| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
time, s390: Get rid of compile warning
dw_apb_timer: constify clocksource name
time: Cleanup old CONFIG_GENERIC_TIME references that snuck in
time: Change jiffies_to_clock_t() argument type to unsigned long
alarmtimers: Fix error handling
clocksource: Make watchdog reset lockless
posix-cpu-timers: Cure SMP accounting oddities
s390: Use direct ktime path for s390 clockevent device
clockevents: Add direct ktime programming function
clockevents: Make minimum delay adjustments configurable
nohz: Remove "Switched to NOHz mode" debugging messages
proc: Consider NO_HZ when printing idle and iowait times
nohz: Make idle/iowait counter update conditional
nohz: Fix update_ts_time_stat idle accounting
cputime: Clean up cputime_to_usecs and usecs_to_cputime macros
alarmtimers: Rework RTC device selection using class interface
alarmtimers: Add try_to_cancel functionality
alarmtimers: Add more refined alarm state tracking
alarmtimers: Remove period from alarm structure
alarmtimers: Remove interval cap limit hack
...
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The parameter's origin type is long. On an i386 architecture, it can
easily be larger than 0x80000000, causing this function to convert it
to a sign-extended u64 type.
Change the type to unsigned long so we get the correct result.
Signed-off-by: hank <pyu@redhat.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: <stable@kernel.org>
[ build fix ]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | |
| |
| |
| |
| |
| |
| | |
commit 8bc0daf (alarmtimers: Rework RTC device selection using class
interface) did not implement required error checks. Add them.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
KGDB needs to trylock watchdog_lock when trying to reset the
clocksource watchdog after the system has been stopped to avoid a
potential deadlock. When the trylock fails TSC usually becomes
unstable.
We can be more clever by using an atomic counter and checking it in
the clocksource_watchdog callback. We restart the watchdog whenever
the counter is > 0 and only decrement the counter when we ran through
a full update cycle.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <johnstul@us.ibm.com>
Acked-by: Jason Wessel <jason.wessel@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1109121326280.2723@ionos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
David reported:
Attached below is a watered-down version of rt/tst-cpuclock2.c from
GLIBC. Just build it with "gcc -o test test.c -lpthread -lrt" or
similar.
Run it several times, and you will see cases where the main thread
will measure a process clock difference before and after the nanosleep
which is smaller than the cpu-burner thread's individual thread clock
difference. This doesn't make any sense since the cpu-burner thread
is part of the top-level process's thread group.
I've reproduced this on both x86-64 and sparc64 (using both 32-bit and
64-bit binaries).
For example:
[davem@boricha build-x86_64-linux]$ ./test
process: before(0.001221967) after(0.498624371) diff(497402404)
thread: before(0.000081692) after(0.498316431) diff(498234739)
self: before(0.001223521) after(0.001240219) diff(16698)
[davem@boricha build-x86_64-linux]$
The diff of 'process' should always be >= the diff of 'thread'.
I make sure to wrap the 'thread' clock measurements the most tightly
around the nanosleep() call, and that the 'process' clock measurements
are the outer-most ones.
---
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <fcntl.h>
#include <string.h>
#include <errno.h>
#include <pthread.h>
static pthread_barrier_t barrier;
static void *chew_cpu(void *arg)
{
pthread_barrier_wait(&barrier);
while (1)
__asm__ __volatile__("" : : : "memory");
return NULL;
}
int main(void)
{
clockid_t process_clock, my_thread_clock, th_clock;
struct timespec process_before, process_after;
struct timespec me_before, me_after;
struct timespec th_before, th_after;
struct timespec sleeptime;
unsigned long diff;
pthread_t th;
int err;
err = clock_getcpuclockid(0, &process_clock);
if (err)
return 1;
err = pthread_getcpuclockid(pthread_self(), &my_thread_clock);
if (err)
return 1;
pthread_barrier_init(&barrier, NULL, 2);
err = pthread_create(&th, NULL, chew_cpu, NULL);
if (err)
return 1;
err = pthread_getcpuclockid(th, &th_clock);
if (err)
return 1;
pthread_barrier_wait(&barrier);
err = clock_gettime(process_clock, &process_before);
if (err)
return 1;
err = clock_gettime(my_thread_clock, &me_before);
if (err)
return 1;
err = clock_gettime(th_clock, &th_before);
if (err)
return 1;
sleeptime.tv_sec = 0;
sleeptime.tv_nsec = 500000000;
nanosleep(&sleeptime, NULL);
err = clock_gettime(th_clock, &th_after);
if (err)
return 1;
err = clock_gettime(my_thread_clock, &me_after);
if (err)
return 1;
err = clock_gettime(process_clock, &process_after);
if (err)
return 1;
diff = process_after.tv_nsec - process_before.tv_nsec;
printf("process: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
process_before.tv_sec, process_before.tv_nsec,
process_after.tv_sec, process_after.tv_nsec, diff);
diff = th_after.tv_nsec - th_before.tv_nsec;
printf("thread: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
th_before.tv_sec, th_before.tv_nsec,
th_after.tv_sec, th_after.tv_nsec, diff);
diff = me_after.tv_nsec - me_before.tv_nsec;
printf("self: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
me_before.tv_sec, me_before.tv_nsec,
me_after.tv_sec, me_after.tv_nsec, diff);
return 0;
}
This is due to us using p->se.sum_exec_runtime in
thread_group_cputime() where we iterate the thread group and sum all
data. This does not take time since the last schedule operation (tick
or otherwise) into account. We can cure this by using
task_sched_runtime() at the cost of having to take locks.
This also means we can (and must) do away with
thread_group_sched_runtime() since the modified thread_group_cputime()
is now more accurate and would deadlock when called from
thread_group_sched_runtime().
Reported-by: David Miller <davem@davemloft.net>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1314874459.7945.22.camel@twins
Cc: stable@kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
There is at least one architecture (s390) with a sane clockevent device
that can be programmed with the equivalent of a ktime. No need to create
a delta against the current time, the ktime can be used directly.
A new clock device function 'set_next_ktime' is introduced that is called
with the unmodified ktime for the timer if the clock event device has the
CLOCK_EVT_FEAT_KTIME bit set.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: john stultz <johnstul@us.ibm.com>
Link: http://lkml.kernel.org/r/20110823133142.815350967@de.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The automatic increase of the min_delta_ns of a clockevents device
should be done in the clockevents code as the minimum delay is an
attribute of the clockevents device.
In addition not all architectures want the automatic adjustment, on a
massively virtualized system it can happen that the programming of a
clock event fails several times in a row because the virtual cpu has
been rescheduled quickly enough. In that case the minimum delay will
erroneously be increased with no way back. The new config symbol
GENERIC_CLOCKEVENTS_MIN_ADJUST is used to enable the automatic
adjustment. The config option is selected only for x86.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: john stultz <johnstul@us.ibm.com>
Link: http://lkml.kernel.org/r/20110823133142.494157493@de.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When performing cpu hotplug tests the kernel printk log buffer gets flooded
with pointless "Switched to NOHz mode..." messages. Especially when afterwards
analyzing a dump this might have removed more interesting stuff out of the
buffer.
Assuming that switching to NOHz mode simply works just remove the printk.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Link: http://lkml.kernel.org/r/20110823112046.GB2540@osiris.boeblingen.de.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
get_cpu_{idle,iowait}_time_us update idle/iowait counters
unconditionally if the given CPU is in the idle loop.
This doesn't work well outside of CPU governors which are singletons
so nobody (except for IRQ) can race with them.
We will need to use both functions from /proc/stat handler to properly
handle nohz idle/iowait times.
Make the update depend on a non NULL last_update_time argument.
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Cc: Dave Jones <davej@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Link: http://lkml.kernel.org/r/11f23179472635ce52e78921d47a20216b872f23.1314172057.git.mhocko@suse.cz
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
update_ts_time_stat currently updates idle time even if we are in
iowait loop at the moment. The only real users of the idle counter
(via get_cpu_idle_time_us) are CPU governors and they expect to get
cumulative time for both idle and iowait times.
The value (idle_sleeptime) is also printed to userspace by print_cpu
but it prints both idle and iowait times so the idle part is misleading.
Let's clean this up and fix update_ts_time_stat to account both counters
properly and update consumers of idle to consider iowait time as well.
If we do this we might use get_cpu_{idle,iowait}_time_us from other
contexts as well and we will get expected values.
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Cc: Dave Jones <davej@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Link: http://lkml.kernel.org/r/e9c909c221a8da402c4da07e4cd968c3218f8eb1.1314172057.git.mhocko@suse.cz
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
This allows cleaner detection of the RTC device being registered, rather
then probing any time someone calls alarmtimer_get_rtcdev.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
There's a number of edge cases when cancelling a alarm, so
to be sure we accurately do so, introduce try_to_cancel, which
returns proper failure errors if it cannot. Also modify cancel
to spin until the alarm is properly disabled.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
In order to allow for functionality like try_to_cancel, add
more refined state tracking (similar to hrtimers).
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
Now that periodic alarmtimers are managed by the handler function,
remove the period value from the alarm structure and let the handlers
manage the interval on their own.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Now that the alarmtimers code has been refactored, the interval
cap limit can be removed.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In order to avoid wasting time expiring and re-adding very high freq
periodic alarmtimers, introduce alarm_forward() which is similar to
hrtimer_forward and moves the timer to the next future expiration time
and returns the number of overruns.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
This patch pushes the periodic alarmtimer re-arming down into the alarmtimer
handler, mimicking how hrtimers handle this.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In order to properly fix the denial of service issue with high freq
periodic alarm timers, we need to push the re-arming logic into the
alarm timer handler, much as the hrtimer code does.
This patch introduces alarmtimer_restart enum and changes the
alarmtimer handler declarations to use it as a return value. Further,
to ease following changes, it extends the alarmtimer handler functions
to also take the time at expiration. No logic is yet modified.
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
| |\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (46 commits)
llist: Add back llist_add_batch() and llist_del_first() prototypes
sched: Don't use tasklist_lock for debug prints
sched: Warn on rt throttling
sched: Unify the ->cpus_allowed mask copy
sched: Wrap scheduler p->cpus_allowed access
sched: Request for idle balance during nohz idle load balance
sched: Use resched IPI to kick off the nohz idle balance
sched: Fix idle_cpu()
llist: Remove cpu_relax() usage in cmpxchg loops
sched: Convert to struct llist
llist: Add llist_next()
irq_work: Use llist in the struct irq_work logic
llist: Return whether list is empty before adding in llist_add()
llist: Move cpu_relax() to after the cmpxchg()
llist: Remove the platform-dependent NMI checks
llist: Make some llist functions inline
sched, tracing: Show PREEMPT_ACTIVE state in trace_sched_switch
sched: Remove redundant test in check_preempt_tick()
sched: Add documentation for bandwidth control
sched: Return unused runtime on group dequeue
...
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Avoid taking locks from debug prints, this avoids latencies on -rt,
and improves reliability of the debug code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The default rt-throttling is a source of never ending questions. Warn
once when we go into throttling so folks have that info in dmesg.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1110051331480.18778@ionos
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Currently every sched_class::set_cpus_allowed() implementation has to
copy the cpumask into task_struct::cpus_allowed, this is pointless,
put this copy in the generic code.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/n/tip-jhl5s9fckd9ptw1fzbqqlrd3@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This task is preparatory for the migrate_disable() implementation, but
stands on its own and provides a cleanup.
It currently only converts those sites required for task-placement.
Kosaki-san once mentioned replacing cpus_allowed with a proper
cpumask_t instead of the NR_CPUS sized array it currently is, that
would also require something like this.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Link: http://lkml.kernel.org/n/tip-e42skvaddos99psip0vce41o@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
rq's idle_at_tick is set to idle/busy during the timer tick
depending on the cpu was idle or not. This will be used later in the load
balance that will be done in the softirq context (which is a process
context in -RT kernels).
For nohz kernels, for the cpu doing nohz idle load balance on behalf of
all the idle cpu's, its rq->idle_at_tick might have a stale value (which is
recorded when it got the timer tick presumably when it is busy).
As the nohz idle load balancing is also being done at the same place
as the regular load balancing, nohz idle load balancing was bailing out
when it sees rq's idle_at_tick not set.
Thus leading to poor system utilization.
Rename rq's idle_at_tick to idle_balance and set it when someone requests
for nohz idle balance on an idle cpu.
Reported-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20111003220934.892350549@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Current use of smp call function to kick the nohz idle balance can deadlock
in this scenario.
1. cpu-A did a generic_exec_single() to cpu-B and after queuing its call single
data (csd) to the call single queue, cpu-A took a timer interrupt. Actual IPI
to cpu-B to process the call single queue is not yet sent.
2. As part of the timer interrupt handler, cpu-A decided to kick cpu-B
for the idle load balancing (sets cpu-B's rq->nohz_balance_kick to 1)
and __smp_call_function_single() with nowait will queue the csd to the
cpu-B's queue. But the generic_exec_single() won't send an IPI to cpu-B
as the call single queue was not empty.
3. cpu-A is busy with lot of interrupts
4. Meanwhile cpu-B is entering and exiting idle and noticed that it has
it's rq->nohz_balance_kick set to '1'. So it will go ahead and do the
idle load balancer and clear its rq->nohz_balance_kick.
5. At this point, csd queued as part of the step-2 above is still locked
and waiting to be serviced on cpu-B.
6. cpu-A is still busy with interrupt load and now it got another timer
interrupt and as part of it decided to kick cpu-B for another idle load
balancing (as it finds cpu-B's rq->nohz_balance_kick cleared in step-4
above) and does __smp_call_function_single() with the same csd that is
still locked.
7. And we get a deadlock waiting for the csd_lock() in the
__smp_call_function_single().
Main issue here is that cpu-B can service the idle load balancer kick
request from cpu-A even with out receiving the IPI and this lead to
doing multiple __smp_call_function_single() on the same csd leading to
deadlock.
To kick a cpu, scheduler already has the reschedule vector reserved. Use
that mechanism (kick_process()) instead of using the generic smp call function
mechanism to kick off the nohz idle load balancing and avoid the deadlock.
[ This issue is present from 2.6.35+ kernels, but marking it -stable
only from v3.0+ as the proposed fix depends on the scheduler_ipi()
that is introduced recently. ]
Reported-by: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: stable@kernel.org # v3.0+
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20111003220934.834943260@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
On -rt we observed hackbench waking all 400 tasks to a single cpu.
This is because of select_idle_sibling()'s interaction with the new
ipi based wakeup scheme.
The existing idle_cpu() test only checks to see if the current task on
that cpu is the idle task, it does not take already queued tasks into
account, nor does it take queued to be woken tasks into account.
If the remote wakeup IPIs come hard enough, there won't be time to
schedule away from the idle task, and would thus keep thinking the cpu
was in fact idle, regardless of the fact that there were already
several hundred tasks runnable.
We couldn't reproduce on mainline, but there's no reason it couldn't
happen.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-3o30p18b2paswpc9ohy2gltp@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Use the generic llist primitives.
We had a private lockless list implementation in the scheduler in the wake-list
code, now that we have a generic llist implementation that provides all required
operations, switch to it.
This patch is not expected to change any behavior.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1315836353.26517.42.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
So we don't have to expose the struct list_node member.
Cc: Huang Ying <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315836348.26517.41.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Use llist in irq_work instead of the lock-less linked list
implementation in irq_work to avoid the code duplication.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-6-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | |\ \
| | | |
| | | |
| | | |
| | | |
| | | | |
Merge reason: pick up the latest fixes.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The caller already checks for nr_running > 1, therefore we don't have
to do so again.
Signed-off-by: Wang Xingchao <xingchao.wang@intel.com>
Reviewed-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1316194552-12019-1-git-send-email-xingchao.wang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Merge reason: We are queueing up a dependent patch.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
When a local cfs_rq blocks we return the majority of its remaining quota to the
global bandwidth pool for use by other runqueues.
We do this only when the quota is current and there is more than
min_cfs_rq_quota [1ms by default] of runtime remaining on the rq.
In the case where there are throttled runqueues and we have sufficient
bandwidth to meter out a slice, a second timer is kicked off to handle this
delivery, unthrottling where appropriate.
Using a 'worst case' antagonist which executes on each cpu
for 1ms before moving onto the next on a fairly large machine:
no quota generations:
197.47 ms /cgroup/a/cpuacct.usage
199.46 ms /cgroup/a/cpuacct.usage
205.46 ms /cgroup/a/cpuacct.usage
198.46 ms /cgroup/a/cpuacct.usage
208.39 ms /cgroup/a/cpuacct.usage
Since we are allowed to use "stale" quota our usage is effectively bounded by
the rate of input into the global pool and performance is relatively stable.
with quota generations [1s increments]:
119.58 ms /cgroup/a/cpuacct.usage
119.65 ms /cgroup/a/cpuacct.usage
119.64 ms /cgroup/a/cpuacct.usage
119.63 ms /cgroup/a/cpuacct.usage
119.60 ms /cgroup/a/cpuacct.usage
The large deficit here is due to quota generations (/intentionally/) preventing
us from now using previously stranded slack quota. The cost is that this quota
becomes unavailable.
with quota generations and quota return:
200.09 ms /cgroup/a/cpuacct.usage
200.09 ms /cgroup/a/cpuacct.usage
198.09 ms /cgroup/a/cpuacct.usage
200.09 ms /cgroup/a/cpuacct.usage
200.06 ms /cgroup/a/cpuacct.usage
By returning unused quota we're able to both stably consume our desired quota
and prevent unintentional overages due to the abuse of slack quota from
previous quota periods (especially on a large machine).
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184758.306848658@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This change introduces statistics exports for the cpu sub-system, these are
added through the use of a stat file similar to that exported by other
subsystems.
The following exports are included:
nr_periods: number of periods in which execution occurred
nr_throttled: the number of periods above in which execution was throttle
throttled_time: cumulative wall-time that any cpus have been throttled for
this group
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Nikhil Rao <ncrao@google.com>
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184758.198901931@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
With the machinery in place to throttle and unthrottle entities, as well as
handle their participation (or lack there of) we can now enable throttling.
There are 2 points that we must check whether it's time to set throttled state:
put_prev_entity() and enqueue_entity().
- put_prev_entity() is the typical throttle path, we reach it by exceeding our
allocated run-time within update_curr()->account_cfs_rq_runtime() and going
through a reschedule.
- enqueue_entity() covers the case of a wake-up into an already throttled
group. In this case we know the group cannot be on_rq and can throttle
immediately. Checks are added at time of put_prev_entity() and
enqueue_entity()
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184758.091415417@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Throttled tasks are invisisble to cpu-offline since they are not eligible for
selection by pick_next_task(). The regular 'escape' path for a thread that is
blocked at offline is via ttwu->select_task_rq, however this will not handle a
throttled group since there are no individual thread wakeups on an unthrottle.
Resolve this by unthrottling offline cpus so that threads can be migrated.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.989000590@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Buddies allow us to select "on-rq" entities without actually selecting them
from a cfs_rq's rb_tree. As a result we must ensure that throttled entities
are not falsely nominated as buddies. The fact that entities are dequeued
within throttle_entity is not sufficient for clearing buddy status as the
nomination may occur after throttling.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.886850167@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
From the perspective of load-balance and shares distribution, throttled
entities should be invisible.
However, both of these operations work on 'active' lists and are not
inherently aware of what group hierarchies may be present. In some cases this
may be side-stepped (e.g. we could sideload via tg_load_down in load balance)
while in others (e.g. update_shares()) it is more difficult to compute without
incurring some O(n^2) costs.
Instead, track hierarchicaal throttled state at time of transition. This
allows us to easily identify whether an entity belongs to a throttled hierarchy
and avoid incorrect interactions with it.
Also, when an entity leaves a throttled hierarchy we need to advance its
time averaging for shares averaging so that the elapsed throttled time is not
considered as part of the cfs_rq's operation.
We also use this information to prevent buddy interactions in the wakeup and
yield_to() paths.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.777916795@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Extend walk_tg_tree to accept a positional argument
static int walk_tg_tree_from(struct task_group *from,
tg_visitor down, tg_visitor up, void *data)
Existing semantics are preserved, caller must hold rcu_lock() or sufficient
analogue.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.677889157@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
At the start of each period we refresh the global bandwidth pool. At this time
we must also unthrottle any cfs_rq entities who are now within bandwidth once
more (as quota permits).
Unthrottled entities have their corresponding cfs_rq->throttled flag cleared
and their entities re-enqueued.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.574628950@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Now that consumption is tracked (via update_curr()) we add support to throttle
group entities (and their corresponding cfs_rqs) in the case where this is no
run-time remaining.
Throttled entities are dequeued to prevent scheduling, additionally we mark
them as throttled (using cfs_rq->throttled) to prevent them from becoming
re-enqueued until they are unthrottled. A list of a task_group's throttled
entities are maintained on the cfs_bandwidth structure.
