On Thu, Dec 10, 2020 at 8:52 PM Prathamesh Kulkarni <prathamesh.kulka...@linaro.org> wrote: > > On Wed, 9 Dec 2020 at 15:52, Hongtao Liu <crazy...@gmail.com> wrote: > > > > On Wed, Dec 9, 2020 at 5:22 PM Prathamesh Kulkarni via Gcc-patches > > <gcc-patches@gcc.gnu.org> wrote: > > > > > > On Wed, 9 Dec 2020 at 00:29, sunil.k.pandey <skpan...@sc.intel.com> wrote: > > > > > > > > On Linux/x86_64, > > > > > > > > 3a6e3ad38a17a03ee0139b49a0946e7b9ded1eb1 is the first bad commit > > > > commit 3a6e3ad38a17a03ee0139b49a0946e7b9ded1eb1 > > > > Author: Prathamesh Kulkarni <prathamesh.kulka...@linaro.org> > > > > Date: Tue Dec 8 14:30:04 2020 +0530 > > > > > > > > gimple-isel: Fold x CMP y ? -1 : 0 to x CMP y [PR97872] > > > > > > > > caused > > > > > > > > FAIL: gcc.target/i386/pr78102.c scan-assembler-times pcmpeqq 3 > > > Hi, > > > This is a known issue with the patch, and discussed here: > > > https://gcc.gnu.org/pipermail/gcc/2020-December/234438.html > > > I guess Hongtao will check in a fix for that soon. > > > > > > > According to https://uops.info/table.html, > > both pcmpeqq and pcmpeqd use only port 1, so i think there's no > > performance difference between > > > > vpcmpeqq %xmm1, %xmm0, %xmm0 > > vpxor %xmm1, %xmm1, %xmm1 > > vpcmpeqq %xmm1, %xmm0, %xmm0 > > > > and > > > > vpcmpeqq %xmm1, %xmm0, %xmm0 > > vpcmpeqd %xmm1, %xmm1, %xmm1 > > vpandn %xmm1, %xmm0, %xmm0 > > > > So fix up testcase as below. > > > > gcc/testsuite > > > > * gcc.target/i386/i386/pr78102.c: Adjust testcase. > > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > gcc/testsuite/gcc.target/i386/pr78102.c | 2 +- > > > > modified gcc/testsuite/gcc.target/i386/pr78102.c > > @@ -1,7 +1,7 @@ > > /* PR target/78102 */ > > /* { dg-do compile } */ > > /* { dg-options "-O2 -mno-sse4.2 -msse4.1" } */ > > -/* { dg-final { scan-assembler-times "pcmpeqq" 3 } } */ > > +/* { dg-final { scan-assembler-times "pcmpeq" 4 } } */ > > > > Ok for trunk? > Thanks for the fix! > Just a small nit - Should it be "pcmpeqq" rather than "pcmpeq" in the > dg-final line ? >
4 pcmpeq would cover both 4 pcmpeqq or 3 pcmpeqq + 1 pcmpeqd(original situation) > Thanks, > Prathamesh > > > > > > > > -- > > BR, > > Hongtao -- BR, Hongtao