Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Jakub Jelinek
On Tue, May 07, 2013 at 02:31:05PM +0200, Christophe Lyon wrote: > On 7 May 2013 14:22, Jakub Jelinek wrote: > > On Tue, May 07, 2013 at 02:18:35PM +0200, Christophe Lyon wrote: > >> Thanks for your review, here is a new version taking your comments into > >> account. > >> > >> Christophe. > >> >

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Christophe Lyon
On 7 May 2013 14:22, Jakub Jelinek wrote: > On Tue, May 07, 2013 at 02:18:35PM +0200, Christophe Lyon wrote: >> Thanks for your review, here is a new version taking your comments into >> account. >> >> Christophe. >> >> 2013-05-06 Christophe Lyon > > Two spaces before <, instead of just one. >

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Jakub Jelinek
On Tue, May 07, 2013 at 02:18:35PM +0200, Christophe Lyon wrote: > Thanks for your review, here is a new version taking your comments into > account. > > Christophe. > > 2013-05-06 Christophe Lyon Two spaces before <, instead of just one. Otherwise it looks good to me, but if it depends on th

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Christophe Lyon
Thanks for your review, here is a new version taking your comments into account. Christophe. 2013-05-06 Christophe Lyon gcc/ * config/arm/arm.c (arm_asan_shadow_offset): New function. (TARGET_ASAN_SHADOW_OFFSET): Define. * config/arm/linux-eabi.h (ASAN_CC1_SPEC): Define. (L

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Jakub Jelinek
On Mon, May 06, 2013 at 08:52:21PM +0200, Christophe Lyon wrote: > --- a/gcc/config/arm/arm.c > +++ b/gcc/config/arm/arm.c > @@ -281,6 +281,8 @@ static unsigned arm_add_stmt_cost (void *data, int count, > > static void arm_canonicalize_comparison (int *code, rtx *op0, rtx *op1, >

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Konstantin Serebryany
The rest of the patch is gcc-specific. Jakub? On Tue, May 7, 2013 at 12:12 PM, Christophe Lyon wrote: > Oops, you are correct, I read the code too quickly and didn't notice > the #endif right after the Thumb bit cancellation (I thought the > function didn't return any value when on ARM). > > A wh

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Christophe Lyon
Oops, you are correct, I read the code too quickly and didn't notice the #endif right after the Thumb bit cancellation (I thought the function didn't return any value when on ARM). A white line before #if defined(__powerpc__) would help readability :-) Forget that part. Thanks, Christophe.

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Evgeniy Stepanov
On Tue, May 7, 2013 at 9:33 AM, Konstantin Serebryany wrote: > +euge...@google.com (our ARM expert) > > Christophe, > > The change in sanitizer_common/sanitizer_stacktrace.cc changes the > logic which is known to work well for us on Android/ARM. > I'd like Evgeniy to comment on it. > > --kcc > > O

Re: [Patch, ARM, v2] Enable libsanitizer

2013-05-07 Thread Konstantin Serebryany
+euge...@google.com (our ARM expert) Christophe, The change in sanitizer_common/sanitizer_stacktrace.cc changes the logic which is known to work well for us on Android/ARM. I'd like Evgeniy to comment on it. --kcc On Mon, May 6, 2013 at 10:52 PM, Christophe Lyon wrote: > Hi, > > Here is a 2nd

[Patch, ARM, v2] Enable libsanitizer

2013-05-06 Thread Christophe Lyon
Hi, Here is a 2nd attempt at enabling libsanitizer on ARM. Compared with the previous version, this patch is more intrusive to workaround some limitations with qemu: * qemu does not work well with threads, so I chose to disable clone-test-1.c and rlimit-mmap-test-1.c when running on a simulator.