[llvm] [AMDGPU] Enable atomic optimizer for divergent i64 and double values (PR #96934)

[Jay Foad via llvm-commits]

================
@@ -755,7 +775,7 @@ void AMDGPUAtomicOptimizerImpl::optimizeAtomic(Instruction &I,
         // of each active lane in the wavefront. This will be our new value
         // which we will provide to the atomic operation.
         Value *const LastLaneIdx = B.getInt32(ST->getWavefrontSize() - 1);
-        assert(TyBitWidth == 32);
+        assert(TyBitWidth == 32 || TyBitWidth == 64);
----------------
jayfoad wrote:

What's the point of having this assertion in this place? Doesn't the whole pass assume 32/64-bit types?

https://github.com/llvm/llvm-project/pull/96934