[llvm] [X86] Use plain load/store instead of cmpxchg16b for atomics with AVX (PR #74275)

[Philip Reames via llvm-commits]

================
@@ -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.
----------------
preames wrote:

Cleanup completed in 943f3e52a0532d1d2b5c743635e1aed15033154b and a0dde7b8d16842005d6f5247dda6f17c77a5906f

https://github.com/llvm/llvm-project/pull/74275