Note: While the machinery for throttling is added in this patch the act of
throttling an entity exceeding its bandwidth is deferred until later within
the series.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.480608533@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Since quota is managed using a global state but consumed on a per-cpu basis
we need to ensure that our per-cpu state is appropriately synchronized.
Most importantly, runtime that is state (from a previous period) should not be
locally consumable.
We take advantage of existing sched_clock synchronization about the jiffy to
efficiently detect whether we have (globally) crossed a quota boundary above.
One catch is that the direction of spread on sched_clock is undefined,
specifically, we don't know whether our local clock is behind or ahead
of the one responsible for the current expiration time.
Fortunately we can differentiate these by considering whether the
global deadline has advanced. If it has not, then we assume our clock to be
"fast" and advance our local expiration; otherwise, we know the deadline has
truly passed and we expire our local runtime.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.379275352@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This patch adds a per-task_group timer which handles the refresh of the global
CFS bandwidth pool.
Since the RT pool is using a similar timer there's some small refactoring to
share this support.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.277271273@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Account bandwidth usage on the cfs_rq level versus the task_groups to which
they belong. Whether we are tracking bandwidth on a given cfs_rq is maintained
under cfs_rq->runtime_enabled.
cfs_rq's which belong to a bandwidth constrained task_group have their runtime
accounted via the update_curr() path, which withdraws bandwidth from the global
pool as desired. Updates involving the global pool are currently protected
under cfs_bandwidth->lock, local runtime is protected by rq->lock.
This patch only assigns and tracks quota, no action is taken in the case that
cfs_rq->runtime_used exceeds cfs_rq->runtime_assigned.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Nikhil Rao <ncrao@google.com>
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.179386821@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Add constraints validation for CFS bandwidth hierarchies.
Validate that:
max(child bandwidth) <= parent_bandwidth
In a quota limited hierarchy, an unconstrained entity
(e.g. bandwidth==RUNTIME_INF) inherits the bandwidth of its parent.
This constraint is chosen over sum(child_bandwidth) as notion of over-commit is
valuable within SCHED_OTHER. Some basic code from the RT case is re-factored
for reuse.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184757.083774572@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
In this patch we introduce the notion of CFS bandwidth, partitioned into
globally unassigned bandwidth, and locally claimed bandwidth.
- The global bandwidth is per task_group, it represents a pool of unclaimed
bandwidth that cfs_rqs can allocate from.
- The local bandwidth is tracked per-cfs_rq, this represents allotments from
the global pool bandwidth assigned to a specific cpu.
Bandwidth is managed via cgroupfs, adding two new interfaces to the cpu subsystem:
- cpu.cfs_period_us : the bandwidth period in usecs
- cpu.cfs_quota_us : the cpu bandwidth (in usecs) that this tg will be allowed
to consume over period above.
Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Nikhil Rao <ncrao@google.com>
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184756.972636699@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Introduce hierarchical task accounting for the group scheduling case in CFS, as
well as promoting the responsibility for maintaining rq->nr_running to the
scheduling classes.
The primary motivation for this is that with scheduling classes supporting
bandwidth throttling it is possible for entities participating in throttled
sub-trees to not have root visible changes in rq->nr_running across activate
and de-activate operations. This in turn leads to incorrect idle and
weight-per-task load balance decisions.
This also allows us to make a small fixlet to the fastpath in pick_next_task()
under group scheduling.
Note: this issue also exists with the existing sched_rt throttling mechanism.
This patch does not address that.
Signed-off-by: Paul Turner <pjt@google.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110721184756.878333391@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Since [sched/cpupri: Remove the vec->lock], member pri_active
of struct cpupri is not needed any more, just remove it. Also
clean stuff related to it.
Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110806001004.GA2207@zhy
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
[ This patch actually compiles. Thanks to Mike Galbraith for pointing
that out. I compiled and booted this patch with no issues. ]
Re-examining the cpupri patch, I see there's a possible race because the
update of the two priorities vec->counts are not protected by a memory
barrier.
When a RT runqueue is overloaded and wants to push an RT task to another
runqueue, it scans the RT priority vectors in a loop from lowest
priority to highest.
