[mlir][vector] Refine Vector to LLVM lowering options #159553

banach-space · 2025-09-18T11:00:54Z

This is a follow-up to #144307, where we removed
vector.matrix_multiply and vector.flat_transpose from the Vector
dialect.

This PR:

Updates comments that were missed in the previous change.
Renames relevant -convert-vector-to-llvm= options:
- vector-contract-lowering=matmul → vector-contract-lowering=llvm
- vector-transpose-lowering=flat_transpose → vector-transpose-lowering=llvm

These new names better reflect the actual transformation target — LLVM
intrinsics — rather than the now-removed abstract operations.

llvmbot · 2025-09-18T11:01:28Z

@llvm/pr-subscribers-mlir-vector

Author: Andrzej Warzyński (banach-space)

Changes

WIP - almost ready!
[mlir][vector] Refine Vector to LLVM lowering options

Full diff: https://github.com/llvm/llvm-project/pull/159553.diff

7 Files Affected:

(modified) mlir/include/mlir/Conversion/Passes.td (+4-4)
(modified) mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td (+8-9)
(modified) mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp (+26-19)
(modified) mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp (+2-2)
(modified) mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir (+3-3)
(modified) mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir (+1-1)
(modified) mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir (+1-1)

diff --git a/mlir/include/mlir/Conversion/Passes.td b/mlir/include/mlir/Conversion/Passes.td
index 1a37d057776e2..aca0963478e63 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -1489,8 +1489,8 @@ def ConvertVectorToLLVMPass : Pass<"convert-vector-to-llvm"> {
            VectorContractLoweringAttr.summary, [{::llvm::cl::values(
            clEnumValN(::mlir::vector::VectorContractLowering::Dot, "dot",
             "Progressively lower to finer grained `vector.contract` and dot-products. (default)"),
-           clEnumValN(::mlir::vector::VectorContractLowering::Matmul, "matmul",
-            "Lower to `vector.matrix_multiply`, maps 1-1 to LLVM matrix intrinsics."),
+           clEnumValN(::mlir::vector::VectorContractLowering::LLVM, "llvm",
+            "Lower directly to `llvm.intr.matrix.multiply`."),
            clEnumValN(::mlir::vector::VectorContractLowering::OuterProduct, "outerproduct",
             "Lower to `vector.outerproduct`."),
            clEnumValN(::mlir::vector::VectorContractLowering::ParallelArith, "parallelarith",
@@ -1502,8 +1502,8 @@ def ConvertVectorToLLVMPass : Pass<"convert-vector-to-llvm"> {
            VectorTransposeLoweringAttr.summary, [{::llvm::cl::values(
            clEnumValN(::mlir::vector::VectorTransposeLowering::EltWise, "eltwise",
             "Lower transpose into element-wise extract and inserts (default)"),
-           clEnumValN(::mlir::vector::VectorTransposeLowering::Flat, "flat",
-            "Lower 2-D transpose to `vector.flat_transpose`, maps 1-1 to LLVM matrix intrinsics"),
+           clEnumValN(::mlir::vector::VectorTransposeLowering::LLVM, "llvm",
+            "Lower 2-D transpose directly to `llvm.intr.matrix.transpose`"),
            clEnumValN(::mlir::vector::VectorTransposeLowering::Shuffle1D, "shuffle1d",
             "Lower 2-D transpose to `vector.shuffle` on 1-D vector."),
            clEnumValN(::mlir::vector::VectorTransposeLowering::Shuffle16x16, "shuffle16x16",
diff --git a/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td b/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
index ef0951ab1d166..cbba44ae4dc8a 100644
--- a/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
+++ b/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
@@ -14,10 +14,9 @@ include "mlir/IR/EnumAttr.td"
 // Lower transpose into element-wise extract and inserts.
 def VectorTransposeLowering_Elementwise:
   I32EnumAttrCase<"EltWise",  0, "eltwise">;
-// Lower 2-D transpose to `vector.flat_transpose`, maps 1-1 to LLVM matrix
-// intrinsics.
-def VectorTransposeLowering_FlatTranspose:
-  I32EnumAttrCase<"Flat",  1, "flat_transpose">;
+// Lower directly to LLVM matrix intrinsics.
+def VectorTransposeLowering_LLVM:
+  I32EnumAttrCase<"LLVM",  1, "llvm">;
 // Lower 2-D transpose to `vector.shuffle` on 1-D vector.
 def VectorTransposeLowering_Shuffle1D:
   I32EnumAttrCase<"Shuffle1D",  2, "shuffle_1d">;
@@ -27,7 +26,7 @@ def VectorTransposeLowering_Shuffle16x16:
 def VectorTransposeLoweringAttr : I32EnumAttr<
     "VectorTransposeLowering",
     "control the lowering of `vector.transpose` operations.",
-    [VectorTransposeLowering_Elementwise, VectorTransposeLowering_FlatTranspose,
+    [VectorTransposeLowering_Elementwise, VectorTransposeLowering_LLVM,
      VectorTransposeLowering_Shuffle1D, VectorTransposeLowering_Shuffle16x16]> {
   let cppNamespace = "::mlir::vector";
 }
@@ -48,9 +47,9 @@ def VectorMultiReductionLoweringAttr: I32EnumAttr<
 
 // Progressively lower to finer grained `vector.contract` and dot-products.
 def VectorContractLowering_Dot: I32EnumAttrCase<"Dot", 0, "dot">;
-// Lower to `vector.matrix_multiply`, maps 1-1 to LLVM matrix intrinsics.
-def VectorContractLowering_Matmul:
-  I32EnumAttrCase<"Matmul", 1, "matmulintrinsics">;
+// Lower directly to LLVM intrinsics. 
+def VectorContractLowering_LLVM:
+  I32EnumAttrCase<"LLVM", 1, "llvm">;
 // Lower to `vector.outerproduct`.
 def VectorContractLowering_OuterProduct:
   I32EnumAttrCase<"OuterProduct", 2, "outerproduct">;
@@ -61,7 +60,7 @@ def VectorContractLowering_ParallelArith:
 def VectorContractLoweringAttr: I32EnumAttr<
     "VectorContractLowering",
     "control the lowering of `vector.contract` operations.",
-  [VectorContractLowering_Dot, VectorContractLowering_Matmul,
+  [VectorContractLowering_Dot, VectorContractLowering_LLVM,
    VectorContractLowering_OuterProduct, VectorContractLowering_ParallelArith]> {
   let cppNamespace = "::mlir::vector";
 }
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
index e7266740894b1..cbce7b006edd0 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
@@ -1987,16 +1987,12 @@ struct VectorScalableStepOpLowering
 ///    %e = add %c, %d
 /// ```
 /// `vector.matrix_multiply` later lowers to `llvm.matrix.multiply`.
-//
-/// This only kicks in when vectorContractLowering is set to Matmul and
-/// the vector.contract op is a row-major matrix multiply.
 class ContractionOpToMatmulOpLowering
     : public vector::MaskableOpRewritePattern<vector::ContractionOp> {
 public:
   using MaskableOpRewritePattern::MaskableOpRewritePattern;
 
   ContractionOpToMatmulOpLowering(
-      vector::VectorContractLowering vectorContractLowering,
       MLIRContext *context, PatternBenefit benefit = 100)
       : MaskableOpRewritePattern<vector::ContractionOp>(context, benefit) {}
 
@@ -2005,23 +2001,22 @@ class ContractionOpToMatmulOpLowering
                             PatternRewriter &rewriter) const override;
 };
 
-/// Progressively lower a `vector.contract %a, %b, %c` with row-major matmul
-/// semantics to:
+/// Lower a qualifying `vector.contract %a, %b, %c` (with row-major matmul
+/// semantics directly into `llvm.intr.matrix.multiply`:
+/// BEFORE:
 /// ```
-///    %mta = maybe_transpose
-///    %mtb = maybe_transpose
-///    %flattened_a = vector.shape_cast %mta
-///    %flattened_b = vector.shape_cast %mtb
-///    %flattened_d = llvm.intr.matrix.multiply %flattened_a, %flattened_b
-///    %mtd = vector.shape_cast %flattened_d
-///    %d = maybe_untranspose %mtd
-///    %e = add %c, %d
+///  %res = vector.contract #matmat_trait %lhs, %rhs, %acc
+///    : vector<2x4xf32>, vector<4x3xf32> into vector<2x3xf32>
 /// ```
-//
-/// This only kicks in when vectorContractLowering is set to `Matmul`.
-/// vector.transpose operations are inserted if the vector.contract op is not a
-/// row-major matrix multiply.
 ///
+/// AFTER:
+/// ```
+///   %lhs = vector.shape_cast %arg0 : vector<2x4xf32> to vector<8xf32>
+///   %rhs = vector.shape_cast %arg1 : vector<4x3xf32> to vector<12xf32>
+///   %matmul = llvm.intr.matrix.multiply %lhs, %rhs
+///   %res = arith.addf %acc, %matmul : vector<2x3xf32>
+/// ```
+//
 /// Scalable vectors are not supported.
 FailureOr<Value> ContractionOpToMatmulOpLowering::matchAndRewriteMaskableOp(
     vector::ContractionOp op, MaskingOpInterface maskOp,
@@ -2116,7 +2111,19 @@ FailureOr<Value> ContractionOpToMatmulOpLowering::matchAndRewriteMaskableOp(
   return res;
 }
 
-/// Lowers vector.transpose to llvm.intr.matrix.transpose
+/// Lowers vector.transpose directly to llvm.intr.matrix.transpose
+///
+/// BEFORE:
+/// ```
+///  %tr = vector.transpose %vec, [1, 0] : vector<2x4xf32> to vector<4x2xf32>
+/// ```
+/// AFTER:
+/// ```
+///  %vec_cs = vector.shape_cast %vec : vector<2x4xf32> to vector<8xf32>
+///  %tr = llvm.intr.matrix.transpose %vec_sc
+///    {columns = 2 : i32, rows = 4 : i32} : vector<8xf32> into vector<8xf32>
+///  %res = vector.shape_cast %tr : vector<8xf32> to vector<4x2xf32>
+/// ```
 class TransposeOpToMatrixTransposeOpLowering
     : public OpRewritePattern<vector::TransposeOp> {
 public:
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
index 0b44ca7ceee42..a65f8ba233b76 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
@@ -70,7 +70,7 @@ void ConvertVectorToLLVMPass::runOnOperation() {
     populateVectorBitCastLoweringPatterns(patterns);
     populateVectorBroadcastLoweringPatterns(patterns);
     populateVectorContractLoweringPatterns(patterns, vectorContractLowering);
-    if (vectorContractLowering == vector::VectorContractLowering::Matmul) {
+    if (vectorContractLowering == vector::VectorContractLowering::LLVM) {
       // This pattern creates a dependency on the LLVM dialect, hence we don't
       // include it in `populateVectorContractLoweringPatterns` that is part of
       // the Vector dialect (and should not depend on LLVM).
@@ -80,7 +80,7 @@ void ConvertVectorToLLVMPass::runOnOperation() {
     populateVectorShapeCastLoweringPatterns(patterns);
     populateVectorInterleaveLoweringPatterns(patterns);
     populateVectorTransposeLoweringPatterns(patterns, vectorTransposeLowering);
-    if (vectorTransposeLowering == vector::VectorTransposeLowering::Flat) {
+    if (vectorTransposeLowering == vector::VectorTransposeLowering::LLVM) {
       // This pattern creates a dependency on the LLVM dialect, hence we don't
       // include it in `populateVectorTransposeLoweringPatterns` that is part of
       // the Vector dialect (and should not depend on LLVM).
diff --git a/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir b/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
index 323d86ac40988..9b18d67e037db 100644
--- a/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
+++ b/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
@@ -13,7 +13,7 @@
 
 // RUN: mlir-opt --convert-vector-to-llvm --dump-pass-pipeline %s 2>&1 | FileCheck %s --check-prefix=DEFAULT
 
-// RUN: mlir-opt --convert-vector-to-llvm='vector-contract-lowering=matmul vector-transpose-lowering=flat' \
+// RUN: mlir-opt --convert-vector-to-llvm='vector-contract-lowering=llvm vector-transpose-lowering=llvm' \
 // RUN:          --dump-pass-pipeline %s 2>&1 | FileCheck %s --check-prefix=NON-DEFAULT
 
