[Experiment] Make resize block sizes tweakable via environment variables. #5301
Conversation
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
mzient
changed the title
|
CI MESSAGE: [12516031]: BUILD STARTED |
|
CI MESSAGE: [12516031]: BUILD PASSED |
|
!build |
|
CI MESSAGE: [12818339]: BUILD STARTED |
|
!build |
|
CI MESSAGE: [12818707]: BUILD STARTED |
|
CI MESSAGE: [12818707]: BUILD FAILED |
|
!build |
|
CI MESSAGE: [13028849]: BUILD STARTED |
|
CI MESSAGE: [13028849]: BUILD FAILED |
Signed-off-by: Michal Zientkiewicz <[email protected]>
Signed-off-by: Kamil Tokarski <[email protected]>
mzient
force-pushed
the
AdjustableResizeBlockSizes
branch
from
2002cc2
to
cc95d44
Compare
|
The tweaks have been used to establish optimum settings. The adjustment is merged as #5382 . |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Category:
Other Performance tuning
Description:
The logical blocks seem to have been too large for resize and resulted in GPU underutilization.
This PR adds two environment variables that control the block sizes:
DALI_RESIZE_CUDA_BLOCK - for the CUDA block
DALI_RESIZE_MAX_ELEMENTS_PER_BLOCK - for the logical block
Previous defaults were:
For 2D: 768 and unlimited (32image_width for vertical pass and/ 32image_height for horizontal pass).
For 3D: 256 and 262,144
Additional information:
Affected modules and functionalities:
Key points relevant for the review:
Tests:
Resampling kernel tests and Python tests for Resize operator
Checklist
Documentation
DALI team only
Requirements
REQ IDs: N/A
JIRA TASK: N/A