[Mlir-commits] [mlir] [mlir][affine][gpu] support unroll dynamic value and apply it to gpu.thread_id op (PR #128113)

[Uday Bondhugula]

================
@@ -43,6 +43,10 @@ void getTripCountMapAndOperands(AffineForOp forOp, AffineMap *map,
 /// constant trip count in non-trivial cases.
 std::optional<uint64_t> getConstantTripCount(AffineForOp forOp);
 
+/// In some scenarios, such as GPU, the number of trip of each thread in the
+/// loop is inconsistent. This function returns the maximum number of trip.
----------------
bondhugula wrote:

A doc comment should describe the utility generically instead of starting a description specific scenarios. Start doc comment with  "Returns ....". 

`getUpperBoundOnTripCount` appears to be a clearer name.

https://github.com/llvm/llvm-project/pull/128113