[PATCH] D27545: Don't assert when redefining a built-in macro in a PCH, PR29119

rnk added a comment. Fix seems reasonable. Comment at: lib/Lex/PPMacroExpansion.cpp:115 + "only built-ins should have an entry here"); +assert(!OldMD->getPrevious() && "builtin should only have a singe entry"); +ED->setPrevious(OldMD); "single

[PATCH] D27545: Don't assert when redefining a built-in macro in a PCH, PR29119

thakis updated this revision to Diff 80687. thakis marked 2 inline comments as done. thakis added a comment. comments https://reviews.llvm.org/D27545 Files: include/clang/Lex/Preprocessor.h lib/Lex/PPMacroExpansion.cpp lib/Serialization/ASTReader.cpp test/PCH/builtin-macro.c Index: tes

[PATCH] D27545: Don't assert when redefining a built-in macro in a PCH, PR29119

thakis added a comment. Thanks! Comment at: test/PCH/builtin-macro.c:29 + +const char s[] = __DATE__ " " __TIME__ " " __TIMESTAMP__; + rnk wrote: > This test doesn't seem to fail if `__DATE__` expands to something. I removed > `-D__DATE__=` from the command li

r289005 - [Driver][Darwin] Disable default stack protector levels in freestanding mode.

Author: bruno Date: Wed Dec 7 18:22:06 2016 New Revision: 289005 URL: http://llvm.org/viewvc/llvm-project?rev=289005&view=rev Log: [Driver][Darwin] Disable default stack protector levels in freestanding mode. Currently -fstack-protector is on by default when using -ffreestanding. Change the defa

[PATCH] D27549: [DebugInfo] Add support for __fp16, float, and double constants.

probinson accepted this revision. probinson added a reviewer: probinson. probinson added a comment. This revision is now accepted and ready to land. LGTM. Comment at: lib/CodeGen/CGDebugInfo.cpp:3765 +InitExpr = + DBuilder.createConstantValueExpression(Init.getFloat

Buildbot numbers for the week of 11/20/2016 - 11/26/2016

Hello everyone, Below are some buildbot numbers for the week of 11/20/2016 - 11/26/2016. Please see the same data in attached csv files: The longest time each builder was red during the last week; "Status change ratio" by active builder (percent of builds that changed the builder status from gre

Buildbot numbers for the week of 11/27/2016 - 12/03/2016

Hello everyone, Below are some buildbot numbers for the last week of 11/27/2016 - 12/03/2016. Please see the same data in attached csv files: The longest time each builder was red during the last week; "Status change ratio" by active builder (percent of builds that changed the builder status fro

[PATCH] D27409: [analyzer] RetainCountChecker: The callback in dispatch_data_create() doesn't free the return symbol.

dcoughlin accepted this revision. dcoughlin added a comment. This revision is now accepted and ready to land. Ship it! :-) https://reviews.llvm.org/D27409 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/

r289011 - Loosen checks for _MSC_FULL_VER under -fms-extensions.

Author: dlj Date: Wed Dec 7 19:11:41 2016 New Revision: 289011 URL: http://llvm.org/viewvc/llvm-project?rev=289011&view=rev Log: Loosen checks for _MSC_FULL_VER under -fms-extensions. Summary: On actual Windows hosts :-) , this could report something other than the fallback, with a non-zero mino

[PATCH] D27555: [libcxx] [test] Fix MSVC warning C6001 "Using uninitialized memory".

