Add Frontend.NumInstructions statistic, populated on macOS by rusage_info_v4 #18658
Conversation
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
|
@swift-ci please test |
|
Great find! |
xedin
approved these changes
Aug 13, 2018
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Outdated suggestions cannot be applied.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.
After all the fussing around I have done trying to get a reliable external tool to read off the PMC instruction counts for a subprocess tree on macOS (a la linux perf(1)) I was somewhat astonished to learn that XNU these days just always tracks every task's PMC-read instruction count (see: https://github.com/apple/darwin-xnu/blob/master/osfmk/kern/kern_monotonic.c and https://github.com/apple/darwin-xnu/blob/master/osfmk/x86_64/monotonic_x86_64.c) and returns it as a standard field in its rusage_info_v4 structure, if you ask via proc_pid_rusage.
So: this patch adds instruction counts to UnifiedStatsCollector / -stats-output-dir. I will be plumbing it through everything else subsequently -- it's a much less-noisy number to use than the time values, while being a decent proxy for it at the granularity we're concerned with -- but this gets the ball rolling.