[PATCH] D23752: [cmake] Respect LLVM_RUNTIMES_LIBDIR_SUFFIX

mgorny abandoned this revision. mgorny added a comment. Very well. I'll look into (ab)using the existing options to achieve the same result. Having stand-alone build systems, this should become easier these days, at least until some major change in runtime handling occurs. https://reviews.llvm

r285386 - Sema: do not warn about unused const vars if main file is a header

Author: erikjv Date: Fri Oct 28 03:28:42 2016 New Revision: 285386 URL: http://llvm.org/viewvc/llvm-project?rev=285386&view=rev Log: Sema: do not warn about unused const vars if main file is a header If we pass a header to libclang, e.g. because it's open in an editor in an IDE, warnings about un

[PATCH] D26067: [openmp] Remove test assumption that invoked binary name contains "clang"

sammccall created this revision. sammccall added a reviewer: bkramer. sammccall added a subscriber: cfe-commits. Remove test assumption that driver-invoked binary name contains "clang" https://reviews.llvm.org/D26067 Files: test/Driver/openmp-offload.c Index: test/Driver/openmp-offload.c ===

[PATCH] D26067: [openmp] Remove test assumption that invoked binary name contains "clang"

sammccall updated this revision to Diff 76172. sammccall added a comment. Use -no-canonical-prefixes so that the invoked binary name will contain "clang" https://reviews.llvm.org/D26067 Files: test/Driver/openmp-offload.c Index: test/Driver/openmp-offload.c =

Re: [libcxx] r285382 - Add __libcpp_version file and __libcpp_library_version function.

Hi, On Fri, Oct 28, 2016 at 9:06 AM, Eric Fiselier via cfe-commits wrote: > Author: ericwf > Date: Fri Oct 28 01:06:50 2016 > New Revision: 285382 > > URL: http://llvm.org/viewvc/llvm-project?rev=285382&view=rev > Log: > Add __libcpp_version file and __libcpp_library_version function. > > This pa

r285388 - [openmp] Remove test assumption that canonical binary name contains "clang"

Author: d0k Date: Fri Oct 28 04:20:02 2016 New Revision: 285388 URL: http://llvm.org/viewvc/llvm-project?rev=285388&view=rev Log: [openmp] Remove test assumption that canonical binary name contains "clang" Patch by Sam McCall! Differential Revision: https://reviews.llvm.org/D26067 Modified:

[PATCH] D26067: [openmp] Remove test assumption that canonical binary name contains "clang"

This revision was automatically updated to reflect the committed changes. Closed by commit rL285388: [openmp] Remove test assumption that canonical binary name contains "clang" (authored by d0k). Changed prior to commit: https://reviews.llvm.org/D26067?vs=76172&id=76174#toc Repository: rL LL

r285390 - Fix MSVC "not all control paths return a value" warning

Author: rksimon Date: Fri Oct 28 05:09:35 2016 New Revision: 285390 URL: http://llvm.org/viewvc/llvm-project?rev=285390&view=rev Log: Fix MSVC "not all control paths return a value" warning Add unreachable after enum switch statement Modified: cfe/trunk/lib/Driver/Action.cpp Modified: cfe/t

r285391 - [Objective-C] Add objc_subclassing_restricted attribute

Author: arphaman Date: Fri Oct 28 05:25:10 2016 New Revision: 285391 URL: http://llvm.org/viewvc/llvm-project?rev=285391&view=rev Log: [Objective-C] Add objc_subclassing_restricted attribute This patch adds an objc_subclassing_restricted attribute into clang. This attribute acts similarly to 'fin

[PATCH] D25993: [Objective-C] Add objc_subclassing_restricted attribute

This revision was automatically updated to reflect the committed changes. Closed by commit rL285391: [Objective-C] Add objc_subclassing_restricted attribute (authored by arphaman). Changed prior to commit: https://reviews.llvm.org/D25993?vs=76019&id=76175#toc Repository: rL LLVM https://rev

[PATCH] D25942: Fix crash in implicit default constructor creation for a template record specialization that has a field declaration with an initializer expression

arphaman abandoned this revision. arphaman added a comment. Abandoning since I discovered that another patch (https://reviews.llvm.org/D13419) that's currently in review fixes this issue. Repository: rL LLVM https://reviews.llvm.org/D25942 ___ c

[PATCH] D13419: Fix several problems at the intersection of template instantiations and visibility

arphaman added a comment. I just noticed that this patch fixes a template crash that I have a patch for at https://reviews.llvm.org/D25942. I will abandon my patch since this patch fixes the other issue. Would you mind adding the test case from my patch to this patch? It's available at https:/

RE: [PATCH] D25343: [OpenCL] Mark group functions as convergent in opencl-c.h

Hi Ettore, Thanks for the example. Up to now we used noduplicate to prevent this erroneous optimisation but using convergent instead would be equally good. And as it's pointed out it is less restrictive to allow more optimisations in LLVM i.e. loop unrolling with convergent operation in it. I

[PATCH] D13419: Fix several problems at the intersection of template instantiations and visibility

arphaman added a comment. In https://reviews.llvm.org/D13419#581356, @loladiro wrote: > Hmm, the rebased version of this introduces new errors in > `test/Modules/cxx-templates.cpp`. That test didn't exist when I wrote this > code and I'm not familiar with templates. @rsmith could you take a loo

[PATCH] D9888: [OPENMP] Driver support for OpenMP offloading

sfantao abandoned this revision. sfantao marked 8 inline comments as done. sfantao added a comment. Hi Jonas, In https://reviews.llvm.org/D9888#581809, @Hahnfeld wrote: > I think these changes have been contributed to trunk in multiple commits so > this can be closed? You're right, this can b

Re: [libcxx] r285382 - Add __libcpp_version file and __libcpp_library_version function.

It appears you're testing fresh headers against a system installed libc++ library. Can you confirm this was your intention? Either way I'll correct the test to handle this. /Eric On Fri, Oct 28, 2016 at 3:22 AM, Ismail Donmez wrote: > Hi, > > On Fri, Oct 28, 2016 at 9:06 AM, Eric Fiselier via

[libcxx] r285392 - Fix test when using an installed libc++

Author: ericwf Date: Fri Oct 28 06:01:12 2016 New Revision: 285392 URL: http://llvm.org/viewvc/llvm-project?rev=285392&view=rev Log: Fix test when using an installed libc++ Modified: libcxx/trunk/test/libcxx/test/config.py libcxx/trunk/test/libcxx/version.pass.cpp Modified: libcxx/trunk/

[PATCH] D25008: [cmake] Split linked libraries into private & public, for linker script

Hahnfeld added a comment. What are the possibilities to proceed here? Repository: rL LLVM https://reviews.llvm.org/D25008 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D25008: [cmake] Split linked libraries into private & public, for linker script

mgorny added a comment. In https://reviews.llvm.org/D25008#581889, @Hahnfeld wrote: > What are the possibilities to proceed here? To be honest, I don't have any good idea besides renaming the library. Keeping two non-interchangeable libraries under the same name is bound to fail. Repository:

[PATCH] D26071: [CodeCompletion] Show block invocation result for block property setters

arphaman created this revision. arphaman added reviewers: manmanren, akyrtzi. arphaman added a subscriber: cfe-commits. arphaman set the repository for this revision to rL LLVM. This patch changes the code completion results for block property setters. Right now we code complete block property se

r285395 - [OpenCL] Diagnose variadic arguments

Author: stulova Date: Fri Oct 28 07:59:39 2016 New Revision: 285395 URL: http://llvm.org/viewvc/llvm-project?rev=285395&view=rev Log: [OpenCL] Diagnose variadic arguments OpenCL disallows using variadic arguments (s6.9.e and s6.12.5 OpenCL v2.0) apart from some exceptions: - printf - enqueue_kern

[clang-tools-extra] r285396 - [include-fixer] Make error message sound less like clang crashed.

Author: d0k Date: Fri Oct 28 08:00:49 2016 New Revision: 285396 URL: http://llvm.org/viewvc/llvm-project?rev=285396&view=rev Log: [include-fixer] Make error message sound less like clang crashed. We suppress all Clang diagnostics (because they would be wrong, include-fixer does custom recovery) b

[PATCH] D25153: preprocessor supports `-dI` flag

elsteveogrande added a comment. Ping, can anyone help with committing this? It's accepted, just needs to be landed. (I don't have commit access.) Thanks! https://reviews.llvm.org/D25153 ___ cfe-commits mailing list cfe-commits@lists.llvm.org htt

[PATCH] D25990: Sema: do not warn about unused const vars if main file is a header

erikjv closed this revision. erikjv added a comment. Committed as r285386. https://reviews.llvm.org/D25990 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D26073: [PPC] Add vec_absd functions to altivec.h

sfertile created this revision. sfertile added reviewers: kbarton, amehsan, lei, jtony, syzaara, nemanjai, echristo. sfertile added a subscriber: cfe-commits. Herald added a subscriber: mehdi_amini. Adds three overloads of vec_absd to altivec .h, as well as matching PPC specific builtins: vect

[PATCH] D22346: [Clang-tidy] CERT-MSC50-CPP (std:rand() )

aaron.ballman added a comment. In https://reviews.llvm.org/D22346#581329, @falho wrote: > Cool! Thank you for the reviews! If you don't have commit privileges, let me know and I'm happy to commit on your behalf. https://reviews.llvm.org/D22346 _

[PATCH] D24799: [XRay] Check in Clang whether XRay supports the target when -fxray-instrument is passed

rSerge added a comment. @dberris , I've checked the .diff file which I was uploading to Phabricator, and it doesn't seem to contain trailing whitespace. Please, see the screenshot: F2536547: Whitespace.jpg So it seems that Phabricator or your downloading tool a

Re: [libcxx] r285382 - Add __libcpp_version file and __libcpp_library_version function.

This breaks code which is using module headers, and then does like this: extern "C" { #include } because: > In file included from include/c++/v1/stdint.h:102: > include/c++/v1/__config:912:35: error: declaration of > '__libcpp_library_version' has a different language linkage > _LIBCPP_FUNC_VIS

[libcxx] r285403 - Explicitly specify extern "C++" on __libcpp_library_version

Author: ericwf Date: Fri Oct 28 10:02:30 2016 New Revision: 285403 URL: http://llvm.org/viewvc/llvm-project?rev=285403&view=rev Log: Explicitly specify extern "C++" on __libcpp_library_version Modified: libcxx/trunk/include/__config Modified: libcxx/trunk/include/__config URL: http://llvm.

[PATCH] D26075: Change from "XFAIL: libcpp-no-exceptions" to "UNSUPPORTED: libcpp-no-exceptions" tests that only check exceptions and nothing else

rogfer01 created this revision. rogfer01 added reviewers: rmaprath, EricWF, mclow.lists. rogfer01 added a subscriber: cfe-commits. This is a follow up of https://reviews.llvm.org/D24562. These tests do not check anything but exceptions, so it makes sense to mark them as UNSUPPORTED under a libra

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

hiraditya updated this revision to Diff 76192. hiraditya added a comment. Addressed Sebastian's comments. https://reviews.llvm.org/D25624 Files: libcxx/include/string libcxx/src/string.cpp Index: libcxx/src/string.cpp === ---

[PATCH] D21845: [Driver][OpenMP] Add specialized action builder for OpenMP offloading actions.

mkuron added a comment. I think `OffloadAction::DeviceDependences::add(..., ..., /*BoundArch=*/nullptr, Action::OFK_OpenMP)` is never sufficient. The invalid `BoundArch` eventually ends up in `NVPTX::Assembler::ConstructJob` and triggers an assert; I don't think there is any code path with Open

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

EricWF added a comment. Additionally _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY is documented here http://libcxx.llvm.org/docs/DesignDocs/VisibilityMacros.html. https://reviews.llvm.org/D25624 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

r285405 - 1. Fixing small types issue (PD|PS) (reduce) .

Author: mzuckerm Date: Fri Oct 28 10:16:03 2016 New Revision: 285405 URL: http://llvm.org/viewvc/llvm-project?rev=285405&view=rev Log: 1. Fixing small types issue (PD|PS) (reduce) . 2. Cosmetic changes Modified: cfe/trunk/lib/Headers/avx512fintrin.h cfe/trunk/test/CodeGen/avx512-reduceIn

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

EricWF added a comment. Actually we don't want to copy `memory.cpp` in this case. Please use `_LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY`. There are only examples in `` you can follow. Make sure to put it on the in-class declaration. https://reviews.llvm.org/D25624 __

r285404 - Change OpenMP offload driver tests so that it doesn't use the full file path during tests.

Author: sfantao Date: Fri Oct 28 10:11:50 2016 New Revision: 285404 URL: http://llvm.org/viewvc/llvm-project?rev=285404&view=rev Log: Change OpenMP offload driver tests so that it doesn't use the full file path during tests. This was causing failures on windows bots. Modified: cfe/trunk/te

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

sebpop added a comment. Other than removing that comment the patch looks good. Thanks! Comment at: libcxx/src/string.cpp:10 +// For keeping the definition of ~basic_string in this translation unit. +#define _LIBCPP_BUILDING_STRING Let's not add this comment h

[PATCH] D21845: [Driver][OpenMP] Add specialized action builder for OpenMP offloading actions.

sfantao added a comment. Hi Michael, In https://reviews.llvm.org/D21845#581988, @mkuron wrote: > I think `OffloadAction::DeviceDependences::add(..., ..., > /*BoundArch=*/nullptr, Action::OFK_OpenMP)` is never sufficient. The invalid > `BoundArch` eventually ends up in `NVPTX::Assembler::Constr

[PATCH] D26082: Incomplete support for Python 3 in libclang python bindings

jbcoe created this revision. jbcoe added reviewers: eliben, compnerd, nemanjai, skalinichev. jbcoe added a subscriber: cfe-commits. jbcoe set the repository for this revision to rL LLVM. This is incomplete and I'm in need of some input. Some test pass in Python 3 now. Python 2 tests pass as befor

r285408 - Define extra variable in OpenMP offloading driver tests.

Author: sfantao Date: Fri Oct 28 10:42:38 2016 New Revision: 285408 URL: http://llvm.org/viewvc/llvm-project?rev=285408&view=rev Log: Define extra variable in OpenMP offloading driver tests. Modified: cfe/trunk/test/Driver/openmp-offload.c Modified: cfe/trunk/test/Driver/openmp-offload.c URL

[PATCH] D23657: Remove some false positives when taking the address of packed members

rogfer01 added a comment. Ping? https://reviews.llvm.org/D23657 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r285410 - [CUDA] [AST] Allow isInlineDefinitionExternallyVisible to be called on functions without bodies.

Author: jlebar Date: Fri Oct 28 11:26:26 2016 New Revision: 285410 URL: http://llvm.org/viewvc/llvm-project?rev=285410&view=rev Log: [CUDA] [AST] Allow isInlineDefinitionExternallyVisible to be called on functions without bodies. Summary: In CUDA compilation, we call isInlineDefinitionExternally

[PATCH] D25640: [CUDA] [AST] Allow isInlineDefinitionExternallyVisible to be called on functions without bodies.

This revision was automatically updated to reflect the committed changes. Closed by commit rL285410: [CUDA] [AST] Allow isInlineDefinitionExternallyVisible to be called on… (authored by jlebar). Changed prior to commit: https://reviews.llvm.org/D25640?vs=76118&id=76208#toc Repository: rL LLV

r285411 - [Preprocessor] Support for '-dI' flag

Author: bruno Date: Fri Oct 28 11:32:10 2016 New Revision: 285411 URL: http://llvm.org/viewvc/llvm-project?rev=285411&view=rev Log: [Preprocessor] Support for '-dI' flag Implement the -dI as supported by GCC: Output ‘#include’ directives in addition to the result of preprocessing. This change ai

[PATCH] D25153: preprocessor supports `-dI` flag

This revision was automatically updated to reflect the committed changes. Closed by commit rL285411: [Preprocessor] Support for '-dI' flag (authored by bruno). Changed prior to commit: https://reviews.llvm.org/D25153?vs=75351&id=76210#toc Repository: rL LLVM https://reviews.llvm.org/D25153

[PATCH] D26073: [PPC] Add vec_absd functions to altivec.h

nemanjai added a comment. This patch LGTM. I'll let @kbarton/@echristo have a look as well and have the final say. https://reviews.llvm.org/D26073 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinf

r285412 - Relax assertion in FunctionDecl::doesDeclarationForceExternallyVisibleDefinition.

Author: jlebar Date: Fri Oct 28 11:46:39 2016 New Revision: 285412 URL: http://llvm.org/viewvc/llvm-project?rev=285412&view=rev Log: Relax assertion in FunctionDecl::doesDeclarationForceExternallyVisibleDefinition. Previously we were asserting that this declaration doesn't have a body *and* won'

r285416 - Revert "[Preprocessor] Support for '-dI' flag"

Author: bruno Date: Fri Oct 28 12:02:10 2016 New Revision: 285416 URL: http://llvm.org/viewvc/llvm-project?rev=285416&view=rev Log: Revert "[Preprocessor] Support for '-dI' flag" This reverts r285411. Tests failing on http://lab.llvm.org:8011/builders/clang-x86-windows-msvc2015/builds/141 Remove

[PATCH] D25153: preprocessor supports `-dI` flag

bruno added a comment. The test is failing on windows: http://lab.llvm.org:8011/builders/clang-x86-windows-msvc2015/builds/141 I reverted the patch for now in r285416, can you take a look? Repository: rL LLVM https://reviews.llvm.org/D25153 ___

[PATCH] D26054: Use `getFileLoc()` instead of `getSpellingLoc()` in the ASTImporter

spyffe updated this revision to Diff 76213. spyffe added a comment. Updated the corresponding comment. https://reviews.llvm.org/D26054 Files: lib/AST/ASTImporter.cpp test/ASTMerge/Inputs/macro.modulemap test/ASTMerge/Inputs/macro1.h test/ASTMerge/Inputs/macro1.m test/ASTMerge/Inputs/m

r285419 - Fixing small problem with avx512-reduceIntrin.c test on some OS.

Author: mzuckerm Date: Fri Oct 28 12:25:26 2016 New Revision: 285419 URL: http://llvm.org/viewvc/llvm-project?rev=285419&view=rev Log: Fixing small problem with avx512-reduceIntrin.c test on some OS. Modified: cfe/trunk/test/CodeGen/avx512-reduceIntrin.c Modified: cfe/trunk/test/CodeGen/avx

[PATCH] D25153: preprocessor supports `-dI` flag

elsteveogrande added inline comments. Comment at: cfe/trunk/test/Preprocessor/dump_include.c:2 +// RUN: %clang_cc1 -w -E -dI -isystem %S -imacros %S/dump_include.h %s -o - | FileCheck %s +// CHECK: {{^}}#__include_macros "/{{([^/]+/)+}}dump_ +// CHECK: {{^}}#include https://revi

[PATCH] D26054: Use `getFileLoc()` instead of `getSpellingLoc()` in the ASTImporter

akyrtzi added a comment. LGTM. https://reviews.llvm.org/D26054 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libcxx] r285427 - Try 2 to add extern C++ to __libcpp_library_version

Author: ericwf Date: Fri Oct 28 13:26:06 2016 New Revision: 285427 URL: http://llvm.org/viewvc/llvm-project?rev=285427&view=rev Log: Try 2 to add extern C++ to __libcpp_library_version Modified: libcxx/trunk/include/__config Modified: libcxx/trunk/include/__config URL: http://llvm.org/viewv

r285428 - Add missing newline at EOF to avoid -Wnewline-eof warnings.

Author: bwilson Date: Fri Oct 28 13:55:50 2016 New Revision: 285428 URL: http://llvm.org/viewvc/llvm-project?rev=285428&view=rev Log: Add missing newline at EOF to avoid -Wnewline-eof warnings. Modified: cfe/trunk/include/clang/Basic/OpenCLImageTypes.def Modified: cfe/trunk/include/clang/Bas

LLVM buildmaster will restarted soon

Hello everyone, LLVM buildmaster will restarted at 1 PM Pacific time today. Thank you for understanding. Thanks Galina ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r285430 - Fix implementation of the likely resolution of core issue 253 to support class

Author: rsmith Date: Fri Oct 28 14:11:18 2016 New Revision: 285430 URL: http://llvm.org/viewvc/llvm-project?rev=285430&view=rev Log: Fix implementation of the likely resolution of core issue 253 to support class based arrays. Patch by Ian Tessier! Differential Review: https://reviews.llvm.org/D25

[PATCH] D25974: Fix implementation of the likely resolution of core issue 253 to support class based arrays.

rsmith closed this revision. rsmith added a comment. Committed as https://reviews.llvm.org/rL285430. https://reviews.llvm.org/D25974 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r285436 - Implement vector count leading/trailing bytes with zero lsb and vector parity

