Quoting Chris Wilson (2020-03-20 17:47:45) > We dropped calling process_csb prior to handling direct submission in > order to avoid the nesting of spinlocks and lift process_csb() and the > majority of the tasklet out of irq-off. However, we do want to avoid > ksoftirqd latency in the fast path, so try and pull the interrupt-bh > local to direct submission if we can acquire the tasklet's lock. > > v2: Tweak the balance to avoid over submitting lite-restores
Sigh. It's a driver problem with lock contention, mostly from aggressive idling. -Chris _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
