Re: more data: SCHED_ULE+PREEMPTION is the problem

2018-04-08 Thread George Mitchell
On 04/07/18 10:18, Peter wrote: > [...] > B. Findings: > > 1. Filesystem > > I could never reproduce this when reading from plain UFS. Only when > reading from ZFS (direct or via l2arc). > [...] My consistent way of reproducing the problem was to run ports/misc/dnetc while trying to

Re: kern.sched.quantum: Creepy, sadistic scheduler

2018-04-08 Thread Julian Elischer
On 7/4/18 10:21 pm, Peter wrote: Julian Elischer wrote: for a single CPU you really should compile a kernel with SMP turned off and 4BSD scheduler. ULE is just trying too hard to do stuff you don't need. Julian, if we agree on this, I am fine. (This implies that SCHED_4BSD will *not* be