Skip to content

Set PyTorch parallelism limits for aarch64/{generic,neoverse_n1,neoverse_v1}, allow more failing tests for A64FX #181

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

Open
bedroge wants to merge 7 commits into
base: main
Choose a base branch
from
Open

Set PyTorch parallelism limits for aarch64/{generic,neoverse_n1,neoverse_v1}, allow more failing tests for A64FX #181

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a556782
set PyTorch CPU limits for aarch64/{generic,neoverse_n1,neoverse_v1}
bedroge Mar 12, 2026
cc9e9e8
use eb_hooks.py from git repo
bedroge Mar 12, 2026
fa76fc6
use 8 cores for PyTorch on A64FX
bedroge Mar 15, 2026
240e649
Merge branch 'main' of github.com:EESSI/software-layer-scripts into m…
bedroge Jun 22, 2026
ae3ae48
allow 41 failing tests for PyTorch 2.9.1 on A64FX
bedroge Jun 22, 2026
3b8d4da
Merge branch 'main' into more_pytorch_fixes
bedroge Sep 7, 2026
bf9dab1
remove hooks override
bedroge Sep 7, 2026
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
20 changes: 15 additions & 5 deletions eb_hooks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1882,11 +1882,15 @@ def pre_test_hook_increase_max_failed_tests_arm_PyTorch(self, *args, **kwargs):
See https://github.com/EESSI/software-layer/issues/461
"""
cpu_target = get_eessi_envvar('EESSI_SOFTWARE_SUBDIR')
if self.name == 'PyTorch' and self.version == '2.1.2':
if get_cpu_architecture() == AARCH64:
self.cfg['max_failed_tests'] = 10
if cpu_target in [CPU_TARGET_CASCADELAKE, CPU_TARGET_ICELAKE, CPU_TARGET_SAPPHIRE_RAPIDS]:
self.cfg['max_failed_tests'] = 4
if self.name == 'PyTorch':
if self.version == '2.1.2':
if get_cpu_architecture() == AARCH64:
self.cfg['max_failed_tests'] = 10
if cpu_target in [CPU_TARGET_CASCADELAKE, CPU_TARGET_ICELAKE, CPU_TARGET_SAPPHIRE_RAPIDS]:
self.cfg['max_failed_tests'] = 4
if self.version == '2.9.1':
if cpu_target == CPU_TARGET_A64FX:
self.cfg['max_failed_tests'] = 41


def pre_test_hook_ignore_failing_tests_OpenBabel_a64fx(self, *args, **kwargs):
Expand Down Expand Up @@ -2491,6 +2495,12 @@ def set_maximum(parallel, max_value):
# so reduce parallelism when running tests (also impacts build)
'*': (divide_by_factor, 2),
},
'PyTorch': {
CPU_TARGET_AARCH64_GENERIC: (divide_by_factor, 2),
CPU_TARGET_A64FX: (set_maximum, 8),
CPU_TARGET_NEOVERSE_N1: (divide_by_factor, 2),
CPU_TARGET_NEOVERSE_V1: (divide_by_factor, 2),
},
'QuantumESPRESSO': {
CPU_TARGET_A64FX: (set_maximum, 6),
},
Expand Down
Loading