[PATCH] D27825: [AVX-512][InstCombine] Teach InstCombine to converted masked vpermv intrinsics into shufflevector instructions

Elena Demikhovsky via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Sat Dec 17 08:39:34 PST 2016

delena added inline comments.

Comment at: lib/Transforms/InstCombine/InstCombineCalls.cpp:1049
+  if (VWidth < 8) {
+    uint32_t Indices[4];
+    for (unsigned i = 0; i != VWidth; ++i)
do we have a good code at the end after all mask shuffles? 


More information about the llvm-commits mailing list