Enforce check to avoid memory corruption#23
Open
danieldugas wants to merge 1 commit intoros-perception:masterfrom
Open
Enforce check to avoid memory corruption#23danieldugas wants to merge 1 commit intoros-perception:masterfrom
danieldugas wants to merge 1 commit intoros-perception:masterfrom
Conversation
k-okada
requested changes
Sep 30, 2021
Contributor
There was a problem hiding this comment.
@danieldugas could you create new PR against melodic-devel branch? we have switched from master to melodic-branch when we changed licence to BSD
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Ubuntu 16.04, ROS Kinetic.
gmapping kept crashing when I published high resolution laser scans, it took me a while to figure out why.
The cause of the crash was especially hard to find since memory corruption caused the program to terminate much later than where the corruption occured. Finally, we traced it down to this.
if the laserscan has more beams than LASER_MAXBEAMS, the memcpy operation here leads to memory corruption.
Of course, there is an assert a few lines above. However it did not get triggered (is NDEBUG defined somewhere?). I believe this is not specific to me, as I am using the debian distributed ros version of openslam_gmapping.
I'm convinced replacing the assert with an inexpensive if, verbose message, and exit(-1) would be one way to ensure this error does not inconvenience anymore users, though this much is up to you.
Cheers