forked from apache/tvm
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[Runtime] PagedKVCache execute data copy on a separate stream (apache…
…#16692) This PR enhances PagedKVCache with the copy stream separation. In detail, for CUDA and ROCm backend, we create a standalone copy stream for the copy of auxiliary data structure from CPU to GPU. Furthermore, we move the copy from BeginForward to Attention, which means it's no longer eagerly executed, instead, becoming lazily executed when Attention computation is needed. By making these changes, we are able to overlap the auxiliary data copy time (on the copy stream) with the model forward computation that happens before the first Attention. As a result, we can hide some of the copy latency. This PR also bumps the version of FlashInfer for the copy stream support.
- Loading branch information
1 parent
484c6c1
commit d50b364
Showing
2 changed files
with
106 additions
and
57 deletions.
There are no files selected for viewing
Submodule flashinfer
updated
53 files
This file contains 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