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

[Uday Bondhugula]

================
@@ -84,6 +85,67 @@ void mlir::affine::getTripCountMapAndOperands(
                             tripCountValueMap.getOperands().end());
 }
 
+/// Replace thread_id with its maximum value, if `replaceWithZero` is true,
+/// thread_id will be replaced by its minimum value 0.
+static void replaceGPUOperands(AffineForOp forOp,
+                               SmallVectorImpl<Value> &operands,
+                               SmallVectorImpl<AffineExpr> &symReplacements,
+                               unsigned numDim, bool replaceWithZero = false) {
+  auto launchOp = forOp->getParentOfType<gpu::LaunchOp>();
+  if (!launchOp)
+    return;
----------------
bondhugula wrote:

This is very specific to gpu.launch and shouldn't be living here. In fact, the approach itself isn't general enough, that's necessitating this kind of transformation.

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