diff options
author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2014-10-21 23:00:20 +0000 |
---|---|---|
committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2014-10-21 23:00:20 +0000 |
commit | d6511b49ace8e3d55ad43beb2cb142022a8032cc (patch) | |
tree | ec95b0252ce02c2c5f0445bdbdb53d0081bd9f4c /llvm/lib/Transforms/Vectorize/BBVectorize.cpp | |
parent | 44e5b4e533a92ff44ae57cf0129398b162e82bf6 (diff) | |
download | bcm5719-llvm-d6511b49ace8e3d55ad43beb2cb142022a8032cc.tar.gz bcm5719-llvm-d6511b49ace8e3d55ad43beb2cb142022a8032cc.zip |
Add minnum / maxnum intrinsics
These are named following the IEEE-754 names for these
functions, rather than the libm fmin / fmax to avoid
possible ambiguities. Some languages may implement something
resembling fmin / fmax which return NaN if either operand is
to propagate errors. These implement the IEEE-754 semantics
of returning the other operand if either is a NaN representing
missing data.
llvm-svn: 220341
Diffstat (limited to 'llvm/lib/Transforms/Vectorize/BBVectorize.cpp')
-rw-r--r-- | llvm/lib/Transforms/Vectorize/BBVectorize.cpp | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/llvm/lib/Transforms/Vectorize/BBVectorize.cpp b/llvm/lib/Transforms/Vectorize/BBVectorize.cpp index 50c3fa41b1d..b4991bc497a 100644 --- a/llvm/lib/Transforms/Vectorize/BBVectorize.cpp +++ b/llvm/lib/Transforms/Vectorize/BBVectorize.cpp @@ -685,6 +685,8 @@ namespace { case Intrinsic::trunc: case Intrinsic::floor: case Intrinsic::fabs: + case Intrinsic::minnum: + case Intrinsic::maxnum: return Config.VectorizeMath; case Intrinsic::bswap: case Intrinsic::ctpop: |