On 24/11/17 6:30 pm, Andriy Gapon wrote:
https://reviews.freebsd.org/D13224
Anyone interested is welcome to join the review.
Thanks.
I see you closed the review.
However don't stop working with that problem..
It's still something that a worth a solution, even if not exactly That
solution.
Maybe a generic idea through the block IO layer of transaction
requirements.
We did face this issue in a BSD4.3 variant in 1990 where a single disk
error
would result in a cascading shockwave of slowdowns that took some time
to resolve.
The issue even comes up in networking too, especially with realtime video
where "don't bother, we missed the window" comes into play.
_______________________________________________
freebsd-geom@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-geom
To unsubscribe, send any mail to "freebsd-geom-unsubscr...@freebsd.org"
