[Mlir-commits] [mlir] [mlir][affine] re-land implement `promoteIfSingleIteration` for `AffineForOp` (PR #72805)

[llvmlistbot at llvm.org]

================
@@ -465,6 +469,23 @@ AffineForOp getForInductionVarOwner(Value val);
 /// AffineParallelOp.
 AffineParallelOp getAffineParallelInductionVarOwner(Value val);
 
+/// Helper to replace uses of loop carried values (iter_args) and loop
+/// yield values while promoting single iteration affine.for ops.
+void replaceIterArgsAndYieldResults(AffineForOp forOp);
+
+/// Returns the trip count of the loop as an affine expression if the latter is
+/// expressible as an affine expression, and nullptr otherwise. The trip count
+/// expression is simplified before returning. This method only utilizes map
+/// composition to construct lower and upper bounds before computing the trip
+/// count expressions.
+void getTripCountMapAndOperands(AffineForOp forOp, AffineMap *tripCountMap,
+                                SmallVectorImpl<Value> *tripCountOperands);
+
+/// Returns the trip count of the loop if it's a constant, std::nullopt
+/// otherwise. This uses affine expression analysis and is able to determine
+/// constant trip count in non-trivial cases.
+std::optional<uint64_t> getConstantTripCount(AffineForOp forOp);
+
----------------
srcarroll wrote:

although it would be orthogonal to this PR, it may be worth thinking about moving these to `AffineForOp` methods since they are being moved to the dialect anyway.  Just a thought, no action necessary.

https://github.com/llvm/llvm-project/pull/72805