[ci] add test workflow for 3090 machine #3561
Open
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
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for running tests on a 3090 machine by introducing new configuration flags, updating test annotations, and adding a dedicated workflow file. The key changes include:
- Injecting environment‑specific command line flags (e.g. “cache‑max‑entry‑count”) for the 3090 machine.
- Updating test functions to include a new @pytest.mark.test_3090 annotation.
- Adding a new daily ETE test workflow file and a dedicated 3090 configuration file.
Reviewed Changes
Copilot reviewed 21 out of 21 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| autotest/utils/run_client_chat.py | Adds extra CLI flag when running on env '3090'. |
| autotest/utils/quantization_utils.py | Adjusts batch-size logic for 3090 and applies conditional cuda_prefix usage. |
| aut 8000 otest/utils/pipeline_chat.py | Updates default parameter handling for extra and injects cache settings for 3090. |
| autotest/utils/config_utils.py | Uses OrderedDict to remove duplicate model entries in quantization config. |
| Various test files | Adds @pytest.mark.test_3090 annotations for tests targeting the 3090 machine. |
| autotest/tools/pipeline/mllm_case.py & llm_case.py | Adds support for passing cache settings via extra configuration. |
| autotest/config.yaml & config-3090.yaml | Introduces env_tag for distinguishing configurations. |
| .github/workflows/daily_ete_test_3090.yml | New workflow configuration for daily testing on the 3090 machine. |
autotest/utils/pipeline_chat.py
Outdated
| @@ -14,7 +14,7 @@ def run_pipeline_chat_test(config, | |||
| model_case, | |||
| backend_type, | |||
| worker_id: str = '', | |||
| extra: object = None, | |||
| extra: object = {}, | |||
There was a problem hiding this comment.
Avoid using a mutable default argument ({}); consider using None as the default and initializing extra inside the function to prevent unintended shared state across calls.
Suggested change
| extra: object = {}, | |
| extra: object = None, |
Copilot uses AI. Check for mistakes.
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for running test workflows on a 3090 machine by updating various test files, configuration files, and adding a new GitHub workflow. Key changes include:
- Inserting environment‐specific logic (checking for env_tag "3090") in the client, quantization, and pipeline modules.
- Adding new pytest marks (@pytest.mark.test_3090) to multiple test cases and updating configuration files to support 3090 settings.
- Introducing a new GitHub workflow file (.github/workflows/daily_ete_test_3090.yml) to orchestrate daily tests on the 3090 machine.
Reviewed Changes
Copilot reviewed 23 out of 23 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| autotest/utils/run_client_chat.py | Added environment check for "3090" to adjust extra cli flags. |
| autotest/utils/quantization_utils.py | Refactored command construction for quantization with environment-specific batch sizes. |
| autotest/utils/pipeline_chat.py | Included test-specific overrides via env_tag "3090" and modified extra parameters for chat tests. |
| autotest/utils/config_utils.py | Updated quantization model list construction to remove duplicate entries using OrderedDict. |
| Various test files under autotest/tools | Added new pytest marker (@pytest.mark.test_3090) for 3090-targeted tests. |
| autotest/config.yaml & autotest/config-3090.yaml | Introduced and updated env_tag settings to distinguish between standard and 3090 environments. |
| .github/workflows/daily_ete_test_3090.yml | Added a comprehensive workflow to run daily E2E tests on the 3090 machine. |
Comments suppressed due to low confidence (1)
autotest/utils/pipeline_chat.py:33
- [nitpick] The extra parameter 'cache-max-entry-count' is set to different values across modules (e.g., 0.7 in run_client_chat, 0.6 and 0.5 in pipeline tests). Verify these values are intentionally different; if not, consider consolidating them to avoid confusion.
if str(config.get('env_tag')) == '3090':
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
add test workflow for 3090 machine