[GPU] Optimze copy tensor with padding #32371
Closed
+76
−8
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
Tensor layout related properties are calculated once and used those cached values during per element offset calculation. [ONNX] Introduce GraphIterator interface in frontend (openvinotoolkit#32325) - *Introduce GraphIterator for ONNX* - *CVS-156050* --------- Signed-off-by: Maxim Vafin <[email protected]>
convert_and_copy_padded_source()
susbhere
force-pushed
the
copy_padded_optimization
branch
from
e130e73 to
7d2720c
Compare
mklimenk
reviewed
Oct 14, 2025
susbhere
force-pushed
the
copy_padded_optimization
branch
from
174e8ba to
f0bd92e
Compare
susbhere
force-pushed
the
copy_padded_optimization
branch
3 times, most recently
from
99c5e9f to
8c8331b
Compare
susbhere
force-pushed
the
copy_padded_optimization
branch
from
8c8331b to
300a281
Compare
Tensor layout related properties are calculated once and used those cached values during per element offset calculation. [ONNX] Introduce GraphIterator interface in frontend (openvinotoolkit#32325) - *Introduce GraphIterator for ONNX* - *CVS-156050* --------- Signed-off-by: Maxim Vafin <[email protected]>
convert_and_copy_padded_source()
susbhere
force-pushed
the
copy_padded_optimization
branch
from
300a281 to
1b0f98e
Compare
yeonbok
changed the title
yeonbok
reviewed
Oct 16, 2025
yeonbok
reviewed
Oct 16, 2025
yeonbok
reviewed
Oct 16, 2025
yeonbok
reviewed
Oct 16, 2025
…e/openvino into copy_padded_optimization
Tensor layout related properties are calculated once and used those cached values during per element offset calculation. [ONNX] Introduce GraphIterator interface in frontend (openvinotoolkit#32325) - *Introduce GraphIterator for ONNX* - *CVS-156050* --------- Signed-off-by: Maxim Vafin <[email protected]>
convert_and_copy_padded_source()
susbhere
force-pushed
the
copy_padded_optimization
branch
from
1b0f98e to
e41a06d
Compare
…e/openvino into copy_padded_optimization
susbhere
force-pushed
the
copy_padded_optimization
branch
from
6d6398e to
088c447
Compare
…ffset calculation to common_utils.cpp
susbhere
force-pushed
the
copy_padded_optimization
branch
from
088c447 to
113dd78
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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 got messed up with multiple merge commits. Closing this and opening new one:
#32461
Details:
cached values during per element offset calculation. This brings ~200x improvement in wait time between two queries for PhiSlica model. That means a user has to wait only for 0.36 sec (instead of 74 sec !!!) between two queries. These numbers are from LNL.
Tickets: