[mlir][vector][spirv] Lower vector.to_elements to SPIR-V
#146618
Merged
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
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
Member
|
@llvm/pr-subscribers-mlir-spirv Author: Eric Feng (efric) ChangesImplement Addresses 145929. Full diff: https://github.com/llvm/llvm-project/pull/146618.diff 2 Files Affected:
diff --git a/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp b/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
index de2af69eba9ec..475fd76c667e6 100644
--- a/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
+++ b/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
@@ -1022,6 +1022,51 @@ struct VectorStepOpConvert final : OpConversionPattern<vector::StepOp> {
}
};
+struct VectorToElementOpConvert final
+ : public OpConversionPattern<vector::ToElementsOp> {
+ using OpConversionPattern::OpConversionPattern;
+
+ LogicalResult
+ matchAndRewrite(vector::ToElementsOp toElementsOp, OpAdaptor adaptor,
+ ConversionPatternRewriter &rewriter) const override {
+
+ Type srcType =
+ getTypeConverter()->convertType(toElementsOp.getSource().getType());
+ if (!srcType)
+ return failure();
+
+ SmallVector<Value> results(toElementsOp->getNumResults());
+ Location loc = toElementsOp.getLoc();
+
+ // Input vectors of size 1 are converted to scalars by the type converter.
+ // We cannot use `spirv::CompositeExtractOp` directly in this case.
+ // For a scalar source, the result is just the scalar itself.
+ if (isa<spirv::ScalarType>(adaptor.getSource().getType())) {
+ results[0] = adaptor.getSource();
+ rewriter.replaceOp(toElementsOp, results);
+ return success();
+ }
+
+ for (auto [idx, element] : llvm::enumerate(toElementsOp.getElements())) {
+ // Create an CompositeExtract operation only for results that are not
+ // dead.
+ if (element.use_empty())
+ continue;
+
+ auto elementType = getTypeConverter()->convertType(element.getType());
+ if (!elementType)
+ return failure();
+ Value result = rewriter.create<spirv::CompositeExtractOp>(
+ loc, elementType, adaptor.getSource(),
+ rewriter.getI32ArrayAttr({static_cast<int32_t>(idx)}));
+ results[idx] = result;
+ }
+
+ rewriter.replaceOp(toElementsOp, results);
+ return success();
+ }
+};
+
} // namespace
#define CL_INT_MAX_MIN_OPS \
spirv::CLUMaxOp, spirv::CLUMinOp, spirv::CLSMaxOp, spirv::CLSMinOp
@@ -1038,7 +1083,7 @@ void mlir::populateVectorToSPIRVPatterns(
VectorBitcastConvert, VectorBroadcastConvert,
VectorExtractElementOpConvert, VectorExtractOpConvert,
VectorExtractStridedSliceOpConvert, VectorFmaOpConvert<spirv::GLFmaOp>,
- VectorFmaOpConvert<spirv::CLFmaOp>, VectorFromElementsOpConvert,
+ VectorFmaOpConvert<spirv::CLFmaOp>, VectorFromElementsOpConvert, VectorToElementOpConvert,
VectorInsertElementOpConvert, VectorInsertOpConvert,
VectorReductionPattern<GL_INT_MAX_MIN_OPS>,
VectorReductionPattern<CL_INT_MAX_MIN_OPS>,
diff --git a/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir b/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
index 4701ac5d96009..99ab0e1dc4eef 100644
--- a/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
+++ b/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
@@ -246,6 +246,41 @@ func.func @extract_dynamic_cst(%arg0 : vector<4xf32>) -> f32 {
// -----
+// CHECK-LABEL: func.func @to_elements_one_element
+// CHECK-SAME: %[[A:.*]]: vector<1xf32>)
