stop calling cnd_timedwait() with a timeout of 0h #1481
Merged
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
edenhill
suggested changes
Oct 23, 2017
src/rdkafka_queue.c
Outdated
| @@ -361,7 +361,8 @@ rd_kafka_op_t *rd_kafka_q_pop_serve (rd_kafka_q_t *rkq, int timeout_ms, | |||
|
|
|||
| /* No op, wait for one */ | |||
| pre = rd_clock(); | |||
| if (cnd_timedwait_ms(&rkq->rkq_cond, | |||
| if (timeout_ms == 0 || | |||
There was a problem hiding this comment.
I think we can simply unlock and return NULL if timeout_ms is 0 instead of trying again, that's what will happen below at the while (timeout_ms != ..NOWAIT) anyway,
Put the check above the pre = .. assignment.
There was a problem hiding this comment.
@edenhill, actually that's what this already does, because in the case cnd_timedwait() returns timedout we just unlock and return null. That said we could avoid the call to clock_gettime() which would save us a couple of microseconds while holding a lock. So I'll make this a separate check.
|
Nice find :) |
Waiting 0ms to be signaled the queue now contains elements doesn't really mean anything. However per the POSIX specification of pthread_cond_timedwait() the lock must be released and reacquired. As a result if we skip calling cnd_timedwait() when the timeout is 0 we improve average and best case latency in various conditions by between 0.1 and 0.2ms.
akhi3030
pushed a commit
to akhi3030/librdkafka
that referenced
this pull request
Nov 13, 2017
…tbsaunde) * stop calling cnd_timedwait() with a timeout of 0 Waiting 0ms to be signaled the queue now contains elements doesn't really mean anything. However per the POSIX specification of pthread_cond_timedwait() the lock must be released and reacquired. As a result if we skip calling cnd_timedwait() when the timeout is 0 we improve average and best case latency in various conditions by between 0.1 and 0.2ms. * fix indent, use define
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.
Waiting 0ms to be signaled the queue now contains elements doesn't
really mean anything. However per the POSIX specification of
pthread_cond_timedwait() the lock must be released and reacquired. As a
result if we skip calling cnd_timedwait() when the timeout is 0 we
improve average and best case latency in various conditions by between
0.1 and 0.2ms.