 // CHECK: builtin.module(
@@ -26,5 +26,5 @@
 // CHECK-SAME: reassociate-fp-reductions={{[aA-zZ0-9]+}}
 // DEFAULT: vector-contract-lowering=dot
 // DEFAULT: vector-transpose-lowering=eltwise
-// NON-DEFAULT: vector-contract-lowering=matmul
-// NON-DEFAULT: vector-transpose-lowering=flat
+// NON-DEFAULT: vector-contract-lowering=llvm
+// NON-DEFAULT: vector-transpose-lowering=llvm
diff --git a/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir b/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
index 3950e54006eec..fd5bbaaadc331 100644
--- a/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
+++ b/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s  --convert-vector-to-llvm='vector-contract-lowering=matmul' | FileCheck %s
+// RUN: mlir-opt %s  --convert-vector-to-llvm='vector-contract-lowering=llvm' | FileCheck %s
 
 #matmat_accesses = [
   affine_map<(i, j, k) -> (i, k)>,
diff --git a/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir b/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
index 94689fa0dfb88..66032014d9307 100644
--- a/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
+++ b/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s --convert-vector-to-llvm='vector-transpose-lowering=flat' --split-input-file | FileCheck %s
+// RUN: mlir-opt %s --convert-vector-to-llvm='vector-transpose-lowering=llvm' --split-input-file | FileCheck %s
 
 // CHECK-LABEL: func @transpose(
 func.func @transpose(%arg0: vector<2x4xf32>) -> vector<4x2xf32> {

llvmbot · 2025-09-18T11:01:29Z

@llvm/pr-subscribers-mlir

Author: Andrzej Warzyński (banach-space)

Changes

WIP - almost ready!
[mlir][vector] Refine Vector to LLVM lowering options

Full diff: https://github.com/llvm/llvm-project/pull/159553.diff

7 Files Affected:

(modified) mlir/include/mlir/Conversion/Passes.td (+4-4)
(modified) mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td (+8-9)
(modified) mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp (+26-19)
(modified) mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp (+2-2)
(modified) mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir (+3-3)
(modified) mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir (+1-1)
(modified) mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir (+1-1)

diff --git a/mlir/include/mlir/Conversion/Passes.td b/mlir/include/mlir/Conversion/Passes.td
index 1a37d057776e2..aca0963478e63 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -1489,8 +1489,8 @@ def ConvertVectorToLLVMPass : Pass<"convert-vector-to-llvm"> {
            VectorContractLoweringAttr.summary, [{::llvm::cl::values(
            clEnumValN(::mlir::vector::VectorContractLowering::Dot, "dot",
             "Progressively lower to finer grained `vector.contract` and dot-products. (default)"),
-           clEnumValN(::mlir::vector::VectorContractLowering::Matmul, "matmul",
-            "Lower to `vector.matrix_multiply`, maps 1-1 to LLVM matrix intrinsics."),
+           clEnumValN(::mlir::vector::VectorContractLowering::LLVM, "llvm",
+            "Lower directly to `llvm.intr.matrix.multiply`."),
            clEnumValN(::mlir::vector::VectorContractLowering::OuterProduct, "outerproduct",
             "Lower to `vector.outerproduct`."),
            clEnumValN(::mlir::vector::VectorContractLowering::ParallelArith, "parallelarith",
@@ -1502,8 +1502,8 @@ def ConvertVectorToLLVMPass : Pass<"convert-vector-to-llvm"> {
            VectorTransposeLoweringAttr.summary, [{::llvm::cl::values(
            clEnumValN(::mlir::vector::VectorTransposeLowering::EltWise, "eltwise",
             "Lower transpose into element-wise extract and inserts (default)"),
-           clEnumValN(::mlir::vector::VectorTransposeLowering::Flat, "flat",
-            "Lower 2-D transpose to `vector.flat_transpose`, maps 1-1 to LLVM matrix intrinsics"),
+           clEnumValN(::mlir::vector::VectorTransposeLowering::LLVM, "llvm",
+            "Lower 2-D transpose directly to `llvm.intr.matrix.transpose`"),
            clEnumValN(::mlir::vector::VectorTransposeLowering::Shuffle1D, "shuffle1d",
             "Lower 2-D transpose to `vector.shuffle` on 1-D vector."),
            clEnumValN(::mlir::vector::VectorTransposeLowering::Shuffle16x16, "shuffle16x16",
diff --git a/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td b/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
index ef0951ab1d166..cbba44ae4dc8a 100644
--- a/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
+++ b/mlir/include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td
@@ -14,10 +14,9 @@ include "mlir/IR/EnumAttr.td"
 // Lower transpose into element-wise extract and inserts.
 def VectorTransposeLowering_Elementwise:
   I32EnumAttrCase<"EltWise",  0, "eltwise">;
-// Lower 2-D transpose to `vector.flat_transpose`, maps 1-1 to LLVM matrix
-// intrinsics.
-def VectorTransposeLowering_FlatTranspose:
-  I32EnumAttrCase<"Flat",  1, "flat_transpose">;
+// Lower directly to LLVM matrix intrinsics.
+def VectorTransposeLowering_LLVM:
+  I32EnumAttrCase<"LLVM",  1, "llvm">;
 // Lower 2-D transpose to `vector.shuffle` on 1-D vector.
 def VectorTransposeLowering_Shuffle1D:
   I32EnumAttrCase<"Shuffle1D",  2, "shuffle_1d">;
@@ -27,7 +26,7 @@ def VectorTransposeLowering_Shuffle16x16:
 def VectorTransposeLoweringAttr : I32EnumAttr<
     "VectorTransposeLowering",
     "control the lowering of `vector.transpose` operations.",
-    [VectorTransposeLowering_Elementwise, VectorTransposeLowering_FlatTranspose,
+    [VectorTransposeLowering_Elementwise, VectorTransposeLowering_LLVM,
      VectorTransposeLowering_Shuffle1D, VectorTransposeLowering_Shuffle16x16]> {
   let cppNamespace = "::mlir::vector";
 }
@@ -48,9 +47,9 @@ def VectorMultiReductionLoweringAttr: I32EnumAttr<
 
