commit 91de08f0782c7d2fbbb95c5b4089fd4e19790f5a
author Simon Pilgrim <llvm-dev@redking.me.uk> Tue Apr 16 19:18:53 2019 +0000
committer Simon Pilgrim <llvm-dev@redking.me.uk> Tue Apr 16 19:18:53 2019 +0000
tree 2b10f54f0612f4a70397cd34197a4775e36a2a60
parent b5612d0a25ac7497e1000c965126a43640afcf5e

[X86][AVX] X86ISD::PERMV/PERMV3 node types can never fold index ops

Improves codegen demonstrated by D60512 - instructions represented by X86ISD::PERMV/PERMV3 can never memory fold the operand used for their index register.

This patch updates the 'isUseOfShuffle' helper into the more capable 'isFoldableUseOfShuffle' that recognises that the op is used for a X86ISD::PERMV/PERMV3 index mask and can't be folded - allowing us to use broadcast/subvector-broadcast ops to reduce the size of the mask constant pool data.

Differential Revision: https://reviews.llvm.org/D60562

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@358516 91177308-0d34-0410-b5e6-96231b3b80d8