g.vassilev wrote: > In D156537#4554052 , @Hahnfeld > wrote: > >> @v.g.vassilev do we have a means to detect this case? In D156897 >>

[PATCH] D157477: [clang-repl] Add test for disambiguation of templates

Hahnfeld created this revision. Hahnfeld added a reviewer: v.g.vassilev. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. This test case was fixed in commit rG2c4620c1

[PATCH] D157480: [clang-repl] Disambiguate global namespace identifiers

Hahnfeld created this revision. Hahnfeld added reviewers: v.g.vassilev, aaron.ballman. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. A double colon starts an identifier name in the global namespace an

[PATCH] D157477: [clang-repl] Additional test for disambiguation of templates

Hahnfeld updated this revision to Diff 548558. Hahnfeld retitled this revision from "[clang-repl] Add test for disambiguation of templates" to "[clang-repl] Additional test for disambiguation of templates". Hahnfeld edited the summary of this revision. CHANGES SINCE LAST ACTION https://reviews.

[PATCH] D157477: [clang-repl] Additional test for disambiguation of templates

Hahnfeld added inline comments. Comment at: clang/test/Interpreter/namespace-template-disambiguate.cpp:7 +namespace NS2 { struct A { public: using S = int; }; } +namespace NS2 { A::S f(A::S a); } + v.g.vassilev wrote: > It might be a good idea to add this test to

[PATCH] D157480: [clang-repl] Disambiguate global namespace identifiers

Hahnfeld updated this revision to Diff 548562. Hahnfeld marked an inline comment as done. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D157480/new/ https://reviews.llvm.org/D157480 Files: clang/lib/Parse/ParseTentative.cpp clang/test/Interpreter/disambiguate-decl-stmt.cpp Index: cl

[PATCH] D157480: [clang-repl] Disambiguate global namespace identifiers

Hahnfeld added inline comments. Comment at: clang/test/Interpreter/global-namespace-disambiguate.cpp:6 +struct A { ~A(); }; +::A::~A() {} + v.g.vassilev wrote: > Can we add this test to `disambiguate-decl-stmt.cpp` instead where we track > these things? done C

[PATCH] D157477: [clang-repl] Additional test for disambiguation of templates

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGda555f750ab2: [clang-repl] Additional test for disambiguation of templates (authored by Hahnfeld). Repository: rG LLVM Github Monorepo CHANGES SI

[PATCH] D156891: [CodeGen] Remove Constant arguments from linkage functions, NFCI.

Hahnfeld added a comment. gentle ping :-) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156891/new/ https://reviews.llvm.org/D156891 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.

[PATCH] D156897: [CodeGen] Clean up access to EmittedDeferredDecls, NFCI.

Hahnfeld added a comment. gentle ping :-) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D156897/new/ https://reviews.llvm.org/D156897 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.

[PATCH] D158413: [Lex] Introduce Preprocessor::LexAll()

Hahnfeld added a comment. ping! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158413/new/ https://reviews.llvm.org/D158413 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/c

[PATCH] D158414: [LexerTest] Use LexAll() in StringifyArgs

Hahnfeld added a comment. ping! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158414/new/ https://reviews.llvm.org/D158414 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/c

[PATCH] D158415: [Lex] Handle repl_input_end in Preprocessor::LexAll()

Hahnfeld added a comment. ping! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158415/new/ https://reviews.llvm.org/D158415 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/c

[PATCH] D158415: [Lex] Handle repl_input_end in Preprocessor::LexAll()

Hahnfeld added a comment. This revision cannot land without https://reviews.llvm.org/D158413 and for that one I really want to have feedback from somebody working in that area... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158415/new/ https://re

[PATCH] D158413: [Lex] Introduce Preprocessor::LexTokensUntilEOF()

Hahnfeld updated this revision to Diff 556791. Hahnfeld retitled this revision from "[Lex] Introduce Preprocessor::LexAll()" to "[Lex] Introduce Preprocessor::LexTokensUntilEOF()". Hahnfeld edited the summary of this revision. Hahnfeld added a comment. Rename to `Preprocessor::LexTokensUntilEOF()

