[PATCH] D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops.

Han-Chung Wang via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Feb 18 14:53:14 PST 2020


hanchung added a comment.

Just few formatting nits while going through the examples.



================
Comment at: mlir/test/Dialect/Linalg/loops.mlir:365
+  affine_map<(i, j) -> (0)>,
+  affine_map<(i,j) -> (i,j)>
+]
----------------
Add a space after comma


================
Comment at: mlir/test/Dialect/Linalg/loops.mlir:376
+
+func @generic_op_zero_rank(%arg0 : memref<f32>, %arg1:  memref<3x4xf32>)
+{
----------------
Remove the space right after %arg0.
And there are two spaces after "%arg1:", remove one.


================
Comment at: mlir/test/Dialect/Linalg/loops.mlir:379
+  linalg.generic #trait_broadcast %arg0, %arg1 {
+    ^bb(%a: f32, %b : f32) :
+      linalg.yield %a : f32
----------------
s/%b :/ %b:


================
Comment at: mlir/test/Dialect/Linalg/roundtrip.mlir:350
+  affine_map<(i, j) -> (0)>,
+  affine_map<(i,j) -> (i,j)>
+]
----------------
Add a space after comma.


================
Comment at: mlir/test/Dialect/Linalg/roundtrip.mlir:361
+
+func @generic_op_zero_rank(%arg0 : tensor<f32>) ->  (tensor<3x4xf32>)
+{
----------------
Remove the space between "%arg0" and ":".


================
Comment at: mlir/test/Dialect/Linalg/roundtrip.mlir:361
+
+func @generic_op_zero_rank(%arg0 : tensor<f32>) ->  (tensor<3x4xf32>)
+{
----------------
hanchung wrote:
> Remove the space between "%arg0" and ":".
There are two spaces after "->", remove one.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74638/new/

https://reviews.llvm.org/D74638





More information about the llvm-commits mailing list