Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-23 Thread Uros Bizjak
On Fri, Sep 23, 2011 at 10:12 AM, Jakub Jelinek wrote: >> > --- gcc/config/i386/i386.c.jj   2011-09-02 16:29:38.0 +0200 >> > +++ gcc/config/i386/i386.c      2011-09-07 21:57:52.0 +0200 >> > @@ -18304,6 +18304,11 @@ ix86_prepare_sse_fp_compare_args (rtx de >> >  { >> >   rtx tmp; >>

Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-23 Thread Jakub Jelinek
On Fri, Sep 23, 2011 at 12:09:18AM +0200, Uros Bizjak wrote: > > --- gcc/config/i386/i386.c.jj   2011-09-02 16:29:38.0 +0200 > > +++ gcc/config/i386/i386.c      2011-09-07 21:57:52.0 +0200 > > @@ -18304,6 +18304,11 @@ ix86_prepare_sse_fp_compare_args (rtx de > >  { > >   rtx tmp; >

Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-22 Thread Uros Bizjak
On Wed, Sep 7, 2011 at 10:01 PM, Jakub Jelinek wrote: > On Wed, Sep 07, 2011 at 09:54:03PM +0200, Uros Bizjak wrote: >> > 2011-09-07  Jakub Jelinek   >> > >> >        PR target/50310 >> >        * config/i386/i386.c (ix86_prepare_sse_fp_compare_args): For >> >        TARGET_AVX return code for LTG

Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-07 Thread Uros Bizjak
On Wed, Sep 7, 2011 at 10:01 PM, Jakub Jelinek wrote: >> > 2011-09-07  Jakub Jelinek   >> > >> >        PR target/50310 >> >        * config/i386/i386.c (ix86_prepare_sse_fp_compare_args): For >> >        TARGET_AVX return code for LTGT and UNEQ. >> >        (ix86_expand_fp_vcond): Handle LTGT an

Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-07 Thread Jakub Jelinek
On Wed, Sep 07, 2011 at 09:54:03PM +0200, Uros Bizjak wrote: > > 2011-09-07  Jakub Jelinek   > > > >        PR target/50310 > >        * config/i386/i386.c (ix86_prepare_sse_fp_compare_args): For > >        TARGET_AVX return code for LTGT and UNEQ. > >        (ix86_expand_fp_vcond): Handle LTGT and

Re: [PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-07 Thread Uros Bizjak
On Wed, Sep 7, 2011 at 8:17 PM, Jakub Jelinek wrote: > The attached testcase ICEs, because the vectorizer assumes that if vcond* > is available, it supports all comparisons, not just a subset of them. > With -mavx vcmpd etc. already support all the needed comparisons (and > several more - we woul

[PATCH] Ensure vcond* expansion doesn't fail on x86 (PR target/50310)

2011-09-07 Thread Jakub Jelinek
Hi! The attached testcase ICEs, because the vectorizer assumes that if vcond* is available, it supports all comparisons, not just a subset of them. With -mavx vcmpd etc. already support all the needed comparisons (and several more - we wouldn't even need to swap the arguments), for SSE the only mi