projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 4bacfe2) | patch
R600: Use new fmad node.
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 20 Feb 2015 22:10:41 +0000 (22:10 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 20 Feb 2015 22:10:41 +0000 (22:10 +0000)
commitbbb748eece06d7bc0e4dbf32cc8212157a4600f8
treefeae35216415db8c41d45e57649ea7bbc0c60a47tree | snapshot
parent4bacfe20954ad20364d01a39bb30c4a7a8b0079dcommit | diff
R600: Use new fmad node.

This enables a few useful combines that used to only
use fma.

Also since v_mad_f32 apparently does not support denormals,
disable the existing cases that are custom handled if they are
requested.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230071 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/R600/AMDGPUISelLowering.cpp diff | blob | history
lib/Target/R600/AMDGPUISelLowering.h diff | blob | history
lib/Target/R600/AMDGPUInstrInfo.td diff | blob | history
lib/Target/R600/AMDGPUInstructions.td diff | blob | history
lib/Target/R600/R600Instructions.td diff | blob | history
lib/Target/R600/SIISelLowering.cpp diff | blob | history
lib/Target/R600/SIInstructions.td diff | blob | history
test/CodeGen/R600/mad-combine.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors