Fix VIRTUAL and PROF interval timers for short intervals, broken at r247903.

Due to the way those timers are implemented, we can't handle very short
intervals. In addition to that mentioned patch c

Fix VIRTUAL and PROF interval timers for short intervals, broken at r247903.

Due to the way those timers are implemented, we can't handle very short
intervals. In addition to that mentioned patch caused math overflows
for short intervals. To avoid that round those intervals to 1 tick.

PR: kern/187668
MFC after: 1 week

show more ...

Merge head r233826 through r256722.

IFC @256277

Approved by: ken (mentor)

Merge from head

IFC @ r254014

IFC @ r253862

- change the SI_SUB_RUN_SCHEDULER sysinits in hv_utilc and
hv_netvsc_drv_freebsd.c to SI_SUB_KTHREAD_IDLE, since the
former is no longer in FreeBSD.
The use of these SYSINITs can pro

IFC @ r253862

- change the SI_SUB_RUN_SCHEDULER sysinits in hv_utilc and
hv_netvsc_drv_freebsd.c to SI_SUB_KTHREAD_IDLE, since the
former is no longer in FreeBSD.
The use of these SYSINITs can probably be removed.

show more ...

Implement compat32 wrappers for the ktimer_* syscalls.

Reported, reviewed and tested by: Petr Salinger <Petr.Salinger@seznam.cz>
Sponsored by: The FreeBSD Foundation
MFC after: 1 week

id_t is 64bit, provide the compat32 wrapper for clock_getcpuclockid2(2).

Reported and tested by: Petr Salinger <Petr.Salinger@seznam.cz>
PR: threads/180652
Sponsored by: The FreeBSD Foundation

IFC @ r253461

IFC @253398

Allow to call clock_gettime() on the clock id for zombie process.

Reported by: Petr Salinger <Petr.Salinger@seznam.cz>
PR: threads/180496
Sponsored by: The FreeBSD Foundation

Merge fresh head.

sync from head

MFC @248461

Make kern_nanosleep() and pause_sbt() to use per-CPU sleep queues.
This removes significant sleep queue lock congestion on multithreaded
microbenchmarks, making them scale to multiple CPUs almost lin

Make kern_nanosleep() and pause_sbt() to use per-CPU sleep queues.
This removes significant sleep queue lock congestion on multithreaded
microbenchmarks, making them scale to multiple CPUs almost linearly.

show more ...

MFC @248093

Reduce minimal time intervals of setitimer(2) from 1/HZ to 1/(16*HZ) by
using callout_reset_sbt() instead of callout_reset(). We can't remove
lower limit completely in this case because of significa

Reduce minimal time intervals of setitimer(2) from 1/HZ to 1/(16*HZ) by
using callout_reset_sbt() instead of callout_reset(). We can't remove
lower limit completely in this case because of significant processing
overhead, caused by unability to use direct callout execution due to using
process mutex in callout handler for sending SEGALRM signal. With support
of periodic events that would allow unprivileged user to abuse the system.

Reviewed by: davide

show more ...

Fix time math overflows and improve zero intervals handling in poll(),
select(), nanosleep() and kevent() functions after calloutng changes.

Reported by: bde

MFcalloutng:
kern_nanosleep() is now converted to use tsleep_sbt(). With this change
nanosleep() and usleep() can handle sub-tick precision for timeouts.
Also, try to help coalesce of events passing

MFcalloutng:
kern_nanosleep() is now converted to use tsleep_sbt(). With this change
nanosleep() and usleep() can handle sub-tick precision for timeouts.
Also, try to help coalesce of events passing as argument to tsleep_bt()
a precision value calculated as a percentage of the sleep time.
This percentage is default 5%, but it can tuned according to users
need via the sysctl interface.

Sponsored by: Google Summer of Code 2012, iXsystems inc.
Tested by: flo, marius, ian, markj, Fabian Keil

show more ...

IFC @247348.

Sync with HEAD.

IFC @ r244983.

Fix a race between kern_setitimer() and realitexpire(), where the
callout is started before kern_setitimer() acquires process mutex, but
looses a race and kern_setitimer() gets the process mutex befo

Fix a race between kern_setitimer() and realitexpire(), where the
callout is started before kern_setitimer() acquires process mutex, but
looses a race and kern_setitimer() gets the process mutex before the
callout. Then, assuming that new specified struct itimerval has
it_interval zero, but it_value non-zero, the callout, after it starts
executing again, clears p->p_realtimer.it_value, but kern_setitimer()
already rescheduled the callout.

As the result of the race, both p_realtimer is zero, and the callout
is rescheduled. Then, in the exit1(), the exit code sees that it_value
is zero and does not even try to stop the callout. This allows the
struct proc to be reused and eventually the armed callout is
re-initialized. The consequence is the corrupted callwheel tailq.

Use process mutex to interlock the callout start, which fixes the race.

Reported and tested by: pho
Reviewed by: jhb
MFC after: 2 weeks

show more ...

MFC @ r241285

IFC @ r242684