[llvm] [CGData][GMF] Skip No Params (PR #116548)

[Kyungwoo Lee via llvm-commits]

================
@@ -159,22 +169,33 @@ static bool isProfitable(
   unsigned InstCount = SFS[0]->InstCount;
   if (InstCount < GlobalMergingMinInstrs)
     return false;
+  double Benefit =
+      InstCount * (StableFunctionCount - 1) * GlobalMergingInstOverhead;
 
-  unsigned ParamCount = SFS[0]->IndexOperandHashMap->size();
-  if (ParamCount > GlobalMergingMaxParams)
-    return false;
-
-  unsigned Benefit = InstCount * (StableFunctionCount - 1);
-  unsigned Cost =
-      (GlobalMergingParamOverhead * ParamCount + GlobalMergingCallOverhead) *
-          StableFunctionCount +
-      GlobalMergingExtraThreshold;
+  double Cost = 0.0;
+  SmallSet<stable_hash, 8> UniqueHashVals;
+  for (auto &SF : SFS) {
+    UniqueHashVals.clear();
+    for (auto &[IndexPair, Hash] : *SF->IndexOperandHashMap)
+      UniqueHashVals.insert(Hash);
+    unsigned ParamCount = UniqueHashVals.size();
+    if (ParamCount > GlobalMergingMaxParams)
+      return false;
----------------
kyulee-com wrote:

Correct. To compute the actual parameters, we need to iterate over the set of candidates with the same hash. By reducing these candidates, we might obtain a smaller set of parameters. However, this adds complexity to repeat these steps until we find a smaller set. In addition, since the overall function hash still matches, we may create unprofitable merging instances that the linker does not fold. I would maintain the current approach for simplicity.

https://github.com/llvm/llvm-project/pull/116548