[Hardware][Intel CPU]use ipex varlen attention to compute prompts for better performance. #5943
Conversation
|
To speed up the CI queue, I've cancelled the distributed tests for the latest CI run in this PR since they won't pass anyway until #5905 has been merged. Now that it has been merged, please merge |
thanks for your heads up! |
There was a problem hiding this comment.
@jikunshang Thanks for submitting the PR! Left some comments.
| @@ -13,9 +13,12 @@ | |||
|
|
|||
| if is_cpu(): | |||
| try: | |||
| from vllm._ipex_ops import ipex_ops | |||
There was a problem hiding this comment.
Shouldn't we add IPEX to the CPU's dependency?
There was a problem hiding this comment.
We will add ipex dependency when its API is stable.
cc @zhouyuan @bigPYJ1151
|
@jikunshang It seems like using IPEX somehow changes the output of the models and thus fails the model tests. Could you please take a look? |
seems only happens on bloom-560m model, I have report this to ipex team. |
|
Let me convert to draft first, while re-open this when ipex fix bloom model related issue. |
| (num_tokens, self.num_heads, self.head_size), | ||
| dtype=query.dtype, | ||
| device=query.device) | ||
| # ipex-cpu provide varlen_attention API |
There was a problem hiding this comment.
LGTM! Nice addition.
| # ipex-cpu provide varlen_attention API | |
| # ipex-cpu provides varlen_attention API |
Ipex provide a
varlen_attentionAPI which could perform better on prompts computation compare totorch.nn.functional.scaled_dot_product_attention. This PR add such support.BEFORE SUBMITTING, PLEASE READ THE CHECKLIST BELOW AND FILL IN THE DESCRIPTION ABOVE
PR Checklist (Click to Expand)
Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.
PR Title and Classification
Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:
[Bugfix]for bug fixes.[CI/Build]for build or continuous integration improvements.[Doc]for documentation fixes and improvements.[Model]for adding a new model or improving an existing model. Model name should appear in the title.[Frontend]For changes on the vLLM frontend (e.g., OpenAI API server,LLMclass, etc.)[Kernel]for changes affecting CUDA kernels or other compute kernels.[Core]for changes in the core vLLM logic (e.g.,LLMEngine,AsyncLLMEngine,Scheduler, etc.)[Hardware][Vendor]for hardware-specific changes. Vendor name should appear in the prefix (e.g.,[Hardware][AMD]).[Misc]for PRs that do not fit the above categories. Please use this sparingly.Note: If the PR spans more than one category, please include all relevant prefixes.
Code Quality
The PR need to meet the following code quality standards:
format.shto format your code.docs/source/if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.Notes for Large Changes
Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with
rfc-requiredand might not go through the PR.What to Expect for the Reviews
The goal of the vLLM team is to be a transparent reviewing machine. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:
action-requiredlabel on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.Thank You
Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone!