projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 82d7608) | patch
[InstCombine][AVX-512] Teach InstCombineCalls how to handle the intrinsics for variab...
authorCraig Topper <craig.topper@gmail.com>
Fri, 18 Nov 2016 06:04:33 +0000 (06:04 +0000)
committerCraig Topper <craig.topper@gmail.com>
Fri, 18 Nov 2016 06:04:33 +0000 (06:04 +0000)
commit1de753f7f547faebedc41b172bfa7dc33765dc79
treebbe621164b6faa24e1faa52fdfc0ebb2c0e1d421tree | snapshot
parent82d760879f62430d7969fac435908ed9d754bcc5commit | diff
[InstCombine][AVX-512] Teach InstCombineCalls how to handle the intrinsics for variable shift with 16-bit elements.

This is a straightforward extension of the existing support for 32/64-bit element types. Just needed to add the additional instrinsics to the switches.

llvm-svn: 287316
llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp diff | blob | history
llvm/test/Transforms/InstCombine/x86-vector-shifts.ll diff | blob | history
Domain: System / Toolchain;