On Thu, Jun 19, 2008 at 12:27:14PM +0200, Peter Zijlstra wrote:
> On Thu, 2008-06-05 at 10:50 +0530, Ankita Garg wrote:
>
> > Thanks Peter for the explanation...
> >
> > I agree with the above and that is the reason why I did not see weird
> > values with cpu_time. But, run_delay still would suffer skews as the end
> > points for delta could be taken on different cpus due to migration (more
> > so on RT kernel due to the push-pull operations). With the below patch,
> > I could not reproduce the issue I had seen earlier. After every dequeue,
> > we take the delta and start wait measurements from zero when moved to a
> > different rq.
>
> OK, so task delay delay accounting is broken because it doesn't take
> migration into account.
>
> What you've done is make it symmetric wrt enqueue, and account it like
>
> cpu0 cpu1
>
> enqueue
> <wait-d1>
> dequeue
> enqueue
> <wait-d2>
> run
>
> Where you add both d1 and d2 to the run_delay,.. right?
>
Thanks for reviewing the patch. The above is exactly what I have done.
> This seems like a good fix, however it looks like the patch will break
> compilation in !CONFIG_SCHEDSTATS && !CONFIG_TASK_DELAY_ACCT, of it
> failing to provide a stub for sched_info_dequeue() in that case.
Fixed. Pl. find the new patch below.
Signed-off-by: Ankita Garg <ankita@in.ibm.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Gregory Haskins <ghaskins@novell.com>
Cc: rostedt@goodmis.org
Cc: suresh.b.siddha@intel.com
Cc: aneesh.kumar@linux.vnet.ibm.com
Cc: dhaval@linux.vnet.ibm.com
Cc: vatsa@linux.vnet.ibm.com
Cc: David Bahi <DBahi@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We have the notion of tracking process-coupling (a.k.a. buddy-wake) via
the p->se.last_wake / p->se.avg_overlap facilities, but it is only used
for cfs to cfs interactions. There is no reason why an rt to cfs
interaction cannot share in establishing a relationhip in a similar
manner.
Because PREEMPT_RT runs many kernel threads as FIFO priority, we often
times have heavy interaction between RT threads waking CFS applications.
This patch offers a substantial boost (50-60%+) in perfomance under those
circumstances.
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Cc: npiggin@suse.de
Cc: rostedt@goodmis.org
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch fixes the following warning:
kernel/sched.c:1667: warning: 'cfs_rq_set_shares' defined but not used
This seems the correct way to fix this; cfs_rq_set_shares() is only used
in a single place, which is also inside #ifdef CONFIG_FAIR_GROUP_SCHED.
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
fix:
kernel/sched.c: In function ‘sched_group_set_shares':
kernel/sched.c:8635: error: implicit declaration of function ‘cfs_rq_set_shares'
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Vegard Nossum reported:
> WARNING: at kernel/lockdep.c:2738 check_flags+0x142/0x160()
which happens due to:
unsigned long long cpu_clock(int cpu)
{
unsigned long long clock;
unsigned long flags;
raw_local_irq_save(flags);
as lower level functions can take locks, we must not do that, use
proper lockdep-annotated irq save/restore.
Reported-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Measurement shows that the difference between cgroup:/ and cgroup:/foo
wake_affine() results is that the latter succeeds significantly more.
Therefore bias the calculations towards failing the test.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Increase the accuracy of the effective_load values.
Not only consider the current increment (as per the attempted wakeup), but
also consider the delta between when we last adjusted the shares and the
current situation.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
It was observed these mults can overflow.
Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We found that the affine wakeup code needs rather accurate load figures
to be effective. The trouble is that updating the load figures is fairly
expensive with group scheduling. Therefore ratelimit the updating.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In case the domain is empty, pretend there is a single task on each cpu, so
that together with the boost logic we end up giving 1/n shares to each
cpu.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The bias given by source/target_load functions can be very large, disable
it by default to get faster convergence.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Priority looses much of its meaning in a hierarchical context. So don't
use it in balance decisions.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently task_h_load() computes the load of a task and uses that to either
subtract it from the total, or add to it.
However, removing or adding a task need not have any effect on the total load
at all. Imagine adding a task to a group that is local to one cpu - in that
case the total load of that cpu is unaffected.
So properly compute addition/removal:
s_i = S * rw_i / \Sum_j rw_j
s'_i = S * (rw_i + wl) / (\Sum_j rw_j + wg)
then s'_i - s_i gives the change in load.
Where s_i is the shares for cpu i, S the group weight, rw_i the runqueue weight
for that cpu, wl the weight we add (subtract) and wg the weight contribution to
the runqueue.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
doing the load balance will change cfs_rq->load.weight (that's the whole point)
but since that's part of the scale factor, we'll scale back with a different
amount.
Weight getting smaller would result in an inflated moved_load which causes
it to stop balancing too soon.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
find_busiest_group() has some assumptions about task weight being in the
NICE_0_LOAD range. Hierarchical task groups break this assumption - fix this
by replacing it with the average task weight, which will adapt the situation.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With hierarchical grouping we can't just compare task weight to rq weight - we
need to scale the weight appropriately.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Remove the fall-back to SCHED_LOAD_SCALE by remembering the previous value of
cpu_avg_load_per_task() - this is useful because of the hierarchical group
model in which task weight can be much smaller.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Finding the least idle cpu is more accurate when done with updated shares.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Re-compute the shares on newidle - so we can make a decision based on
recent data.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
While thinking about the previous patch - I realized that using per domain
aggregate load values in load_balance_fair() is wrong. We should use the
load value for that CPU.
By not needing per domain hierarchical load values we don't need to store
per domain aggregate shares, which greatly simplifies all the math.
It basically falls apart in two separate computations:
- per domain update of the shares
- per CPU update of the hierarchical load
Also get rid of the move_group_shares() stuff - just re-compute the shares
again after a successful load balance.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We only need to know the task_weight of the busiest rq - nothing to do
if there are no tasks there.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We used to try and contain the loss of 'shares' by playing arithmetic
games. Replace that by noticing that at the top sched_domain we'll
always have the full weight in shares to distribute.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The idea was to balance groups until we've reached the global goal, however
Vatsa rightly pointed out that we might never reach that goal this way -
hence take out this logic.
[ the initial rationale for this 'feature' was to promote max concurrency
within a group - it does not however affect fairness ]
Reported-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
It was observed that in __update_group_shares_cpu()
rq_weight > aggregate()->rq_weight
This is caused by forks/wakeups in between the initial aggregate pass and
locking of the RQs for load balance. To avoid this situation partially re-do
the aggregation once we have the RQs locked (which avoids new tasks from
appearing).
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Keeping the aggregate on the first cpu of the sched domain has two problems:
- it could collide between different sched domains on different cpus
- it could slow things down because of the remote accesses
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
show all the schedstats in /debug/sched_debug as well.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Uncouple buddy selection from wakeup granularity.
The initial idea was that buddies could run ahead as far as a normal task
can - do this by measuring a pair 'slice' just as we do for a normal task.
This means we can drop the wakeup_granularity back to 5ms.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
with sched_clock_cpu() being reasonably in sync between cpus (max 1 jiffy
difference) use this to provide cpu_clock().
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ok, so why are we in this mess, it was:
1/w
but now we mixed that rw in the mix like:
rw/w
rw being \Sum w suggests: fiddling w, we should also fiddle rw, humm?
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
calc_delta_asym() is supposed to do the same as calc_delta_fair() except
linearly shrink the result for negative nice processes - this causes them
to have a smaller preemption threshold so that they are more easily preempted.
The problem is that for task groups se->load.weight is the per cpu share of
the actual task group weight; take that into account.
Also provide a debug switch to disable the asymmetry (which I still don't
like - but it does greatly benefit some workloads)
This would explain the interactivity issues reported against group scheduling.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In file included from /mnt/build/linux-2.6/kernel/sched.c:1496:
/mnt/build/linux-2.6/kernel/sched_rt.c: In function '__enable_runtime':
/mnt/build/linux-2.6/kernel/sched_rt.c:339: warning: unused variable 'rd'
/mnt/build/linux-2.6/kernel/sched_rt.c: In function 'requeue_rt_entity':
/mnt/build/linux-2.6/kernel/sched_rt.c:692: warning: unused variable 'queue'
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
- Fix warning reported by sparse
kernel/kgdb.c:1502:6: warning: symbol 'kgdb_console_write' was not declared.
