[clang] [llvm] [llvm][AMDGPU] Fold `llvm.amdgcn.wavefrontsize` early (PR #114481)

[Joseph Huber via cfe-commits]

================
@@ -1024,6 +1024,15 @@ GCNTTIImpl::instCombineIntrinsic(InstCombiner &IC, IntrinsicInst &II) const {
     }
     break;
   }
+  case Intrinsic::amdgcn_wavefrontsize: {
+    // TODO: this is a workaround for the pseudo-generic target one gets with no
+    // specified mcpu, which spoofs its wave size to 64; it should be removed.
+    if ((ST->getCPU().empty() || ST->getCPU() == "generic") &&
+        !ST->getFeatureString().contains("+wavefrontsize"))
----------------
jhuber6 wrote:

I see this comment
```
  // Don't crash on invalid devices.
  if (WavefrontSizeLog2 == 0)
    WavefrontSizeLog2 = 5;
```
So might break something.

https://github.com/llvm/llvm-project/pull/114481