projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 845fe8a) | patch
[X86] Use __builtin_convertvector to implement _mm512_cvtps_pd.
authorCraig Topper <craig.topper@intel.com>
Mon, 14 May 2018 04:05:06 +0000 (04:05 +0000)
committerCraig Topper <craig.topper@intel.com>
Mon, 14 May 2018 04:05:06 +0000 (04:05 +0000)
commitdaaf105f8603050e036627324dd36f6882e081d1
tree3241d5b1899110a96e308cf63d6763d3dc3c2126tree | snapshot
parent845fe8a7087c68d8e88ac41229efa12c390d1491commit | diff
[X86] Use __builtin_convertvector to implement _mm512_cvtps_pd.

If we're using default rounding mode we can let __builtin_convertvector to generate an fpextend. This matches 128 and 256 bit.

If we're using the version that takes an explicit rounding mode argument we would need to look at the immediate to see if its CUR_DIRECTION.

llvm-svn: 332210
clang/lib/Headers/avx512fintrin.h diff | blob | history
clang/test/CodeGen/avx512f-builtins.c diff | blob | history
Domain: System / Toolchain;