[llvm-bugs] [Bug 21138] New vector shuffle lowering produces an insertf128 + blend instead of a single insertf128

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=21138

Simon Pilgrim <llvm-dev at redking.me.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Fixed By Commit(s)|                            |339818
             Status|REOPENED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #9 from Simon Pilgrim <llvm-dev at redking.me.uk> ---
This was fixed by rL339818

AVX1:
PR21138: # @PR21138
  vperm2f128 $49, %ymm1, %ymm0, %ymm2 # ymm2 = ymm0[2,3],ymm1[2,3]
  vinsertf128 $1, %xmm1, %ymm0, %ymm0
  vshufps $221, %ymm2, %ymm0, %ymm0 # ymm0 =
ymm0[1,3],ymm2[1,3],ymm0[5,7],ymm2[5,7]
  retq

AVX2:
PR21138: # @PR21138
  vshufps $221, %ymm1, %ymm0, %ymm0 # ymm0 =
ymm0[1,3],ymm1[1,3],ymm0[5,7],ymm1[5,7]
  vpermpd $216, %ymm0, %ymm0 # ymm0 = ymm0[0,2,1,3]
  retq

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20180818/4de51e99/attachment.html>