]> granicus.if.org Git - clang/commit
projects / clang / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 41e20a3) | patch
[X86] Rewrite the add/mul/or/and reduction intrinsics to make better use of other...
authorCraig Topper <craig.topper@intel.com>
Thu, 21 Jun 2018 16:41:28 +0000 (16:41 +0000)
committerCraig Topper <craig.topper@intel.com>
Thu, 21 Jun 2018 16:41:28 +0000 (16:41 +0000)
commit0be86b8c934425669a4da497cbc08425370fda0a
tree12a2d324bee1b0856206bea2bf0a9a996f055c11tree | snapshot
parent41e20a3f297fbf794689979fad39a2cc77e5aafbcommit | diff
[X86] Rewrite the add/mul/or/and reduction intrinsics to make better use of other intrinsics and remove undef shuffle indices.

Similar to what was done to max/min recently.

These already reduced the vector width to 256 and 128 bit as we go unlike the original max/min code.

Differential Revision: https://reviews.llvm.org/D48346

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@335253 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Headers/avx512fintrin.h diff | blob | history
test/CodeGen/avx512-reduceIntrin.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.