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

Groups > linux.kernel > #1231075 > unrolled thread

Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned

Back to article view | Back to linux.kernel

Contents

  Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned Greg Thelen <gthelen@google.com> - 2015-09-23 02:50 +0200
    Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned Andrew Morton <akpm@linux-foundation.org> - 2015-09-23 06:10 +0200
      Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned Greg Thelen <gthelen@google.com> - 2015-09-23 09:30 +0200
        Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned Michal Hocko <mhocko@kernel.org> - 2015-09-25 17:30 +0200

#1231075 — Re: [PATCH] memcg: make mem_cgroup_read_stat() unsigned

Andrew Morton wrote:

> On Tue, 22 Sep 2015 15:16:32 -0700 Greg Thelen <gthelen@google.com> wrote:
>
>> mem_cgroup_read_stat() returns a page count by summing per cpu page
>> counters.  The summing is racy wrt. updates, so a transient negative sum
>> is possible.  Callers don't want negative values:
>> - mem_cgroup_wb_stats() doesn't want negative nr_dirty or nr_writeback.
>> - oom reports and memory.stat shouldn't show confusing negative usage.
>> - tree_usage() already avoids negatives.
>>
>> Avoid returning negative page counts from mem_cgroup_read_stat() and
>> convert it to unsigned.
>
> Someone please remind me why this code doesn't use the existing
> percpu_counter library which solved this problem years ago.
>
>>   for_each_possible_cpu(cpu)
>
> and which doesn't iterate across offlined CPUs.

I found [1] and [2] discussing memory layout differences between:
a) existing memcg hand rolled per cpu arrays of counters
vs
b) array of generic percpu_counter
The current approach was claimed to have lower memory overhead and
better cache behavior.

I assume it's pretty straightforward to create generic
percpu_counter_array routines which memcg could use.  Possibly something
like this could be made general enough could be created to satisfy
vmstat, but less clear.

[1] http://www.spinics.net/lists/cgroups/msg06216.html
[2] https://lkml.org/lkml/2014/9/11/1057
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1231128

On Tue, 22 Sep 2015 17:42:13 -0700 Greg Thelen <gthelen@google.com> wrote:

> Andrew Morton wrote:
> 
> > On Tue, 22 Sep 2015 15:16:32 -0700 Greg Thelen <gthelen@google.com> wrote:
> >
> >> mem_cgroup_read_stat() returns a page count by summing per cpu page
> >> counters.  The summing is racy wrt. updates, so a transient negative sum
> >> is possible.  Callers don't want negative values:
> >> - mem_cgroup_wb_stats() doesn't want negative nr_dirty or nr_writeback.
> >> - oom reports and memory.stat shouldn't show confusing negative usage.
> >> - tree_usage() already avoids negatives.
> >>
> >> Avoid returning negative page counts from mem_cgroup_read_stat() and
> >> convert it to unsigned.
> >
> > Someone please remind me why this code doesn't use the existing
> > percpu_counter library which solved this problem years ago.
> >
> >>   for_each_possible_cpu(cpu)
> >
> > and which doesn't iterate across offlined CPUs.
> 
> I found [1] and [2] discussing memory layout differences between:
> a) existing memcg hand rolled per cpu arrays of counters
> vs
> b) array of generic percpu_counter
> The current approach was claimed to have lower memory overhead and
> better cache behavior.
> 
> I assume it's pretty straightforward to create generic
> percpu_counter_array routines which memcg could use.  Possibly something
> like this could be made general enough could be created to satisfy
> vmstat, but less clear.
> 
> [1] http://www.spinics.net/lists/cgroups/msg06216.html
> [2] https://lkml.org/lkml/2014/9/11/1057

That all sounds rather bogus to me.  __percpu_counter_add() doesn't
modify struct percpu_counter at all except for when the cpu-local
counter overflows the configured batch size.  And for the memcg
application I suspect we can set the batch size to INT_MAX...


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1231189

Andrew Morton wrote:

> On Tue, 22 Sep 2015 17:42:13 -0700 Greg Thelen <gthelen@google.com> wrote:
>
>> Andrew Morton wrote:
>> 
>> > On Tue, 22 Sep 2015 15:16:32 -0700 Greg Thelen <gthelen@google.com> wrote:
>> >
>> >> mem_cgroup_read_stat() returns a page count by summing per cpu page
>> >> counters.  The summing is racy wrt. updates, so a transient negative sum
>> >> is possible.  Callers don't want negative values:
>> >> - mem_cgroup_wb_stats() doesn't want negative nr_dirty or nr_writeback.
>> >> - oom reports and memory.stat shouldn't show confusing negative usage.
>> >> - tree_usage() already avoids negatives.
>> >>
>> >> Avoid returning negative page counts from mem_cgroup_read_stat() and
>> >> convert it to unsigned.
>> >
>> > Someone please remind me why this code doesn't use the existing
>> > percpu_counter library which solved this problem years ago.
>> >
>> >>   for_each_possible_cpu(cpu)
>> >
>> > and which doesn't iterate across offlined CPUs.
>> 
>> I found [1] and [2] discussing memory layout differences between:
>> a) existing memcg hand rolled per cpu arrays of counters
>> vs
>> b) array of generic percpu_counter
>> The current approach was claimed to have lower memory overhead and
>> better cache behavior.
>> 
>> I assume it's pretty straightforward to create generic
>> percpu_counter_array routines which memcg could use.  Possibly something
>> like this could be made general enough could be created to satisfy
>> vmstat, but less clear.
>> 
>> [1] http://www.spinics.net/lists/cgroups/msg06216.html
>> [2] https://lkml.org/lkml/2014/9/11/1057
>
> That all sounds rather bogus to me.  __percpu_counter_add() doesn't
> modify struct percpu_counter at all except for when the cpu-local
> counter overflows the configured batch size.  And for the memcg
> application I suspect we can set the batch size to INT_MAX...

Nod.  The memory usage will be a bit larger, but the code reuse is
attractive.  I dusted off Vladimir's
https://lkml.org/lkml/2014/9/11/710.  Next step is to benchmark it
before posting.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1232890

On Wed 23-09-15 00:21:33, Greg Thelen wrote:
> 
> Andrew Morton wrote:
> 
> > On Tue, 22 Sep 2015 17:42:13 -0700 Greg Thelen <gthelen@google.com> wrote:
[...]
> >> I assume it's pretty straightforward to create generic
> >> percpu_counter_array routines which memcg could use.  Possibly something
> >> like this could be made general enough could be created to satisfy
> >> vmstat, but less clear.
> >> 
> >> [1] http://www.spinics.net/lists/cgroups/msg06216.html
> >> [2] https://lkml.org/lkml/2014/9/11/1057
> >
> > That all sounds rather bogus to me.  __percpu_counter_add() doesn't
> > modify struct percpu_counter at all except for when the cpu-local
> > counter overflows the configured batch size.  And for the memcg
> > application I suspect we can set the batch size to INT_MAX...
> 
> Nod.  The memory usage will be a bit larger, but the code reuse is
> attractive.  I dusted off Vladimir's
> https://lkml.org/lkml/2014/9/11/710.  Next step is to benchmark it
> before posting.

I am definitely in favor of using generic per-cpu counters.

-- 
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web