Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-05 Thread Edwin Vane
Committed by Rafael Espíndola. http://llvm-reviews.chandlerc.com/D91 ___ cfe-commits mailing list cfe-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-02 Thread Evgeniy Stepanov
lgtm Comment at: lib/Driver/Tools.cpp:5768 @@ -5765,1 +5767,3 @@ + const bool isAndroid = +ToolChain.getTriple().getEnvironment() == llvm::Triple::Android; Nit: you've got the same line broken in a different way in the first chunk above.

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-02 Thread Edwin Vane
Fixed line break nit. These drive me nuts too! Hi eugenis, http://llvm-reviews.chandlerc.com/D91 CHANGE SINCE LAST DIFF http://llvm-reviews.chandlerc.com/D91?vs=246id=250#toc Files: lib/Driver/Tools.cpp test/Driver/clang-translation.c test/Driver/linux-ld.c Index:

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-02 Thread Rafael Espíndola
lgtm2. Do you have commit access? On 2 November 2012 08:45, Edwin Vane edwin.v...@intel.com wrote: Fixed line break nit. These drive me nuts too! Hi eugenis, http://llvm-reviews.chandlerc.com/D91 CHANGE SINCE LAST DIFF http://llvm-reviews.chandlerc.com/D91?vs=246id=250#toc Files:

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-02 Thread Vane, Edwin
: Friday, November 02, 2012 2:37 PM To: reviews+d91+public+6009407b19b79...@llvm-reviews.chandlerc.com Cc: euge...@google.com; Vane, Edwin; cfe-commits@cs.uiuc.edu Subject: Re: [cfe-commits] [PATCH] Fix x86 android support in clang lgtm2. Do you have commit access? On 2 November 2012 08:45, Edwin Vane

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-02 Thread Rafael Espíndola
On 2 November 2012 15:48, Vane, Edwin edwin.v...@intel.com wrote: No. I have a few revisions in Phab marked as accepted and wasn't sure what to do next. One revision Chandler indicated as good (D94) except it hasn’t been accepted yet and I'm stuck on that as well. I have committed this one.

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Evgeniy Stepanov
LGTM with the default cpu test Comment at: lib/Driver/Tools.cpp:1091 @@ -1090,2 +1090,3 @@ ArgStringList CmdArgs) const { + const bool isAndroid = getToolChain().getTriple().getEnvironment() == llvm::Triple::Android; if

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Edwin Vane
Instead of using 'atom' as default cpu name, after some internal consultation, 'core2' is a better common denominator for x86 android. Added tests to test/Driver/clang-translation.c. Note that gcc doesn't appear to do anything special if you don't explicitly specify which CPU to target

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Edwin Vane
Just want to make sure the new default CPUName is ok. http://llvm-reviews.chandlerc.com/D91 ___ cfe-commits mailing list cfe-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Rafael Espíndola
On 1 November 2012 15:59, Edwin Vane edwin.v...@intel.com wrote: Instead of using 'atom' as default cpu name, after some internal consultation, 'core2' is a better common denominator for x86 android. Added tests to test/Driver/clang-translation.c. Why? Can you add a comment saying why

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Edwin Vane
Sorry for the spam. Uploaded the wrong patch file (i.e. the old one again) last time. Hi eugenis, http://llvm-reviews.chandlerc.com/D91 CHANGE SINCE LAST DIFF http://llvm-reviews.chandlerc.com/D91?vs=245id=246#toc Files: lib/Driver/Tools.cpp test/Driver/clang-translation.c

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Rafael Espíndola
lgtm On 1 November 2012 16:29, Edwin Vane edwin.v...@intel.com wrote: Sorry for the spam. Uploaded the wrong patch file (i.e. the old one again) last time. Hi eugenis, http://llvm-reviews.chandlerc.com/D91 CHANGE SINCE LAST DIFF http://llvm-reviews.chandlerc.com/D91?vs=245id=246#toc

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-11-01 Thread Edwin Vane
Addressed reviewer comments. Hi eugenis, http://llvm-reviews.chandlerc.com/D91 CHANGE SINCE LAST DIFF http://llvm-reviews.chandlerc.com/D91?vs=244id=245#toc Files: lib/Driver/Tools.cpp test/Driver/clang-translation.c test/Driver/linux-ld.c Index: lib/Driver/Tools.cpp

[cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Edwin Vane
Hi eugenis, - -Bsymbolic must be added for x86 as well. - Default CPU name also set to 'atom' for x86 android. http://llvm-reviews.chandlerc.com/D91 Files: lib/Driver/Tools.cpp test/Driver/linux-ld.c Index: lib/Driver/Tools.cpp

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Rafael Espíndola
-CPUName = pentium4; +CPUName = isAndroid ? atom : pentium4; This matches gcc behavior? Can you add a test for this part? Cheers, Rafael ___ cfe-commits mailing list cfe-commits@cs.uiuc.edu

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Rafael Espíndola
On 31 October 2012 15:12, Vane, Edwin edwin.v...@intel.com wrote: -CPUName = pentium4; +CPUName = isAndroid ? atom : pentium4; This matches gcc behavior? Can you add a test for this part? REV: Sure but what test case would I adjust? These options only seem to show up for

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Rafael Espíndola
REV: I know *how* I'd write the test, I'm just not sure which actual test suite it belongs to. For example, test/Driver/linux-ld.c has linking tests for android but I couldn't find any existing tests for testing compilation. If you could point me at the right test suite I'll add tests

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Vane, Edwin
On 31 October 2012 15:12, Vane, Edwin edwin.v...@intel.com wrote: -CPUName = pentium4; +CPUName = isAndroid ? atom : pentium4; This matches gcc behavior? Can you add a test for this part? REV: Sure but what test case would I adjust? These options only seem to show up for

Re: [cfe-commits] [PATCH] Fix x86 android support in clang

2012-10-31 Thread Vane, Edwin
-CPUName = pentium4; +CPUName = isAndroid ? atom : pentium4; This matches gcc behavior? Can you add a test for this part? REV: Sure but what test case would I adjust? These options only seem to show up for compiling and I couldn't find a test suite for testing android