From: Pekka Enberg <penberg@cs.helsinki.fi>
To: Christoph Lameter <cl@linux-foundation.org>
Cc: Tejun Heo <tj@kernel.org>,
linux-kernel@vger.kernel.org, Mel Gorman <mel@csn.ul.ie>,
Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Subject: Re: [this_cpu_xx V8 03/16] Use this_cpu operations in slub
Date: Sun, 20 Dec 2009 11:11:15 +0200 [thread overview]
Message-ID: <4B2DEA33.7000800@cs.helsinki.fi> (raw)
In-Reply-To: <20091218222649.091070364@quilx.com>
Christoph Lameter wrote:
> Using per cpu allocations removes the needs for the per cpu arrays in the
> kmem_cache struct. These could get quite big if we have to support systems
> with thousands of cpus. The use of this_cpu_xx operations results in:
>
> 1. The size of kmem_cache for SMP configuration shrinks since we will only
> need 1 pointer instead of NR_CPUS. The same pointer can be used by all
> processors. Reduces cache footprint of the allocator.
>
> 2. We can dynamically size kmem_cache according to the actual nodes in the
> system meaning less memory overhead for configurations that may potentially
> support up to 1k NUMA nodes / 4k cpus.
>
> 3. We can remove the diddle widdle with allocating and releasing of
> kmem_cache_cpu structures when bringing up and shutting down cpus. The cpu
> alloc logic will do it all for us. Removes some portions of the cpu hotplug
> functionality.
>
> 4. Fastpath performance increases since per cpu pointer lookups and
> address calculations are avoided.
>
> V7-V8
> - Convert missed get_cpu_slab() under CONFIG_SLUB_STATS
>
> Cc: Pekka Enberg <penberg@cs.helsinki.fi>
> Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Patches 3-6 applied.
next prev parent reply other threads:[~2009-12-20 9:11 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-18 22:26 [this_cpu_xx V8 00/16] Per cpu atomics in core allocators, cleanup and more this_cpu_ops Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 01/16] this_cpu_ops: page allocator conversion Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 02/16] this_cpu ops: Remove pageset_notifier Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 03/16] Use this_cpu operations in slub Christoph Lameter
2009-12-20 9:11 ` Pekka Enberg [this message]
2009-12-18 22:26 ` [this_cpu_xx V8 04/16] SLUB: Get rid of dynamic DMA kmalloc cache allocation Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 05/16] this_cpu: Remove slub kmem_cache fields Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 06/16] Make slub statistics use this_cpu_inc Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 07/16] Module handling: Use this_cpu_xx to dynamically allocate counters Christoph Lameter
2009-12-21 7:47 ` Tejun Heo
2009-12-21 7:59 ` Tejun Heo
2009-12-21 8:19 ` Tejun Heo
2009-12-21 23:28 ` Rusty Russell
2009-12-22 0:02 ` Tejun Heo
2009-12-22 16:17 ` Christoph Lameter
2009-12-22 15:58 ` Christoph Lameter
2009-12-22 15:57 ` Christoph Lameter
2009-12-23 2:07 ` Tejun Heo
2010-01-04 17:22 ` Christoph Lameter
2010-01-04 17:51 ` Mathieu Desnoyers
2009-12-18 22:26 ` [this_cpu_xx V8 08/16] Remove cpu_local_xx macros Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 09/16] Allow arch to provide inc/dec functionality for each size separately Christoph Lameter
2009-12-21 7:25 ` Tejun Heo
2009-12-22 15:56 ` Christoph Lameter
2009-12-23 2:08 ` Tejun Heo
2009-12-18 22:26 ` [this_cpu_xx V8 10/16] Support generating inc/dec for this_cpu_inc/dec Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 11/16] Generic support for this_cpu_cmpxchg Christoph Lameter
2009-12-19 14:45 ` Mathieu Desnoyers
2009-12-22 15:54 ` Christoph Lameter
2009-12-22 17:24 ` Mathieu Desnoyers
2010-01-04 17:21 ` Christoph Lameter
2010-01-05 22:29 ` Mathieu Desnoyers
2010-01-05 22:35 ` Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 12/16] Add percpu cmpxchg operations Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 13/16] Generic support for this_cpu_xchg Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 14/16] x86 percpu xchg operation Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 15/16] Generic support for this_cpu_add_return() Christoph Lameter
2009-12-18 22:26 ` [this_cpu_xx V8 16/16] x86 support for this_cpu_add_return Christoph Lameter
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=4B2DEA33.7000800@cs.helsinki.fi \
--to=penberg@cs.helsinki.fi \
--cc=cl@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mel@csn.ul.ie \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome