[Mac OS] Default --device to cpu on macOS by AntoineRichard · Pull Request #8190 · isaac-sim/IsaacLab

AntoineRichard · 2026-09-30T13:55:49Z

Description

--device defaults to cuda:0. On macOS, which has no CUDA, every command without an explicit --device cpu fails at simulation start (Torch not compiled with CUDA enabled in SimulationContext._init_usd_physics_scene).

The default is now cpu on macOS and stays cuda:0 elsewhere. It checks sys.platform rather than torch.cuda.is_available() because the launcher defers importing torch until after Kit starts (see the comment in _resolve_device_settings). An explicit --device still wins, and device_explicit is untouched, so the XR device logic is unaffected.

launch_simulation logs a warning when it applies this default (No --device given; using 'cpu' because macOS has no CUDA.), so the switch is visible. Passing --device explicitly suppresses it. Hosts that request a CUDA device without having CUDA are handled separately by #8188, which falls back to the CPU with its own warning.

Part of the experimental macOS (Apple silicon, CPU-only) support series.

Type of change

New feature (non-breaking change which adds functionality)

Validation

On macOS arm64, rebased on develop with [Core] Decide visualizers and the device once in launch_simulation #8125: isaaclab train --rl_library rsl_rl --task Isaac-Cartpole-Direct physics=newton_mjwarp --max_iterations 2 (no --device) logs the warning, reports Device cpu, and trains to completion. This also needs [Mac OS] [RL] Run RL agents on the simulation device and raise without CUDA #8188.
With --device cpu, no warning is logged.

Checklist

I have read and understood the contribution guidelines
I have run the pre-commit checks with uv run isaaclab --format (all hooks pass except check-git-lfs-pointers, which needs git-lfs installed locally)
I have made corresponding changes to the documentation
My changes generate no new warnings
I have validated the changed behavior with focused coverage and applied the test-audit criteria to any test changes
I have added a changelog fragment under source/<pkg>/changelog.d/ for every touched package (do not edit CHANGELOG.rst or bump package versions — CI handles that)
I have added my name to the CONTRIBUTORS.md or my name already exists there

greptile-apps · 2026-09-30T14:01:02Z

Confidence Score: 5/5

[Medium risk] Changes the default compute device on macOS.

This PR appears safe to merge as the scoped launcher-default change.

Summary

The PR makes the launcher’s --device default cpu on macOS while retaining cuda:0 elsewhere.

Updates argument help and the changelog.
Adds a parametrized test for the three platform defaults and the unset explicit-device marker.

_{Reviews (1) · Last reviewed commit: "Default --device to cpu on macOS"}

isaaclab-review-bot

Isaac Lab Review Bot

The change correctly makes the --device parser default cpu on macOS while retaining cuda:0 elsewhere, preserves explicit-device tracking, and documents the behavior in CLI help and the package changelog. The new regression test contains one redundant non-Darwin parameter row that should be removed.

Design and architecture: The platform-dependent default remains at the existing launcher argument-registration boundary and uses sys.platform, avoiding an early torch import as required by launcher initialization. No new helper, flag, or configuration surface is introduced.
API: The --device option retains its type and _StoreAndMarkExplicit action. Explicit values still take precedence, default values remain unmarked as explicit, and the help text now describes both platform defaults.
Compatibility and deprecation: Breaking changes: none identified. Linux and Windows retain the existing cuda:0 default, while macOS changes to the usable CPU default. No launcher attribute, environment-variable, or explicit-device contract is removed, and the change has a package changelog fragment.
Implementation: The default expression is evaluated when launcher arguments are registered, and its value flows through the existing device-resolution paths without changing device_explicit, XR pinning, or physics-device handling. The implementation is appropriately localized.
Style consistency: The comment explains the non-obvious reason for checking sys.platform instead of torch availability, and the updated docstring and help wording match the changed behavior and surrounding launcher style.
Test quality: The added parser-boundary test protects the platform-specific default and verifies that a default is not marked explicit. However, the linux and win32 cases exercise the same sys.platform != "darwin" branch with the same expected result, violating the test-audit requirement that parameter rows cover distinct paths. Keep the Darwin case and one representative non-Darwin case.

Minor fixes needed. Posted 1 actionable finding inline.

Automated review; human maintainers own approval decisions.

isaaclab-review-bot · 2026-09-30T14:03:10Z

    return launcher


+@pytest.mark.parametrize(("platform", "expected"), [("linux", "cuda:0"), ("win32", "cuda:0"), ("darwin", "cpu")])


🔵 Suggestion · Test Quality — Redundant non-Darwin parametrize row

The default is chosen by the single predicate sys.platform == "darwin", so the linux and win32 rows execute the identical else branch and assert the identical value; neither can fail without the other. The test-audit guidance rejects parametrize rows whose axis value selects no distinct code path. Keep darwin plus one representative non-Darwin platform.

StafaH · 2026-09-30T15:23:24Z



+@pytest.mark.parametrize(("platform", "expected"), [("linux", "cuda:0"), ("win32", "cuda:0"), ("darwin", "cpu")])
+def test_device_defaults_to_cpu_on_macos(platform, expected, monkeypatch):


We can drop the test

hujc7 · 2026-09-30T18:19:04Z

run-ci

…s-cpu-device-default

AntoineRichard · 2026-10-02T08:29:47Z

run-ci

…t CUDA (#8188) # Description Hosts without CUDA, such as macOS, could not train with RSL-RL even with `--device cpu`. The agent configuration's device (`cuda:0` by default) was used to build the policy, so the run failed with `AssertionError: Torch not compiled with CUDA enabled`. This PR uses the device `launch_simulation` already resolves (since #8125), and sets it in each RL library's shared agent-config setup: - **Clear error without CUDA.** `_resolve_device` in `isaaclab.app.sim_launcher` raises `RuntimeError: Device 'cuda:0' requires CUDA, which is not available; pass --device cpu.` when the resolved device is a CUDA device and CUDA is unavailable. Before, the run failed later inside PyTorch. On macOS, #8190 defaults `--device` to `cpu`. - **The agent runs on the resolved device, set in one shared setup per library.** Each library's config setup takes the device (`args_cli.device`, which `launch_simulation` resolves) as an explicit argument. Every workflow calls it after launch, before the agent is built or its config is saved. This is unconditional, not only for `--distributed`: - **RSL-RL:** the existing `update_rsl_rl_cfg(agent_cfg, args_cli)` sets `agent_cfg.device = args_cli.device`, for train, play, and benchmark train. Export and benchmark play, whose CLIs don't define the RSL-RL training arguments, set the device directly. - **RL-Games:** the new `update_rl_games_cfg(agent_cfg, args_cli)` mirrors `update_rsl_rl_cfg`: it sets `device` / `device_name` and the seed for train, play, export, and both benchmarks. RL-Games export gains `--seed`, matching RSL-RL export, so every RL-Games workflow can call it. - **Stable-Baselines3:** the new `update_sb3_cfg(agent_cfg, args_cli)` mirrors `update_rsl_rl_cfg`: it sets the device and seed for train, play, and both benchmarks. Both `PPO.load()` calls pass `device=agent_cfg["device"]`, so checkpoint loading follows the same setup. skrl and TorchRL already used the simulation's device. The saved agent config records the device actually used, which addresses the bot findings. Closes #8193. Part of the experimental macOS (Apple silicon, CPU-only) support series. ## Type of change - Bug fix (non-breaking change which fixes an issue) - Breaking change (existing functionality will not work without user modification) **Breaking:** outside distributed runs, the RSL-RL, RL-Games and SB3 CLI workflows used to ignore `--device` for the agent: | Command | Before | After | |---|---|---| | `--device cpu` | Simulation on CPU, agent on `cuda:0` | Both on the CPU | | `--device cuda:1` | Simulation on GPU 1, agent on GPU 0 (see #2148) | Both on GPU 1 | The agent configuration's `device` is now overwritten by these workflows. The wrappers still support running the simulation and the agent on different devices for custom scripts; `source/isaaclab_tasks/test/core/test_rl_device_separation.py` covers that and is unaffected. `isaaclab_rl` gets a `.major` fragment. ## Validation On macOS arm64, CPU-only: | Run | Result | |---|---| | RSL-RL train, `--device cpu` | Trains | | SB3 train, `--device cpu` | Trains (`Using cpu device`) | | RL-Games train, `--device cpu` (512 envs) | Completes 2 epochs | | RSL-RL train, `--device cuda:0` | Stops with `RuntimeError: Device 'cuda:0' requires CUDA, which is not available; pass --device cpu.` | `source/isaaclab_rl/test/test_entrypoints.py`, `test_entrypoints_common.py`, and `source/isaaclab/test/app/test_launch_simulation_require_kit.py`: 83 passed. The one failure is `test_skrl_training_parser_leaves_algorithm_implicit`, which needs `skrl`, not installed locally. `test_run_summary_reports_the_launch_resolved_visualizers_and_device` now launches on `cpu`: it only used `cuda:0` as a label and still checks the device resolved after launch. ## Checklist - [x] I have read and understood the [contribution guidelines](https://isaac-sim.github.io/IsaacLab/main/source/refs/contributing.html) - [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with `uv run isaaclab --format` (all hooks pass except `check-git-lfs-pointers`, which needs git-lfs installed locally) - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [ ] I have validated the changed behavior with focused coverage and applied the [test-audit criteria](https://github.com/isaac-sim/IsaacLab/blob/develop/skills/developer/test-audit/SKILL.md) to any test changes - [x] I have added a changelog fragment under `source/<pkg>/changelog.d/` for every touched package (do **not** edit `CHANGELOG.rst` or bump package versions — CI handles that) - [x] I have added my name to the `CONTRIBUTORS.md` or my name already exists there

Default --device to cpu on macOS

450c2be

github-actions Bot added the isaac-lab Related to Isaac Lab team label Sep 30, 2026

AntoineRichard mentioned this pull request Sep 30, 2026

[Mac OS] Support macOS arm64 (CPU-only) in the uv environment #8191

Open

7 tasks done

AntoineRichard marked this pull request as ready for review September 30, 2026 13:58

AntoineRichard requested a review from a team September 30, 2026 13:58

AntoineRichard requested review from kellyguo11 and ooctipus as code owners September 30, 2026 13:58

AntoineRichard changed the title ~~Default --device to cpu on macOS~~ [Mac OS] Default --device to cpu on macOS Sep 30, 2026

isaaclab-review-bot Bot reviewed Sep 30, 2026

View reviewed changes

StafaH approved these changes Sep 30, 2026

View reviewed changes

Drop the macOS device default test

e6e562c

isaaclab-bot Bot added ci:run-docker Trigger the on-demand Docker and GPU CI workflow and removed ci:run-docker Trigger the on-demand Docker and GPU CI workflow labels Sep 30, 2026

hujc7 approved these changes Sep 30, 2026

View reviewed changes

AntoineRichard added 2 commits October 1, 2026 17:52

Merge remote-tracking branch 'origin/develop' into antoiner/feat/maco…

5ee5a39

…s-cpu-device-default

Warn when macOS defaults --device to cpu

6ba0189

AntoineRichard requested review from fatimaanes and pbarejko as code owners October 1, 2026 15:54

isaaclab-bot Bot added ci:run-docker Trigger the on-demand Docker and GPU CI workflow and removed ci:run-docker Trigger the on-demand Docker and GPU CI workflow labels Oct 2, 2026

AntoineRichard enabled auto-merge (squash) October 2, 2026 08:52

AntoineRichard mentioned this pull request Oct 5, 2026

[Mac OS] [RL] Run RL agents on the simulation device and raise without CUDA #8188

Merged

5 of 7 tasks

kellyguo11 approved these changes Oct 5, 2026

View reviewed changes

AntoineRichard merged commit 13a958a into isaac-sim:develop Oct 5, 2026
58 checks passed

AntoineRichard mentioned this pull request Oct 6, 2026

[Bug Fix] Add missing sys import to simulation launcher #8306

Closed

8 tasks done

Navigation Menu

Repository navigation

[Mac OS] Default --device to cpu on macOS - #8190

[Mac OS] Default --device to cpu on macOS#8190
AntoineRichard merged 4 commits into
isaac-sim:developfrom
AntoineRichard:antoiner/feat/macos-cpu-device-default

AntoineRichard commented Sep 30, 2026 •

edited

Loading

Uh oh!

greptile-apps Bot commented Sep 30, 2026 •

edited

Loading

Uh oh!

isaaclab-review-bot Bot left a comment

Uh oh!

isaaclab-review-bot Bot Sep 30, 2026

Uh oh!

StafaH Sep 30, 2026

Uh oh!

hujc7 commented Sep 30, 2026

Uh oh!

AntoineRichard commented Oct 2, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

		return launcher


		@pytest.mark.parametrize(("platform", "expected"), [("linux", "cuda:0"), ("win32", "cuda:0"), ("darwin", "cpu")])



		@pytest.mark.parametrize(("platform", "expected"), [("linux", "cuda:0"), ("win32", "cuda:0"), ("darwin", "cpu")])
		def test_device_defaults_to_cpu_on_macos(platform, expected, monkeypatch):

Conversation

AntoineRichard commented Sep 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Type of change

Validation

Checklist

Uh oh!

greptile-apps Bot commented Sep 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Confidence Score: 5/5

Summary

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

Isaac Lab Review Bot

Uh oh!

isaaclab-review-bot Bot Sep 30, 2026

Choose a reason for hiding this comment

Uh oh!

StafaH Sep 30, 2026

Choose a reason for hiding this comment

Uh oh!

hujc7 commented Sep 30, 2026

Uh oh!

AntoineRichard commented Oct 2, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

AntoineRichard commented Sep 30, 2026 •

edited

Loading

greptile-apps Bot commented Sep 30, 2026 •

edited

Loading