[clang-tools-extra] r246224 - Reverting r246209 while I investigate a build bot failure: http://lab.llvm.org:8011/builders/clang-x86_64-debian-fast/builds/30516

Author: aaronballman Date: Thu Aug 27 17:19:50 2015 New Revision: 246224 URL: http://llvm.org/viewvc/llvm-project?rev=246224&view=rev Log: Reverting r246209 while I investigate a build bot failure: http://lab.llvm.org:8011/builders/clang-x86_64-debian-fast/builds/30516 Removed: clang-tools-e

Re: [PATCH] D11740: ABI versioning macros for libc++

EricWF added a comment. In http://reviews.llvm.org/D11740#234575, @eugenis wrote: > Yes, not being able to use headers in the libcxx source tree is quite > unpleasant. It can be fixed by providing a __config_version in libcxx/include > with the default version values. Or, in the approach of >

r246225 - Use an explicit assignment.

Author: echristo Date: Thu Aug 27 17:20:03 2015 New Revision: 246225 URL: http://llvm.org/viewvc/llvm-project?rev=246225&view=rev Log: Use an explicit assignment. Modified: cfe/trunk/lib/CodeGen/CGCall.cpp Modified: cfe/trunk/lib/CodeGen/CGCall.cpp URL: http://llvm.org/viewvc/llvm-project/c

r246228 - [X86] Use AVX features instead of ABI to init. SimdDefaultAlign.

Author: ab Date: Thu Aug 27 17:24:56 2015 New Revision: 246228 URL: http://llvm.org/viewvc/llvm-project?rev=246228&view=rev Log: [X86] Use AVX features instead of ABI to init. SimdDefaultAlign. The ABI string only exists to communicate with TargetCodeGenInfo. Concretely, since we only used "avx*"

Re: [PATCH] D12390: Also enable the avx/avx512 ABIs for i386, not just x86_64.

This revision was automatically updated to reflect the committed changes. Closed by commit rL246228: [X86] Use AVX features instead of ABI to init. SimdDefaultAlign. (authored by ab). Changed prior to commit: http://reviews.llvm.org/D12390?vs=33280&id=33363#toc Repository: rL LLVM http://re

Re: [PATCH] D12081: Add use-nullptr check to clang-tidy.

aaron.ballman added a subscriber: aaron.ballman. aaron.ballman added a comment. While working on r246209, one of the build bots ran into an issue (commented below) that has me slightly perplexed. The build break can be found at: http://lab.llvm.org:8011/builders/clang-x86_64-debian-fast/builds/3

r246229 - [X86] Conditionalize Darwin MaxVectorAlign on the presence of AVX.

Author: ab Date: Thu Aug 27 17:30:38 2015 New Revision: 246229 URL: http://llvm.org/viewvc/llvm-project?rev=246229&view=rev Log: [X86] Conditionalize Darwin MaxVectorAlign on the presence of AVX. There's no point in using a larger alignment if we have no instructions that would benefit from it.

Re: [PATCH] D12389: Conditionalize X86 Darwin MaxVectorAlign on the presence of AVX.

This revision was automatically updated to reflect the committed changes. Closed by commit rL246229: [X86] Conditionalize Darwin MaxVectorAlign on the presence of AVX. (authored by ab). Changed prior to commit: http://reviews.llvm.org/D12389?vs=33278&id=33365#toc Repository: rL LLVM http://

Re: [PATCH] D12390: Use 32/64 SimdDefaultAlign when AVX/AVX512 is enabled, even on i386.

ab added a comment. r246228 using the features Thanks for the reviews! Repository: rL LLVM http://reviews.llvm.org/D12390 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Re: [PATCH] D10724: Bump X86 Darwin MaxVectorAlign to 64, for AVX512.

This revision was automatically updated to reflect the committed changes. Closed by commit rL246230: [X86] Bump Darwin MaxVectorAlign to 64 when AVX512 is enabled. (authored by ab). Changed prior to commit: http://reviews.llvm.org/D10724?vs=28433&id=33367#toc Repository: rL LLVM http://revi

r246230 - [X86] Bump Darwin MaxVectorAlign to 64 when AVX512 is enabled.

Author: ab Date: Thu Aug 27 17:42:12 2015 New Revision: 246230 URL: http://llvm.org/viewvc/llvm-project?rev=246230&view=rev Log: [X86] Bump Darwin MaxVectorAlign to 64 when AVX512 is enabled. Without this, 64-byte vector types (__m512), specified to be 64-byte aligned in the AVX512 draft SysV ABI

Re: [PATCH] D12355: [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*)

EricWF marked an inline comment as done. Comment at: include/string:3802 @@ -3799,2 +3801,3 @@ +return _Traits::compare(__rhs.data(), __lhs, _VSTD::min(__lhs_len, __rhs.size())) == 0; } mclow.lists wrote: > Also, you should not call `traits::compare` here.

Re: [PATCH] D12355: [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*)

EricWF updated this revision to Diff 33368. EricWF added a comment. Remove unnecessary call to `std::min`. http://reviews.llvm.org/D12355 Files: include/string Index: include/string === --- include/string +++ include/string @@ -

Re: [PATCH] D11740: ABI versioning macros for libc++

eugenis added a comment. In http://reviews.llvm.org/D11740#234610, @EricWF wrote: > In http://reviews.llvm.org/D11740#234575, @eugenis wrote: > > > Yes, not being able to use headers in the libcxx source tree is quite > > unpleasant. It can be fixed by providing a __config_version in > > libcxx

Re: [PATCH] D12312: Emiting invariant.group.barrier and adding -fstrict-vptrs

rjmccall added inline comments. Comment at: include/clang/Driver/Options.td:990 @@ -988,2 +989,3 @@ "value range">; +def fstrict_vptrs: Flag<["-"], "fstrict-vptrs">, Group, Flags<[CC1Option]>; def fstrict_overflow : Flag<["-"], "fstrict-overflow">, Group; ---

Re: [PATCH] D12181: [sanitizer] Add -fsanitize-trap-function.

rsmith added a comment. In http://reviews.llvm.org/D12181#234590, @jmgao wrote: > Ping, I think @samsonov was waiting on @rsmith's feedback on the following: > > In http://reviews.llvm.org/D12181#229493, @jmgao wrote: > > > The goal is to be able to give a useful fsanitize-specific error message

Re: [PATCH] D11740: ABI versioning macros for libc++

EricWF added a comment. In http://reviews.llvm.org/D11740#234642, @eugenis wrote: > OK. Then _LIBCPP_ABI_UNSTABLE won't bump the ABI version (as seen in library > soname and header path)? Yeah. That was how I imagined it. http://reviews.llvm.org/D11740

Re: [PATCH] D12312: Emiting invariant.group.barrier and adding -fstrict-vptrs

On Thu, Aug 27, 2015 at 3:49 PM, John McCall wrote: > rjmccall added inline comments. > > > Comment at: include/clang/Driver/Options.td:990 > @@ -988,2 +989,3 @@ > "value range">; > +def fstrict_vptrs: Flag<["-"], "fstrict-vptrs">, Group, > Flags<[CC1Option]>; > def

r246231 - CGDebugInfo: Instead of uniquing RetainedTypes, just refrain from retaining

Author: adrian Date: Thu Aug 27 17:56:46 2015 New Revision: 246231 URL: http://llvm.org/viewvc/llvm-project?rev=246231&view=rev Log: CGDebugInfo: Instead of uniquing RetainedTypes, just refrain from retaining them more than once. (NFC) Modified: cfe/trunk/lib/CodeGen/CGDebugInfo.cpp Modified

Re: r246210 - CGDebugInfo: Factor out a getOrCreateStandaloneType() method.

> On Aug 27, 2015, at 3:06 PM, David Blaikie wrote: > > > > On Thu, Aug 27, 2015 at 2:35 PM, Adrian Prantl > wrote: > >> On Aug 27, 2015, at 2:25 PM, David Blaikie > > wrote: >> >> >> >> On Thu, Aug 27, 2015 at 2:21 PM, Adrian Prantl vi

Re: [PATCH] D12312: Emiting invariant.group.barrier and adding -fstrict-vptrs

Prazek added inline comments. Comment at: include/clang/Driver/Options.td:990 @@ -988,2 +989,3 @@ "value range">; +def fstrict_vptrs: Flag<["-"], "fstrict-vptrs">, Group, Flags<[CC1Option]>; def fstrict_overflow : Flag<["-"], "fstrict-overflow">, Group; -

Re: [PATCH] D12081: Add use-nullptr check to clang-tidy.

alexfh added a comment. In http://reviews.llvm.org/D12081#234614, @aaron.ballman wrote: > While working on r246209, one of the build bots ran into an issue (commented > below) that has me slightly perplexed. The build break can be found at: > http://lab.llvm.org:8011/builders/clang-x86_64-debia

Re: [PATCH] D12081: Add use-nullptr check to clang-tidy.

On Thu, Aug 27, 2015 at 7:20 PM, Alexander Kornienko wrote: > alexfh added a comment. > > In http://reviews.llvm.org/D12081#234614, @aaron.ballman wrote: > >> While working on r246209, one of the build bots ran into an issue (commented >> below) that has me slightly perplexed. The build break can

Re: [PATCH] D12081: Add use-nullptr check to clang-tidy.

On Fri, Aug 28, 2015 at 1:23 AM, Aaron Ballman wrote: > On Thu, Aug 27, 2015 at 7:20 PM, Alexander Kornienko > wrote: > > alexfh added a comment. > > > > In http://reviews.llvm.org/D12081#234614, @aaron.ballman wrote: > > > >> While working on r246209, one of the build bots ran into an issue > (

r246237 - Fix macro backtrace printing.

Author: rtrieu Date: Thu Aug 27 18:38:45 2015 New Revision: 246237 URL: http://llvm.org/viewvc/llvm-project?rev=246237&view=rev Log: Fix macro backtrace printing. Sometimes, a macro that expands to another macro name will not be printed in the macro backtrace. This patch finds the missed macro e

[clang-tools-extra] r246238 - [clang-tidy] Documented the reason to run the test in C++98 mode.

Author: alexfh Date: Thu Aug 27 18:43:39 2015 New Revision: 246238 URL: http://llvm.org/viewvc/llvm-project?rev=246238&view=rev Log: [clang-tidy] Documented the reason to run the test in C++98 mode. Modified: clang-tools-extra/trunk/test/clang-tidy/modernize-use-nullptr-basic.cpp Modified:

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

hfinkel added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + The code below for OPT_mstackrealign uses -mstackrealign as the name of the backend

Re: [PATCH] D11740: ABI versioning macros for libc++

eugenis added a comment. There is a bit of a problem with testing libc++. If the library is built for the non-default ABI, we can not build tests against headers in libc++ source. And the headers in the build directory are only updated when cmake is re-run. I guess the latter can be fixed by up

Re: [PATCH] D11740: ABI versioning macros for libc++

EricWF added a comment. In http://reviews.llvm.org/D11740#234760, @eugenis wrote: > There is a bit of a problem with testing libc++. If the library is built for > the non-default ABI, we can not build tests against headers in libc++ source. > And the headers in the build directory are only upda

Re: [PATCH] D12355: [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*)

EricWF updated this revision to Diff 33380. EricWF added a comment. Call `basic_string::compare` as requested by @mclow.lists. http://reviews.llvm.org/D12355 Files: include/string Index: include/string === --- include/string +++

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

echristo added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + hfinkel wrote: > The code below for OPT_mstackrealign uses -mstackrealign as the nam

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

hfinkel added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + echristo wrote: > hfinkel wrote: > > The code below for OPT_mstackrealign uses -mstac

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

echristo added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + hfinkel wrote: > echristo wrote: > > hfinkel wrote: > > > The code below for OPT_mst

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

ahatanak added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + echristo wrote: > hfinkel wrote: > > echristo wrote: > > > hfinkel wrote: > > > > Th

Re: [PATCH] D11740: ABI versioning macros for libc++

eugenis updated this revision to Diff 33381. eugenis added a comment. Remove minor version, added abi_unstable,. Keeping __config_version until the other change lands. Tests use headers from the build directory. http://reviews.llvm.org/D11740 Files: CMakeLists.txt docs/Abi.rst docs/Buildi

Re: [PATCH] D11815: Pass subtarget feature "force-align-stack"

hfinkel added inline comments. Comment at: lib/Driver/Tools.cpp:4232 @@ +4231,3 @@ + false)) +CmdArgs.push_back(Args.MakeArgString("-force-align-stack")); + ahatanak wrote: > echristo wrote: > > hfinkel wrote: > > > echristo wrote: > > > > hf

Re: [PATCH] D11963: Create a __config_site file to capture configuration decisions.

