[Parallel_libs-commits] [PATCH] D24951: Import/adapt the SLEEF vector math-function library as an LLVM runtime
Hal Finkel via Parallel_libs-commits
parallel_libs-commits at lists.llvm.org
Fri Oct 7 12:36:21 PDT 2016
hfinkel updated this revision to Diff 73970.
hfinkel added a comment.
Herald added a subscriber: modocache.
The library name has been updated to be vmath. The function names now all look like:
__llvm_<func>_<type>[_<abi>]
So, for example, for sin the library contains on my x86_64 system:
__llvm_sin_f64
__llvm_sin_u1_f64
__llvm_sin_f32
__llvm_sin_u1_f32
__llvm_sin_u1_v2f64_sse2
__llvm_sin_v2f64_sse2
__llvm_sin_u1_v4f32_sse2
__llvm_sin_v4f32_sse2
__llvm_sin_u1_v4f64_avx
__llvm_sin_v4f64_avx
__llvm_sin_u1_v8f32_avx
__llvm_sin_v8f32_avx
__llvm_sin_u1_v4f64_avx2
__llvm_sin_v4f64_avx2
__llvm_sin_u1_v8f32_avx2
__llvm_sin_v8f32_avx2
https://reviews.llvm.org/D24951
Files:
CMakeLists.txt
vmath/CMakeLists.txt
vmath/include/__vmath.def
vmath/include/vmath.h
vmath/lib/CMakeLists.txt
vmath/lib/avx.h
vmath/lib/avx2.h
vmath/lib/dd.h
vmath/lib/df.h
vmath/lib/dp-avx.cpp
vmath/lib/dp-avx2.cpp
vmath/lib/dp-scalar.cpp
vmath/lib/dp-sse2.cpp
vmath/lib/dp.cpp
vmath/lib/fma4.h
vmath/lib/isa.h
vmath/lib/neon.h
vmath/lib/nonnumber.h
vmath/lib/sp-avx.cpp
vmath/lib/sp-avx2.cpp
vmath/lib/sp-neon.cpp
vmath/lib/sp-scalar.cpp
vmath/lib/sp-sse2.cpp
vmath/lib/sp.cpp
vmath/lib/sse2.h
vmath/unittests/CMakeLists.txt
vmath/unittests/VMathTestDouble.cpp
vmath/unittests/VMathTestDoubleAVX.cpp
vmath/unittests/VMathTestDoubleAVX2.cpp
vmath/unittests/VMathTestDoubleSSE2.cpp
vmath/unittests/VMathTestSingle.cpp
vmath/unittests/VMathTestSingleAVX.cpp
vmath/unittests/VMathTestSingleAVX2.cpp
vmath/unittests/VMathTestSingleNEON.cpp
vmath/unittests/VMathTestSingleSSE2.cpp
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D24951.73970.patch
Type: text/x-patch
Size: 342827 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/parallel_libs-commits/attachments/20161007/9fd8081f/attachment-0001.bin>
More information about the Parallel_libs-commits
mailing list