Skip to content

Reenable some mxfp tests for XPU #5379

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 9 commits into from
Oct 27, 2025
+12 −2
Merged

Reenable some mxfp tests for XPU #5379

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
6ad64c8
Skip mxfp tests same as CUDA, skip large block tests for PVC and add …
HBN-MichalSzy Oct 23, 2025
3d95ac1
Skip CONST_SCALE with large blocks for MAX1000
HBN-MichalSzy Oct 23, 2025
66cb3f3
narrow down num_warps=8 cases
HBN-MichalSzy Oct 23, 2025
56e7e16
Merge remote-tracking branch 'origin/main' into dev/mxfp_matmul_tests
HBN-MichalSzy Oct 23, 2025
4d21baa
add test_mxfp8_mxfp4_matmul to mxfp_tests
HBN-MichalSzy Oct 24, 2025
aa27676
Merge remote-tracking branch 'origin/main' into dev/mxfp_matmul_tests
HBN-MichalSzy Oct 24, 2025
8744b95
cr changes
HBN-MichalSzy Oct 24, 2025
a8efff7
cr changes
HBN-MichalSzy Oct 26, 2025
b061cca
Merge branch 'main' into dev/mxfp_matmul_tests
HBN-MichalSzy Oct 26, 2025
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
10 changes: 9 additions & 1 deletion python/test/unit/language/test_matmul.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1217,7 +1217,13 @@ def test_mxfp8_mxfp4_matmul(M, N, K, BLOCK_M, BLOCK_N, BLOCK_K, NUM_STAGES, B_TR
if (A_DATA_TYPE == 'float4' and not WITH_A_SCALE) or (B_DATA_TYPE == 'float4' and not WITH_B_SCALE):
pytest.skip("Float4 without scale is tested in test_block_scale_fp4")
elif is_xpu():
pytest.xfail("XPU does not natively support scaled mxfp8 & mxfp4 matmul")
if not (WITH_A_SCALE and WITH_B_SCALE):
pytest.skip("None scale has not been tested on XPU backend")
if not (A_DATA_TYPE == "float8e5" and B_DATA_TYPE == "float4"):
pytest.skip(f"(A: {A_DATA_TYPE}, B: {B_DATA_TYPE}) has not been tested on XPU backend")
if (BLOCK_M, BLOCK_N, BLOCK_K) == (128, 256,
256) and CONST_SCALE and "GPU Max 1100" in torch.xpu.get_device_name():
pytest.skip("XPU Max 1100 does not fit in memory large block size for CONST_SCALE mxfp matmul")
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To avoid being tied to the GPU name.

      # check maximum shared memory
        if triton.runtime.driver.active.utils.get_device_properties(
                triton.runtime.driver.active.get_current_device())["max_shared_mem"] <= 196608:
            pytest.xfail("XPU: Not enough shared memory")

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done.

if not PACK_B_ALONG_K and B_DATA_TYPE != "float4":
pytest.xfail("Pack along K can only be False for float4")

Expand Down Expand Up @@ -1288,6 +1294,8 @@ def create_operand(dtype: str, size0: int, size1: int, k_dim: int, transpose: bo
kernel_kwargs = {}
if is_hip():
kernel_kwargs["matrix_instr_nonkdim"] = nonKDim
if is_xpu() and (128, 256, 256) == (BLOCK_M, BLOCK_N, BLOCK_K) and not CONST_SCALE and not PACK_B_ALONG_K:
kernel_kwargs["num_warps"] = 8
out = mxfp8_mxfp4_matmul[grid](a, b, output, a_scale, b_scale, M, N, K, stride_scale, a.stride(0), a.stride(1),
b.stride(0), b.stride(1), output.stride(0), output.stride(1), not CONST_SCALE,
dtype_converter[A_DATA_TYPE], dtype_converter[B_DATA_TYPE], BLOCK_M, BLOCK_N,
Expand Down
2 changes: 1 addition & 1 deletion scripts/test-triton.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -395,7 +395,7 @@ run_mxfp_tests() {
cd $TRITON_PROJ/python/test/unit

TRITON_DISABLE_LINE_INFO=1 TRITON_TEST_SUITE=mxfp \
run_pytest_command -vvv -n ${PYTEST_MAX_PROCESSES:-8} --device xpu intel/test_mxfp_matmul.py
run_pytest_command -vvv -n ${PYTEST_MAX_PROCESSES:-8} --device xpu intel/test_mxfp_matmul.py language/test_matmul.py::test_mxfp8_mxfp4_matmul
}

run_scaled_dot_tests() {
Expand Down
Loading