STL_MSFT created this revision. STL_MSFT added reviewers: EricWF, mclow.lists. STL_MSFT added a subscriber: cfe-commits. [libcxx] [test] Fix MSVC warning C6001 "Using uninitialized memory". /analyze sees array::size() being called on arrays with garbage-inited doubles, and complains. It doesn't k

[PATCH] D27557: Update the default of the Mozilla coding style

sylvestre.ledru created this revision. sylvestre.ledru added a reviewer: klimek. sylvestre.ledru added a subscriber: cfe-commits. I also proposed the change in Firefox .clang-format file: https://bugzilla.mozilla.org/show_bug.cgi?id=1322321 https://reviews.llvm.org/D27557 Files: lib/Format/Fo

[PATCH] D27545: Don't assert when redefining a built-in macro in a PCH, PR29119

rsmith added inline comments. Comment at: lib/Lex/PPMacroExpansion.cpp:110-112 +// FIXME: shouldIgnoreMacro() in ASTWriter also stops at macros from the +// predefines buffer in module builds. Do we need to splice to those here +// too? If I remember

r289018 - [Headers] Enable #include_next on Darwin

Author: bruno Date: Wed Dec 7 20:13:56 2016 New Revision: 289018 URL: http://llvm.org/viewvc/llvm-project?rev=289018&view=rev Log: [Headers] Enable #include_next on Darwin Allows darwin targets to provide additional definitions and implementation specifc values for float.h rdar://problem/219614

[PATCH] D22296: CodeGen: New vtable group representation: struct of vtable arrays.

pcc updated this revision to Diff 80709. pcc added a comment. Refresh https://reviews.llvm.org/D22296 Files: clang/include/clang/AST/VTableBuilder.h clang/lib/AST/VTableBuilder.cpp clang/lib/CodeGen/CGCXX.cpp clang/lib/CodeGen/CGVTT.cpp clang/lib/CodeGen/CGVTables.cpp clang/lib/Code

[PATCH] D22296: CodeGen: New vtable group representation: struct of vtable arrays.

pcc added a reviewer: rjmccall. pcc added a comment. John, may I ask you to take a look? https://reviews.llvm.org/D22296 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D24431: CodeGen: Start using inrange annotations on vtable getelementptr.

pcc added a reviewer: rjmccall. pcc added a comment. John, may I ask you to take a look? https://reviews.llvm.org/D24431 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r289019 - [c++1z] P0003R5: Removing dynamic exception specifications.

Author: rsmith Date: Wed Dec 7 20:49:07 2016 New Revision: 289019 URL: http://llvm.org/viewvc/llvm-project?rev=289019&view=rev Log: [c++1z] P0003R5: Removing dynamic exception specifications. We continue to support dynamic exception specifications in C++1z as an extension, but produce an error-b

r289021 - [c++1z] P0490R0, NB comment GB 20: if std::tuple_size is complete, use the

Author: rsmith Date: Wed Dec 7 21:24:55 2016 New Revision: 289021 URL: http://llvm.org/viewvc/llvm-project?rev=289021&view=rev Log: [c++1z] P0490R0, NB comment GB 20: if std::tuple_size is complete, use the tuple-like interpretation of decomposition declaration even if there is no ::value member.

[PATCH] D27535: [analyzer] Add ObjCPropertyChecker - check for autosynthesized copy-properties of mutable types.

dcoughlin added a comment. This looks great to me other than the idiom I mentioned inline (which is common, as I have found to my chagrin). Comment at: lib/StaticAnalyzer/Checkers/ObjCPropertyChecker.cpp:12 +// Currently finds only one kind of issue: +// - Find autosynthesiz

[PATCH] D25417: [libcxxabi] Fix alignment of allocated exceptions in 32 bit builds

EricWF added a comment. I ran into some test failures on OS X while attempting to commit this. The `unwind.h` header provided on OS X hasn't upstreamed the libunwind fix from r276215 , meaning `__cxa_exception` is still under-aligned and therefore so is the exc

[PATCH] D13134: [analyzer] Add keyboard shortcuts to report.html

dcoughlin commandeered this revision. dcoughlin edited reviewers, added: skomski; removed: dcoughlin. dcoughlin added a comment. Commandeering and closing this revision as it has been more than a year. Repository: rL LLVM https://reviews.llvm.org/D13134

[PATCH] D27564: [libcxx] Fix __compressed_pair so it doesn't copy the argument multiple times

EricWF created this revision. EricWF added a reviewer: mclow.lists. EricWF added subscribers: cfe-commits, loladiro. __compressed_pair takes and passes it's constructor arguments by value. This causes arguments to be moved 3 times instead of once. This patch addresses that issue and fixes `const

r288892 - Compilation database test: don't try to output to CWD

Author: sammccall Date: Wed Dec 7 03:19:07 2016 New Revision: 288892 URL: http://llvm.org/viewvc/llvm-project?rev=288892&view=rev Log: Compilation database test: don't try to output to CWD Summary: Write output from compilation database test to %T rather than the working dir. Sometimes CWD isn't

[PATCH] D27564: [libcxx] Fix __compressed_pair so it doesn't copy the argument multiple times

EricWF added a comment. Ideally I would like to re-write `__compressed_pair` so that it doesn't need 4 complete specializations but I'm not sure the change would be non-ABI breaking. https://reviews.llvm.org/D27564 ___ cfe-commits mailing list cfe-

r289022 - [DOXYGEN] Improved doxygen comments for avxintrin.h intrinsics.

Author: kromanova Date: Wed Dec 7 22:09:17 2016 New Revision: 289022 URL: http://llvm.org/viewvc/llvm-project?rev=289022&view=rev Log: [DOXYGEN] Improved doxygen comments for avxintrin.h intrinsics. Tagged parameter names with \a doxygen command to display them in italics. Formatted comments to

[PATCH] D27565: [libcxx] Alternative fix for compressed pair

EricWF created this revision. EricWF added a reviewer: mclow.lists. EricWF added a subscriber: cfe-commits. I would rather have this fix than https://reviews.llvm.org/D27564, but I'm not yet sure it isn't ABI breaking. https://reviews.llvm.org/D27565 Files: include/__hash_table include/mem

[PATCH] D27565: [libcxx] Alternative fix for compressed pair

EricWF updated this revision to Diff 80715. EricWF added a comment. Remove ever more dead code. https://reviews.llvm.org/D27565 Files: include/__hash_table include/memory include/string test/std/utilities/memory/unique.ptr/unique.ptr.runtime/unique.ptr.runtime.ctor/default01.fail.cpp

[PATCH] D22296: CodeGen: New vtable group representation: struct of vtable arrays.

rjmccall added a comment. This seems reasonable to me. Just a few representational / API requests. Comment at: clang/include/clang/AST/VTableBuilder.h:222 + typedef llvm::DenseMap> + AddressPointsMapTy; Please use a struct instead of std::pair. =

[PATCH] D24431: CodeGen: Start using inrange annotations on vtable getelementptr.

rjmccall accepted this revision. rjmccall added a comment. This revision is now accepted and ready to land. I'm not sure I like this IR design, but this use of it seems fine. LGTM. https://reviews.llvm.org/D24431 ___ cfe-commits mailing list cfe-co

[PATCH] D27565: [libcxx] Alternative fix for compressed pair

EricWF added a comment. Grr... OK so this is currently ABI breaking because it re-arranges the layout of `__compressed_pair` when `T2` is empty but `T1` is not. I might be able to work around this with some metaprogramming. https://reviews.llvm.org/D27565 ___

[PATCH] D27566: Fix PR30323: numeric_limits::max_digits10 when using 16 bit ints.

EricWF created this revision. EricWF added a reviewer: mclow.lists. EricWF added a subscriber: cfe-commits. Also see https://llvm.org/bugs/show_bug.cgi?id=30323 https://reviews.llvm.org/D27566 Files: include/limits Index: include/limits ==

[PATCH] D27564: [libcxx] Fix __compressed_pair so it doesn't copy the argument multiple times

loladiro added a comment. I take it this supersedes https://reviews.llvm.org/D24372? I apologize for not getting around to commiting that yet, but if it does supersede that revision, we should probably keep the tests that we have there. https://reviews.llvm.org/D27564 __

[PATCH] D27429: [Chrono][Darwin] On Darwin use CLOCK_UPTIME_RAW instead of CLOCK_MONOTONIC

EricWF added a comment. I'm not sure `CLOCK_UPTIME_RAW` meets the requirements of `steady_clock`. The manpage for `clock_gettime` on OS X specifies `CLOCK_UPTIME_RAW` as: > CLOCK_UPTIME_RAW clock that increments monotonically, in the same manner as > CLOCK_MONOTONIC_RAW, but that does not in

[libcxx] r289028 - Fix _LIBCPP_VERSION tests with modules on Darwin

Author: ericwf Date: Thu Dec 8 00:37:41 2016 New Revision: 289028 URL: http://llvm.org/viewvc/llvm-project?rev=289028&view=rev Log: Fix _LIBCPP_VERSION tests with modules on Darwin Modified: libcxx/trunk/include/module.modulemap Modified: libcxx/trunk/include/module.modulemap URL: http://l

[PATCH] D27540: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 3/7.

EricWF added a comment. MSVC seriously emits a warning for `void foo(short); foo(0);` because the literal `0` is an int? If so you really should fix that in MSVC; That's a bogus warning. Does it emit a warning for `short x = 3;`? https://reviews.llvm.org/D27540 _

[PATCH] D27540: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 3/7.

EricWF added a comment. > MSVC seriously emits a warning for void foo(short); foo(0); because the > literal 0 is an int? Oh my goodness it does... Your compiler is bad and it should feel bad. It's not like you can write `foo(0s)`. https://reviews.llvm.org/D27540 ___

[PATCH] D27542: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 5/7.

EricWF accepted this revision. EricWF added a comment. This revision is now accepted and ready to land. LGTM. https://reviews.llvm.org/D27542 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-

[PATCH] D27543: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 6/7.

EricWF added a comment. LGTM. https://reviews.llvm.org/D27543 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D27123: Add AVR target and toolchain to Clang

saaadhu updated this revision to Diff 80721. saaadhu added a comment. Make defines for CHAR16_TYPE, {U,}INT_{LEAST,FAST}16_TYPE use int instead of short. {U,}INT16_TYPE still gets defined as short though - lib/Frontend/InitPreprocessor.cpp::DefineExactWidthIntType does not use TargetInfo::getI

[PATCH] D27544: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 7/7.

EricWF accepted this revision. EricWF added a comment. This revision is now accepted and ready to land. LGTM. Comment at: test/std/re/re.traits/value.pass.cpp:119 } for (int c = 'g'; c < 0x; ++c) { Can't we just make `c` a `wchar_

[PATCH] D27541: [libcxx] [test] Fix MSVC warning C4244 "conversion from 'X' to 'Y', possible loss of data", part 4/7.

EricWF added a comment. Waiting for a response on https://reviews.llvm.org/D27540 before reviewing this. https://reviews.llvm.org/D27541 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commi

[PATCH] D27555: [libcxx] [test] Fix MSVC warning C6001 "Using uninitialized memory".

EricWF added a comment. Are you planning on fixing this in your compiler? https://reviews.llvm.org/D27555 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libcxx] r289029 - Fix PR30323: numeric_limits::max_digits10 when using 16 bit ints.

Author: ericwf Date: Thu Dec 8 01:30:01 2016 New Revision: 289029 URL: http://llvm.org/viewvc/llvm-project?rev=289029&view=rev Log: Fix PR30323: numeric_limits::max_digits10 when using 16 bit ints. Summary: Also see https://llvm.org/bugs/show_bug.cgi?id=30323 Reviewers: mclow.lists Subscribers

[PATCH] D27566: Fix PR30323: numeric_limits::max_digits10 when using 16 bit ints.

This revision was automatically updated to reflect the committed changes. Closed by commit rL289029: Fix PR30323: numeric_limits::max_digits10 when using 16 bit ints. (authored by EricWF). Changed prior to commit: https://reviews.llvm.org/D27566?vs=80716&id=80722#toc Repository: rL LLVM htt

<    1   2