Re: [PATCH v5 1/3] sched: Stop nohz stats when decayed

From: Vincent Guittot
Date: Fri Feb 16 2018 - 12:03:28 EST


On 16 February 2018 at 13:53, Valentin Schneider
<valentin.schneider@xxxxxxx> wrote:
> On 02/14/2018 03:26 PM, Vincent Guittot wrote:
>> Stopped the periodic update of blocked load when all idle CPUs have fully
>> decayed. We introduce a new nohz.has_blocked that reflect if some idle
>> CPUs has blocked load that have to be periodiccally updated. nohz.has_blocked
>> is set everytime that a Idle CPU can have blocked load and it is then clear
>> when no more blocked load has been detected during an update. We don't need
>> atomic operation but only to make cure of the right ordering when updating
>> nohz.idle_cpus_mask and nohz.has_blocked.
>>
>> Suggested-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
>> Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
>> ---
>> kernel/sched/fair.c | 122 ++++++++++++++++++++++++++++++++++++++++++---------
>> kernel/sched/sched.h | 1 +
>> 2 files changed, 102 insertions(+), 21 deletions(-)
>>
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index 7af1fa9..5a6835e 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>>
>> [...]
>>
>> -static void update_nohz_stats(struct rq *rq)
>> +static bool update_nohz_stats(struct rq *rq)
>> {
>> #ifdef CONFIG_NO_HZ_COMMON
>> unsigned int cpu = rq->cpu;
>>
>> + if (!rq->has_blocked_load)
>> + return false;
>> +
>> if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask))
>> - return;
>> + return false;
>>
>> if (!time_after(jiffies, rq->last_blocked_load_update_tick))
>> - return;
>> + return true;
>>
>> update_blocked_averages(cpu);
>> +
>> + return rq->has_blocked_load;
>> +#else
>> + return false;
>> #endif
>> }
>>
>
> (Wrongly thought that this bit was in a different patch, comment should have
> been squashed in previous reply...)
>
> I've been thinking about this, and it's a messy one if we want to
> skip CPUs in idle_balance() / clear the nohz.has_blocked_flag.
>
> AFAICT, the rq->has_blocked_load flag can be wrongly cleared: if we're
> calling update_nohz_stats() for CPU A, but CPU A got out/in of
> idle really quickly in that same timeframe, I'm not sure you can guarantee
> the clearing of rq->has_blocked_load done in update_blocked_averages() will
> always end up in memory before the setting of the flag in
> nohz_balance_enter_idle().

Not sure it's a problem in this case because the clear done in
update_blocked_averages() only happens if there is no load on the rq
and new load can't be added in the mean time

>
> I was going to say we don't have this problem in _nohz_idle_balance() but
> actually I think we do. We have the checking of nohz.idle_cpus_mask after the
> smp_mb(), which makes sure the clear of nohz.has_blocked will never
> overwrite the set in nohz_balance_enter_idle(), but it doesn't
> guarantee the same for the rq flag. So we can have nohz CPUs with blocked
> load but with rq->has_blocked_load set to false. Which isn't a problem now

I don't think that we can have this case. Or at least I can't find a
sequence leading to this state.

Have you got a particular sequence in mind ?

> but it is if we want to use the flag to skip CPUs.
>
> Am I correct or am I going crazy ? There's a comment about this in
> nohz_balance_enter_idle() but I'm confused now:
>
> /*
> * Can be set safely without rq->lock held
> * If a clear happens, it will have evaluated last additions because
> * rq->lock is held during the check and the clear
> */
> rq->has_blocked_load = 1;