When we queue or dequeue an RT task that changes a runqueue's highest
priority task, we update the vectors to show that a runqueue is rated at
a different priority. To do this, we first set the new priority mask,
and increment the vec->count, and then set the old priority mask by
decrementing the vec->count.
If we are lowering the runqueue's RT priority rating, it will trigger a
RT pull, and we do not care if we miss pushing to this runqueue or not.
But if we raise the priority, but the priority is still lower than an RT
task that is looking to be pushed, we must make sure that this runqueue
is still seen by the push algorithm (the loop).
Because the loop reads from lowest to highest, and the new priority is
set before the old one is cleared, we will either see the new or old
priority set and the vector will be checked.
But! Since there's no memory barrier between the updates of the two, the
old count may be decremented first before the new count is incremented.
This means the loop may see the old count of zero and skip it, and also
the new count of zero before it was updated. A possible runqueue that
the RT task could move to could be missed.
A conditional memory barrier is placed between the vec->count updates
and is only called when both updates are done.
The smp_wmb() has also been changed to smp_mb__before_atomic_inc/dec(),
as they are not needed by archs that already synchronize
atomic_inc/dec().
The smp_rmb() has been moved to be called at every iteration of the loop
so that the race between seeing the two updates is visible by each
iteration of the loop, as an arch is free to optimize the reading of
memory of the counters in the loop.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1312547269.18583.194.camel@gandalf.stny.rr.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
sched/cpupri: Remove the vec->lock
The cpupri vec->lock has been showing up as a top contention
lately. This is because of the RT push/pull logic takes an
agressive approach for migrating RT tasks. The cpupri logic is
in place to improve the performance of the push/pull when dealing
with large number CPU machines.
The problem though is a vec->lock is required, where a vec is a
global per RT priority structure. That is, if there are lots of
RT tasks at the same priority, every time they are added or removed
from the RT queue, this global vec->lock is taken. Now that more
kernel threads are becoming RT (RCU boost and threaded interrupts)
this is becoming much more of an issue.
There are two variables that are being synced by the vec->lock.
The cpupri bitmask, and the vec->counter. The cpupri bitmask
is one bit per priority. If a RT priority vec has a process queued,
then the vec->count is > 0 and the cpupri bitmask is set for that
RT priority.
If the cpupri bitmask gets out of sync with the vec->counter, we could
end up pushing a low proirity RT task to a high priority queue.
That RT task that could have run immediately could be queued on a
run queue with a higher priority task indefinitely.
The solution is not to use the cpupri bitmask and just look at the
vec->count directly when doing a pull. The cpupri bitmask is just
a fast way to scan the RT priorities when a pull is made. Instead
of using the bitmask, and just examine all RT priorities, and
look at the vec->counts, we could eliminate the vec->lock. The
scan of RT tasks is to find a run queue that we can push an RT task
to, and we do not push to a high priority queue, thus the scan only
needs to go from 1 to RT task->prio, and not all 100 RT priorities.
The push algorithm, which does the scan of RT priorities (and
scan of the bitmask) only happens when we have an overloaded RT run
queue (more than one RT task queued). The grabbing of the vec->lock
happens every time any RT task is queued or dequeued on the run
queue for that priority. The slowing down of the scan by not using
a bitmask is negligible by the speed up of removing the vec->lock
contention, and replacing it with an atomic counter and memory barrier.
To prove this, I wrote a patch that times both the loop and the code
that grabs the vec->locks. I passed the patches to various people
(and companies) to test and show the results. I let everyone choose
their own load to test, giving different loads on the system,
for various different setups.
Here's some of the results: (snipping to a few CPUs to not make
this change log huge, but the results were consistent across
the entire system).
System 1 (24 CPUs)
Before patch:
CPU: Name Count Max Min Average Total
---- ---- ----- --- --- ------- -----
[...]
