https://bugs.llvm.org/show_bug.cgi?id=23116
Sanjay Patel <[email protected]> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |FIXED
Status|NEW |RESOLVED
--- Comment #11 from Sanjay Patel <[email protected]> ---
After:
https://reviews.llvm.org/rG056d31dd2a04 (and several other patches as noted)
And with:
$ clang -O2 23116.c -S -o - -fno-signed-zeros -ffinite-math-only -mavx
We finally have the ideal output with the minimal relaxed FP settings:
vextractf128 $1, %ymm0, %xmm1
vmaxps %xmm1, %xmm0, %xmm0
vpermilpd $1, %xmm0, %xmm1 ## xmm1 = xmm0[1,0]
vmaxps %xmm1, %xmm0, %xmm0
vmovshdup %xmm0, %xmm1 ## xmm1 = xmm0[1,1,3,3]
vmaxss %xmm1, %xmm0, %xmm0
vzeroupper
retq
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