[PATCH] D158413: [Lex] Introduce Preprocessor::LexTokensUntilEOF()

Hahnfeld marked 2 inline comments as done. Hahnfeld added inline comments. Comment at: clang/lib/Lex/Preprocessor.cpp:998 +std::vector Preprocessor::LexAll() { + std::vector toks; aaron.ballman wrote: > v.g.vassilev wrote: > > Shouldn't we take the results as

[PATCH] D158413: [Lex] Introduce Preprocessor::LexTokensUntilEOF()

Hahnfeld updated this revision to Diff 556834. Hahnfeld marked an inline comment as done. Hahnfeld added a comment. Handle all of `tok::unknown, tok::eof, tok::eod`. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158413/new/ https://reviews.llvm.org/D158413 Files: clang/include/clang/

[PATCH] D158413: [Lex] Introduce Preprocessor::LexTokensUntilEOF()

Hahnfeld added inline comments. Comment at: clang/lib/Lex/Preprocessor.cpp:1000 + std::vector toks; + while (1) { +Token tok; aaron.ballman wrote: > v.g.vassilev wrote: > > aaron.ballman wrote: > > > Hahnfeld wrote: > > > > aaron.ballman wrote: > > > > > I'

[PATCH] D158414: [LexerTest] Use LexTokensUntilEOF() in StringifyArgs

Hahnfeld updated this revision to Diff 556848. Hahnfeld retitled this revision from "[LexerTest] Use LexAll() in StringifyArgs" to "[LexerTest] Use LexTokensUntilEOF() in StringifyArgs". Hahnfeld edited the summary of this revision. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158414/new

[PATCH] D158415: [Lex] Handle repl_input_end in Preprocessor::LexTokensUntilEOF()

Hahnfeld updated this revision to Diff 556849. Hahnfeld retitled this revision from "[Lex] Handle repl_input_end in Preprocessor::LexAll()" to "[Lex] Handle repl_input_end in Preprocessor::LexTokensUntilEOF()". Hahnfeld added a comment. Rebase on D158413 CHAN

[PATCH] D154328: [AST] Add API to iterate already loaded specializations

Hahnfeld created this revision. Hahnfeld added reviewers: erichkeane, aaron.ballman. Herald added a project: All. Hahnfeld requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. These new functions allow to look at specializations without triggeri

[PATCH] D137787: [CodeGen] Relax assertion on generating destructor call

Hahnfeld abandoned this revision. Hahnfeld added a comment. apparently not needed anymore downstream... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D137787/new/ https://reviews.llvm.org/D137787 ___ cfe-

[PATCH] D134813: Properly print unnamed TagDecl objects in diagnostics

Hahnfeld added a comment. Out of curiosity, not sure if it's worth fixing because it's easy enough to work around: I think after this change, it's not possible anymore to call `printName(raw_ostream &OS)` on a (statically typed) `TagDecl` / `EnumDecl` because it's hidden by `TagDecl::printName(

[PATCH] D134813: Properly print unnamed TagDecl objects in diagnostics

Hahnfeld added a comment. In D134813#4482808 , @aaron.ballman wrote: > In D134813#4482674 , @Hahnfeld > wrote: > >> Out of curiosity, not sure if it's worth fixing because it's easy enough to >> work around: I

[PATCH] D134813: Properly print unnamed TagDecl objects in diagnostics

Hahnfeld added a comment. In D134813#4482822 , @aaron.ballman wrote: > In D134813#4482819 , @Hahnfeld > wrote: > >> Thanks! Note that the same probably holds true (but I didn't test) for all >> other classes th

[PATCH] D41416: [modules] [pch] Do not deserialize all lazy template specializations when looking for one.

Hahnfeld added a comment. In D41416#4487516 , @ChuanqiXu wrote: > But some local in tree tests got failed. I took some time to investigate them > but I didn't get insights : ( If the failures involve template template parameters, please note that you ne

[PATCH] D41416: [modules] [pch] Do not deserialize all lazy template specializations when looking for one.

Hahnfeld added inline comments. Comment at: clang/include/clang/AST/DeclTemplate.h:786-788 +uint32_t DeclID = ~0U; +unsigned ODRHash = ~0U; +bool IsPartial = false; ChuanqiXu wrote: > Maybe this can save some space. Can you elaborate why this would sa

[PATCH] D154856: [MemProf] Use new option/pass for profile feedback and matching

Hahnfeld added a comment. Hi @tejohnson, I'm seeing crashes after this revision: Failed Tests (2): LLVM-Unit :: Passes/./PluginsTests/0/2 LLVM-Unit :: Passes/./PluginsTests/1/2 A stack trace looks as follows: #0 0x7fba1f899667 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int)

[PATCH] D154856: [MemProf] Use new option/pass for profile feedback and matching

Hahnfeld added a comment. Ok, turns out I had old test binaries around that got automatically picked up by `lit`. Sorry for the noise. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D154856/new/ https://reviews.llvm.org/D154856

[PATCH] D158413: [Lex] Introduce Preprocessor::LexAll()

Hahnfeld added a comment. ping :) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158413/new/ https://reviews.llvm.org/D158413 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D158414: [LexerTest] Use LexAll() in StringifyArgs

Hahnfeld added a comment. ping :) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158414/new/ https://reviews.llvm.org/D158414 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D158415: [Lex] Handle repl_input_end in Preprocessor::LexAll()

Hahnfeld added a comment. ping :) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158415/new/ https://reviews.llvm.org/D158415 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org

[PATCH] D138792: [AArch64] Improve TargetParser API

Hahnfeld added a comment. Hi, I bisected this change to lead to a couple of test failures when building with `LLVM_LINK_LLVM_DYLIB`. In the past, this had to do with global variable initialization order, but nothing immediately jumps to my eye in this patch. Is `AARCH64_ARCH` used to define glo

[PATCH] D75914: systemz: allow configuring default CLANG_SYSTEMZ_ARCH

Hahnfeld added inline comments. Comment at: clang/CMakeLists.txt:311 + "SystemZ Default Arch") +add_definitions( -DCLANG_SYSTEMZ_DEFAULT_ARCH="${CLANG_SYSTEMZ_DEFAULT_ARCH}") + Passing values like this is unusual for CMake and causes all source files to be reco

[PATCH] D75001: [OpenMP][cmake] ignore warning on unknown CUDA version

Hahnfeld added a comment. In D75001#1887876 , @jdoerfert wrote: > I like this way better. I was hoping we could do it in our cmake only :) > > Give others a day or so to comment before your commit but I'm fine with this. Well, that doesn't really work if

[PATCH] D53250: [ToolChain] Use default linker if the toolchain uses a custom one

Hahnfeld added a comment. Herald added a project: clang. Going through my old revisions, is this still something that we want? Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D53250/new/ https://reviews.llvm.org/D53250

[PATCH] D70804: [Frontend] Allow OpenMP offloading to aarch64

Hahnfeld added a comment. In D70804#1763443 , @bryanpkc wrote: > In D70804#1763289 , @ABataev wrote: > > > Tests are required > > > The tests for this were added in D30644 , > but t

[PATCH] D71134: [OpenMP] Require trivially copyable type for mapping

Hahnfeld created this revision. Hahnfeld added reviewers: ABataev, jdoerfert. Herald added subscribers: cfe-commits, guansong. Herald added a project: clang. A trivially copyable type provides a trivial copy constructor and a trivial copy assignment operator. This is enough for the runtime to memc

[PATCH] D71134: [OpenMP] Require trivially copyable type for mapping

Hahnfeld marked 2 inline comments as done. Hahnfeld added inline comments. Comment at: clang/lib/Sema/SemaOpenMP.cpp:14917 if (FullCheck && !SemaRef.CurContext->isDependentContext() && - !QTy.isTrivialType(SemaRef.Context)) + !QTy.isTriviallyCopyableType(SemaRef.Cont

[PATCH] D71134: [OpenMP] Require trivially copyable type for mapping

This revision was automatically updated to reflect the committed changes. Hahnfeld marked an inline comment as done. Closed by commit rG071dca24cea9: [OpenMP] Require trivially copyable type for mapping (authored by Hahnfeld). Changed prior to commit: https://reviews.llvm.org/D71134?vs=232610&i

[PATCH] D71340: [VE,#3] Runtime libraries

Hahnfeld added a comment. In D71340#1779446 , @simoll wrote: > To clarify: this patch is not intended to be committed as-is but gives an > overview of the required changes. It's the a patch of a series for the VE > backend

[PATCH] D75001: [OpenMP][cmake] ignore warning on unknown CUDA version

Hahnfeld added a comment. In D75001#1922789 , @jdoerfert wrote: > In D75001#1891284 , @Hahnfeld wrote: > > > In D75001#1887876 , @jdoerfert > > wrote: > > > > > I like this

[PATCH] D45212: [HIP] Let CUDA toolchain support HIP language mode and amdgpu

Hahnfeld added inline comments. Comment at: include/clang/Basic/Cuda.h:61 + GFX900, + GFX902, LAST, rjmccall wrote: > yaxunl wrote: > > rjmccall wrote: > > > Does this actually have anything to do with HIP? You have a lot of > > > changes in this patch whi

[PATCH] D45212: [HIP] Let CUDA toolchain support HIP language mode and amdgpu

Hahnfeld added a comment. Can this revision be split further? The summary mentions many things that might make up multiple independent changes... Comment at: lib/Driver/ToolChains/Cuda.cpp:263 +// HIP needs c++11. +CC1Args.push_back("-std=c++11"); +// Skip CUDA inc

[PATCH] D42922: [CUDA] Register relocatable GPU binaries

Hahnfeld updated this revision to Diff 141685. Hahnfeld added a comment. Sorry for the long delay. This update rebases the patch against current trunk and adapts the regression test. https://reviews.llvm.org/D42922 Files: lib/CodeGen/CGCUDANV.cpp test/CodeGenCUDA/device-stub.cu Index: tes

[PATCH] D42922: [CUDA] Register relocatable GPU binaries

Hahnfeld updated this revision to Diff 141698. Hahnfeld added a comment. Correct test check prefix. https://reviews.llvm.org/D42922 Files: lib/CodeGen/CGCUDANV.cpp test/CodeGenCUDA/device-stub.cu Index: test/CodeGenCUDA/device-stub.cu ===

[PATCH] D45449: [CUDA] Document recent changes

Hahnfeld created this revision. Hahnfeld added reviewers: tra, jlebar, rsmith. Herald added a subscriber: cfe-commits. Hahnfeld added a dependency: D42922: [CUDA] Register relocatable GPU binaries. - Finding installations via ptxas binary - Relocatable device code Repository: rC Clang https:/

[PATCH] D45449: [CUDA] Document recent changes

Hahnfeld added a comment. In https://reviews.llvm.org/D45449#1062642, @tra wrote: > We could use specific instructions for what exactly one needs to do in order > to use relocatable device code in practice. Could be a separate patch. Maybe I should add an example to https://llvm.org/docs/Compi

[PATCH] D45212: [HIP] Let CUDA toolchain support HIP language mode and amdgpu

Hahnfeld added a comment. In https://reviews.llvm.org/D45212#1063186, @yaxunl wrote: > In https://reviews.llvm.org/D45212#1060628, @Hahnfeld wrote: > > > Can this revision be split further? The summary mentions many things that > > might make up multiple independent changes... > > > I separated

[PATCH] D39457: [OPENMP] Current status of OpenMP support.

