[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-08-02 Thread Michael Kruse via Phabricator via cfe-commits
This revision was automatically updated to reflect the committed changes. Closed by commit rC338810: [CodeGen] Emit parallel_loop_access for each loop in the loop stack. (authored by Meinersbur, committed by ). Changed prior to commit: https://reviews.llvm.org/D48808?vs=155288&id=158912#toc Re

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-08-01 Thread Hal Finkel via Phabricator via cfe-commits
hfinkel accepted this revision. hfinkel added a comment. This revision is now accepted and ready to land. LGTM Repository: rC Clang https://reviews.llvm.org/D48808 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-26 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur added a comment. ping Repository: rC Clang https://reviews.llvm.org/D48808 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-17 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur added a comment. ping Repository: rC Clang https://reviews.llvm.org/D48808 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-12 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur updated this revision to Diff 155288. Meinersbur added a comment. - Hoist I->mayReadOrWriteMemory() condition. Repository: rC Clang https://reviews.llvm.org/D48808 Files: lib/CodeGen/CGLoopInfo.cpp test/CodeGenCXX/pragma-loop-safety-nested.cpp test/CodeGenCXX/pragma-loop-saf

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur updated this revision to Diff 153803. Meinersbur added a comment. - Adapt test file message. Repository: rC Clang https://reviews.llvm.org/D48808 Files: lib/CodeGen/CGLoopInfo.cpp test/CodeGenCXX/pragma-loop-safety-nested.cpp test/CodeGenCXX/pragma-loop-safety-outer.cpp In

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur updated this revision to Diff 153769. Meinersbur added a comment. - Allow multiple parallel annotations Repository: rC Clang https://reviews.llvm.org/D48808 Files: lib/CodeGen/CGLoopInfo.cpp test/CodeGenCXX/pragma-loop-safety-nested.cpp test/CodeGenCXX/pragma-loop-safety-out

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Hal Finkel via Phabricator via cfe-commits
hfinkel added a comment. In https://reviews.llvm.org/D48808#1149862, @Meinersbur wrote: > In https://reviews.llvm.org/D48808#1149845, @hfinkel wrote: > > > We specifically defined the metadata to support nested loops. The LangRef > > says, "The llvm.mem.parallel_loop_access metadata refers to a

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur added a comment. In https://reviews.llvm.org/D48808#1149845, @hfinkel wrote: > We specifically defined the metadata to support nested loops. The LangRef > says, "The llvm.mem.parallel_loop_access metadata refers to a loop > identifier, or metadata containing a list of loop identifier

Re: [PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Hal Finkel via cfe-commits
On 07/02/2018 12:27 PM, Tyler Nowicki wrote: > Hi Michael, Hal, > > Sorry it has been a while since I looked at this. My memory is a > little fuzzy. The intent of 'assume_safety' is to tell LAA to > skip dependency checking on loads and stores so the vectorizer doesn't > stop as soon as it sees bo

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Hal Finkel via Phabricator via cfe-commits
hfinkel added a comment. In https://reviews.llvm.org/D48808#1149828, @Meinersbur wrote: > In https://reviews.llvm.org/D48808#1149549, @hfinkel wrote: > > > In https://reviews.llvm.org/D48808#1149534, @ABataev wrote: > > > > > > > > > > > Michael, can you please add a test with two inner loops, on

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur added a comment. In https://reviews.llvm.org/D48808#1149549, @hfinkel wrote: > In https://reviews.llvm.org/D48808#1149534, @ABataev wrote: > > > > > > Michael, can you please add a test with two inner loops, one where more than > one is annotated, and one where only the outer loop is

Re: [PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Tyler Nowicki via cfe-commits
Hi Michael, Hal, Sorry it has been a while since I looked at this. My memory is a little fuzzy. The intent of 'assume_safety' is to tell LAA to skip dependency checking on loads and stores so the vectorizer doesn't stop as soon as it sees both in a loop. At the time 'assume_safety' was implemented

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur added a comment. In https://reviews.llvm.org/D48808#1149534, @ABataev wrote: > I don't think that this is the intended behavior of the `#pragma clang loop`. > it is better to ask the author of this pragma is this correct or not. I understand it as the intended behavior of the `assum

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Hal Finkel via Phabricator via cfe-commits
hfinkel added a comment. In https://reviews.llvm.org/D48808#1149534, @ABataev wrote: > I don't think that this is the intended behavior of the `#pragma clang loop`. > it is better to ask the author of this pragma is this correct or not. It is the intended behavior that the memory accesses are

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-07-02 Thread Alexey Bataev via Phabricator via cfe-commits
ABataev added a comment. I don't think that this is the intended behavior of the `#pragma clang loop`. it is better to ask the author of this pragma is this correct or not. Repository: rC Clang https://reviews.llvm.org/D48808 ___ cfe-commits mai

[PATCH] D48808: [CodeGen] Emit parallel_loop_access for each loop in the loop stack.

2018-06-30 Thread Michael Kruse via Phabricator via cfe-commits
Meinersbur created this revision. Meinersbur added reviewers: ABataev, hfinkel. Emit !llvm.mem.parallel_loop_access metadata for memory accesses even if the parallel loop is not the top on the loop stack. Fixes llvm.org/PR37558. Repository: rC Clang https://reviews.llvm.org/D48808 Files: