Re: r222393 - [CMake] Always include the Clang repo version, just like the autoconf build.

Nice! But .svn/entries doesn't seem to be touched when I `svn up` my llvm checkout. On Wed, Nov 19, 2014 at 2:03 PM, Jordan Rose wrote: > Author: jrose > Date: Wed Nov 19 16:03:48 2014 > New Revision: 222393 > > URL: http://llvm.org/viewvc/llvm-project?rev=222393&view=rev > Log: > [CMake] Alway

Re: r222393 - [CMake] Always include the Clang repo version, just like the autoconf build.

On Wed, Nov 19, 2014 at 5:24 PM, Nico Weber wrote: > Nice! > > But .svn/entries doesn't seem to be touched when I `svn up` my llvm > checkout. > (sorry, I tried to paste the following but triggered "send message" somehow.) llvm-rw$ ls -l .svn/entries -rw-r--r-- 1

Re: r222393 - [CMake] Always include the Clang repo version, just like the autoconf build.

gt; more tonight. > > Jordan > > > On Nov 19, 2014, at 17:24, Nico Weber wrote: > > Nice! > > But .svn/entries doesn't seem to be touched when I `svn up` my llvm > checkout. > > On Wed, Nov 19, 2014 at 2:03 PM, Jordan Rose > wrote: > >> Au

Re: r222393 - [CMake] Always include the Clang repo version, just like the autoconf build.

PM, Jordan Rose > wrote: > >> Yup, just found that described online. .svn/entries seems to be the 1.6 >> version, .svn/wc.db the 1.7 one. As long as wc.db shows up first in the >> search, we should be good. >> >> Jordan >> >> >> On Nov 19, 2014,

r222839 - clang-format: Tweak -style=Chromium for Java files.

Author: nico Date: Wed Nov 26 10:43:18 2014 New Revision: 222839 URL: http://llvm.org/viewvc/llvm-project?rev=222839&view=rev Log: clang-format: Tweak -style=Chromium for Java files. For Java, don't do any of the deviations from Google Style that Chromium style does for C++. Chromium's Java foll

r222964 - Fix typo.

Author: nico Date: Sat Nov 29 17:57:35 2014 New Revision: 222964 URL: http://llvm.org/viewvc/llvm-project?rev=222964&view=rev Log: Fix typo. Modified: cfe/trunk/lib/AST/VTableBuilder.cpp Modified: cfe/trunk/lib/AST/VTableBuilder.cpp URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/AST

r223056 - Add a test for devirtualization of virtual operator calls.

Author: nico Date: Mon Dec 1 11:48:04 2014 New Revision: 223056 URL: http://llvm.org/viewvc/llvm-project?rev=223056&view=rev Log: Add a test for devirtualization of virtual operator calls. There was no test coverage for this before: Modifiying EmitCXXOperatorMemberCallee() to not call CanDevirtu

Re: [PATCH] Allow clang-cl to automatically use lld instead of forcing the MSVC linker

Comment at: lib/Driver/Tools.cpp:7990 @@ +7989,3 @@ + } else { +ld_value = "link"; + } Somewhat tangential: It would be nice if clang could default to lld-link if link.exe isn't found. If I'm using clang-cl targeting Windows on non-Windows, and lld-link is

[patch] add support for has_feature(cxx_alignof) and has_feature(c_alignof)

r142020 added support for has_feature(cxx_alignas). This patch does the same for alignof. clang-alignof.patch Description: Binary data ___ cfe-commits mailing list cfe-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Re: r176887 - Revert r166370 and r166540 now that Xcode 4.6 has been available for a while.

Now is fine. Thanks for waiting :-) On Dec 1, 2014 1:43 PM, "Bob Wilson" wrote: > How about now? I’d like to clean up that FIXME and Xcode 4.6 has been out > for a long time…. > > On Nov 11, 2013, at 1:07 PM, Nico Weber wrote: > > On Mon, Nov 11, 2013 at 12:17 PM,

r223149 - Wrap to 80 columns. No behavior change.

Author: nico Date: Tue Dec 2 14:41:18 2014 New Revision: 223149 URL: http://llvm.org/viewvc/llvm-project?rev=223149&view=rev Log: Wrap to 80 columns. No behavior change. Modified: cfe/trunk/lib/CodeGen/CGExprCXX.cpp Modified: cfe/trunk/lib/CodeGen/CGExprCXX.cpp URL: http://llvm.org/viewvc/

Re: [patch] add support for has_feature(cxx_alignof) and has_feature(c_alignof)

ping! On Mon, Dec 1, 2014 at 1:07 PM, Nico Weber wrote: > r142020 added support for has_feature(cxx_alignas). This patch does the > same for alignof. > ___ cfe-commits mailing list cfe-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/lis

r223185 - Fix incorrect codegen for devirtualized calls to virtual overloaded operators.

Author: nico Date: Tue Dec 2 19:21:41 2014 New Revision: 223185 URL: http://llvm.org/viewvc/llvm-project?rev=223185&view=rev Log: Fix incorrect codegen for devirtualized calls to virtual overloaded operators. Consider this program: struct A { virtual void operator-() { printf("base\n"

Re: r223185 - Fix incorrect codegen for devirtualized calls to virtual overloaded operators.

D'oh, forgot to say "Over-the-shoulder-reviewed by rnk" in the CL description. On Tue, Dec 2, 2014 at 5:21 PM, Nico Weber wrote: > Author: nico > Date: Tue Dec 2 19:21:41 2014 > New Revision: 223185 > > URL: http://llvm.org/viewvc/llvm-project?rev=223185&view=

r223186 - Add support for has_feature(cxx_alignof) and has_feature(c_alignof).

Author: nico Date: Tue Dec 2 19:25:49 2014 New Revision: 223186 URL: http://llvm.org/viewvc/llvm-project?rev=223186&view=rev Log: Add support for has_feature(cxx_alignof) and has_feature(c_alignof). r142020 added support for has_feature(cxx_alignas). This does the same for alignof. Modified:

Re: [patch] add support for has_feature(cxx_alignof) and has_feature(c_alignof)

. Maybe "... to determine if > support for the ``alignof`` keyword is enabled."? > > Likewise for _Alignof. > > Otherwise, this seems fine to me. > > On Tue, Dec 2, 2014 at 4:40 PM, Nico Weber wrote: > >> ping! >> >> On Mon, Dec 1, 2014 at 1:07

Re: clang-apply-replacements doesn't handle windows newlines correctly

On Wed, Dec 3, 2014 at 6:43 PM, Nikola Smiljanic wrote: > I think it'll only happen when you run it on Windows. Writing files in > text mode will translate each \n into CRLF on this silly platform :) I'm > honestly surprised that nobody noticed this before. I filed a bug for > clang-modernize for

Re: r220714 - Frontend: Define __EXCEPTIONS if -fexceptions is passed

This doesn't look right. Isn't cc1 -fexceptions passed if either ObjC or C++ exceptions are enabled? As is, it looks like this breaks this lovely bit of code from breakpad in a .mm file: #ifndef __EXCEPTIONS // This file uses C++ try/catch (but shouldn't). Duplicate the macros from // allowing t

Re: r220714 - Frontend: Define __EXCEPTIONS if -fexceptions is passed

Possible, but http://llvm.org/viewvc/llvm-project?view=revision&revision=139496 suggests that at least Doug thinks that this is working as intended (that's the revision that added the test that got its expectations inverted here). On Thu, Dec 4, 2014 at 5:00 PM, Richard Smith wrote: > This looks

Re: r223455 - Driver: Objective-C should respect -fno-exceptions

On Fri, Dec 5, 2014 at 12:56 AM, David Majnemer wrote: > Author: majnemer > Date: Fri Dec 5 02:56:55 2014 > New Revision: 223455 > > URL: http://llvm.org/viewvc/llvm-project?rev=223455&view=rev > Log: > Driver: Objective-C should respect -fno-exceptions > > Clang attempted to replicate a GCC bug

Re: r223455 - Driver: Objective-C should respect -fno-exceptions

On Fri, Dec 5, 2014 at 11:41 AM, David Majnemer wrote: > > > On Fri, Dec 5, 2014 at 11:25 AM, Nico Weber wrote: > >> On Fri, Dec 5, 2014 at 12:56 AM, David Majnemer > > wrote: >> >>> Author: majnemer >>> Date: Fri Dec 5 02:56:55 2014 >>>

Re: r223455 - Driver: Objective-C should respect -fno-exceptions

On Fri, Dec 5, 2014 at 11:55 AM, David Majnemer wrote: > > > On Fri, Dec 5, 2014 at 11:47 AM, Nico Weber wrote: > >> On Fri, Dec 5, 2014 at 11:41 AM, David Majnemer > > wrote: >> >>> >>> >>> On Fri, Dec 5, 2014 at 11:25 AM, Nico Webe

Re: r223455 - Driver: Objective-C should respect -fno-exceptions

On Fri, Dec 5, 2014 at 1:00 PM, David Majnemer wrote: > > > On Fri, Dec 5, 2014 at 12:37 PM, Nico Weber wrote: > >> On Fri, Dec 5, 2014 at 11:55 AM, David Majnemer > > wrote: >> >>> >>> >>> On Fri, Dec 5, 2014 at 11:47 AM, Nico Webe

r223714 - Add a test for MS-ABI this adjustment for virtual calls to member operators.

Author: nico Date: Mon Dec 8 17:25:55 2014 New Revision: 223714 URL: http://llvm.org/viewvc/llvm-project?rev=223714&view=rev Log: Add a test for MS-ABI this adjustment for virtual calls to member operators. They too were fixed by r223185, despite the commit message saying otherwise. Add a test t

r223850 - clang-format: Add a test for PR19603 which seems fixed (maybe by r221338?).

Author: nico Date: Tue Dec 9 17:22:35 2014 New Revision: 223850 URL: http://llvm.org/viewvc/llvm-project?rev=223850&view=rev Log: clang-format: Add a test for PR19603 which seems fixed (maybe by r221338?). Modified: cfe/trunk/unittests/Format/FormatTest.cpp Modified: cfe/trunk/unittests/For

r223861 - cmake: Make SVNVersion.inc step depend on GetSVN.cmake.

Author: nico Date: Tue Dec 9 18:03:37 2014 New Revision: 223861 URL: http://llvm.org/viewvc/llvm-project?rev=223861&view=rev Log: cmake: Make SVNVersion.inc step depend on GetSVN.cmake. This way, the step generating SVNVersion.inc gets rerun every time someone changes GetSVN.cmake (which is the

Re: r223860 - Updated the AST importer to support importing

On Tue, Dec 9, 2014 at 4:00 PM, Sean Callanan wrote: > Author: spyffe > Date: Tue Dec 9 18:00:37 2014 > New Revision: 223860 > > URL: http://llvm.org/viewvc/llvm-project?rev=223860&view=rev > Log: > Updated the AST importer to support importing > LinkageSpecDecls. This is relevant when LLDB > w

Re: [PATCH] No memcpy for copy ctor with -fsanitize-address-field-padding=1

Sorry about the slow review. Rather than adding all these checks before calling EmitAggregateCopy(), is it maybe possible to do this check in EmitAggregateCopy() itself? Else there will always be cases that you miss. For example: $ cat foo.cc struct ClassWithTrivialCopy { ClassWithTrivialCop

Re: r221562 - [Objective-C Sema]. Issue availability/deprecated warning when

This is causing http://llvm.org/bugs/show_bug.cgi?id=21587 Looks like Sema::CollectMultipleMethodsInGlobalPool() should probably also look at stuff from an external sema source? On Fri, Nov 7, 2014 at 3:51 PM, Fariborz Jahanian wrote: > Author: fjahanian > Date: Fri Nov 7 17:51:15 2014 > New R

Re: r221562 - [Objective-C Sema]. Issue availability/deprecated warning when

12, 2014 at 1:54 PM, Nico Weber wrote: > This is causing http://llvm.org/bugs/show_bug.cgi?id=21587 > > Looks like Sema::CollectMultipleMethodsInGlobalPool() should probably also > look at stuff from an external sema source? > > On Fri, Nov 7, 2014 at 3:51 PM, Fariborz Jahani

Re: [PATCH] String plus char

Looks good to me, please land. On Mon, Dec 15, 2014 at 1:45 AM, Anders Rönnholm < anders.ronnh...@evidente.se> wrote: > > Ping. Is it ok to commit this one? > > //Anders > > Från: Anders Rönnholm > Skickat: den 30 april 2014 08

Re: r224012 - Emit warning if define or undef reserved identifier or keyword.

You can look at the warnings this produces for Chromium in a clang-cl build here: http://build.chromium.org/p/chromium.fyi/builders/CrWinClang/builds/832/steps/compile/logs/stdio (search for "warning(clang)"). Examples: It warns on libopenjpeg defining restrict to nothing if C99 isn't enabled:

Re: r224012 - Emit warning if define or undef reserved identifier or keyword.

I agree that off by default warnings are a "warning smell". But this doesn't look like a warning that's supposed to find bugs, it's more a language compliance thing, like -pedantic. Maybe it should be an ExtWarn. But is someone is actively implementing the suggested heuristics, we can wait and see

Re: r221562 - [Objective-C Sema]. Issue availability/deprecated warning when

On Mon, Dec 15, 2014 at 11:43 AM, jahanian wrote: > > > On Dec 12, 2014, at 3:34 PM, Nico Weber wrote: > > Ah, I think I found it. You indented it like this: > > if (Method) { > if (ObjCMethodDecl *BestMethod = > SelectBestMethod(Sel, ArgsI

Re: r224012 - Emit warning if define or undef reserved identifier or keyword.

10:24:22AM -0800, Nico Weber wrote: > > For the "final(a,b,c)" define: should we warn on that? It doesn't change > > the meaning of the keyword "final" since that doesn't have parameters. > > For C++11 and newer: IMO yes. Otherwise, no. > > Joerg

r224371 - Move -Wkeyword-macro into -pedantic, remove -Wreserved-id-macro.

Author: nico Date: Tue Dec 16 15:16:10 2014 New Revision: 224371 URL: http://llvm.org/viewvc/llvm-project?rev=224371&view=rev Log: Move -Wkeyword-macro into -pedantic, remove -Wreserved-id-macro. As discussed on the post-commit review thread for r224012, -Wkeyword-macro fires mostly on headers tr

[patch] Don't drop attributes when checking explicit specializations.

Don't drop attributes when checking explicit specializations. Consider a template class with attributes on a method, and an explicit specialization of that method: template struct A { void foo() final; }; template <> void A<0>::foo() {} In this example, the attribute

Re: r224012 - Emit warning if define or undef reserved identifier or keyword.

On Wed, Dec 17, 2014 at 7:05 PM, Richard Smith wrote: > > On Tue, Dec 16, 2014 at 1:17 PM, Nico Weber wrote: >> >> Ok, I moved -Wkeyword-macro into -pedantic in r224371. I also removed >> -Wreserved-id-macro in the same revision, as it's not covered by the >>

Re: [patch] Don't drop attributes when checking explicit specializations.

(+Jordan as he wrote r176728) On Wed, Dec 17, 2014 at 6:58 PM, Nico Weber wrote: > Don't drop attributes when checking explicit specializations. > > Consider a template class with attributes on a method, and an explicit > specialization of that method: > > t

Re: r224012 - Emit warning if define or undef reserved identifier or keyword.

On Thu, Dec 18, 2014 at 2:01 PM, Richard Smith wrote: > On Wed, Dec 17, 2014 at 7:58 PM, Nico Weber wrote: >> >> On Wed, Dec 17, 2014 at 7:05 PM, Richard Smith >> wrote: >>> >>> On Tue, Dec 16, 2014 at 1:17 PM, Nico Weber wrote: >>>> >>

Re: [patch] Don't drop attributes when checking explicit specializations.

On Thu, Dec 18, 2014 at 9:58 PM, John McCall wrote: > > On Dec 18, 2014, at 6:18 PM, Richard Smith wrote: > On Wed, Dec 17, 2014 at 6:58 PM, Nico Weber wrote: >> >> Don't drop attributes when checking explicit specializations. >> >> Consider a template cla

r224651 - Don't drop attributes when checking explicit specializations.

Author: nico Date: Fri Dec 19 17:52:45 2014 New Revision: 224651 URL: http://llvm.org/viewvc/llvm-project?rev=224651&view=rev Log: Don't drop attributes when checking explicit specializations. Consider a template class with attributes on a method, and an explicit specialization of that method:

Re: [patch] Don't drop attributes when checking explicit specializations.

ty >> >> I'm fine with this if there's no behavior change here. >> > > In both GCC and Clang, with and without this patch, A<0>::foo takes its > visibility from the attribute. > > >> On Thu, Dec 18, 2014 at 11:42 PM, Nico Weber wrote: >> &

[patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

Hi, the attached patch disables trigraphs by default in clang-cl mode. To do so, I'm renaming the cc1 trigraph flag to -ftrigraphs, and I'm adding a -fno-trigraphs flag. (This requires updating a handful of cc1 tests, and translating -trigraphs to -ftrigraphs in the driver.) The driver grows pars

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

Forgot to say: Fixes PR21974. On Fri, Dec 19, 2014 at 6:07 PM, Nico Weber wrote: > Hi, > > the attached patch disables trigraphs by default in clang-cl mode. To do > so, I'm renaming the cc1 trigraph flag to -ftrigraphs, and I'm adding a > -fno-trigraphs flag. (This re

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

Thanks! All done. On Fri, Dec 19, 2014 at 6:47 PM, Richard Smith wrote: > On Fri, Dec 19, 2014 at 6:08 PM, Nico Weber wrote: >> >> Forgot to say: Fixes PR21974. >> >> On Fri, Dec 19, 2014 at 6:07 PM, Nico Weber wrote: >> >>> Hi, >>> >

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

in CompilerInvocation based on the language > options. It seems we currently disable them in gnu mode for example. > Maybe we should do the same if -fms-extensions or -fms-compatibility > if enabled? > > On Sat, Dec 20, 2014 at 11:16 AM, Nico Weber wrote: > > Thanks! All done.

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

>From my understanding, /Zc: are sema-level flags. /GF- maps to -fwritable-strings and seems to be the codegen-level flag. On Sun, Dec 21, 2014 at 6:07 PM, David Majnemer wrote: > Should we map /Zc:strictStrings- to fwritable-strings? > > > On Sunday, December 21, 2014, Ni

r224695 - Wrap to 80 columns. NfC.

Author: nico Date: Sun Dec 21 23:21:03 2014 New Revision: 224695 URL: http://llvm.org/viewvc/llvm-project?rev=224695&view=rev Log: Wrap to 80 columns. NfC. Modified: cfe/trunk/lib/AST/DeclObjC.cpp Modified: cfe/trunk/lib/AST/DeclObjC.cpp URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/li

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

On Mon, Dec 22, 2014 at 1:38 AM, Hans Wennborg wrote: > On Sun, Dec 21, 2014 at 5:24 PM, Nico Weber wrote: > > That's much nicer, thanks. I couldn't figure out how to alias a clang-cl > > option to a cc1 option, so I added 4 lines to explicitly translate >

r224714 - Add test coverage for cc1's trigraph option handling.

Author: nico Date: Mon Dec 22 11:56:50 2014 New Revision: 224714 URL: http://llvm.org/viewvc/llvm-project?rev=224714&view=rev Log: Add test coverage for cc1's trigraph option handling. Added: cfe/trunk/test/Frontend/trigraphs.cpp Added: cfe/trunk/test/Frontend/trigraphs.cpp URL: http://llvm

r224718 - Rename test.cc files to test.cpp.

Author: nico Date: Mon Dec 22 12:13:07 2014 New Revision: 224718 URL: http://llvm.org/viewvc/llvm-project?rev=224718&view=rev Log: Rename test.cc files to test.cpp. The lit.cfg files only add .cpp to suffixes, so these tests used to never run, oops. (Also tweak to of these tests in minor ways to

Re: r217836 - PR19692: Add (passing) regression test.

Is my.cc used for anything? I'm asking because I just renamed a few cc files to cpp in r224718. my.cc is the last .cc file that remains in the test/ directory, but I don't understand how (or if) it's used. On Mon, Sep 15, 2014 at 3:14 PM, Richard Smith wrote: > Author: rsmith > Date: Mon Sep 15

r224719 - Disable trigraphs in microsoft mode by default. Matches cl.exe.

Author: nico Date: Mon Dec 22 12:35:03 2014 New Revision: 224719 URL: http://llvm.org/viewvc/llvm-project?rev=224719&view=rev Log: Disable trigraphs in microsoft mode by default. Matches cl.exe. The default value of Opts.Trigraphs now no longer depends solely on the language input kind, so move t

r224747 - Add a triple to try and get this test passing on the ARM bots.

Author: nico Date: Mon Dec 22 19:07:10 2014 New Revision: 224747 URL: http://llvm.org/viewvc/llvm-project?rev=224747&view=rev Log: Add a triple to try and get this test passing on the ARM bots. Modified: cfe/trunk/test/CodeGen/builtin-recursive.cpp Modified: cfe/trunk/test/CodeGen/builtin-re

Re: r224718 - Rename test.cc files to test.cpp.

builds/2054 > > Cheers, > Renato > On 22 Dec 2014 19:14, "Nico Weber" wrote: > >> Author: nico >> Date: Mon Dec 22 12:13:07 2014 >> New Revision: 224718 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=224718&view=rev >> Log: >&

Re: r217836 - PR19692: Add (passing) regression test.

On Tue, Dec 23, 2014 at 11:18 AM, Richard Smith wrote: > On 22 Dec 2014 10:15, "Nico Weber" wrote: > > > > Is my.cc used for anything? I'm asking because I just renamed a few cc > files to cpp in r224718. my.cc is the last .cc file that remains in the > tes

r224784 - Remove unused test input.

Author: nico Date: Tue Dec 23 13:30:13 2014 New Revision: 224784 URL: http://llvm.org/viewvc/llvm-project?rev=224784&view=rev Log: Remove unused test input. Removed: cfe/trunk/test/Modules/Inputs/pr19692/my.cc Removed: cfe/trunk/test/Modules/Inputs/pr19692/my.cc URL: http://llvm.org/viewvc/

r224790 - Add driver flags -ftrigraphs, -fno-trigraphs.

Author: nico Date: Tue Dec 23 16:32:37 2014 New Revision: 224790 URL: http://llvm.org/viewvc/llvm-project?rev=224790&view=rev Log: Add driver flags -ftrigraphs, -fno-trigraphs. -trigraphs is now an alias for -ftrigraphs. -fno-trigraphs makes it possible to explicitly disable trigraphs, which cou

r224791 - clang-cl: Various changes to /Zc: handling.

Author: nico Date: Tue Dec 23 16:55:34 2014 New Revision: 224791 URL: http://llvm.org/viewvc/llvm-project?rev=224791&view=rev Log: clang-cl: Various changes to /Zc: handling. * /Zc:trigraphs and /Zc:trigraphs- are now honored * /Zc:strictStrings is now honored * /Zc:auto is now honored/ignored (c

Re: [patch, clang-cl] Implement /Zc:trigraphs support, disable trigraphs by default

Thanks, landed in that form in r224790 / r224791. On Tue, Dec 23, 2014 at 1:16 PM, Richard Smith wrote: > On Mon, Dec 22, 2014 at 8:40 AM, Nico Weber wrote: > >> On Mon, Dec 22, 2014 at 1:38 AM, Hans Wennborg wrote: >> >>> On Sun, Dec 21, 2014 at 5:24 PM, Nico Webe

Re: r221933 - Objective-C. Fixes a regression caused by implementation

On Thu, Nov 13, 2014 at 3:37 PM, Argyrios Kyrtzidis wrote: > > On Nov 13, 2014, at 2:27 PM, Fariborz Jahanian > wrote: > > /// ObjCMethodList - a linked list of methods with different signatures. > struct ObjCMethodList { > ObjCMethodDecl *Method; > + /// \brief count of methods with same sig

r224873 - Add more test coverage for the Objective-C deprected selector warning.

Author: nico Date: Fri Dec 26 19:05:55 2014 New Revision: 224873 URL: http://llvm.org/viewvc/llvm-project?rev=224873&view=rev Log: Add more test coverage for the Objective-C deprected selector warning. I broke this case in a local patch I'm writing, and there was no test to stop me. Now there is

r224875 - Make the test from r224873 actually pass.

Author: nico Date: Fri Dec 26 21:38:18 2014 New Revision: 224875 URL: http://llvm.org/viewvc/llvm-project?rev=224875&view=rev Log: Make the test from r224873 actually pass. The behavior looks incorrect to me, but the test is supposed to document current behavior for now. Modified: cfe/trunk/

r224876 - Address review feedback on r221933.

Author: nico Date: Fri Dec 26 21:58:08 2014 New Revision: 224876 URL: http://llvm.org/viewvc/llvm-project?rev=224876&view=rev Log: Address review feedback on r221933. Remove ObjCMethodList::Count, instead store a "has more than one decl" bit in the low bit of the ObjCMethodDecl pointer, using a P

Re: r221933 - Objective-C. Fixes a regression caused by implementation

On Thu, Dec 25, 2014 at 10:41 PM, Nico Weber wrote: > On Thu, Nov 13, 2014 at 3:37 PM, Argyrios Kyrtzidis > wrote: > >> >> On Nov 13, 2014, at 2:27 PM, Fariborz Jahanian >> wrote: >> >> /// ObjCMethodList - a linked list of methods with differe

r224881 - Objective-C: Tweak unavailability warning.

Author: nico Date: Sat Dec 27 01:09:37 2014 New Revision: 224881 URL: http://llvm.org/viewvc/llvm-project?rev=224881&view=rev Log: Objective-C: Tweak unavailability warning. Don't warn when a selector has an unavailable and an available variant, and the first also has an implementation. Modifie

r224892 - Objective-C: Serialize "more than one decl" state of ObjCMethodList.

Author: nico Date: Sat Dec 27 16:14:15 2014 New Revision: 224892 URL: http://llvm.org/viewvc/llvm-project?rev=224892&view=rev Log: Objective-C: Serialize "more than one decl" state of ObjCMethodList. This fixes PR21587, what r221933 fixed for regular programs is now also fixed for decls coming fr

r224893 - Remove unneeded include of DeclCXX.h from libParse.

Author: nico Date: Sat Dec 27 19:52:28 2014 New Revision: 224893 URL: http://llvm.org/viewvc/llvm-project?rev=224893&view=rev Log: Remove unneeded include of DeclCXX.h from libParse. Modified: cfe/trunk/lib/Parse/ParseAST.cpp Modified: cfe/trunk/lib/Parse/ParseAST.cpp URL: http://llvm.org/v

r224894 - Add stub sections about Parse, Sema, CodeGen to the internals manual.

Author: nico Date: Sat Dec 27 20:07:26 2014 New Revision: 224894 URL: http://llvm.org/viewvc/llvm-project?rev=224894&view=rev Log: Add stub sections about Parse, Sema, CodeGen to the internals manual. I'd be interested if the paragraph on Parse not knowing much about AST is something folks agree

r224895 - Fix markup from r224894.

Author: nico Date: Sat Dec 27 20:12:59 2014 New Revision: 224895 URL: http://llvm.org/viewvc/llvm-project?rev=224895&view=rev Log: Fix markup from r224894. Modified: cfe/trunk/docs/InternalsManual.rst Modified: cfe/trunk/docs/InternalsManual.rst URL: http://llvm.org/viewvc/llvm-project/cfe/

Re: r224892 - Objective-C: Serialize "more than one decl" state of ObjCMethodList.

On Sat, Dec 27, 2014 at 2:14 PM, Nico Weber wrote: > Author: nico > Date: Sat Dec 27 16:14:15 2014 > New Revision: 224892 > > URL: http://llvm.org/viewvc/llvm-project?rev=224892&view=rev > Log: > Objective-C: Serialize "more than one decl" state of ObjCMetho

r224906 - Don't crash on surprising tokens in default parameter template lists.

Author: nico Date: Sun Dec 28 17:24:02 2014 New Revision: 224906 URL: http://llvm.org/viewvc/llvm-project?rev=224906&view=rev Log: Don't crash on surprising tokens in default parameter template lists. Fixes this snippet from SLi's afl fuzzer output: class { i (x = <, enum This parsed i

r224915 - Don't crash on malformed attributes in an incorrect location.

Author: nico Date: Mon Dec 29 00:56:50 2014 New Revision: 224915 URL: http://llvm.org/viewvc/llvm-project?rev=224915&view=rev Log: Don't crash on malformed attributes in an incorrect location. r168626 added nicer diagnostics for attributes in the wrong places, such as after the `final` on a class

Re: r224945 - Parse: Ignore '::' in 'struct :: {'

Looks like this caused the following on SLi's bot: *Was:* OK *New:* Assertion `is() && "Invalid accessor called"' failed. *Cases:* (num=2) 298299 (reduced)

r224958 - Crash even less on malformed attributes in an incorrect location.

Author: nico Date: Mon Dec 29 15:56:22 2014 New Revision: 224958 URL: http://llvm.org/viewvc/llvm-project?rev=224958&view=rev Log: Crash even less on malformed attributes in an incorrect location. This is a follow-up to r224915. This adds a bit more line noise to the tests added in that revision

[patch] Remove an assert that was added in r220085 that isn't true for invalid inputs

Hi, r220085 moved some existing code into a new function, added another caller to that function, and added a new assert to the existing code. This assert fires on invalid inputs (see test in patch). Removing it seems to work, and the assert wasn't there before. Nico clang-qualname-assert.patch

r224974 - Don't crash on an invalid trailing return type on a function before a '...'

Author: nico Date: Mon Dec 29 20:06:40 2014 New Revision: 224974 URL: http://llvm.org/viewvc/llvm-project?rev=224974&view=rev Log: Don't crash on an invalid trailing return type on a function before a '...' clang tries to produce a helpful diagnostic for the traiilng '...', but the code that r216

r225004 - Remove a comment that appears a second time 22 lines further down.

Author: nico Date: Tue Dec 30 14:13:37 2014 New Revision: 225004 URL: http://llvm.org/viewvc/llvm-project?rev=225004&view=rev Log: Remove a comment that appears a second time 22 lines further down. Modified: cfe/trunk/lib/Sema/SemaCXXScopeSpec.cpp Modified: cfe/trunk/lib/Sema/SemaCXXScopeSpe

Re: r225000 - Reapply "DebugInfo: Generalize debug info location handling"

Probably just needs a triple? On Dec 31, 2014 11:25 AM, "Aaron Ballman" wrote: > On Wed, Dec 31, 2014 at 9:36 AM, İsmail Dönmez wrote: > > Hi, > > > > On Tue, Dec 30, 2014 at 9:39 PM, David Blaikie > wrote: > >> > >> +// CHECK-LABEL: define > >> +__complex double f11() { > >> + __complex doubl

r225122 - Wrap to 80 columns, no behavior change.

Author: nico Date: Sat Jan 3 18:47:22 2015 New Revision: 225122 URL: http://llvm.org/viewvc/llvm-project?rev=225122&view=rev Log: Wrap to 80 columns, no behavior change. Modified: cfe/trunk/lib/Parse/Parser.cpp Modified: cfe/trunk/lib/Parse/Parser.cpp URL: http://llvm.org/viewvc/llvm-proje

r225124 - Crash less enthusiasticially on _Atomic or __restrict__ on invalid types.

Author: nico Date: Sat Jan 3 22:53:10 2015 New Revision: 225124 URL: http://llvm.org/viewvc/llvm-project?rev=225124&view=rev Log: Crash less enthusiasticially on _Atomic or __restrict__ on invalid types. Many places in Sema cannot handle isNull() types. This is fine, because in most places the

r225125 - Document that GetTypeForDeclarator() cannot return a null type.

Author: nico Date: Sat Jan 3 23:29:21 2015 New Revision: 225125 URL: http://llvm.org/viewvc/llvm-project?rev=225125&view=rev Log: Document that GetTypeForDeclarator() cannot return a null type. Also add a few asserts for this. The existing code assumes this in a bunch of places already (see e.g

r225128 - Wrap to 80 columns, no behavior change.

Author: nico Date: Sun Jan 4 01:59:35 2015 New Revision: 225128 URL: http://llvm.org/viewvc/llvm-project?rev=225128&view=rev Log: Wrap to 80 columns, no behavior change. Modified: cfe/trunk/lib/Sema/SemaType.cpp Modified: cfe/trunk/lib/Sema/SemaType.cpp URL: http://llvm.org/viewvc/llvm-pro

r225129 - Fix indentation. No behavior change.

Author: nico Date: Sun Jan 4 02:07:54 2015 New Revision: 225129 URL: http://llvm.org/viewvc/llvm-project?rev=225129&view=rev Log: Fix indentation. No behavior change. Modified: cfe/trunk/lib/Parse/ParseStmt.cpp Modified: cfe/trunk/lib/Parse/ParseStmt.cpp URL: http://llvm.org/viewvc/llvm-pr

r225140 - Remove an assert that's not true on invalid code.

Author: nico Date: Sun Jan 4 14:32:12 2015 New Revision: 225140 URL: http://llvm.org/viewvc/llvm-project?rev=225140&view=rev Log: Remove an assert that's not true on invalid code. r185773 added an assert that checked that a CXXUnresolvedConstructExpr either has a valid rparen, or exactly one arg

Re: r225141 - clang-format: [Java] Change a few flags for Google's Java style.

We (chromium) definitely want AllowShortIfStatementsOnASingleLine true since https://source.android.com/source/code-style.html explicitly allows this, but if Google style doesn't want AlignTrailingComments for Java I'd guess we don't want that either. On Sun, Jan 4, 2015 at 12:40 PM, Daniel Jasper

Re: r217302 - Add error, recovery and fixit for "~A::A() {...}".

Sorry for the somewhat delayed review. Two points below. On Fri, Sep 5, 2014 at 7:06 PM, Richard Smith wrote: > Author: rsmith > Date: Fri Sep 5 21:06:12 2014 > New Revision: 217302 > > URL: http://llvm.org/viewvc/llvm-project?rev=217302&view=rev > Log: > Add error, recovery and fixit for "~A::

Re: r225140 - Remove an assert that's not true on invalid code.

Jan 2015 20:37, "Nico Weber" wrote: > >> Author: nico >> Date: Sun Jan 4 14:32:12 2015 >> New Revision: 225140 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=225140&view=rev >> Log: >> Remove an assert that's

Re: [libcxxabi] r225187 - Make the Unwinder -Werror clean.

On Mon, Jan 5, 2015 at 10:12 AM, Dan Albert wrote: > Anyone mind if I make -Werror on by default? > Since you don't know which compiler people will use when building this, that seems like a bad idea to me. clang usually builds with a warning or two for me on OS X for example (due to a since-fixe

Re: r225000 - Reapply "DebugInfo: Generalize debug info location handling"

On Mon, Jan 5, 2015 at 9:07 PM, Jiangning Liu wrote: > Hi David, > > This commit introduced a regression for one of my benchmarks, and the > failure is an assertion failure. > > The failure point is at > > LexicalScope *LexicalScopes::getOrCreateRegularScope(MDNode *Scope) { > .. > if (!Par

r225324 - Fix grammar-o in comment.

Author: nico Date: Tue Jan 6 17:54:59 2015 New Revision: 225324 URL: http://llvm.org/viewvc/llvm-project?rev=225324&view=rev Log: Fix grammar-o in comment. Modified: cfe/trunk/lib/Sema/SemaDeclCXX.cpp Modified: cfe/trunk/lib/Sema/SemaDeclCXX.cpp URL: http://llvm.org/viewvc/llvm-project/cfe

r225340 - Fix comment typo.

Author: nico Date: Tue Jan 6 23:25:05 2015 New Revision: 225340 URL: http://llvm.org/viewvc/llvm-project?rev=225340&view=rev Log: Fix comment typo. Modified: cfe/trunk/lib/CodeGen/ItaniumCXXABI.cpp Modified: cfe/trunk/lib/CodeGen/ItaniumCXXABI.cpp URL: http://llvm.org/viewvc/llvm-project/c

r225361 - Revert r225085, it caused PR22096.

Author: nico Date: Wed Jan 7 12:23:08 2015 New Revision: 225361 URL: http://llvm.org/viewvc/llvm-project?rev=225361&view=rev Log: Revert r225085, it caused PR22096. PR22096 has several test cases that assert that look fairly different. I'm adding one of those as an automated test, but when relan

Re: r225000 - Reapply "DebugInfo: Generalize debug info location handling"

On Mon, Jan 5, 2015 at 9:57 PM, Nico Weber wrote: > On Mon, Jan 5, 2015 at 9:07 PM, Jiangning Liu > wrote: > >> Hi David, >> >> This commit introduced a regression for one of my benchmarks, and the >> failure is an assertion failure. >> >&g

r225363 - clang-format: Let Chromium use the Google default for AlignTrailingComments.

Author: nico Date: Wed Jan 7 12:41:10 2015 New Revision: 225363 URL: http://llvm.org/viewvc/llvm-project?rev=225363&view=rev Log: clang-format: Let Chromium use the Google default for AlignTrailingComments. r225141 changed the defaults of AllowShortIfStatementsOnASingleLine and AlignTrailingComm

r225364 - Slightly simplify the test from r225361.

Author: nico Date: Wed Jan 7 12:47:51 2015 New Revision: 225364 URL: http://llvm.org/viewvc/llvm-project?rev=225364&view=rev Log: Slightly simplify the test from r225361. Shorter and doesn't need -O2 -- but still suboptimal as it's still doing -emit-obj. dblaikie says he'll improve this when he

r225409 - Add help text for mmacosx-version-min=, mios-version-min=.

Author: nico Date: Wed Jan 7 16:40:55 2015 New Revision: 225409 URL: http://llvm.org/viewvc/llvm-project?rev=225409&view=rev Log: Add help text for mmacosx-version-min=, mios-version-min=. I keep forgetting the exact spelling of -macosx-version-min=, and now I can run `bin/clang --help | grep ve

r225414 - Wrap to 80 columns. No behavior change.

Author: nico Date: Wed Jan 7 17:50:05 2015 New Revision: 225414 URL: http://llvm.org/viewvc/llvm-project?rev=225414&view=rev Log: Wrap to 80 columns. No behavior change. Modified: cfe/trunk/lib/Sema/SemaDeclAttr.cpp Modified: cfe/trunk/lib/Sema/SemaDeclAttr.cpp URL: http://llvm.org/viewvc/

[patch/rfc] An opt-in warning for functions whose availability(introduced) is newer than the deployment target

Hi, the Mac OS X and iOS SDKs add new functions in new releases. Apple recommends using the newest SDK and setting the deployment target to whatever old OS version one wants to support, and only calling new functions after checking that they are available at runtime. In practice, we (Chromium) ge

  1   2   3   4   5   6   7   8   9   10   >