> What do you mean no tests require it? For instance, all of the ones > that currently pass with with vect_perm? Current implementation of vect_perm doesn't check for SSSE3 - so any x86 target is supposed to support permutation.
> Just leave vect_perm alone for now. That may not be absolutely > correct either, but it's the good temporary solution that involves > the minimal amount of churn. Ok, those were just attempts to adjust dg-scans in slp-perm-9.c, in which one more loop was vectorized when compiled with -mavx2. In fact, just SSSE3 isn't enough for vectorization of this loop - it seems that vector size also matters, so I undid changes in vect_perm and just add a vect-size check to the test - could you please check if the changes are ok? On 17 December 2011 02:17, Richard Henderson <r...@redhat.com> wrote: > On 12/16/2011 09:44 AM, Michael Zolotukhin wrote: >>> Why? SSSE3 *really can* do arbitrary permutation. If you say that >>> isn't enough, then there's something wrong in the back end, and all >>> you're doing is papering over a bug. >> Yes, sure, SSSE3 is enough for that. I checked for AVX in vect_perm >> just because there is no check for SSSE3 and no tests currently >> require exactly SSSE3 (without AVX) - that's not absolutely correct >> and I suggest it as a temporary solution - we can weaken this check >> when a real case will be met (in which SSSE3 is needed without AVX). > > What do you mean no tests require it? For instance, all of the ones > that currently pass with with vect_perm? > > Just leave vect_perm alone for now. That may not be absolutely > correct either, but it's the good temporary solution that involves > the minimal amount of churn. > > > r~ -- --- Best regards, Michael V. Zolotukhin, Software Engineer Intel Corporation.
vec-tests-avx2_fixes-6.patch
Description: Binary data
