DEVICE/API: Simplify createGpuXferReq to single-step API - WIP #894
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
michal-shalev
requested review from
a team,
brminich,
gleon99 and
yosefe
as code owners
|
👋 Hi michal-shalev! Thank you for contributing to ai-dynamo/nixl. Your PR reviewers will review your contribution then trigger the CI to test your changes. 🚀 |
michal-shalev
changed the title
Signed-off-by: Michal Shalev <[email protected]>
michal-shalev
force-pushed
the
nixl-single-step-gpu-xfer-creation
branch
from
b0fe40c to
c3d2b1e
Compare
itayalroy
reviewed
Oct 12, 2025
| nixl_status_t | ||
| createGpuXferReq(const nixlXferReqH &req_hndl, nixlGpuXferReqH &gpu_req_hndl) const; | ||
| createGpuXferReq(const nixl_xfer_dlist_t &local_descs, | ||
| const nixl_xfer_dlist_t &remote_descs, |
There was a problem hiding this comment.
I think it's confusing to have local_descs and remote_descs with different lengths, which implies that the last len(remote_descs) - len(local_descs) remote descs are inline only.
Instead, I think we should enforce len(local_descs) == len(remote_descs) and introduce a third parameter- inline_descs which will describe remote memory that has no corresponding src buffer (and therefore can only be written into with inline data)
Signed-off-by: Michal Shalev <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
THIS PR IS PENDING openucx/ucx#10945
What?
Replace two-step GPU transfer request creation with a single
createGpuXferReqAPI that takes descriptor lists directly and returns both transfer request and GPU request handles.Why?
Simplify the API by reducing GPU transfer request creation from two separate function calls to a single call.
How?
Updated
createGpuXferReqto substitute calls tocreateXferReq+createGpuXferReq. Removed the oldcreateGpuXferReqmethod and updated tests accordingly.