[llvm] [AMDGPU][True16][CodeGen] flat/global/scratch load/store pseudo for true16 (PR #127945)

Brox Chen via llvm-commits llvm-commits at lists.llvm.org
Fri Feb 21 10:26:53 PST 2025


================
@@ -255,6 +255,12 @@ class FLAT_Store_Pseudo <string opName, RegisterClass vdataClass,
   let enabled_saddr = EnableSaddr;
 }
 
+multiclass FLAT_Store_Pseudo_t16<string opName> {
+  def "" : FLAT_Store_Pseudo<opName, VGPR_32>;
+  let OtherPredicates = [HasTrue16BitInsts] in
----------------
broxigarchen wrote:

seems the predicate class is not compatible here. reverted to previous version

https://github.com/llvm/llvm-project/pull/127945


More information about the llvm-commits mailing list