From: Nick Piggin <nickpiggin@yahoo.com.au>
To: "David S. Miller" <davem@davemloft.net>
Cc: anton@samba.org, akpm@osdl.org, torvalds@osdl.org,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC][PATCH 0/3] beat kswapd with the proverbial clue-bat
Date: Mon, 06 Sep 2004 18:55:16 +1000 [thread overview]
Message-ID: <413C25F4.8030501@yahoo.com.au> (raw)
In-Reply-To: <20040905203331.7a2a2fad.davem@davemloft.net>
David S. Miller wrote:
> On Sun, 05 Sep 2004 20:09:30 +1000
> Nick Piggin <nickpiggin@yahoo.com.au> wrote:
>
>
>>Yeah I had seen a few, surprisingly few though. Sorry I'm a bit clueless
>>about networking - I suppose there is a good reason for the 16K MTU? My
>>first thought might be that a 4K one could be better on CPU cache as well
>>as lighter on the mm. I know the networking guys know what they're doing
>>though...
>
>
> It's better to get as long a stride as possible for the copy
> from userspace, and yes as you get larger you run into cache
> issues. 16K turned out the be the break point considering those
> two attributes when I did my testing.
>
OK. Makes sense.
> Just fool around with ifconfig lo mtu XXX and TCP bandwidth tests.
> See what you come up with.
>
Thanks, I'll give that a try. I don't nearly have access to a
representitive range of architectures, but if I see anything
interesting on what I've got, I'll ping you.
WARNING: multiple messages have this Message-ID (diff)
From: Nick Piggin <nickpiggin@yahoo.com.au>
To: "David S. Miller" <davem@davemloft.net>
Cc: anton@samba.org, akpm@osdl.org, torvalds@osdl.org,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC][PATCH 0/3] beat kswapd with the proverbial clue-bat
Date: Mon, 06 Sep 2004 18:55:16 +1000 [thread overview]
Message-ID: <413C25F4.8030501@yahoo.com.au> (raw)
In-Reply-To: <20040905203331.7a2a2fad.davem@davemloft.net>
David S. Miller wrote:
> On Sun, 05 Sep 2004 20:09:30 +1000
> Nick Piggin <nickpiggin@yahoo.com.au> wrote:
>
>
>>Yeah I had seen a few, surprisingly few though. Sorry I'm a bit clueless
>>about networking - I suppose there is a good reason for the 16K MTU? My
>>first thought might be that a 4K one could be better on CPU cache as well
>>as lighter on the mm. I know the networking guys know what they're doing
>>though...
>
>
> It's better to get as long a stride as possible for the copy
> from userspace, and yes as you get larger you run into cache
> issues. 16K turned out the be the break point considering those
> two attributes when I did my testing.
>
OK. Makes sense.
> Just fool around with ifconfig lo mtu XXX and TCP bandwidth tests.
> See what you come up with.
>
Thanks, I'll give that a try. I don't nearly have access to a
representitive range of architectures, but if I see anything
interesting on what I've got, I'll ping you.
--
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:"aart@kvack.org"> aart@kvack.org </a>
next prev parent reply other threads:[~2004-09-06 8:55 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-05 5:44 [RFC][PATCH 0/3] beat kswapd with the proverbial clue-bat Nick Piggin
2004-09-05 5:44 ` Nick Piggin
2004-09-05 5:45 ` [RFC][PATCH 1/3] account free buddy areas Nick Piggin
2004-09-05 5:46 ` [RFC][PATCH 2/3] alloc-order watermarks Nick Piggin
2004-09-05 5:47 ` [RFC][PATCH 3/3] teach kswapd about watermarks Nick Piggin
2004-09-05 6:04 ` David S. Miller
2004-09-05 6:04 ` David S. Miller
2004-09-05 6:20 ` Nick Piggin
2004-09-05 6:20 ` Nick Piggin
2004-09-05 5:50 ` [RFC][PATCH 2/3] alloc-order watermarks Nick Piggin
2004-09-05 5:50 ` Nick Piggin
2004-09-05 6:13 ` [RFC][PATCH 1/3] account free buddy areas Nick Piggin
2004-09-05 6:13 ` Nick Piggin
2004-09-05 6:02 ` [RFC][PATCH 0/3] beat kswapd with the proverbial clue-bat David S. Miller
2004-09-05 6:02 ` David S. Miller
2004-09-05 6:16 ` Nick Piggin
2004-09-05 6:16 ` Nick Piggin
2004-09-05 10:13 ` Nick Piggin
2004-09-05 10:13 ` Nick Piggin
2004-09-05 17:24 ` Linus Torvalds
2004-09-05 17:24 ` Linus Torvalds
2004-09-05 17:36 ` Martin J. Bligh
2004-09-05 17:36 ` Martin J. Bligh
2004-09-05 17:37 ` Arjan van de Ven
2004-09-05 17:58 ` Linus Torvalds
2004-09-05 17:58 ` Linus Torvalds
2004-09-05 18:41 ` Arjan van de Ven
2004-09-06 1:35 ` Nick Piggin
2004-09-06 1:35 ` Nick Piggin
2004-09-15 13:27 ` Jörn Engel
2004-09-15 13:27 ` Jörn Engel
2004-09-15 13:29 ` Arjan van de Ven
2004-09-15 13:34 ` Jörn Engel
2004-09-15 13:34 ` Jörn Engel
2004-09-15 13:39 ` Arjan van de Ven
2004-09-15 14:18 ` Jörn Engel
2004-09-15 14:18 ` Jörn Engel
2004-09-06 1:09 ` Nick Piggin
2004-09-06 1:09 ` Nick Piggin
2004-09-05 6:09 ` Andrew Morton
2004-09-05 6:09 ` Andrew Morton
2004-09-05 6:26 ` Nick Piggin
2004-09-05 6:26 ` Nick Piggin
2004-09-05 6:27 ` Anton Blanchard
2004-09-05 6:27 ` Anton Blanchard
2004-09-05 10:09 ` Nick Piggin
2004-09-05 10:09 ` Nick Piggin
2004-09-06 3:33 ` David S. Miller
2004-09-06 3:33 ` David S. Miller
2004-09-06 8:55 ` Nick Piggin [this message]
2004-09-06 8:55 ` Nick Piggin
2004-09-05 16:49 ` Linus Torvalds
2004-09-05 16:49 ` Linus Torvalds
2004-09-06 0:54 ` Nick Piggin
2004-09-06 0:54 ` Nick Piggin
2004-09-06 1:49 ` Nick Piggin
2004-09-06 1:49 ` Nick Piggin
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=413C25F4.8030501@yahoo.com.au \
--to=nickpiggin@yahoo.com.au \
--cc=akpm@osdl.org \
--cc=anton@samba.org \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=torvalds@osdl.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.