tree 02ec7f59d9b33287390f6b1871d15cbee0a6a178
parent 3e53c14be52f1a5c36952b3a3861202e3f201fc6
author Philip Reames <listmail@philipreames.com> 1550176877 +0000
committer Philip Reames <listmail@philipreames.com> 1550176877 +0000

Canonicalize all integer "idempotent" atomicrmw ops

For "idempotent" atomicrmw instructions which we can't simply turn into load, canonicalize the operation and constant. This reduces the matching needed elsewhere in the optimizer, but doesn't directly impact codegen.

For any architecture where OR/Zero is not a good default choice, you can extend the AtomicExpand lowerIdempotentRMWIntoFencedLoad mechanism. I reviewed X86 to make sure this works well, haven't audited other backends.

Differential Revision: https://reviews.llvm.org/D58244


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@354058 91177308-0d34-0410-b5e6-96231b3b80d8