[llvm] [AMDGPU] Classify FLAT instructions as VMEM (PR #137148)

Robert Imschweiler via llvm-commits llvm-commits at lists.llvm.org
Tue May 6 04:01:51 PDT 2025


================
@@ -2420,29 +2420,27 @@ bool SchedGroup::canAddMI(const MachineInstr &MI) const {
     Result = true;
 
   else if (((SGMask & SchedGroupMask::VMEM) != SchedGroupMask::NONE) &&
-           (TII->isVMEM(MI) || (TII->isFLAT(MI) && !TII->isDS(MI))))
+           TII->isVMEM(MI))
----------------
ro-i wrote:

Otherwise, it should check for `!isLDSDMA` if we specifically want to avoid DS accesses, I think

https://github.com/llvm/llvm-project/pull/137148


More information about the llvm-commits mailing list