Should it be static?
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
This patch addresses a very sporadic pi-futex related failure in
highly threaded java apps on large SMP systems.
David Holmes reported that the pi_state consistency check in
lookup_pi_state triggered with his test application. This means that
the kernel internal pi_state and the user space futex variable are out
of sync. First we assumed that this is a user space data corruption,
but deeper investigation revieled that the problem happend because the
pi-futex code is not handling a fault in the futex_lock_pi path when
the user space variable needs to be fixed up.
The fault happens when a fork mapped the anon memory which contains
the futex readonly for COW or the page got swapped out exactly between
the unlock of the futex and the return of either the new futex owner
or the task which was the expected owner but failed to acquire the
kernel internal rtmutex. The current futex_lock_pi() code drops out
with an inconsistent in case it faults and returns -EFAULT to user
space. User space has no way to fixup that state.
When we wrote this code we thought that we could not drop the hash
bucket lock at this point to handle the fault.
After analysing the code again it turned out to be wrong because there
are only two tasks involved which might modify the pi_state and the
user space variable:
- the task which acquired the rtmutex
- the pending owner of the pi_state which did not get the rtmutex
Both tasks drop into the fixup_pi_state() function before returning to
user space. The first task which acquired the hash bucket lock faults
in the fixup of the user space variable, drops the spinlock and calls
futex_handle_fault() to fault in the page. Now the second task could
acquire the hash bucket lock and tries to fixup the user space
variable as well. It either faults as well or it succeeds because the
first task already faulted the page in.
One caveat is to avoid a double fixup. After returning from the fault
handling we reacquire the hash bucket lock and check whether the
pi_state owner has been modified already.
Reported-by: David Holmes <david.holmes@sun.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: David Holmes <david.holmes@sun.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/futex.c | 93 ++++++++++++++++++++++++++++++++++++++++++++-------------
1 file changed, 73 insertions(+), 20 deletions(-)
* 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
softlockup: fix NMI hangs due to lock race - 2.6.26-rc regression
rcupreempt: remove export of rcu_batches_completed_bh
cpuset: limit the input of cpuset.sched_relax_domain_level
Simplify the code and fix the boundary condition of
wait_for_completion_timeout(,0).
We can kill the first __remove_wait_queue() as well.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
It seems that the current implementaton of wait_for_completion_timeout()
has a small problem under very high load for the common pattern:
if (!wait_for_completion_timeout(&done, timeout))
/* handle failure */
because the implementation very roughly does (lots of code deleted to
show the basic flow):
static inline long __sched
do_wait_for_common(struct completion *x, long timeout, int state)
{
if (x->done)
return timeout;
do {
timeout = schedule_timeout(timeout);
if (!timeout)
return timeout;
} while (!x->done);
return timeout;
}
so if the system is very busy and x->done is not set when
do_wait_for_common() is entered, it is possible that the first call to
schedule_timeout() returns 0 because the task doing wait_for_completion
doesn't get rescheduled for a long time, even if it is woken up early
enough.
In this case, wait_for_completion_timeout() returns 0 without even
checking x->done again, and the code above falls into its failure case
purely for scheduler reasons, even if the hardware event or whatever was
being waited for happened early enough.
It would make sense to add an extra test to do_wait_for() in the timeout
case and return 1 if x->done is actually set.
A quick audit (not exhaustive) of wait_for_completion_timeout() callers
seems to indicate that no one actually cares about the return value in
the success case -- they just test for 0 (timed out) versus non-zero
(wait succeeded).
Signed-off-by: Ingo Molnar <mingo@elte.hu>
So if the group ever gets throttled, it will never wake up again.
Reported-by: "Daniel K." <dk@uw.no>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Daniel K. <dk@uw.no>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
So if the group ever gets throttled, it will never wake up again.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: "Daniel K." <dk@uw.no>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reported-by: "Daniel K." <dk@uw.no>