Hahnfeld added a comment. Do we want to add documentation about which target devices Clang can generate code for? Comment at: docs/OpenMPSupport.rst:46 + +* #pragma omp distribute parallel for [simd]: :partial:`Partial`. No full codegen support. + I think tha

[PATCH] D39504: [OpenMP] Extend "Avoid VLAs for reduction" optimization to VLAs as base

Hahnfeld created this revision. We can generate constant sized arrays whenever the array section has constant length, even if the base expression itself is a VLA. https://reviews.llvm.org/D39504 Files: lib/Sema/SemaOpenMP.cpp test/OpenMP/for_reduction_codegen.cpp Index: test/OpenMP/for_red

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld created this revision. Herald added a subscriber: jholewinski. Some target devices (e.g. Nvidia GPUs) don't support dynamic stack allocation and hence no VLAs. Print errors with description instead of failing in the backend or generating code that doesn't work. This patch handles explici

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > The way you've written this makes it sound l

[PATCH] D39504: [OpenMP] Extend "Avoid VLAs for reduction" optimization to VLAs as base

This revision was automatically updated to reflect the committed changes. Closed by commit rL317207: [OpenMP] Extend "Avoid VLAs for reduction" optimization to VLAs as base (authored by Hahnfeld). Changed prior to commit: https://reviews.llvm.org/D39504?vs=121169&id=121297#toc Repository: rL

[PATCH] D38976: [OpenMP] Add implicit data sharing support when offloading to NVIDIA GPUs using OpenMP device offloading

Hahnfeld added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp:2393 + + // Level0(CGF); + Is this meant to be a comment or should there be something happening here? Comment at: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp:2396 + CGF

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld marked 3 inline comments as done. Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > H

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld marked an inline comment as done. Hahnfeld added a subscriber: gtbercea. Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; }

[PATCH] D39708: [OMPT] Fix test cancel_parallel.c

Hahnfeld created this revision. If a parallel region is cancelled, execution resumes at the end of the structured block. That is why this test cannot use the "normal" macros that print right after inserting the label. Instead it previously printed the addresses before the pragma and swapped the ch

[PATCH] D39821: Add CLANG_DEFAULT_OBJCOPY to allow Clang to use llvm-objcopy for dwarf fission

Hahnfeld added inline comments. Comment at: CMakeLists.txt:236 +set(CLANG_DEFAULT_OBJCOPY "objcopy" CACHE STRING + "Default objcopy runtime to use.") + Nit: `runtime` -> `executable` https://reviews.llvm.org/D39821 __

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > ABataev wrote: > > ABataev wrote: > > > Hahn

[PATCH] D39821: Add CLANG_DEFAULT_OBJCOPY to allow Clang to use llvm-objcopy for dwarf fission

Hahnfeld added a comment. I think the current tests will only pass iff the executable name ends with `objcopy` - can we make this more reliable? Repository: rL LLVM https://reviews.llvm.org/D39821 ___ cfe-commits mailing list cfe-commits@lists.l

[PATCH] D39821: Add CLANG_DEFAULT_OBJCOPY to allow Clang to use llvm-objcopy for dwarf fission

Hahnfeld added a comment. In https://reviews.llvm.org/D39821#922026, @jakehehrlich wrote: > In https://reviews.llvm.org/D39821#921703, @Hahnfeld wrote: > > > I think the current tests will only pass iff the executable name ends with > > `objcopy` - can we make this more reliable? > > > If I read

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > Hahnfeld wrote: > > rjmccall wrote: > > > AB

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > Hahnfeld wrote: > > rjmccall wrote: > > > Ha

[PATCH] D39505: [OpenMP] Show error if VLAs are not supported

Hahnfeld added inline comments. Comment at: include/clang/Basic/TargetInfo.h:944 + /// \brief Whether target supports variable-length arrays. + bool isVLASupported() const { return VLASupported; } + rjmccall wrote: > ABataev wrote: > > Hahnfeld wrote: > > > rjm

<    1   2   3   4   5   6   >