projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 91c066a) | patch
R600/SI: Use V_FRACT_F64 for faster 64-bit floor on SI
authorMarek Olsak <marek.olsak@amd.com>
Tue, 24 Mar 2015 13:40:15 +0000 (13:40 +0000)
committerMarek Olsak <marek.olsak@amd.com>
Tue, 24 Mar 2015 13:40:15 +0000 (13:40 +0000)
commit3f05a5e0ad0982569148d39cac04a0c8b87ec62d
treeb966957040a13a53646f74b1886c283e537b90f4tree | snapshot
parent91c066ae153c135b79f8b330ec07e9b96dfd1a02commit | diff
R600/SI: Use V_FRACT_F64 for faster 64-bit floor on SI

Other f64 opcodes not supported on SI can be lowered in a similar way.

v2: use complex VOP3 patterns

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233076 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/R600/SIISelLowering.cpp diff | blob | history
lib/Target/R600/SIInstrInfo.cpp diff | blob | history
lib/Target/R600/SIInstrInfo.td diff | blob | history
lib/Target/R600/SIInstructions.td diff | blob | history
test/CodeGen/R600/ffloor.f64.ll diff | blob | history
test/CodeGen/R600/llvm.AMDGPU.fract.f64.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors