Skip to content

Add a lit test for the gpu pipeline #420

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Dec 13, 2024
Merged

Add a lit test for the gpu pipeline #420

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
78bf70a
Add a lit test for the gpu pipeline
kurapov-peter Dec 13, 2024
af0a1a5
move to gpu folder to use proper lit config
kurapov-peter Dec 13, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions test/mlir/test/gc/Transforms/GPU/gpu-pipeline.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
// RUN: gc-opt %s --gc-gpu-pipeline -split-input-file | FileCheck %s

// Regression check for infinite loop in the linalg-to-xegpu pass
// CHECK-DAG: @gcGpuOclKernel_corner_shape_matmul_f16_kernel_SPIRV
// CHECK-DAG: @createGcGpuOclKernel_corner_shape_matmul_f16_kernel
// CHECK-DAG: @corner_shape_matmul_f16
module @fragment_name attributes {} {
func.func @corner_shape_matmul_f16(%arg0: memref<521x521xf16>, %arg1: memref<521x521xf16>, %arg2: memref<521x521xf16>) {
%0 = bufferization.to_tensor %arg0 restrict : memref<521x521xf16>
%1 = bufferization.to_tensor %arg1 restrict : memref<521x521xf16>
%2 = tensor.empty() : tensor<521x521xf16>
%cst = arith.constant 0.000000e+00 : f16
%3 = linalg.fill ins(%cst : f16) outs(%2 : tensor<521x521xf16>) -> tensor<521x521xf16>
%4 = linalg.matmul_transpose_b ins(%0, %1 : tensor<521x521xf16>, tensor<521x521xf16>) outs(%3 : tensor<521x521xf16>) -> tensor<521x521xf16>
bufferization.materialize_in_destination %4 in restrict writable %arg2 : (tensor<521x521xf16>, memref<521x521xf16>) -> ()
return
}
}
Loading