[Mlir-commits] [mlir] [GPUToXeVMPipeline][Pipeline] Modify pipeline to add `convert-vector-to-llvm`. (PR #166204)

[Md Abdullah Shahneous Bari]

https://github.com/mshahneo updated https://github.com/llvm/llvm-project/pull/166204

>From 5afb320c5422b4b0cd9d7bce8bc090e7ad2f345d Mon Sep 17 00:00:00 2001
From: "Shahneous Bari, Md Abdullah" <md.abdullah.shahneous.bari at intel.com>
Date: Wed, 29 Oct 2025 20:27:44 +0000
Subject: [PATCH] Add `convert-vector-to-llvm` pass to the pipeline.

The `convert-vector-to-llvm` pass applies a set of vector transformation
patterns that are not included in the standard `convert-to-llvm` pass interface.
These additional transformations are required to properly lower
MLIR vector operations. Since not all vector ops have direct
`llvm` dialect lowering, many of them must first be progressively
rewritten into simpler or more canonical vector ops, which are then
lowered to `llvm`. Therefore, running `convert-vector-to-llvm` is
necessary to ensure a complete and correct lowering of vector operations
to the `llvm` dialect.
---
 mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp b/mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp
index b097d3a0c9686..38313dc3c01d5 100644
--- a/mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp
+++ b/mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp
@@ -111,8 +111,11 @@ void buildPostGPUCommonPassPipeline(
     pm.addPass(createGpuToLLVMConversionPass(gpuToLLVMOptions));
   }
   pm.addPass(createLowerAffinePass());
+  pm.addPass(createConvertVectorToLLVMPass());
   pm.addPass(createConvertToLLVMPass());
   pm.addPass(createReconcileUnrealizedCastsPass());
+  pm.addNestedPass<gpu::GPUModuleOp>(createCanonicalizerPass());
+  pm.addNestedPass<gpu::GPUModuleOp>(createCSEPass());
   // gpu-module-to-binary
   {
     GpuModuleToBinaryPassOptions gpuToModuleBinOptions;