[mlir][vector] Skip uniform vectorization for non scalar type #128294
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@llvm/pr-subscribers-mlir-affine @llvm/pr-subscribers-mlir Author: Kai Sasaki (Lewuathe) ChangesThe code to vectorize the uniform type effectively assume the scalar type such as integer, float and index type. We should not regard other types as vectorizable uniform type because the following vectorization cannot treat the type properlty for now. Fixes #128277 Full diff: https://github.com/llvm/llvm-project/pull/128294.diff 2 Files Affected:
diff --git a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
index 71e9648a5e00f..7e69bce34aea4 100644
--- a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
+++ b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
@@ -1106,6 +1106,10 @@ static bool isUniformDefinition(Value value,
if (!loop.isDefinedOutsideOfLoop(value))
return false;
}
+
+ if (!isa<IndexType, IntegerType, FloatType>(value.getType()))
+ return false;
+
return true;
}
diff --git a/mlir/test/Dialect/Affine/SuperVectorize/vectorize_1d.mlir b/mlir/test/Dialect/Affine/SuperVectorize/vectorize_1d.mlir
index 9244604128cb7..388d9ff423626 100644
--- a/mlir/test/Dialect/Affine/SuperVectorize/vectorize_1d.mlir
+++ b/mlir/test/Dialect/Affine/SuperVectorize/vectorize_1d.mlir
@@ -682,5 +682,21 @@ func.func @vec_vecdim_reduction_rejected(%in: memref<256x512xf32>, %out: memref<
return
}
+// -----
+
+// Non scalar type is not regarded as the uniform in the vectorization
+
// CHECK-LABEL: @vec_vecdim_reduction_rejected
// CHECK-NOT: vector
+
+func.func @vec_non_scalar_type() {
+ %idx0 = index.constant 0
+ %alloc_82 = memref.alloc() : memref<1xi64>
+ affine.for %arg0 = 0 to 78 {
+ %dim_191 = memref.dim %alloc_82, %idx0 : memref<1xi64>
+ }
+ return
+}
+
+// CHECK-LABEL: @vec_non_scalar_type
+// CHECK-NOT: vector
|
CoTinker
reviewed
Feb 22, 2025
CoTinker
reviewed
Feb 22, 2025
The code to vectorize the uniform type effectively assume the scalar type such as integer, float and index type. We should not regard other types as vectorizable uniform type because the following vectorization cannot treat the type properlty for now. See: llvm#128277
Lewuathe
force-pushed
the
skip-vecortization-for-non-scalar-type
branch
from
05ceaf0 to
f67d451
Compare
|
@CoTinker Thank you for the review! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The code to vectorize the uniform type effectively assume the scalar type such as integer, float and index type. We should not regard other types as vectorizable uniform type because the following vectorization cannot treat the type properlty for now.
Fixes #128277