[NPUW] Extend remote memory check in set_tensor() with sliced tensor case #32025
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
github-actions
bot
added
category: NPU
esmirno
reviewed
Sep 9, 2025
| if (tensor->is_continuous()) { | ||
| m_input_allocated.insert(tensor->data()); | ||
| } else { | ||
| LOG_WARN("Strided remote tensor is not supported on the device! Expect worse performance due " |
Contributor
There was a problem hiding this comment.
ok in case of strided tensor - shouldn't we clear m_input_allocated - so undo what we did in line 198?
Contributor
Author
There was a problem hiding this comment.
Good point. It doesn't occur in the use-case I'm considering, but generally it could happen. I've removed the old check
esmirno
approved these changes
Sep 10, 2025
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
Merged
via the queue into
openvinotoolkit:master
with commit Sep 10, 2025
e9ab56b
198 of 200 checks passed
dmatveev
reviewed
Sep 15, 2025
Contributor
dmatveev
left a comment
dmatveev
left a comment
There was a problem hiding this comment.
Some post-merge comments here
| // ZeroRemoteTensor and ZeroHostTensor should guarantee the correct memory allocation | ||
| m_input_allocated.insert(tensor->data()); | ||
| // Also we can get a strided remote tensor. In this case the copy cannot be avoided for now. | ||
| if (m_npuw_model->global_mem_device() == "NPU") { |
Contributor
There was a problem hiding this comment.
not sure if this check is right here.
Comment on lines
+197
to
+198
| auto remote_ctx = | ||
| m_npuw_model->get_plugin()->get_core()->get_default_context(m_npuw_model->global_mem_device())._ptr; |
Contributor
There was a problem hiding this comment.
Having this on every set_tensor may be expensive.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Follow-up on #31792