derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
69 lines
3.0 KiB
MLIR
69 lines
3.0 KiB
MLIR
// Make sure that signed extended multiplication produces expected results
|
|
// with and without expansion to primitive mul/add ops for WebGPU.
|
|
|
|
// RUN: mlir-opt %s -test-vulkan-runner-pipeline \
|
|
// RUN: | mlir-runner - \
|
|
// RUN: --shared-libs=%mlir_vulkan_runtime,%mlir_runner_utils \
|
|
// RUN: --entry-point-result=void | FileCheck %s
|
|
|
|
// RUN: mlir-opt %s -test-vulkan-runner-pipeline=spirv-webgpu-prepare \
|
|
// RUN: | mlir-runner - \
|
|
// RUN: --shared-libs=%mlir_vulkan_runtime,%mlir_runner_utils \
|
|
// RUN: --entry-point-result=void | FileCheck %s
|
|
|
|
// CHECK: [0, 1, -2, 1, 1048560, -87620295, -131071, 560969770]
|
|
// CHECK: [0, 0, -1, 0, 0, -1, 0, -499807318]
|
|
module attributes {
|
|
gpu.container_module,
|
|
spirv.target_env = #spirv.target_env<
|
|
#spirv.vce<v1.4, [Shader], [SPV_KHR_storage_buffer_storage_class]>, #spirv.resource_limits<>>
|
|
} {
|
|
gpu.module @kernels {
|
|
gpu.func @kernel_add(%arg0 : memref<8xi32>, %arg1 : memref<8xi32>, %arg2 : memref<8xi32>, %arg3 : memref<8xi32>)
|
|
kernel attributes { spirv.entry_point_abi = #spirv.entry_point_abi<workgroup_size = [1, 1, 1]>} {
|
|
%0 = gpu.block_id x
|
|
%lhs = memref.load %arg0[%0] : memref<8xi32>
|
|
%rhs = memref.load %arg1[%0] : memref<8xi32>
|
|
%low, %hi = arith.mulsi_extended %lhs, %rhs : i32
|
|
memref.store %low, %arg2[%0] : memref<8xi32>
|
|
memref.store %hi, %arg3[%0] : memref<8xi32>
|
|
gpu.return
|
|
}
|
|
}
|
|
|
|
func.func @main() {
|
|
%buf0 = memref.alloc() : memref<8xi32>
|
|
%buf1 = memref.alloc() : memref<8xi32>
|
|
%buf2 = memref.alloc() : memref<8xi32>
|
|
%buf3 = memref.alloc() : memref<8xi32>
|
|
%i32_0 = arith.constant 0 : i32
|
|
|
|
// Initialize output buffers.
|
|
%buf4 = memref.cast %buf2 : memref<8xi32> to memref<?xi32>
|
|
%buf5 = memref.cast %buf3 : memref<8xi32> to memref<?xi32>
|
|
call @fillResource1DInt(%buf4, %i32_0) : (memref<?xi32>, i32) -> ()
|
|
call @fillResource1DInt(%buf5, %i32_0) : (memref<?xi32>, i32) -> ()
|
|
|
|
%idx_0 = arith.constant 0 : index
|
|
%idx_1 = arith.constant 1 : index
|
|
%idx_8 = arith.constant 8 : index
|
|
|
|
// Initialize input buffers.
|
|
%lhs_vals = arith.constant dense<[0, 1, -1, -1, 65535, 65535, -65535, 2088183954]> : vector<8xi32>
|
|
%rhs_vals = arith.constant dense<[0, 1, 2, -1, 16, -1337, -65535, -1028001427]> : vector<8xi32>
|
|
vector.store %lhs_vals, %buf0[%idx_0] : memref<8xi32>, vector<8xi32>
|
|
vector.store %rhs_vals, %buf1[%idx_0] : memref<8xi32>, vector<8xi32>
|
|
|
|
gpu.launch_func @kernels::@kernel_add
|
|
blocks in (%idx_8, %idx_1, %idx_1) threads in (%idx_1, %idx_1, %idx_1)
|
|
args(%buf0 : memref<8xi32>, %buf1 : memref<8xi32>, %buf2 : memref<8xi32>, %buf3 : memref<8xi32>)
|
|
%buf_low = memref.cast %buf4 : memref<?xi32> to memref<*xi32>
|
|
%buf_hi = memref.cast %buf5 : memref<?xi32> to memref<*xi32>
|
|
call @printMemrefI32(%buf_low) : (memref<*xi32>) -> ()
|
|
call @printMemrefI32(%buf_hi) : (memref<*xi32>) -> ()
|
|
return
|
|
}
|
|
func.func private @fillResource1DInt(%0 : memref<?xi32>, %1 : i32)
|
|
func.func private @printMemrefI32(%ptr : memref<*xi32>)
|
|
}
|