 // Progressively lower to finer grained `vector.contract` and dot-products.
 def VectorContractLowering_Dot: I32EnumAttrCase<"Dot", 0, "dot">;
-// Lower to `vector.matrix_multiply`, maps 1-1 to LLVM matrix intrinsics.
-def VectorContractLowering_Matmul:
-  I32EnumAttrCase<"Matmul", 1, "matmulintrinsics">;
+// Lower directly to LLVM intrinsics. 
+def VectorContractLowering_LLVM:
+  I32EnumAttrCase<"LLVM", 1, "llvm">;
 // Lower to `vector.outerproduct`.
 def VectorContractLowering_OuterProduct:
   I32EnumAttrCase<"OuterProduct", 2, "outerproduct">;
@@ -61,7 +60,7 @@ def VectorContractLowering_ParallelArith:
 def VectorContractLoweringAttr: I32EnumAttr<
     "VectorContractLowering",
     "control the lowering of `vector.contract` operations.",
-  [VectorContractLowering_Dot, VectorContractLowering_Matmul,
+  [VectorContractLowering_Dot, VectorContractLowering_LLVM,
    VectorContractLowering_OuterProduct, VectorContractLowering_ParallelArith]> {
   let cppNamespace = "::mlir::vector";
 }
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
index e7266740894b1..cbce7b006edd0 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
@@ -1987,16 +1987,12 @@ struct VectorScalableStepOpLowering
 ///    %e = add %c, %d
 /// ```
 /// `vector.matrix_multiply` later lowers to `llvm.matrix.multiply`.
-//
-/// This only kicks in when vectorContractLowering is set to Matmul and
-/// the vector.contract op is a row-major matrix multiply.
 class ContractionOpToMatmulOpLowering
     : public vector::MaskableOpRewritePattern<vector::ContractionOp> {
 public:
   using MaskableOpRewritePattern::MaskableOpRewritePattern;
 
   ContractionOpToMatmulOpLowering(
-      vector::VectorContractLowering vectorContractLowering,
       MLIRContext *context, PatternBenefit benefit = 100)
       : MaskableOpRewritePattern<vector::ContractionOp>(context, benefit) {}
 
@@ -2005,23 +2001,22 @@ class ContractionOpToMatmulOpLowering
                             PatternRewriter &rewriter) const override;
 };
 
-/// Progressively lower a `vector.contract %a, %b, %c` with row-major matmul
-/// semantics to:
+/// Lower a qualifying `vector.contract %a, %b, %c` (with row-major matmul
+/// semantics directly into `llvm.intr.matrix.multiply`:
+/// BEFORE:
 /// ```
-///    %mta = maybe_transpose
-///    %mtb = maybe_transpose
-///    %flattened_a = vector.shape_cast %mta
-///    %flattened_b = vector.shape_cast %mtb
-///    %flattened_d = llvm.intr.matrix.multiply %flattened_a, %flattened_b
-///    %mtd = vector.shape_cast %flattened_d
-///    %d = maybe_untranspose %mtd
-///    %e = add %c, %d
+///  %res = vector.contract #matmat_trait %lhs, %rhs, %acc
+///    : vector<2x4xf32>, vector<4x3xf32> into vector<2x3xf32>
 /// ```
-//
-/// This only kicks in when vectorContractLowering is set to `Matmul`.
-/// vector.transpose operations are inserted if the vector.contract op is not a
-/// row-major matrix multiply.
 ///
+/// AFTER:
+/// ```
+///   %lhs = vector.shape_cast %arg0 : vector<2x4xf32> to vector<8xf32>
+///   %rhs = vector.shape_cast %arg1 : vector<4x3xf32> to vector<12xf32>
+///   %matmul = llvm.intr.matrix.multiply %lhs, %rhs
+///   %res = arith.addf %acc, %matmul : vector<2x3xf32>
+/// ```
+//
 /// Scalable vectors are not supported.
 FailureOr<Value> ContractionOpToMatmulOpLowering::matchAndRewriteMaskableOp(
     vector::ContractionOp op, MaskingOpInterface maskOp,
@@ -2116,7 +2111,19 @@ FailureOr<Value> ContractionOpToMatmulOpLowering::matchAndRewriteMaskableOp(
   return res;
 }
 
-/// Lowers vector.transpose to llvm.intr.matrix.transpose
+/// Lowers vector.transpose directly to llvm.intr.matrix.transpose
+///
+/// BEFORE:
+/// ```
+///  %tr = vector.transpose %vec, [1, 0] : vector<2x4xf32> to vector<4x2xf32>
+/// ```
+/// AFTER:
+/// ```
+///  %vec_cs = vector.shape_cast %vec : vector<2x4xf32> to vector<8xf32>
+///  %tr = llvm.intr.matrix.transpose %vec_sc
+///    {columns = 2 : i32, rows = 4 : i32} : vector<8xf32> into vector<8xf32>
+///  %res = vector.shape_cast %tr : vector<8xf32> to vector<4x2xf32>
+/// ```
 class TransposeOpToMatrixTransposeOpLowering
     : public OpRewritePattern<vector::TransposeOp> {
 public:
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
index 0b44ca7ceee42..a65f8ba233b76 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
@@ -70,7 +70,7 @@ void ConvertVectorToLLVMPass::runOnOperation() {
     populateVectorBitCastLoweringPatterns(patterns);
     populateVectorBroadcastLoweringPatterns(patterns);
     populateVectorContractLoweringPatterns(patterns, vectorContractLowering);
-    if (vectorContractLowering == vector::VectorContractLowering::Matmul) {
+    if (vectorContractLowering == vector::VectorContractLowering::LLVM) {
       // This pattern creates a dependency on the LLVM dialect, hence we don't
       // include it in `populateVectorContractLoweringPatterns` that is part of
       // the Vector dialect (and should not depend on LLVM).
@@ -80,7 +80,7 @@ void ConvertVectorToLLVMPass::runOnOperation() {
     populateVectorShapeCastLoweringPatterns(patterns);
     populateVectorInterleaveLoweringPatterns(patterns);
     populateVectorTransposeLoweringPatterns(patterns, vectorTransposeLowering);
-    if (vectorTransposeLowering == vector::VectorTransposeLowering::Flat) {
+    if (vectorTransposeLowering == vector::VectorTransposeLowering::LLVM) {
       // This pattern creates a dependency on the LLVM dialect, hence we don't
       // include it in `populateVectorTransposeLoweringPatterns` that is part of
       // the Vector dialect (and should not depend on LLVM).
diff --git a/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir b/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
index 323d86ac40988..9b18d67e037db 100644
--- a/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
+++ b/mlir/test/Conversion/VectorToLLVM/pass-option-serialization.mlir
@@ -13,7 +13,7 @@
 
