[PATCH] Improve SSE4.1+ float extraction from V4SFmode

2011-09-19 Thread Jakub Jelinek
Hi! As mentioned in an earlier posting, for -msse4.1 and above on the gcc.target/i386/{sse2,avx}-extract-1.c testcases we generate inefficient code when trying to extract SFmode from V4SFmode, unless it is the element 0. The problem is that the sse4_1_extractps pattern matches, but doesn't have a

Re: [PATCH] Improve SSE4.1+ float extraction from V4SFmode

2011-09-19 Thread Richard Henderson
On 09/19/2011 04:33 AM, Jakub Jelinek wrote: * config/i386/sse.md (*sse4_1_extractps): Change into define_insn_and_split, add =x 0 n and =x x n alternatives and split them after reload. Ok. r~

Re: [PATCH] Improve SSE4.1+ float extraction from V4SFmode

2011-09-19 Thread Uros Bizjak
On Mon, Sep 19, 2011 at 1:33 PM, Jakub Jelinek ja...@redhat.com wrote: As mentioned in an earlier posting, for -msse4.1 and above on the gcc.target/i386/{sse2,avx}-extract-1.c testcases we generate inefficient code when trying to extract SFmode from V4SFmode, unless it is the element 0. The