Skip to content

[SYCL][E2E] Fix lit for arch selection and l0v2 adapter #20728

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 6 commits into from
Nov 28, 2025
Merged

[SYCL][E2E] Fix lit for arch selection and l0v2 adapter #20728

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
b262dc6
fix lit for arch selection and l0v2 adapter
mateuszpn Nov 21, 2025
08db888
apply comments
mateuszpn Nov 24, 2025
c5fa726
fix affinity mask setting
mateuszpn Nov 25, 2025
ed8add4
some features were missing when using arch-based selection
mateuszpn Nov 25, 2025
770fe63
apply comments
mateuszpn Nov 26, 2025
0382494
Merge branch 'intel:sycl' into lit-fix-v2
mateuszpn Nov 27, 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 sycl/test-e2e/format.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -334,6 +334,14 @@ def get_extra_env(sycl_devices):
expanded = "env"

extra_env = get_extra_env([parsed_dev_name])
backend, device = parsed_dev_name.split(":", 1)
Copy link
Contributor

@sarnex sarnex Nov 21, 2025

Choose a reason for hiding this comment

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

The example of the intended lit string after the change in the PR description is

llvm-lit --param "sycl_devices=level_zero_v2:arch-intel_gpu_mtl_u" sycl/test-e2e

but I was expecting something like

llvm-lit --param "sycl_devices='level_zero_v1:gpu;level_zero_v2:gpu" sycl/test-e2e

Can you clarify on the intended use case? Thanks

Copy link
Contributor Author

@mateuszpn mateuszpn Nov 24, 2025

Choose a reason for hiding this comment

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

The goal is to force tests to run on specified device, eg. integrated gpu, even when other devices are present. The option was available before (see lit.cpg.py:901). Although it is rarely used, it is still perfect for testing integrated GPUs. Unfortunately, due to a bug, it did not work with the _v1 or _v2 postfixes in the backend name. The PR fixes this bug.

device_selector = parsed_dev_name
if backend == "level_zero" and device.isdigit():
# level_zero:0 maps to a selector for the first GPU, so pin
# execution to the requested device via ZE_AFFINITY_MASK.
extra_env.append(f"ZE_AFFINITY_MASK={device}")
device_selector = f"{backend}:0"

if extra_env:
expanded += " {}".format(" ".join(extra_env))

Expand All @@ -343,7 +351,7 @@ def get_extra_env(sycl_devices):
expanded += " env UR_LOADER_USE_LEVEL_ZERO_V2=0"

expanded += " ONEAPI_DEVICE_SELECTOR={} {}".format(
parsed_dev_name, test.config.run_launcher
device_selector, test.config.run_launcher
)
cmd = directive.command.replace("%{run}", expanded)
# Expand device-specific condtions (%if ... %{ ... %}).
Expand Down
7 changes: 5 additions & 2 deletions sycl/test-e2e/lit.cfg.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -917,12 +917,14 @@ def get_sycl_ls_verbose(sycl_device, env):

env = copy.copy(llvm_config.config.environment)

backend_for_selector = backend.replace("_v2", "").replace("_v1", "")

# Find all available devices under the backend
env["ONEAPI_DEVICE_SELECTOR"] = backend + ":*"
env["ONEAPI_DEVICE_SELECTOR"] = backend_for_selector + ":*"

detected_architectures = []

platform_devices = remove_level_zero_suffix(backend + ":*")
platform_devices = backend_for_selector + ":*"

for line in get_sycl_ls_verbose(platform_devices, env).stdout.splitlines():
if re.match(r" *Architecture:", line):
Expand Down Expand Up @@ -1112,6 +1114,7 @@ def get_sycl_ls_verbose(sycl_device, env):
features.update(sg_size_features)
features.update(architecture_feature)
features.update(device_family)
features.update(aspects)
Copy link
Contributor

@aelovikov-intel aelovikov-intel Nov 25, 2025

Choose a reason for hiding this comment

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

Why this line? Isn't 1120 enough?

Copy link
Contributor Author

@mateuszpn mateuszpn Nov 26, 2025

Choose a reason for hiding this comment

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

When using arch-selection, the device is not :gpu and 'gpu' is not added to the features. The questioned line solves this in simple way, but with overhead - adds more then only 'gpu'. Replaced it with clean solution


be, dev = sycl_device.split(":")
features.add(dev.replace("fpga", "accelerator"))
Expand Down