[Inductor Intel GPU backend Upstream] Reuse inductor test for Intel GPU (PART 1) (#122866)

Reuse Inductor test suite for Intel GPU including:
test_torchinductor.py
test_triton_wrapper.py
test_metrics.py
test_codecache.py
test_codegen_triton.py
test_kernel_benchmark.py
test_triton_heuristics.py

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122866
Approved by: https://github.com/EikanWang, https://github.com/jansel
This commit is contained in:
xinan.lin
2024-05-07 00:25:00 -07:00
committed by PyTorch MergeBot
parent 4dd33a1c2b
commit 78a1693266
10 changed files with 184 additions and 52 deletions

View File

@ -15,6 +15,7 @@ from torch.testing._internal.common_utils import (
skipIfTorchDynamo,
TEST_CUDA,
TEST_PRIVATEUSE1,
TEST_XPU,
)
from torch.utils.cpp_extension import CUDA_HOME, ROCM_HOME
@ -34,7 +35,7 @@ def remove_build_path():
@unittest.skipIf(
IS_ARM64 or not IS_LINUX or TEST_CUDA or TEST_PRIVATEUSE1 or TEST_ROCM,
IS_ARM64 or not IS_LINUX or TEST_CUDA or TEST_PRIVATEUSE1 or TEST_ROCM or TEST_XPU,
"Only on linux platform and mutual exclusive to other backends",
)
@torch.testing._internal.common_utils.markDynamoStrictTest