[Mlir-commits] [mlir] [MLIR][Linalg] Prevent PackOp canonicalization crash on zero tile factor (PR #185624)

[Renato Golin]

================
@@ -0,0 +1,23 @@
+// RUN: mlir-opt %s --inline --sccp --canonicalize -split-input-file -verify-diagnostics
+
+func.func @get_tile_size() -> index {
+  %c0 = arith.constant 0 : index
+  return %c0 : index
+}
+
+func.func private @use(%A: tensor<?x16x?x1xi32>)
+
+func.func @pack(%A: tensor<7x16xi32>) {
+  %c1 = arith.constant 1 : index
+  %pad_val = arith.constant 123 : i32
+  %tile_size = func.call @get_tile_size() : () -> index
+  %empty = tensor.empty(%c1, %tile_size) : tensor<?x16x?x1xi32>
+  // expected-error @below {{invalid zero tile factor}}
----------------
rengolin wrote:

You don't seem to be adding this error message, so it's coming from somewhere else. This PR does not look like the right fix. Please try to understand the problem and the fix first.

https://github.com/llvm/llvm-project/pull/185624