Add a inplace concat custom op based on CUDA VMM API #9126
Closed
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Outdated suggestions cannot be applied.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.
PR types
Performance optimization
PR changes
Others
Description
这一PR尝试为当前的大模型推理过程增加基于CUDA VMM API的inplace concat支持(原理类似于vAttention),从而避免在每一个解码步都复制一次整个KV Cache。
该功能暂时只实现了自定义算子,未来还需要增加相关的pass以自动适配其他模型。
目前这一PR在llama模型上应用了这一方案,在3072 input+1024 output的情况下大约有10%的提升。
目前主要的思路是:
vtensor_reserve_one_token自定义算子的语义大致如下:
目前可能存在的问题:
Moved to #9320