[Mlir-commits] [llvm] [mlir] [MLIR][AMDGPU] Adding Vector to AMDGPU conversion lowering (PR #131803)

[Zhuoran Yin]

================
@@ -0,0 +1,18 @@
+add_mlir_conversion_library(MLIRVectorToAMDGPU
+  VectorToAMDGPU.cpp
+
+  ADDITIONAL_HEADER_DIRS
+  ${MLIR_MAIN_INCLUDE_DIR}/mlir/Conversion/VectorToAMDGPU
+
+  DEPENDS
+  MLIRConversionPassIncGen
+
+  LINK_COMPONENTS
+  Core
+
+  LINK_LIBS PUBLIC
+  MLIRAMDGPUDialect
+  MLIRVectorDialect
+  MLIRPass
+  MLIRTransforms
+  )
----------------
jerryyin wrote:

This is added and tested locally.

It took me a while to realize the base commit of this branch is bad and keep erroring out on vector transform tablegen target. I have to cherry-pick the base commit to tip of llvm-project to be able to build and verify my change.

https://github.com/llvm/llvm-project/pull/131803