[0.9.1] fix dbo error info #2358
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request refactors the error handling for when both DBO and speculative decoding are enabled, moving the check to an earlier stage in the configuration validation. This is a good improvement for user experience. However, a critical typo was introduced in vllm_ascend/platform.py that will cause a runtime error and needs to be fixed.
| if enforce_eager or compilation_config.level == CompilationLevel.NO_COMPILATION: | ||
| logger.info("Compilation disabled, using eager mode by default") | ||
| compilation_config.level = CompilationLevel.NO_COMPILATION | ||
| elif compilation_config.level != CompilationLevel.PIECEWISE: | ||
| logger.warning( | ||
| "NPU does not support %s compilation level. Setting level to NO_COMPILATION", | ||
| compilation_config.level) | ||
| compilation_config.level = CompilationLevel.NO_COMPILATION | ||
| compilation_config.level = CompilationLevel.NfvO_COMPILATION |
There was a problem hiding this comment.
There appears to be a typo in the compilation level. NfvO_COMPILATION should be NO_COMPILATION. This will likely cause an AttributeError at runtime.
| compilation_config.level = CompilationLevel.NfvO_COMPILATION | |
| compilation_config.level = CompilationLevel.NO_COMPILATION |
Signed-off-by: 赵江江 <[email protected]>
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
What this PR does / why we need it?
Does this PR introduce any user-facing change?
How was this patch tested?