EricWF updated this revision to Diff 33384. EricWF added a comment. Copy the headers to the build directory during every build instead of just during CMake configuration. http://reviews.llvm.org/D11963 Files: CMakeLists.txt cmake/Modules/CopyLibcxxHeaders.cmake cmake/Modules/HandleLibcxx

Re: [PATCH] D12135: [libcxx] Mark most test/std/future tests as UNSUPPORTED in C++03

EricWF added a comment. Gentle ping. @mclow.lists I don't want a full review for this. Just a nod that I'm doing the right thing by not modifying . http://reviews.llvm.org/D12135 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.

Re: [PATCH] D11380: Implement LFTS searchers. Boyer_Moore and Boyer_Moore_Horspool

EricWF added a comment. @mclow.lists I'm holding off on more review until the current comments are addressed. http://reviews.llvm.org/D11380 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-

Re: [PATCH] D12381: [Static Analyzer] Merge the Objective-C Generics Checker into Dynamic Type Propagation checker.

zaks.anna added a comment. Partial review in-line. Comment at: lib/StaticAnalyzer/Checkers/Checkers.td:456 @@ -455,2 +455,3 @@ def ObjCGenericsChecker : Checker<"ObjCGenerics">, HelpText<"Check for incorrect usages of parameterized types.">, + DescFile<"DynamicTypePropagati

Re: [PATCH] D12247: [libc++] remove possible trailing padding from aligned_storage

EricWF added a comment. @mclow.lists I have also tested this patch using libabigail and it couldn't spot any problems. http://reviews.llvm.org/D12247 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/list

[PATCH] D12422: Allow TLS vars in dllimport/export functions; only inline dllimport functions when safe (PR24593)

hans created this revision. hans added a reviewer: majnemer. hans added a subscriber: cfe-commits. This seems to match the MS behaviour. http://reviews.llvm.org/D12422 Files: lib/CodeGen/CodeGenModule.cpp lib/Sema/SemaDecl.cpp test/CodeGenCXX/dllimport.cpp test/SemaCXX/dllexport.cpp te

[libcxx] r246250 - Creating release candidate final from release_370 branch

Author: hans Date: Thu Aug 27 20:43:57 2015 New Revision: 246250 URL: http://llvm.org/viewvc/llvm-project?rev=246250&view=rev Log: Creating release candidate final from release_370 branch Added: libcxx/tags/RELEASE_370/final/ (props changed) - copied from r246249, libcxx/branches/rele

[libcxxabi] r246251 - Creating release candidate final from release_370 branch

Author: hans Date: Thu Aug 27 20:44:00 2015 New Revision: 246251 URL: http://llvm.org/viewvc/llvm-project?rev=246251&view=rev Log: Creating release candidate final from release_370 branch Added: libcxxabi/tags/RELEASE_370/final/ (props changed) - copied from r246250, libcxxabi/branche

[libunwind] r246257 - Creating release candidate final from release_370 branch

Author: hans Date: Thu Aug 27 20:44:18 2015 New Revision: 246257 URL: http://llvm.org/viewvc/llvm-project?rev=246257&view=rev Log: Creating release candidate final from release_370 branch Added: libunwind/tags/RELEASE_370/final/ (props changed) - copied from r246256, libunwind/branche

Re: [PATCH] D12181: [sanitizer] Add -fsanitize-trap-function.

jmgao added a comment. With #1, it seems unfortunate to not be able to distinguish between a sanitize inserted __builtin_trap and code manually calling it. (Would there be an -fsanitize-trap=trap? :-) With #2, we're worried about the generated code being noticeably worse in the unexceptional ca

r246259 - Merge the two feature map setting functions into a single function

Author: echristo Date: Thu Aug 27 21:13:58 2015 New Revision: 246259 URL: http://llvm.org/viewvc/llvm-project?rev=246259&view=rev Log: Merge the two feature map setting functions into a single function and replace all callers. Modified: cfe/trunk/include/clang/Basic/TargetInfo.h cfe/trunk

r246260 - Instead of duplicating code, call the base implementation.

Author: echristo Date: Thu Aug 27 21:14:00 2015 New Revision: 246260 URL: http://llvm.org/viewvc/llvm-project?rev=246260&view=rev Log: Instead of duplicating code, call the base implementation. Modified: cfe/trunk/lib/Basic/Targets.cpp Modified: cfe/trunk/lib/Basic/Targets.cpp URL: http://l

Re: r246229 - [X86] Conditionalize Darwin MaxVectorAlign on the presence of AVX.

Hi Ahmed, A quick note: I think this is going to fail in the presence of the target attribute. I.e. if someone decorates a function with __attribute__((target("avx512"))) this is unlikely to catch that. Also, should we do this for all of the x86 OSes? -eric On Thu, Aug 27, 2015 at 3:31 PM Ahmed

Re: [PATCH] D12355: [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*)

mclow.lists added a comment. The first change LGTM. The second one needs to match it. Comment at: include/string:3816 @@ -3808,2 +3815,3 @@ +return __lhs.compare(0, _String::npos, __rhs, __rhs_len) == 0; } You'll want to do the same as above here. http:

Re: [PATCH] D11380: Implement LFTS searchers. Boyer_Moore and Boyer_Moore_Horspool

mclow.lists added a comment. Two comments answered. One still remaining. Comment at: include/experimental/functional:256 @@ +255,3 @@ + +public: // TODO private: +_RandomAccessIterator1 __first_; EricWF wrote: > Is this for testing? Well, for debuggi

r246263 - PR24597: Fix in-place evaluation of call expressions to provide a proper "this"

Author: rsmith Date: Thu Aug 27 21:43:42 2015 New Revision: 246263 URL: http://llvm.org/viewvc/llvm-project?rev=246263&view=rev Log: PR24597: Fix in-place evaluation of call expressions to provide a proper "this" pointer to an RVO construction of a returned object. Modified: cfe/trunk/lib/AST

[libcxx] r246266 - [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*)

Author: ericwf Date: Thu Aug 27 22:02:37 2015 New Revision: 246266 URL: http://llvm.org/viewvc/llvm-project?rev=246266&view=rev Log: [libcxx] Optimize away unneeded length calculation in basic_string::compare(const char*) Summary: This patch optimizes basic_string::compare to use strcmp when the

Re: [PATCH] D12299: [libcxx] ABI-Breaking Fix for ALL undefined behavior in .

EricWF added a comment. I've spent some time testing this patch to see if it causes ABI problem. Here is what I did: 1. Add external instantiations of std::list into the libc++ dylib for the types used in the tests. 2. Compile `libcxx-old.dylib` with the external instantations but without this

Re: [PATCH] D12135: [libcxx] Mark most test/std/future tests as UNSUPPORTED in C++03

mclow.lists accepted this revision. mclow.lists added a comment. This revision is now accepted and ready to land. This looks very mechanical, except for the changes to test_allocator.h, and that's a bug fix. http://reviews.llvm.org/D12135 ___ cfe-c

[libcxx] r246270 - Fix bug in test_allocator that used the wrong value to represent object state

Author: ericwf Date: Fri Aug 28 00:00:25 2015 New Revision: 246270 URL: http://llvm.org/viewvc/llvm-project?rev=246270&view=rev Log: Fix bug in test_allocator that used the wrong value to represent object state Modified: libcxx/trunk/test/support/test_allocator.h Modified: libcxx/trunk/test

Re: [PATCH] D12173: [libcxx] Constrain unique_ptr::operator=(unique_ptr) in C++03 mode

mclow.lists accepted this revision. mclow.lists added a comment. This revision is now accepted and ready to land. This is just making the two halves of the #ifdef consistent. LGTM. http://reviews.llvm.org/D12173 ___ cfe-commits mailing list cfe-comm

Re: [PATCH] D12135: [libcxx] Mark most test/std/future tests as UNSUPPORTED in C++03

EricWF updated this revision to Diff 33395. EricWF added a comment. Reuploading without test_allocator change. http://reviews.llvm.org/D12135 Files: test/libcxx/thread/futures/version.pass.cpp test/std/thread/futures/futures.promise/alloc_ctor.pass.cpp test/std/thread/futures/futures.prom

[libcxx] r246271 - [libcxx] Mark most test/std/future tests as UNSUPPORTED in C++03

Author: ericwf Date: Fri Aug 28 00:06:04 2015 New Revision: 246271 URL: http://llvm.org/viewvc/llvm-project?rev=246271&view=rev Log: [libcxx] Mark most test/std/future tests as UNSUPPORTED in C++03 Summary: This patch marks *most* tests for `std::promise`, `std::future` and `std::shared_future`

[libcxx] r246272 - [libcxx] Constrain unique_ptr::operator=(unique_ptr) in C++03 mode

