[Mlir-commits] [mlir] [mlir][affine] Make --affine-super-vectorize fail if no vector size is specified (PR #114530)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir-affine

Author: Clément Fournier (oowekyala)

<details>
<summary>Changes</summary>

Fix #<!-- -->114528 

---
Full diff: https://github.com/llvm/llvm-project/pull/114530.diff


2 Files Affected:

- (modified) mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp (+6) 
- (added) mlir/test/Dialect/Affine/SuperVectorize/invalid-no-size.mlir (+9) 


``````````diff
diff --git a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
index 6bb8dfecba0ec5..6a6c98e7544679 100644
--- a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
+++ b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
@@ -1763,6 +1763,12 @@ void Vectorize::runOnOperation() {
     return signalPassFailure();
   }
 
+  if (vectorSizes.empty()) {
+    f.emitError("Vector rank cannot be zero, specify pass option ")
+        << vectorSizes.getArgStr();
+    return signalPassFailure();
+  }
+
   if (vectorizeReductions && vectorSizes.size() != 1) {
     f.emitError("Vectorizing reductions is supported only for 1-D vectors.");
     return signalPassFailure();
diff --git a/mlir/test/Dialect/Affine/SuperVectorize/invalid-no-size.mlir b/mlir/test/Dialect/Affine/SuperVectorize/invalid-no-size.mlir
new file mode 100644
index 00000000000000..8805c4327c6ef3
--- /dev/null
+++ b/mlir/test/Dialect/Affine/SuperVectorize/invalid-no-size.mlir
@@ -0,0 +1,9 @@
+// RUN: mlir-opt %s -verify-diagnostics --affine-super-vectorize
+
+// expected-error at +1 {{Vector rank cannot be zero, specify pass option virtual-vector-size}} 
+func.func @with_zero_vector_size(%arg0: memref<21x12x12xi1>) {
+  affine.for %arg1 = 0 to 84 step 4294967295 {
+  }
+  return
+}
+

``````````

</details>


https://github.com/llvm/llvm-project/pull/114530