feat: add NVIDIA Jetson platform support #11
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- README.md: Fix typo 'vllm-cl' -> 'vllm-cli' in Basic Usage section - Makefile: Add missing targets to .PHONY declaration (test-cov, format-check, type-check, ci-local, ci-matrix, pre-commit) - scripts/test_ci_locally.sh: Fix pytest-cov detection using Python import instead of shell command - scripts/test_ci_locally.sh: Add Python 3.11+ tomllib compatibility for pyproject.toml validation - docs/profiles.md: Fix grammar 'defaults configuration' -> 'default configuration' - docs/profiles.md: Update VLLM_ATTENTION_BACKEND value to match actual profile usage
- Fix nvidia-smi parsing to handle [N/A] values on Jetson devices - Add PyTorch fallback when nvidia-smi returns incomplete data - Add Jetson architectures to capability mapping (SM 8.7, 7.2, 10.x) - Update UI warning message with Jetson-specific guidance Tested on Jetson AGX Orin with CUDA 12.9. Related: Similar fix applied to vllm-playground in PR Chen-zexi#8 micytao/vllm-playground#8
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Add support for NVIDIA Jetson platforms (Orin, Thor, Xavier) which were previously not detected as GPUs.
Problem
On Jetson devices,
nvidia-smireturns[N/A]for memory and utilization fields:This caused
int("[N/A]")to raiseValueError, and the PyTorch fallback was never triggered.Solution
_safe_int()helper to handle[N/A]values safelyTesting
Tested on Jetson AGX Orin with CUDA 12.9 - GPU now detected correctly via PyTorch fallback.
Related
Similar fix was applied to vllm-playground: PR #8