[PATCH] D75052: [MLIR][GPU] Properly model step in parallel loop to gpu conversion.

[Alex Zinenko via Phabricator via llvm-commits]

ftynse accepted this revision.
ftynse added inline comments.
This revision is now accepted and ready to land.


================
Comment at: mlir/lib/Conversion/LoopsToGPU/LoopsToGPU.cpp:537
   Value constantBound = {};
   if (AffineMinOp minOp =
           dyn_cast_or_null<AffineMinOp>(upperBound.getDefiningOp())) {
----------------
Nit: could you early-return instead?


================
Comment at: mlir/lib/Conversion/LoopsToGPU/LoopsToGPU.cpp:655
+          // affine expression ceilDiv (upperBound - lowerBound) step. We use
+          // affine here so that it composes nicely with the provided map.
+          AffineMap stepMap =
----------------
Nit "we use affine apply" ?


================
Comment at: mlir/test/Conversion/LoopsToGPU/parallel_loop.mlir:18
 
-// CHECK:       #map0 = affine_map<(d0) -> (d0)>
-// CHECK:       module {
+// CHECK:       #map0 = affine_map<()[s0, s1, s2] -> (s0 - s1 ceildiv s2)>
+// CHECK:       #map1 = affine_map<(d0)[s0, s1] -> (d0 * s0 + s1)>
----------------
Let's not match on attribute names here, i.e. prefer `#[[MAP0:.*]] = affine_map<...`


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75052/new/

https://reviews.llvm.org/D75052