Author: ericwf Date: Fri Aug 28 00:07:06 2015 New Revision: 246272 URL: http://llvm.org/viewvc/llvm-project?rev=246272&view=rev Log: [libcxx] Constrain unique_ptr::operator=(unique_ptr) in C++03 mode Summary: This patch properly constrains the converting assignment operator in C++03. It also fix

[libcxx] r246273 - Remove empty file that arcanist created

Author: ericwf Date: Fri Aug 28 00:18:13 2015 New Revision: 246273 URL: http://llvm.org/viewvc/llvm-project?rev=246273&view=rev Log: Remove empty file that arcanist created Removed: libcxx/trunk/test/std/thread/futures/version.pass.cpp Removed: libcxx/trunk/test/std/thread/futures/version.pa

Re: [PATCH] D12312: Emiting invariant.group.barrier and adding -fstrict-vptrs

rjmccall added inline comments. Comment at: lib/CodeGen/CGClass.cpp:1279 @@ +1278,3 @@ + if (CGM.getCodeGenOpts().StrictVPtrs && BaseVPtrsInitialized) +CXXThisValue = Builder.CreateInvariantGroupBarrier(LoadCXXThis()); + Prazek wrote: > rjmccall wrote: > > Pr

[libcxx] r246275 - Finally get the test suite passing in C++03!!

Author: ericwf Date: Fri Aug 28 00:46:17 2015 New Revision: 246275 URL: http://llvm.org/viewvc/llvm-project?rev=246275&view=rev Log: Finally get the test suite passing in C++03!! After months of work there are only 4 tests still failing in C++03. This patch fixes those tests. All of the libc++ b

Re: [PATCH] D12313: Introduce __builtin_nontemporal_store and __builtin_nontemporal_load.

mzolotukhin updated this revision to Diff 33396. mzolotukhin added a comment. Address review remarks: - Remove typed versions - indeed, we don't need them. - Allow vector types. - Properly handle bool-type (promote i1 to i8). - Check arguments number. - Simplify SemaBuiltinNontemporalOverloaded (

Re: [PATCH] D12358: [Analyzer] Handling constant bound loops

zaks.anna added a comment. > I accept that my current patch is not a comprehensive solution to the problem > and that it may introduce > false positives, however I do think it is an > improvement, where it is preferable to have false positives > over doing no analysis after the loop. We try

Re: [PATCH] D12313: Introduce __builtin_nontemporal_store and __builtin_nontemporal_load.

mzolotukhin added a comment. Hi Richard, Hal, and others, I updated the patch according to review remarks - now we support vector and boolean types too! Could you please take a look? Thanks, Michael Comment at: lib/CodeGen/CGBuiltin.cpp:128-129 @@ +127,4 @@ + Val = CGF.EmitT

r246278 - [OPENMP 4.0] Codegen for array sections.

Author: abataev Date: Fri Aug 28 01:09:05 2015 New Revision: 246278 URL: http://llvm.org/viewvc/llvm-project?rev=246278&view=rev Log: [OPENMP 4.0] Codegen for array sections. Added codegen for array section in 'depend' clause of 'task' directive. It emits to pointers, one for the begin of array s

Re: [PATCH] D12313: Introduce __builtin_nontemporal_store and __builtin_nontemporal_load.

hfinkel added a comment. Thanks, but I still have this question: > Plus, I don't understand why you're implementing another place in CodeGen > that generates IR to load and store scalar values. Can't you enhance > EmitLoadOfScalar/EmitStoreOfScalar and then use those functions? If nothing > el

Re: [PATCH] D12313: Introduce __builtin_nontemporal_store and __builtin_nontemporal_load.

mzolotukhin added a comment. > I still have this question: Oops, sorry - I missed that. I'll check how can we reuse that logic. Thanks, Michael http://reviews.llvm.org/D12313 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.ll

Re: [PATCH] D12358: [Analyzer] Handling constant bound loops

krememek added a comment. In http://reviews.llvm.org/D12358#234949, @zaks.anna wrote: > > I accept that my current patch is not a comprehensive solution to the > > problem and that it may introduce > false positives, however I do think it > > is an improvement, where it is preferable to have fa

Re: [PATCH] D12358: [Analyzer] Handling constant bound loops

krememek added a comment. I think the functionality started here by doing something clever with loops is complex enough to warrant putting this into a separate .cpp file. We can keep this here for now, but this seems like complexity that is going to naturally creep up and make this file even m

<    1   2