Shekharrajak commented on PR #23468:
URL: https://github.com/apache/kafka/pull/23468#issuecomment-5700590242
Thanks for checking @apoorvmittal10 .
Not claiming a throughput improvement. The concrete benefit is that the old
path calls `nextFetchOffset()` inside `canAcquireRecords()` and again after
eligibility succeeds.
When an AVAILABLE record is found, findNextFetchOffset remains true, so both
calls scan cachedState and acquire the write lock; only the “no AVAILABLE
record” case clears the flag and makes the second call cheap for CPU
usage/cost.
```
canAcquireRecords() -> cachedState scan 1
nextFetchOffset() -> cachedState scan 2
```
With `nextFetchOffsetIfAcquirable` only 1 cacheState entry scan. The tests
validate the single-call behavior.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
