Significant speedups in disk and network I/O
From a mail message from Theo de Raadt to the OpenBSD tech mailing list:
In the last two weeks we have commited a few things which significantly
speed up disk and network io. i will mention two things.
some of you can start cranking the kern.bufcachepercent sysctl to 90.
try it out. be impressed, but don't report deadlocks or lockups;
however real crashes are worth knowing about. the lockup situations
are largely understood and will be slowly worked out, which is why
this is not close to being the default.
the other change is that interrupt servicing times have gone down
substantially. there are some broken drivers we will need to find
and fix, so please watch your vmstat -i numbers carefully, and for
this circumstance, if you get machine lockups at boot please report
those very carefully in a PR (with an old working dmesg).
You don't need to be a genius to debug a pf.conf firewall ruleset, you just need the guts to run tcpdump