[llvm] [AArch64] Combine getActiveLaneMask with vector_extract (PR #81139)
David Sherwood via llvm-commits
llvm-commits at lists.llvm.org
Tue Apr 9 08:33:52 PDT 2024
================
@@ -1873,8 +1873,8 @@ void AArch64TargetLowering::addTypeForNEON(MVT VT) {
bool AArch64TargetLowering::shouldExpandGetActiveLaneMask(EVT ResVT,
EVT OpVT) const {
- // Only SVE has a 1:1 mapping from intrinsic -> instruction (whilelo).
- if (!Subtarget->hasSVE())
+ // Only SVE/SME has a 1:1 mapping from intrinsic -> instruction (whilelo).
+ if (!Subtarget->hasSVEorSME())
----------------
david-arm wrote:
Can you revert this change for now due to a missing capability in the backend? Basically, the SME feature doesn't necessarily imply that SVE is always available. You could add a TODO here that we should also enable this for SME with SVE.
https://github.com/llvm/llvm-project/pull/81139
More information about the llvm-commits
mailing list