+// CHECK: %[[ELEM0:.*]] = builtin.unrealized_conversion_cast %[[A]] : vector<1xf32> to f32
+// CHECK: return %[[ELEM0]] : f32
+func.func @to_elements_one_element(%a: vector<1xf32>) -> (f32) {
+ %0:1 = vector.to_elements %a : vector<1xf32>
+ return %0#0 : f32
+}
+
+// CHECK-LABEL: func.func @to_elements_no_dead_elements
+// CHECK-SAME: %[[A:.*]]: vector<4xf32>)
+// CHECK: %[[ELEM0:.*]] = spirv.CompositeExtract %[[A]][0 : i32] : vector<4xf32>
+// CHECK: %[[ELEM1:.*]] = spirv.CompositeExtract %[[A]][1 : i32] : vector<4xf32>
+// CHECK: %[[ELEM2:.*]] = spirv.CompositeExtract %[[A]][2 : i32] : vector<4xf32>
+// CHECK: %[[ELEM3:.*]] = spirv.CompositeExtract %[[A]][3 : i32] : vector<4xf32>
+// CHECK: return %[[ELEM0]], %[[ELEM1]], %[[ELEM2]], %[[ELEM3]] : f32, f32, f32, f32
+func.func @to_elements_no_dead_elements(%a: vector<4xf32>) -> (f32, f32, f32, f32) {
+ %0:4 = vector.to_elements %a : vector<4xf32>
+ return %0#0, %0#1, %0#2, %0#3 : f32, f32, f32, f32
+}
+
+// CHECK-LABEL: func.func @to_elements_dead_elements
+// CHECK-SAME: %[[A:.*]]: vector<4xf32>)
+// CHECK-NOT: spirv.CompositeExtract %[[A]][0 : i32]
+// CHECK: %[[ELEM1:.*]] = spirv.CompositeExtract %[[A]][1 : i32] : vector<4xf32>
+// CHECK-NOT: spirv.CompositeExtract %[[A]][2 : i32]
+// CHECK: %[[ELEM3:.*]] = spirv.CompositeExtract %[[A]][3 : i32] : vector<4xf32>
+// CHECK: return %[[ELEM1]], %[[ELEM3]] : f32, f32
+func.func @to_elements_dead_elements(%a: vector<4xf32>) -> (f32, f32) {
+ %0:4 = vector.to_elements %a : vector<4xf32>
+ return %0#1, %0#3 : f32, f32
+}
+
+// -----
+
// CHECK-LABEL: @from_elements_0d
// CHECK-SAME: %[[ARG0:.+]]: f32
// CHECK: %[[RETVAL:.+]] = builtin.unrealized_conversion_cast %[[ARG0]]
|
Member
|
@llvm/pr-subscribers-mlir Author: Eric Feng (efric) ChangesImplement Addresses 145929. Full diff: https://github.com/llvm/llvm-project/pull/146618.diff 2 Files Affected:
diff --git a/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp b/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
index de2af69eba9ec..475fd76c667e6 100644
--- a/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
+++ b/mlir/lib/Conversion/VectorToSPIRV/VectorToSPIRV.cpp
@@ -1022,6 +1022,51 @@ struct VectorStepOpConvert final : OpConversionPattern<vector::StepOp> {
}
};
+struct VectorToElementOpConvert final
+ : public OpConversionPattern<vector::ToElementsOp> {
+ using OpConversionPattern::OpConversionPattern;
+
+ LogicalResult
+ matchAndRewrite(vector::ToElementsOp toElementsOp, OpAdaptor adaptor,
+ ConversionPatternRewriter &rewriter) const override {
+
+ Type srcType =
+ getTypeConverter()->convertType(toElementsOp.getSource().getType());
+ if (!srcType)
+ return failure();
+
+ SmallVector<Value> results(toElementsOp->getNumResults());
+ Location loc = toElementsOp.getLoc();
+
+ // Input vectors of size 1 are converted to scalars by the type converter.
+ // We cannot use `spirv::CompositeExtractOp` directly in this case.
+ // For a scalar source, the result is just the scalar itself.
+ if (isa<spirv::ScalarType>(adaptor.getSource().getType())) {
+ results[0] = adaptor.getSource();
+ rewriter.replaceOp(toElementsOp, results);
+ return success();
+ }
+
+ for (auto [idx, element] : llvm::enumerate(toElementsOp.getElements())) {
+ // Create an CompositeExtract operation only for results that are not
+ // dead.
