[llvm] [AArch64] Disconnect FeatureUseScalarIncVL from FeatureSVE2. (PR #114568)

David Green via llvm-commits llvm-commits at lists.llvm.org
Tue Nov 5 10:32:41 PST 2024


================
@@ -797,6 +794,9 @@ def FeatureUseFixedOverScalableIfEqualCost: SubtargetFeature<"use-fixed-over-sca
   "UseFixedOverScalableIfEqualCost", "true",
   "Prefer fixed width loop vectorization over scalable if the cost-model assigns equal costs">;
 
+def FeatureDontUseScalarIncVL : SubtargetFeature<"dont-use-scalar-inc-vl",
----------------
davemgreen wrote:

Yeah the string matching isn't the greatest. We only do it once though, and I believe this is the only way to check if a feature was specified. I assume that it was made a target feature so that we could add it to certain cpus if it is useful. I'm not 100% sure we won't find that would still be useful in some cases, but we have progressed a bit in time now and it hasn't come up yet.

https://github.com/llvm/llvm-project/pull/114568


More information about the llvm-commits mailing list