projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c36a8d2) | patch
LoopVectorize: Math functions only read rounding mode
authorArnold Schwaighofer <aschwaighofer@apple.com>
Mon, 1 Jul 2013 00:54:44 +0000 (00:54 +0000)
committerArnold Schwaighofer <aschwaighofer@apple.com>
Mon, 1 Jul 2013 00:54:44 +0000 (00:54 +0000)
commitc14380d1953acd5a735e1d8bcf18d0f96876da74
treeb10d536e874fee69c772bf198cfae53df04fa311tree | snapshot
parentc36a8d2b3d6e543de8d9f210ecd39a9a0641d826commit | diff
LoopVectorize: Math functions only read rounding mode

Math functions are mark as readonly because they read the floating point
rounding mode. Because we don't vectorize loops that would contain function
calls that set the rounding mode it is safe to ignore this memory read.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185299 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/Vectorize/LoopVectorize.cpp diff | blob | history
test/Transforms/LoopVectorize/funcall.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors