Get suspends from logind to ignore watchdog then #29
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
On every resume, the watchdog thread happily reports that there has been a deadlock since the canary hasn't responded since before the computer resumed. Get notifications from logind over dbus to tell when suspend happens and ignore the one watchdog violation that might happen then. Depending on how threads are scheduled on resume (canary scheduled first), the watchdog after resume might not happen and the next real deadlock will take two watchdog cycles to hit, which is not horrible in my mind but might be fixable by having the ignore condition have a timeout for N seconds after resume.
On my personal computer, the scheduling race condition mentioned in the last commit happens about ~1/4 of the time. That's too often to ignore. Only make the resume-ignore condition apply for a few seconds after resume. Additionally, this feature can now be disabled by setting a threshold of zero.
|
Annoying that this has been open for almost a year, especially for such a simple fix. |
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Outdated suggestions cannot be applied.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.
On every resume, the watchdog thread happily reports that there has been
a deadlock since the canary hasn't responded since before the computer
resumed. Get notifications from logind over dbus to tell when suspend
happens and ignore the one watchdog violation that might happen then.
On start-suspend, this ignores a single timeout indefinitely. On resume, that
ignore gets a timeout of 2 seconds in the future (configurable), if it hasn't
already been tripped. At any point if a real deadlock occurs, it will take a max
of two watchdog periods to recognize it. If the configurable timeout is zero,
no ignores happen and the extra dbus work is skipped.
This code is not Y2038 safe... but neither is the old code.
Addresses #13, which has apparently been around for almost 10 years in some form.