 // RUN: mlir-opt --convert-vector-to-llvm --dump-pass-pipeline %s 2>&1 | FileCheck %s --check-prefix=DEFAULT
 
-// RUN: mlir-opt --convert-vector-to-llvm='vector-contract-lowering=matmul vector-transpose-lowering=flat' \
+// RUN: mlir-opt --convert-vector-to-llvm='vector-contract-lowering=llvm vector-transpose-lowering=llvm' \
 // RUN:          --dump-pass-pipeline %s 2>&1 | FileCheck %s --check-prefix=NON-DEFAULT
 
 // CHECK: builtin.module(
@@ -26,5 +26,5 @@
 // CHECK-SAME: reassociate-fp-reductions={{[aA-zZ0-9]+}}
 // DEFAULT: vector-contract-lowering=dot
 // DEFAULT: vector-transpose-lowering=eltwise
-// NON-DEFAULT: vector-contract-lowering=matmul
-// NON-DEFAULT: vector-transpose-lowering=flat
+// NON-DEFAULT: vector-contract-lowering=llvm
+// NON-DEFAULT: vector-transpose-lowering=llvm
diff --git a/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir b/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
index 3950e54006eec..fd5bbaaadc331 100644
--- a/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
+++ b/mlir/test/Dialect/Vector/vector-contract-to-matrix-intrinsics-transforms.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s  --convert-vector-to-llvm='vector-contract-lowering=matmul' | FileCheck %s
+// RUN: mlir-opt %s  --convert-vector-to-llvm='vector-contract-lowering=llvm' | FileCheck %s
 
 #matmat_accesses = [
   affine_map<(i, j, k) -> (i, k)>,
diff --git a/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir b/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
index 94689fa0dfb88..66032014d9307 100644
--- a/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
+++ b/mlir/test/Dialect/Vector/vector-transpose-to-matrix-intrinsics-transform.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s --convert-vector-to-llvm='vector-transpose-lowering=flat' --split-input-file | FileCheck %s
+// RUN: mlir-opt %s --convert-vector-to-llvm='vector-transpose-lowering=llvm' --split-input-file | FileCheck %s
 
 // CHECK-LABEL: func @transpose(
 func.func @transpose(%arg0: vector<2x4xf32>) -> vector<4x2xf32> {

github-actions · 2025-09-18T11:05:32Z

✅ With the latest revision this PR passed the C/C++ code formatter.

This is a follow-up to llvm#144307, where we removed `vector.matrix_multiply` and `vector.flat_transpose` from the Vector dialect. This PR: * Updates comments that were missed in the previous change. * Renames relevant `-convert-vector-to-llvm=` options: - `vector-contract-lowering=matmul` → `vector-contract-lowering=llvm` - `vector-transpose-lowering=flat_transpose` → `vector-transpose-lowering=llvm` These new names better reflect the actual transformation target — LLVM intrinsics — rather than the now-removed abstract operations.

kuhar

Seems like you also need to update python bindings

banach-space · 2025-09-18T18:42:46Z

Seems like you also need to update python bindings

Just had to update a Python test: commit

kuhar

LGTM but let's wait for some other review

kuhar · 2025-09-19T01:16:48Z

mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp

-/// semantics to:
+/// Lower a qualifying `vector.contract %a, %b, %c` (with row-major matmul
+/// semantics directly into `llvm.intr.matrix.multiply`:
+/// BEFORE:


nit/optional: you can tag code blocks with mlir, e.g.:

```mlir %res = ... ```

also below

dcaballe · 2025-09-19T17:34:23Z

mlir/include/mlir/Conversion/Passes.td

-           clEnumValN(::mlir::vector::VectorContractLowering::Matmul, "matmul",
-            "Lower to `vector.matrix_multiply`, maps 1-1 to LLVM matrix intrinsics."),
+           clEnumValN(::mlir::vector::VectorContractLowering::LLVM, "llvm",
+            "Lower directly to `llvm.intr.matrix.multiply`."),


llvm only could be misleading as that's not the only option we have to lower to llvm. What about ~llvm-matmul?

Good point!

Since this is about lowering to "LLVM intrinsics", and the intrinsics are:

llvm.intr.matrix.multiply (i.e. llvm.intr.<...>)

llvm.intr.matrix.transpose (i.e. llvm.intr.<...>),

let me rename this as llvmintr.

LLVM intrinsic sounds like the best name for me

dcaballe · 2025-09-19T17:34:33Z

mlir/include/mlir/Conversion/Passes.td

-           clEnumValN(::mlir::vector::VectorTransposeLowering::Flat, "flat",
-            "Lower 2-D transpose to `vector.flat_transpose`, maps 1-1 to LLVM matrix intrinsics"),
+           clEnumValN(::mlir::vector::VectorTransposeLowering::LLVM, "llvm",
+            "Lower 2-D transpose directly to `llvm.intr.matrix.transpose`"),


same, in general, throughout the file

dcaballe

SG, thanks!

banach-space requested review from dcaballe, nicolasvasilache and Groverkss as code owners September 18, 2025 11:00

llvmbot added mlir:vectorops mlir mlir:vector labels Sep 18, 2025

banach-space force-pushed the andrzej/vector/rename_pass_optoins branch from 3439e0a to 1b1bc23 Compare September 18, 2025 13:53

banach-space changed the title ~~andrzej/vector/rename pass optoins~~ [mlir][vector] Refine Vector to LLVM lowering options Sep 18, 2025

banach-space requested a review from kuhar September 18, 2025 13:54

banach-space force-pushed the andrzej/vector/rename_pass_optoins branch from 1b1bc23 to 060f098 Compare September 18, 2025 13:54

kuhar reviewed Sep 18, 2025

View reviewed changes

Fix Python test

9b36aca

banach-space requested review from ftynse, makslevental, stellaraccident and rolfmorel as code owners September 18, 2025 18:41

kuhar approved these changes Sep 19, 2025

View reviewed changes

dcaballe reviewed Sep 19, 2025

View reviewed changes

Addressing comments from Jakub and Diego

1269750

dcaballe approved these changes Sep 22, 2025

View reviewed changes

banach-space merged commit 3a1111d into llvm:main Sep 23, 2025
9 checks passed

banach-space deleted the andrzej/vector/rename_pass_optoins branch September 23, 2025 15:30

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[mlir][vector] Refine Vector to LLVM lowering options #159553

[mlir][vector] Refine Vector to LLVM lowering options #159553

Uh oh!

banach-space commented Sep 18, 2025 •

edited

Loading

Uh oh!

llvmbot commented Sep 18, 2025

Uh oh!

llvmbot commented Sep 18, 2025

Uh oh!

github-actions bot commented Sep 18, 2025 •

edited

Loading

Uh oh!

kuhar left a comment

Uh oh!

banach-space commented Sep 18, 2025

Uh oh!

kuhar left a comment

Uh oh!

kuhar Sep 19, 2025

Uh oh!

dcaballe Sep 19, 2025

Uh oh!

banach-space Sep 22, 2025

Uh oh!

kuhar Sep 22, 2025

Uh oh!

dcaballe Sep 19, 2025

Uh oh!

dcaballe left a comment

Uh oh!

Uh oh!

Uh oh!

[mlir][vector] Refine Vector to LLVM lowering options #159553

[mlir][vector] Refine Vector to LLVM lowering options #159553

Uh oh!

Conversation

banach-space commented Sep 18, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

llvmbot commented Sep 18, 2025

Uh oh!

llvmbot commented Sep 18, 2025

Uh oh!

github-actions bot commented Sep 18, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kuhar left a comment

Choose a reason for hiding this comment

Uh oh!

banach-space commented Sep 18, 2025

Uh oh!

kuhar left a comment

Choose a reason for hiding this comment

Uh oh!

kuhar Sep 19, 2025

Choose a reason for hiding this comment

Uh oh!

dcaballe Sep 19, 2025

Choose a reason for hiding this comment

Uh oh!

banach-space Sep 22, 2025

Choose a reason for hiding this comment

Uh oh!

kuhar Sep 22, 2025

Choose a reason for hiding this comment

Uh oh!

dcaballe Sep 19, 2025

Choose a reason for hiding this comment

Uh oh!

dcaballe left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

banach-space commented Sep 18, 2025 •

edited

Loading

github-actions bot commented Sep 18, 2025 •

edited

Loading