projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 51be368) | patch
AMDGPU: Minor cleanups to help with f16 support
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 21 Aug 2015 23:49:51 +0000 (23:49 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 21 Aug 2015 23:49:51 +0000 (23:49 +0000)
commitfbb7ef0f362c72b34f2d35d202b5cd3c00fb6ae5
tree38f619b1dbc3039ed01909e73db8cd2a19eddc06tree | snapshot
parent51be368e3fb558e6d678db7c6e9f6840c1cf7aa5commit | diff
AMDGPU: Minor cleanups to help with f16 support

The main change is inverting the condition for the
operand class classes so that VT.Size == 16 uses VGPR_32
instead of 64.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@245764 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/SIInstrInfo.td diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors