[llvm] [X86] Use plain load/store instead of cmpxchg16b for atomics with AVX (PR #74275)
James Y Knight via llvm-commits
llvm-commits at lists.llvm.org
Mon Dec 4 06:40:54 PST 2023
================
@@ -3831,17 +3831,14 @@ void DAGTypeLegalizer::ExpandIntRes_XROUND_XRINT(SDNode *N, SDValue &Lo,
void DAGTypeLegalizer::ExpandIntRes_LOAD(LoadSDNode *N,
SDValue &Lo, SDValue &Hi) {
if (N->isAtomic()) {
- // It's typical to have larger CAS than atomic load instructions.
----------------
jyknight wrote:
I had thought the intent was to migrate only for 'unordered' atomics. @preames for input as original author of this code.
https://github.com/llvm/llvm-project/pull/74275
More information about the llvm-commits
mailing list