From: Eric Dumazet <dada1@cosmosbay.com>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Paul Jackson <pj@sgi.com>,
akpm@osdl.org, linux-kernel@vger.kernel.org,
Simon Derr <Simon.Derr@bull.net>, Andi Kleen <ak@suse.de>,
Christoph Lameter <clameter@sgi.com>
Subject: Re: [PATCH] Cpuset: rcu optimization of page alloc hook
Date: Mon, 12 Dec 2005 10:06:58 +0100 [thread overview]
Message-ID: <439D3DB2.9010900@cosmosbay.com> (raw)
In-Reply-To: <439D3AD5.3080403@yahoo.com.au>
Nick Piggin a écrit :
> Eric Dumazet wrote:
>
>> Paul Jackson a écrit :
>>
>>> +
>>> +static kmem_cache_t *cpuset_cache;
>>> +
>>
>>
>>
>> Hi Paul
>>
>> Please do use __read_mostly for new kmem_cache :
>>
>> static kmem_cache_t *cpuset_cache __read_mostly;
>>
>> If not, the pointer can sit in the midle of a highly modified cache
>> line, and multiple CPUS will have memory cache misses to access the
>> cpuset_cache, while slab code/data layout itself is very NUMA/SMP
>> friendly.
>>
>
> Is it a good idea for all kmem_cache_t? If so, can we move
> __read_mostly to the type definition?
>
>
Well this question was already asked.
You cannot move __read_mostly to the type itself as some kmem_cache_t are
included in some data structures.
struct {
...
kmem_cache_t *slab;
...
};
And in this case you cannot automatically add __read_mostly
Eric
next prev parent reply other threads:[~2005-12-12 9:07 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-11 23:31 [PATCH] Cpuset: rcu optimization of page alloc hook Paul Jackson
2005-12-12 3:29 ` Andi Kleen
2005-12-12 6:11 ` Paul Jackson
2005-12-12 6:21 ` Andi Kleen
2005-12-12 6:50 ` Paul Jackson
2005-12-12 8:49 ` Eric Dumazet
2005-12-12 8:54 ` Nick Piggin
2005-12-12 9:06 ` Eric Dumazet [this message]
2005-12-12 9:11 ` Andrew Morton
2005-12-12 9:38 ` Nick Piggin
2005-12-12 10:02 ` Paul Jackson
2005-12-12 10:12 ` Andrew Morton
2005-12-13 15:53 ` Paul Jackson
2005-12-13 16:31 ` Eric Dumazet
2005-12-13 17:42 ` Christoph Lameter
2005-12-13 17:56 ` Eric Dumazet
2005-12-13 18:07 ` Christoph Lameter
2005-12-13 21:03 ` Paul Jackson
2005-12-13 21:16 ` Christoph Lameter
2005-12-13 21:38 ` Eric Dumazet
2005-12-13 22:23 ` Paul Jackson
2005-12-13 22:29 ` Christoph Lameter
2005-12-14 3:54 ` Paul Jackson
2005-12-14 4:02 ` Andi Kleen
2005-12-14 4:06 ` Paul Jackson
2005-12-14 8:06 ` Eric Dumazet
2005-12-14 8:40 ` Paul Jackson
2005-12-13 20:08 ` Paul Jackson
2005-12-13 20:29 ` Eric Dumazet
2005-12-13 22:35 ` Paul Jackson
2005-12-13 21:44 ` Paul Jackson
2005-12-13 17:37 ` 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=439D3DB2.9010900@cosmosbay.com \
--to=dada1@cosmosbay.com \
--cc=Simon.Derr@bull.net \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=clameter@sgi.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nickpiggin@yahoo.com.au \
--cc=pj@sgi.com \
/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.