From: Leonard Crestez <lcrestez@ixiacom.com>
To: Tejun Heo <tj@kernel.org>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Christoph Lameter <cl@linux-foundation.org>,
Sorin Dumitru <sdumitru@ixiacom.com>
Subject: Re: [RFC v2] percpu: Add a separate function to merge free areas
Date: Thu, 4 Dec 2014 22:10:18 +0200 [thread overview]
Message-ID: <5480BFAA.2020106@ixiacom.com> (raw)
In-Reply-To: <20141204175713.GE2995@htj.dyndns.org>
On 12/04/2014 07:57 PM, Tejun Heo wrote:
> Hello,
>
> On Wed, Dec 03, 2014 at 12:33:59AM +0200, Leonard Crestez wrote:
>> It seems that free_percpu performance is very bad when working with small
>> objects. The easiest way to reproduce this is to allocate and then free a large
>> number of percpu int counters in order. Small objects (reference counters and
>> pointers) are common users of alloc_percpu and I think this should be fast.
>> This particular issue can be encountered with very large number of net_device
>> structs.
>
> Do you actually experience this with an actual workload? The thing is
> allocation has the same quadratic complexity. If this is actually an
> issue (which can definitely be the case), I'd much prefer implementing
> a properly scalable area allocator than mucking with the current
> implementation.
Yes, we are actually experiencing issues with this. We create lots of virtual
net_devices and routes, which means lots of percpu counters/pointers. In particular
we are getting worse performance than in older kernels because the net_device refcnt
is now a percpu counter. We could turn that back into a single integer but this
would negate an upstream optimization.
We are working on top of linux_3.10. We already pulled some allocation optimizations.
At least for simple allocation patterns pcpu_alloc does not appear to be unreasonably
slow.
Having a "properly scalable" percpu allocator would be quite nice indeed.
Regards,
Leonard
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Leonard Crestez <lcrestez@ixiacom.com>
To: Tejun Heo <tj@kernel.org>
Cc: <linux-mm@kvack.org>, <linux-kernel@vger.kernel.org>,
Christoph Lameter <cl@linux-foundation.org>,
Sorin Dumitru <sdumitru@ixiacom.com>
Subject: Re: [RFC v2] percpu: Add a separate function to merge free areas
Date: Thu, 4 Dec 2014 22:10:18 +0200 [thread overview]
Message-ID: <5480BFAA.2020106@ixiacom.com> (raw)
In-Reply-To: <20141204175713.GE2995@htj.dyndns.org>
On 12/04/2014 07:57 PM, Tejun Heo wrote:
> Hello,
>
> On Wed, Dec 03, 2014 at 12:33:59AM +0200, Leonard Crestez wrote:
>> It seems that free_percpu performance is very bad when working with small
>> objects. The easiest way to reproduce this is to allocate and then free a large
>> number of percpu int counters in order. Small objects (reference counters and
>> pointers) are common users of alloc_percpu and I think this should be fast.
>> This particular issue can be encountered with very large number of net_device
>> structs.
>
> Do you actually experience this with an actual workload? The thing is
> allocation has the same quadratic complexity. If this is actually an
> issue (which can definitely be the case), I'd much prefer implementing
> a properly scalable area allocator than mucking with the current
> implementation.
Yes, we are actually experiencing issues with this. We create lots of virtual
net_devices and routes, which means lots of percpu counters/pointers. In particular
we are getting worse performance than in older kernels because the net_device refcnt
is now a percpu counter. We could turn that back into a single integer but this
would negate an upstream optimization.
We are working on top of linux_3.10. We already pulled some allocation optimizations.
At least for simple allocation patterns pcpu_alloc does not appear to be unreasonably
slow.
Having a "properly scalable" percpu allocator would be quite nice indeed.
Regards,
Leonard
next prev parent reply other threads:[~2014-12-04 20:10 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-12-02 22:33 [RFC v2] percpu: Add a separate function to merge free areas Leonard Crestez
2014-12-02 22:33 ` Leonard Crestez
2014-12-04 17:57 ` Tejun Heo
2014-12-04 17:57 ` Tejun Heo
2014-12-04 20:10 ` Leonard Crestez [this message]
2014-12-04 20:10 ` Leonard Crestez
2014-12-04 20:28 ` Christoph Lameter
2014-12-04 20:28 ` Christoph Lameter
2014-12-04 20:45 ` Tejun Heo
2014-12-04 20:45 ` Tejun Heo
2014-12-04 20:52 ` Al Viro
2014-12-04 20:52 ` Al Viro
2014-12-04 21:15 ` Christoph Lameter
2014-12-04 21:15 ` Christoph Lameter
2014-12-04 21:19 ` Tejun Heo
2014-12-04 21:19 ` Tejun Heo
2014-12-04 21:20 ` Christoph Lameter
2014-12-04 21:20 ` Christoph Lameter
2014-12-05 6:25 ` Konstantin Khlebnikov
2014-12-05 6:25 ` Konstantin Khlebnikov
2014-12-04 20:42 ` Tejun Heo
2014-12-04 20:42 ` Tejun Heo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5480BFAA.2020106@ixiacom.com \
--to=lcrestez@ixiacom.com \
--cc=cl@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=sdumitru@ixiacom.com \
--cc=tj@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.