Fix ringbuffer hardware race condition bug. #3240
Closed
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Outdated suggestions cannot be applied.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.
When attempting to get a snapshot of the completion count and position, the current ringbuffer implementation uses a critical section. This does indeed prevent the completion_count from updating but does not prevent the position from updating since that is fetched directly from the DMA register.
When the code is called right at the end of the DMA cycle, the position that is read may be that of the next cycle, while the completion_count is still that of the previous cycle. As a result it looks to the code as if we went backwards and an overrun error is emitted.
In order to fix this we remove the critical section and get two readings, comparing the two in order to deduce that a wrap-around occured.