[PATCH] D52677: [AMDGPU] Match v_swap_b32

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec updated this revision to Diff 170763.
rampitec added a comment.

Relaxed match to allow multiple uses of temp register and keep initial mov to temp:

mov t, x
mov x, y
mov y, t

>
=

mov t, x (t is potentially dead and move eliminated)
v_swap_b32 x, y


https://reviews.llvm.org/D52677

Files:
  lib/Target/AMDGPU/AMDGPUSubtarget.h
  lib/Target/AMDGPU/SIShrinkInstructions.cpp
  test/CodeGen/AMDGPU/v_swap_b32.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D52677.170763.patch
Type: text/x-patch
Size: 22841 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181023/8317bc71/attachment.bin>