Skip to content

Arm backend: Update failures to slice operator for int16x8 #15104

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 4 commits into from
Oct 16, 2025
Merged

Arm backend: Update failures to slice operator for int16x8 #15104

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
ef93d92
Arm backend: Update failures to slice operator for int16x8
SaoirseARM Oct 13, 2025
71fbef4
Arm backend: Remove pytest.mark.fails fomr testcases
SaoirseARM Oct 13, 2025
d8716ff
Merge branch 'main' into toupstream/slice_op
SaoirseARM Oct 16, 2025
a627e26
Fix linter issue
SaoirseARM Oct 16, 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
18 changes: 4 additions & 14 deletions backends/arm/test/ops/test_slice.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,6 @@

from typing import Tuple

import pytest
import torch
from executorch.backends.arm.quantizer.arm_quantizer import (
get_symmetric_a16w8_quantization_config,
Expand Down Expand Up @@ -78,8 +77,8 @@ def test_slice_tensor_tosa_INT_nhwc(test_data: torch.Tensor):


x_fails = {
"ones_slice_3": "MLETORCH-1402: Slice operator has incorrect number of inputs",
"ones_slice_4": "MLETORCH-1402: Slice operator has incorrect number of inputs",
"ones_slice_3": "MLETORCH-1402: Compiler limitation when passing more than 255 char as argument to FVP.",
"ones_slice_4": "MLETORCH-1402: Compiler limitation when passing more than 255 char as argument to FVP.",
}


Expand Down Expand Up @@ -153,9 +152,6 @@ def get_symmetric_a16w8_slice_quantizer(per_channel_quantization=False):


@common.parametrize("test_data", test_data_suite)
@pytest.mark.xfail(
reason="missing int16 slice ops support; fails at TOSA reference model with Unsupported operation type or rank. See: https://github.com/pytorch/executorch/issues/13976"
)
def test_slice_tensor_16a8w_tosa_INT(test_data: torch.Tensor):
"""Test slice operation with 16A8W quantization (16-bit activations, 8-bit weights)"""
per_channel_quantization = False
Expand All @@ -179,11 +175,8 @@ def test_slice_tensor_16a8w_tosa_INT(test_data: torch.Tensor):
pipeline.run()


@common.parametrize("test_data", test_data_suite)
@common.parametrize("test_data", test_data_suite, x_fails)
@common.XfailIfNoCorstone300
@pytest.mark.xfail(
reason="Vela compilation fails with 'Invalid arguments' for int16 slice operations"
)
def test_slice_tensor_16a8w_u55_INT16(test_data: torch.Tensor):
"""Test slice operation with 16A8W quantization on U55 (16-bit activations, 8-bit weights)"""
per_channel_quantization = False
Expand All @@ -206,11 +199,8 @@ def test_slice_tensor_16a8w_u55_INT16(test_data: torch.Tensor):
pipeline.run()


@common.parametrize("test_data", test_data_suite)
@common.parametrize("test_data", test_data_suite, x_fails)
@common.XfailIfNoCorstone320
@pytest.mark.xfail(
reason="Vela compilation fails with 'Invalid arguments' for int16 slice operations"
)
def test_slice_tensor_16a8w_u85_INT16(test_data: torch.Tensor):
"""Test slice operation with 16A8W quantization on U85 (16-bit activations, 8-bit weights)"""
per_channel_quantization = False
Expand Down
Loading