Author: nemanjai Date: Fri Oct 28 14:49:03 2016 New Revision: 285436 URL: http://llvm.org/viewvc/llvm-project?rev=285436&view=rev Log: Implement vector count leading/trailing bytes with zero lsb and vector parity builtins - clang portion This patch corresponds to review: https://reviews.llvm.org/

[PATCH] D26002: Implement vector count leading/trailing bytes with zero lsb and vector parity builtins - clang portion

nemanjai added a comment. Committed revision 285436. Zaara, please close this review if there are no buildbot failures. https://reviews.llvm.org/D26002 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/lis

r285437 - PR30831: Teach template type diffing to cope with TemplateSpecializationTypes

Author: rsmith Date: Fri Oct 28 14:54:43 2016 New Revision: 285437 URL: http://llvm.org/viewvc/llvm-project?rev=285437&view=rev Log: PR30831: Teach template type diffing to cope with TemplateSpecializationTypes that desugar to non-TSTs (such as injected-class-names). Modified: cfe/trunk/lib/A

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

hiraditya updated this revision to Diff 76240. https://reviews.llvm.org/D25624 Files: libcxx/include/string Index: libcxx/include/string === --- libcxx/include/string +++ libcxx/include/string @@ -1834,6 +1834,7 @@ #endif // _L

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

EricWF accepted this revision. EricWF added inline comments. Comment at: libcxx/include/string:1837 template +inline _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY basic_string<_CharT, _Traits, _Allocator>::~basic_string() The attribute should appear on the first d

r285439 - [PPC] add float and double overloads for vec_orc and vec_nand in altivec.h

Author: nemanjai Date: Fri Oct 28 15:04:53 2016 New Revision: 285439 URL: http://llvm.org/viewvc/llvm-project?rev=285439&view=rev Log: [PPC] add float and double overloads for vec_orc and vec_nand in altivec.h This patch corresponds to review https://reviews.llvm.org/D25950. Committing on behalf

[libcxx] r285445 - Fix Clang 3.6 build error

Author: ericwf Date: Fri Oct 28 15:19:36 2016 New Revision: 285445 URL: http://llvm.org/viewvc/llvm-project?rev=285445&view=rev Log: Fix Clang 3.6 build error Modified: libcxx/trunk/src/experimental/filesystem/path.cpp Modified: libcxx/trunk/src/experimental/filesystem/path.cpp URL: http://

r285446 - Fix handling of constructor inherited through multiple levels of virtual base class.

Author: rsmith Date: Fri Oct 28 15:20:58 2016 New Revision: 285446 URL: http://llvm.org/viewvc/llvm-project?rev=285446&view=rev Log: Fix handling of constructor inherited through multiple levels of virtual base class. Modified: cfe/trunk/include/clang/AST/DeclCXX.h cfe/trunk/test/CXX/spe

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

hiraditya added inline comments. Comment at: libcxx/include/string:1837 template +inline _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY basic_string<_CharT, _Traits, _Allocator>::~basic_string() EricWF wrote: > The attribute should appear on the first declaration n

[PATCH] D13330: Implement __attribute__((unique_instantiation))

loladiro added a comment. I meant `template __attribute__((unique_instantiation)) int * static_separate_template::a_static_field;` of course, though we probably need a better diagnostic for the other spelling (which applies the attribute to the static_separate_template). I'll look into adding

[libcxx] r285456 - Added 'inline' attribute to basic_string's destructor

Author: hiraditya Date: Fri Oct 28 16:27:24 2016 New Revision: 285456 URL: http://llvm.org/viewvc/llvm-project?rev=285456&view=rev Log: Added 'inline' attribute to basic_string's destructor Author: laxmansole Reviewers: howard.hinnant mclow.lists Subscribers: EricWF, flyingforyou, eva

r285457 - [Sema] Delay partial availability diagnostics, just like deprecated

Author: epilk Date: Fri Oct 28 16:39:27 2016 New Revision: 285457 URL: http://llvm.org/viewvc/llvm-project?rev=285457&view=rev Log: [Sema] Delay partial availability diagnostics, just like deprecated This is done so that the following compiles with no warnings: int fn(type_10_12) __attribute__((a

[PATCH] D25284: AvailabilityAttrs: Delay partial availability diagnostics

This revision was automatically updated to reflect the committed changes. Closed by commit rL285457: [Sema] Delay partial availability diagnostics, just like deprecated (authored by epilk). Changed prior to commit: https://reviews.llvm.org/D25284?vs=74738&id=76250#toc Repository: rL LLVM ht

r285458 - Make a function static. NFC.

Author: gbiv Date: Fri Oct 28 16:42:06 2016 New Revision: 285458 URL: http://llvm.org/viewvc/llvm-project?rev=285458&view=rev Log: Make a function static. NFC. Modified: cfe/trunk/lib/Sema/SemaOverload.cpp Modified: cfe/trunk/lib/Sema/SemaOverload.cpp URL: http://llvm.org/viewvc/llvm-projec

[PATCH] D13330: Implement __attribute__((unique_instantiation))

loladiro updated this revision to Diff 76256. loladiro added a comment. Fix for the corner case I found and add it as a test. Repository: rL LLVM https://reviews.llvm.org/D13330 Files: include/clang/Basic/Attr.td include/clang/Basic/AttrDocs.td include/clang/Basic/DiagnosticSemaKinds.t

[PATCH] D13419: Fix several problems at the intersection of template instantiations and visibility

loladiro updated this revision to Diff 76257. loladiro added a comment. Add the expected-error annotation Repository: rL LLVM https://reviews.llvm.org/D13419 Files: lib/AST/Decl.cpp lib/AST/DeclCXX.cpp test/CodeGenCXX/visibility.cpp test/Modules/cxx-templates.cpp Index: test/Modules

[PATCH] D13419: Fix several problems at the intersection of template instantiations and visibility

rsmith added a comment. Please factor out the fix for `getTemplateInstantiationPattern` and commit it separately. https://reviews.llvm.org/D25942 has a testcase for that portion of this change. Comment at: lib/AST/Decl.cpp:1057-1058 +RD->getTemplateInstantiationPatter

[libcxx] r285466 - Revert addition of __libcpp_library_version

Author: ericwf Date: Fri Oct 28 17:37:24 2016 New Revision: 285466 URL: http://llvm.org/viewvc/llvm-project?rev=285466&view=rev Log: Revert addition of __libcpp_library_version Modified: libcxx/trunk/include/__config libcxx/trunk/lib/abi/CHANGELOG.TXT libcxx/trunk/lib/abi/x86_64-apple

[PATCH] D13330: Implement __attribute__((unique_instantiation))

rsmith added a comment. I think this attribute is poorly named. Explicit instantiation definitions are *already* required to be globally unique; see [temp.spec]/5.1: "For a given template and a given set of template-arguments, an explicit instantiation definition shall appear at most once in a

[libcxx] r285469 - Remove files missed in r285466

Author: ericwf Date: Fri Oct 28 17:54:24 2016 New Revision: 285469 URL: http://llvm.org/viewvc/llvm-project?rev=285469&view=rev Log: Remove files missed in r285466 Removed: libcxx/trunk/src/libcpp_version.cpp libcxx/trunk/test/libcxx/version.pass.cpp Removed: libcxx/trunk/src/libcpp_vers

[PATCH] D26105: Allow CaseStmt to be initialized with a SubStmt

khazem created this revision. khazem added reviewers: a.sidorin, xazax.hun, doug.gregor, rsmith. khazem added subscribers: phosek, cfe-commits, smklein. The CaseStmt constructor now takes an optional SubStmt argument to initialize its SubExprs field with. This fixes an issue where the ASTImporter

[PATCH] D26105: Allow CaseStmt to be initialized with a SubStmt

khazem updated this revision to Diff 76272. khazem added a comment. (added more context) https://reviews.llvm.org/D26105 Files: include/clang/AST/Stmt.h lib/AST/ASTImporter.cpp Index: lib/AST/ASTImporter.cpp === --- lib/AST/A

[PATCH] D26108: Add -Wnullability-completeness-on-arrays.

jordan_rose created this revision. jordan_rose added a reviewer: doug.gregor. jordan_rose added a subscriber: cfe-commits. jordan_rose set the repository for this revision to rL LLVM. jordan_rose added a dependency: D25850: Accept nullability annotations (_Nullable) on array parameters. This is a

[PATCH] D26109: Warn when 'assume_nonnull' infers nullability within an array.

jordan_rose created this revision. jordan_rose added a reviewer: doug.gregor. jordan_rose added a subscriber: cfe-commits. jordan_rose set the repository for this revision to rL LLVM. jordan_rose added a dependency: D25850: Accept nullability annotations (_Nullable) on array parameters. ...or wit

[PATCH] D25940: [analyzer] LibraryFunctions: Fix errors due to different integral types and typedefs on different architectures.

zaks.anna added a comment. How about this imperfect solution that will work quite well in practice? For the ssize_t case, where type size cannot be used, we check the function name, # of arguments , and check that the functions are coming from the system header. https://reviews.llvm.org/D25940

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. I don't understand: 1. The motivation for this change 2. Why is this correct? Can we restart the discussion here (I'm reverting to fix the LTO build failure here: http://lab.llvm.org:8080/green/job/clang-stage2-configure-Rlto_build/10737/console ) https://review

[libcxx] r285485 - Revert "Added 'inline' attribute to basic_string's destructor"

Author: mehdi_amini Date: Fri Oct 28 19:50:02 2016 New Revision: 285485 URL: http://llvm.org/viewvc/llvm-project?rev=285485&view=rev Log: Revert "Added 'inline' attribute to basic_string's destructor" This reverts commit r285456, which broke LTO bootstrap on Darwin. Modified: libcxx/trunk/in

Re: [libcxx] r285456 - Added 'inline' attribute to basic_string's destructor

Hi Adtiya, I had to revert in r285485 to unbreak this build bot: http://lab.llvm.org:8080/green/job/clang-stage2-configure-Rlto_build/10737/ (LTO bootstrap of clang). Let’s iterate on the revision on Phabricator to fi

[PATCH] D23765: Fix for clang PR 29087

twoh marked an inline comment as not done. twoh added a comment. Ping https://reviews.llvm.org/D23765 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

sebpop added a comment. In https://reviews.llvm.org/D25624#582731, @mehdi_amini wrote: > I don't understand: > > 1. The motivation for this change This is a change for performance: we have seen some benchmarks where inlining the string dtor brings performance up by 5%: from what I remember, th

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. Trying to figure out: it seems you are trying to address what I reported here: https://llvm.org/bugs/show_bug.cgi?id=26498 ; right? > Except for inline functions, declarations with types deduced > from their initializer or return value (7.1.6.4), const > variables

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. In https://reviews.llvm.org/D25624#582752, @mehdi_amini wrote: > (The commit message is confusing, it mentions "Currently basic_string's > destructor is not getting inlined. So adding 'inline' attribute to > ~basic_string()", please add the quote of the standard and

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. FYI reproduced locally, and investigating now. https://reviews.llvm.org/D25624 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. I see a decl: declare hidden void @_ZNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEED1Ev(%"class.std::__1::basic_string"*) unnamed_addr #9 align 2 (Note the hidden which prevent from finding it in the dyib) And a use (after inlining): %1024 = ca

[PATCH] D13419: Fix several problems at the intersection of template instantiations and visibility

loladiro added inline comments. Comment at: lib/AST/DeclCXX.cpp:1349-1354 + while (!CTD->isMemberSpecialization()) { +auto *NewCTD = CTD->getInstantiatedFromMemberTemplate(); +if (!NewCTD) break; CTD = NewCTD; } rsm

Re: [libcxx] r281681 - [libc++] Add _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY to support GCC ABI compatibility

Hi Eric, I’m trying to make sense of _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY here. If I understand correctly, such a symbol will be present and visible in the dyliib, but emitted as available_externally and visibility hidden in the client code. I don’t understand the hidden visibility part,

[PATCH] D25624: Added 'inline' attribute to basic_string's destructor

mehdi_amini added a comment. I believe the issue is _LIBCPP_EXTERN_TEMPLATE_INLINE_VISIBILITY itself. I asked here: http://lists.llvm.org/pipermail/cfe-commits/Week-of-Mon-20161024/175454.html https://reviews.llvm.org/D25624 ___ cfe-commits maili

[PATCH] D13330: Implement __attribute__((unique_instantiation))

loladiro planned changes to this revision. loladiro added a comment. Thanks for the review! Sorry, it's been a while since I wrote this code, so I'm not fluent in all the details, but I've replied below. I am not particularly partial to the name, so whatever you feel is best is ok with me. ==