i#5843 scheduler: Only consider long-latency syscalls blocking #6458
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
Rather than context switching on every syscall labeled maybe-blocking, the scheduler uses the now-available syscall latency to decide whether the syscall should block and result in a context switch. Adds two new command line options, -sched_syscall_switch_us (default 500us) and -sched_blocking_switch_us (default 100us), and corresponding scheduler_t inputs, to control the latency thresholds. To avoid relying too much on the maybe-blocking labels, we do consider a very-high-latency syscall not marked as maybe-blocking to block. Adds a new unit test. Tested in a large proprietary app where this reduces the context switch rate from ~100x too high down to ~10x too high. The next step of adding i/o wait times should further improve the representativeness. Issue: #5843
|
ub22 is off.simple #6416 |
abhinav92003
approved these changes
Nov 16, 2023
…nts and rename vars to make clearer.
derekbruening
added a commit
that referenced
this pull request
Nov 17, 2023
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Outdated suggestions cannot be applied.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.
Rather than context switching on every syscall labeled maybe-blocking, the scheduler uses the now-available syscall latency to decide whether the syscall should block and result in a context switch.
Adds two new command line options, -sched_syscall_switch_us (default 500us) and -sched_blocking_switch_us (default 100us), and corresponding scheduler_t inputs, to control the latency thresholds. To avoid relying too much on the maybe-blocking labels, we do consider a very-high-latency syscall not marked as maybe-blocking to block.
Adds a new schedule_stats unit test.
Tested in a large proprietary app where this reduces the context switch rate from ~100x too high down to ~10x too high. The next step of adding i/o wait times should further improve the representativeness.
Issue: #5843