[QDP] PyTorch CUDA stream‑aware encode for GPU tensors #930
+341
−33
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
Signed-off-by: 400Ping <fourhundredping@gmail.com>
400Ping
changed the title
400Ping
changed the title
Contributor
|
@400Ping thanks for the patch! |
Member
Author
|
@CheyuWu PTAL |
Signed-off-by: 400Ping <fourhundredping@gmail.com>
CheyuWu
approved these changes
Jan 27, 2026
Contributor
|
I think in this pr should add tests for the feature. like I mentioned previously. |
viiccwen
reviewed
Jan 29, 2026
Contributor
viiccwen
left a comment
viiccwen
left a comment
There was a problem hiding this comment.
overall LGTM!
left comments.
cursor
bot
force-pushed
the
qdp/pytorch-direct-gpu-stream-sync
branch
from
cd52809 to
4b71ad0
Compare
400Ping
force-pushed
the
qdp/pytorch-direct-gpu-stream-sync
branch
from
cd52809 to
4b71ad0
Compare
Signed-off-by: 400Ping <jiekaichang@apache.org>
Signed-off-by: 400Ping <jiekaichang@apache.org>
Signed-off-by: 400Ping <jiekaichang@apache.org>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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Outdated suggestions cannot be applied.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.
Purpose of PR
lib.rs: obtain PyTorch’s current CUDA stream pointer and route CUDA-tensor encoding through stream-aware core APIs to prevent cross-stream races.lib.rs: addencode_from_gpu_ptr_with_streamandencode_batch_from_gpu_ptr_with_stream, and synchronize on the provided stream; keep default-stream entry points intact.amplitude.rs: add a stream-aware GPU norm path so normalization and kernels run on the same stream before host reads.Related Issues or PRs
Related to #726
Changes Made
Breaking Changes
Checklist