Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1431180 > unrolled thread

Re: [PATCH 2/2] ipc/sem: sem_lock with hysteresis

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH 2/2] ipc/sem: sem_lock with hysteresis Manfred Spraul <manfred@colorfullife.com> - 2016-06-25 19:40 +0200
    Re: [PATCH 2/2] ipc/sem: sem_lock with hysteresis Davidlohr Bueso <dave@stgolabs.net> - 2016-06-28 20:00 +0200

#1431180 — Re: [PATCH 2/2] ipc/sem: sem_lock with hysteresis

On 06/21/2016 10:29 PM, Davidlohr Bueso wrote:
> On Sat, 18 Jun 2016, Manfred Spraul wrote:
>
>> sysv sem has two lock modes: One with per-semaphore locks, one lock mode
>> with a single big lock for the whole array.
>> When switching from the per-semaphore locks to the big lock, all
>> per-semaphore locks must be scanned for ongoing operations.
>>
>> The patch adds a hysteresis for switching from the big lock to the per
>> semaphore locks. This reduces how often the per-semaphore locks must
>> be scanned.
>
> Isn't this very arbitrary depending on the workload? Ie the other way 
> around:
> when we have a lot more simple ops going on not so good. While I'm 
> more worried
> about combinations that could cause enough complex ops to always delay 
> taking
> the finer grained lock, this change also obviously makes simple ops 
> more expensive
> on newly created segments.
I

Entering complex mode requires a scan of sem_base[].sem_lock.
>         for (i = 0; i < sma->sem_nsems; i++) {
>                 sem = sma->sem_base + i;
>                 spin_unlock_wait(&sem->lock);
This is what the patch tries to avoid.
>
> In general I don't trust magic numbers much. What sort of numbers have 
> you seen
> with this patch? Is this a real concern (particularly because a lot of 
> the sem->lock
> work was because real world workloads were doing a lot more simple ops 
> afaicr)?
>
With a microbenchmark: As much improvement as you want :-)

- Only simple ops: patch has no impact (the first 10 semops do not matter)
- sleeping complex ops: patch has no impact, we are always in complex mode
- not sleeping complex ops: depends on the size of the array.
With a 4.000 semaphore array, I see an improvement of factor 20.

There is obviously one case where the patch causes a slowdown:
- complex op, then 11 simple ops, then repeat.


Perhaps: set COMPLEX_MODE_ENTER to 1 or 2, then allow to configure it 
from user space.
Or do not merge the patch and wait until someone come with a profile 
that shows complexmode_enter().

--
     Manfred

[toc] | [next] | [standalone]

#1433190

On Sat, 25 Jun 2016, Manfred Spraul wrote:

>- Only simple ops: patch has no impact (the first 10 semops do not matter)

Agreed.

>- sleeping complex ops: patch has no impact, we are always in complex mode
>- not sleeping complex ops: depends on the size of the array.
>With a 4.000 semaphore array, I see an improvement of factor 20.
>
>There is obviously one case where the patch causes a slowdown:
>- complex op, then 11 simple ops, then repeat.

Yeah, you reset the counter to COMPLEX_MODE_ENTER every time we do a
complexmode_enter(), so if you have interleaving of complex and simple
ops you could end up always taking the big lock. This is my main concern
and not an unusual scenario at all.

I wonder if we could be a bit less aggressive if already in complex_mode
and do something like:

if (sma->complex_mode > 0) {
   WRITE_ONCE(sma->complex_mode, min(complex_mode + 1, COMPLEX_MODE_ENTER));
   return;
}

and still be constrained by COMPLEX_MODE_ENTER... of course that has its own
additional overhead, albeit less contention on the big lock :/

>
>Perhaps: set COMPLEX_MODE_ENTER to 1 or 2, then allow to configure it
>from user space.

Nah, I don't think it's a good idea to expose such internals to userspace.

>Or do not merge the patch and wait until someone come with a profile
>that shows complexmode_enter().

Testing/benchmarking this patch is on my todo list mainly because I'm lacking
a decent box to test it on. But I'm not conformable having this one without
any numbers for say at least a rdbms benchmark. I'm very eager to add the
first patch (complex_mode) once the nits are settled, as it fixes a real bug.

Thanks,
Davidlohr

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web