Do the advisory aborts of the in-flight requests before flushing the vnodes. It should mostly eliminate the waits due to requests busying the mp. Reported by: rew
Details
Details
Diff Detail
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
Comment Actions
this doesn't address the behavior I observed on my test but I can see that this is worth doing still
it seems the waits I was experiencing is when the autofs request had failed..the request is deallocated, and then autofs retries the request according to an_retries
the following patch: https://reviews.freebsd.org/D58691 seems to address the waits I observed