projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 521c9dc) | patch
[PowerPC] Implement combineRepeatedFPDivisors
authorHal Finkel <hfinkel@anl.gov>
Mon, 24 Nov 2014 23:45:21 +0000 (23:45 +0000)
committerHal Finkel <hfinkel@anl.gov>
Mon, 24 Nov 2014 23:45:21 +0000 (23:45 +0000)
commit5d6f18565358decf831d9abe415e7380419d39da
tree145d3c48f96ffccfceac3f8a41013fc52ec250d2tree | snapshot
parent521c9dc7d807c7716d69f86a902b47100e34d508commit | diff
[PowerPC] Implement combineRepeatedFPDivisors

This does not matter on newer cores (where we can use reciprocal estimates in
fast-math mode anyway), but for older cores this allows us to generate better
fast-math code where we have multiple FDIVs with a common divisor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222710 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/PowerPC/PPCISelLowering.cpp diff | blob | history
lib/Target/PowerPC/PPCISelLowering.h diff | blob | history
test/CodeGen/PowerPC/fdiv-combine.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors