[llvm] AMDGPU: Add codegen for atomicrmw operations usub_cond and usub_sat (PR #141068)

[Matt Arsenault via llvm-commits]

================
@@ -1089,7 +1089,31 @@ multiclass DSAtomicRetNoRetPat_mc<DS_Pseudo inst, DS_Pseudo noRetInst,
   }
 }
 
+multiclass DSAtomicRetNoRetPatCondSub_mc<DS_Pseudo inst, DS_Pseudo noRetInst,
+                                         ValueType vt, string frag> {
+  let OtherPredicates = [LDSRequiresM0Init] in {
+    def : DSAtomicRetPat<inst, vt,
+                         !cast<PatFrag>(frag#"_local_m0_"#vt)>;
+    def : DSAtomicRetPat<noRetInst, vt,
+                         !cast<PatFrag>(frag#"_local_m0_noret_"#vt), /* complexity */ 1>;
+  }
+
+  let OtherPredicates = [NotLDSRequiresM0Init] in {
+    def : DSAtomicRetPat<!cast<DS_Pseudo>(!cast<string>(inst)#"_gfx9"), vt,
+                         !cast<PatFrag>(frag#"_local_"#vt)>;
+    def : DSAtomicRetPat<!cast<DS_Pseudo>(!cast<string>(noRetInst)#"_gfx9"), vt,
+                         !cast<PatFrag>(frag#"_local_noret_"#vt), /* complexity */ 1>;
+  }
 
+  let OtherPredicates = [HasGDS] in {
----------------
arsenm wrote:

GDS case isn't tested, but GDS atomic support is mostly missing for every other operator 

https://github.com/llvm/llvm-project/pull/141068