+ if (element.use_empty())
+ continue;
+
+ auto elementType = getTypeConverter()->convertType(element.getType());
+ if (!elementType)
+ return failure();
+ Value result = rewriter.create<spirv::CompositeExtractOp>(
+ loc, elementType, adaptor.getSource(),
+ rewriter.getI32ArrayAttr({static_cast<int32_t>(idx)}));
+ results[idx] = result;
+ }
+
+ rewriter.replaceOp(toElementsOp, results);
+ return success();
+ }
+};
+
} // namespace
#define CL_INT_MAX_MIN_OPS \
spirv::CLUMaxOp, spirv::CLUMinOp, spirv::CLSMaxOp, spirv::CLSMinOp
@@ -1038,7 +1083,7 @@ void mlir::populateVectorToSPIRVPatterns(
VectorBitcastConvert, VectorBroadcastConvert,
VectorExtractElementOpConvert, VectorExtractOpConvert,
VectorExtractStridedSliceOpConvert, VectorFmaOpConvert<spirv::GLFmaOp>,
- VectorFmaOpConvert<spirv::CLFmaOp>, VectorFromElementsOpConvert,
+ VectorFmaOpConvert<spirv::CLFmaOp>, VectorFromElementsOpConvert, VectorToElementOpConvert,
VectorInsertElementOpConvert, VectorInsertOpConvert,
VectorReductionPattern<GL_INT_MAX_MIN_OPS>,
VectorReductionPattern<CL_INT_MAX_MIN_OPS>,
diff --git a/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir b/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
index 4701ac5d96009..99ab0e1dc4eef 100644
--- a/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
+++ b/mlir/test/Conversion/VectorToSPIRV/vector-to-spirv.mlir
@@ -246,6 +246,41 @@ func.func @extract_dynamic_cst(%arg0 : vector<4xf32>) -> f32 {
// -----
+// CHECK-LABEL: func.func @to_elements_one_element
+// CHECK-SAME: %[[A:.*]]: vector<1xf32>)
+// CHECK: %[[ELEM0:.*]] = builtin.unrealized_conversion_cast %[[A]] : vector<1xf32> to f32
+// CHECK: return %[[ELEM0]] : f32
+func.func @to_elements_one_element(%a: vector<1xf32>) -> (f32) {
+ %0:1 = vector.to_elements %a : vector<1xf32>
+ return %0#0 : f32
+}
+
+// CHECK-LABEL: func.func @to_elements_no_dead_elements
+// CHECK-SAME: %[[A:.*]]: vector<4xf32>)
+// CHECK: %[[ELEM0:.*]] = spirv.CompositeExtract %[[A]][0 : i32] : vector<4xf32>
+// CHECK: %[[ELEM1:.*]] = spirv.CompositeExtract %[[A]][1 : i32] : vector<4xf32>
+// CHECK: %[[ELEM2:.*]] = spirv.CompositeExtract %[[A]][2 : i32] : vector<4xf32>
+// CHECK: %[[ELEM3:.*]] = spirv.CompositeExtract %[[A]][3 : i32] : vector<4xf32>
+// CHECK: return %[[ELEM0]], %[[ELEM1]], %[[ELEM2]], %[[ELEM3]] : f32, f32, f32, f32
+func.func @to_elements_no_dead_elements(%a: vector<4xf32>) -> (f32, f32, f32, f32) {
+ %0:4 = vector.to_elements %a : vector<4xf32>
+ return %0#0, %0#1, %0#2, %0#3 : f32, f32, f32, f32
+}
+
+// CHECK-LABEL: func.func @to_elements_dead_elements
+// CHECK-SAME: %[[A:.*]]: vector<4xf32>)
+// CHECK-NOT: spirv.CompositeExtract %[[A]][0 : i32]
+// CHECK: %[[ELEM1:.*]] = spirv.CompositeExtract %[[A]][1 : i32] : vector<4xf32>
+// CHECK-NOT: spirv.CompositeExtract %[[A]][2 : i32]
+// CHECK: %[[ELEM3:.*]] = spirv.CompositeExtract %[[A]][3 : i32] : vector<4xf32>
+// CHECK: return %[[ELEM1]], %[[ELEM3]] : f32, f32
+func.func @to_elements_dead_elements(%a: vector<4xf32>) -> (f32, f32) {
+ %0:4 = vector.to_elements %a : vector<4xf32>
+ return %0#1, %0#3 : f32, f32
+}
+
+// -----
+
// CHECK-LABEL: @from_elements_0d
// CHECK-SAME: %[[ARG0:.+]]: f32
// CHECK: %[[RETVAL:.+]] = builtin.unrealized_conversion_cast %[[ARG0]]
|
kuhar
reviewed
Jul 2, 2025
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
Member
|
@efric I'll merge this once you format the code |
|
@efric Congratulations on having your first Pull Request (PR) merged into the LLVM Project! Your changes will be combined with recent changes from other authors, then tested by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR. Please check whether problems have been caused by your change specifically, as the builds can include changes from many authors. It is not uncommon for your change to be included in a build that fails due to someone else's changes, or infrastructure issues. How to do this, and the rest of the post-merge process, is covered in detail here. If your change does cause a problem, it may be reverted, or you can revert it yourself. This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again. If you don't get any reports, no action is required from you. Your changes are working as expected, well done! |
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/207/builds/3278 Here is the relevant piece of the build log for the reference |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Implement
vector.to_elementslowering to SPIR-V.Fixes: 145929.