diff options
| author | Sang Ik Lee <sang.ik.lee@intel.com> | 2023-12-05 13:55:24 -0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-12-05 16:55:24 -0500 |
| commit | 7fc792cba7663b2aa54f259515319d74a5625be0 (patch) | |
| tree | 003801c6819f6cf9272edc026dd38049926dbabb /mlir/test/lit.site.cfg.py.in | |
| parent | 2284771fa02fa3c75b07fafc37c559cf6deddd59 (diff) | |
| download | llvm-7fc792cba7663b2aa54f259515319d74a5625be0.zip llvm-7fc792cba7663b2aa54f259515319d74a5625be0.tar.gz llvm-7fc792cba7663b2aa54f259515319d74a5625be0.tar.bz2 | |
[MLIR] Enable GPU Dialect to SYCL runtime integration (#71430)
GPU Dialect lowering to SYCL runtime is driven by spirv.target_env
attached to gpu.module. As a result of this, spirv.target_env remains as
an input to LLVMIR Translation.
A SPIRVToLLVMIRTranslation without any actual translation is added to
avoid an unregistered error in mlir-cpu-runner.
SelectObjectAttr.cpp is updated to
1) Pass binary size argument to getModuleLoadFn
2) Pass parameter count to getKernelLaunchFn
This change does not impact CUDA and ROCM usage since both
mlir_cuda_runtime and mlir_rocm_runtime are already updated to accept
and ignore the extra arguments.
Diffstat (limited to 'mlir/test/lit.site.cfg.py.in')
| -rw-r--r-- | mlir/test/lit.site.cfg.py.in | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/mlir/test/lit.site.cfg.py.in b/mlir/test/lit.site.cfg.py.in index 146e844..c0fa1b8 100644 --- a/mlir/test/lit.site.cfg.py.in +++ b/mlir/test/lit.site.cfg.py.in @@ -31,6 +31,7 @@ config.run_rocm_tests = @MLIR_ENABLE_ROCM_CONVERSIONS@ config.enable_rocm_runner = @MLIR_ENABLE_ROCM_RUNNER@ config.gpu_compilation_format = "@MLIR_GPU_COMPILATION_TEST_FORMAT@" config.rocm_test_chipset = "@ROCM_TEST_CHIPSET@" +config.enable_sycl_runner = @MLIR_ENABLE_SYCL_RUNNER@ config.enable_spirv_cpu_runner = @MLIR_ENABLE_SPIRV_CPU_RUNNER@ config.enable_vulkan_runner = @MLIR_ENABLE_VULKAN_RUNNER@ config.enable_bindings_python = @MLIR_ENABLE_BINDINGS_PYTHON@ |