cpu 20: loop 3057 1.766 0.061 0.642 1963.170
vec 6782949 90.469 0.089 0.414 2811760.503
cpu 21: loop 2617 1.723 0.062 0.641 1679.074
vec 6782810 90.499 0.089 0.291 1978499.900
cpu 22: loop 2212 1.863 0.063 0.699 1547.160
vec 6767244 85.685 0.089 0.435 2949676.898
cpu 23: loop 2320 2.013 0.062 0.594 1380.265
vec 6781694 87.923 0.088 0.431 2928538.224
After patch:
cpu 20: loop 2078 1.579 0.061 0.533 1108.006
vec 6164555 5.704 0.060 0.143 885185.809
cpu 21: loop 2268 1.712 0.065 0.575 1305.248
vec 6153376 5.558 0.060 0.187 1154960.469
cpu 22: loop 1542 1.639 0.095 0.533 823.249
vec 6156510 5.720 0.060 0.190 1172727.232
cpu 23: loop 1650 1.733 0.068 0.545 900.781
vec 6170784 5.533 0.060 0.167 1034287.953
All times are in microseconds. The 'loop' is the amount of time spent
doing the loop across the priorities (before patch uses bitmask).
the 'vec' is the amount of time in the code that requires grabbing
the vec->lock. The second patch just does not have the vec lock, but
encompasses the same code.
Amazingly the loop code even went down on average. The vec code went
from .5 down to .18, that's more than half the time spent!
Note, more than one test was run, but they all had the same results.
System 2 (64 CPUs)
Before patch:
CPU: Name Count Max Min Average Total
---- ---- ----- --- --- ------- -----
cpu 60: loop 0 0 0 0 0
vec 5410840 277.954 0.084 0.782 4232895.727
cpu 61: loop 0 0 0 0 0
vec 4915648 188.399 0.084 0.570 2803220.301
cpu 62: loop 0 0 0 0 0
vec 5356076 276.417 0.085 0.786 4214544.548
cpu 63: loop 0 0 0 0 0
vec 4891837 170.531 0.085 0.799 3910948.833
After patch:
cpu 60: loop 0 0 0 0 0
vec 5365118 5.080 0.021 0.063 340490.267
cpu 61: loop 0 0 0 0 0
vec 4898590 1.757 0.019 0.071 347903.615
cpu 62: loop 0 0 0 0 0
vec 5737130 3.067 0.021 0.119 687108.734
cpu 63: loop 0 0 0 0 0
vec 4903228 1.822 0.021 0.071 348506.477
The test run during the measurement did not have any (very few,
from other CPUs) RT tasks pushing. But this shows that it helped
out tremendously with the contention, as the contention happens
because the vec->lock is taken only on queuing at an RT priority,
and different CPUs that queue tasks at the same priority will
have contention.
I tested on my own 4 CPU machine with the following results:
Before patch:
CPU: Name Count Max Min Average Total
---- ---- ----- --- --- ------- -----
cpu 0: loop 2377 1.489 0.158 0.588 1398.395
vec 4484 770.146 2.301 4.396 19711.755
cpu 1: loop 2169 1.962 0.160 0.576 1250.110
vec 4425 152.769 2.297 4.030 17834.228
cpu 2: loop 2324 1.749 0.155 0.559 1299.799
vec 4368 779.632 2.325 4.665 20379.268
cpu 3: loop 2325 1.629 0.157 0.561 1306.113
vec 4650 408.782 2.394 4.348 20222.577
After patch:
CPU: Name Count Max Min Average Total
---- ---- ----- --- --- ------- -----
cpu 0: loop 2121 1.616 0.113 0.636 1349.189
vec 4303 1.151 0.225 0.421 1811.966
cpu 1: loop 2130 1.638 0.178 0.644 1372.927
vec 4627 1.379 0.235 0.428 1983.648
cpu 2: loop 2056 1.464 0.165 0.637 1310.141
vec 4471 1.311 0.217 0.433 1937.927
cpu 3: loop 2154 1.481 0.162 0.601 1295.083
vec 4236 1.253 0.230 0.425 1803.008
This was running my migrate.c code that can be found at:
http://lwn.net/Articles/425763/
The migrate code does stress the RT tasks a bit. This shows that
the loop did increase a little after the patch, but not by much.
The vec code dropped dramatically. From 4.3us down to .42us.
That's a 10x improvement!
Tested-by: Mike Galbraith <mgalbraith@suse.de>
Tested-by: Luis Claudio R. Gonçalves <lgoncalv@redhat.com>
Tested-by: Matthew Hank Sabins<msabins@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Gregory Haskins <gregory.haskins@gmail.com>
Acked-by: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Chris Mason <chris.mason@oracle.com>
Link: http://lkml.kernel.org/r/1312317372.18583.101.camel@gandalf.stny.rr.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
