Fix unpack from sr300 cuda implementation - #8896
Merged
Merged
Conversation
ev-mp
approved these changes
May 5, 2021
ev-mp
left a comment
Contributor
There was a problem hiding this comment.
@andrusza2 , thank you very much for the contribution!
This resolves the issue with incorrect distances reported with CUDA for SR300.
This was referenced May 14, 2021
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 fixes a bug with incorrect depth image on the SR300 series cameras when librealsense is built with CUDA [/issues/8897].
The functions
unpack_z16_y8_from_sr300_inziandunpack_z16_y16_from_sr300_inziuse the source data pointer (in) - the first half of the data is bit-shifted and stored indest[1], the second half is directly copied todest[0]. In the CPU version, the pointer is moved in a loop and after its execution it points toin + count- so the copying works as it should. In the CUDA version, the pointer is never moved, so bad data is copied todest[0], resulting in an incorrect depth image. Simply moving the pointer to the right place fixes the problem.