From: Jens Axboe <axboe@suse.de>
To: Zlatko Calusic <zlatko.calusic@iskon.hr>
Cc: Linus Torvalds <torvalds@transmeta.com>,
Marcelo Tosatti <marcelo@conectiva.com.br>,
linux-mm@kvack.org, lkml <linux-kernel@vger.kernel.org>
Subject: Re: xmm2 - monitor Linux MM active/inactive lists graphically
Date: Fri, 26 Oct 2001 16:39:58 +0200 [thread overview]
Message-ID: <20011026163958.C3324@suse.de> (raw)
In-Reply-To: <Pine.LNX.4.31.0110250920270.2184-100000@cesium.transmeta.com> <dnr8rqu30x.fsf@magla.zg.iskon.hr>
In-Reply-To: <dnr8rqu30x.fsf@magla.zg.iskon.hr>
On Fri, Oct 26 2001, Zlatko Calusic wrote:
> Linus Torvalds <torvalds@transmeta.com> writes:
>
> > On 25 Oct 2001, Zlatko Calusic wrote:
> > >
> > > Yes, I definitely have DMA turned ON. All parameters are OK. :)
> >
> > I suspect it may just be that "queue_nr_requests"/"batch_count" is
> > different in -ac: what happens if you tweak them to the same values?
> >
>
> Next test:
>
> block: 1024 slots per queue, batch=341
That's way too much, batch should just stay around 32, that is fine.
> Still very spiky, and during the write disk is uncapable of doing any
> reads. IOW, no serious application can be started before writing has
> finished. Shouldn't we favour reads over writes? Or is it just that
> the elevator is not doing its job right, so reads suffer?
You are probably just seeing starvation due to the very long queues.
--
Jens Axboe
WARNING: multiple messages have this Message-ID (diff)
From: Jens Axboe <axboe@suse.de>
To: Zlatko Calusic <zlatko.calusic@iskon.hr>
Cc: Linus Torvalds <torvalds@transmeta.com>,
Marcelo Tosatti <marcelo@conectiva.com.br>,
linux-mm@kvack.org, lkml <linux-kernel@vger.kernel.org>
Subject: Re: xmm2 - monitor Linux MM active/inactive lists graphically
Date: Fri, 26 Oct 2001 16:39:58 +0200 [thread overview]
Message-ID: <20011026163958.C3324@suse.de> (raw)
In-Reply-To: <dnr8rqu30x.fsf@magla.zg.iskon.hr>
On Fri, Oct 26 2001, Zlatko Calusic wrote:
> Linus Torvalds <torvalds@transmeta.com> writes:
>
> > On 25 Oct 2001, Zlatko Calusic wrote:
> > >
> > > Yes, I definitely have DMA turned ON. All parameters are OK. :)
> >
> > I suspect it may just be that "queue_nr_requests"/"batch_count" is
> > different in -ac: what happens if you tweak them to the same values?
> >
>
> Next test:
>
> block: 1024 slots per queue, batch=341
That's way too much, batch should just stay around 32, that is fine.
> Still very spiky, and during the write disk is uncapable of doing any
> reads. IOW, no serious application can be started before writing has
> finished. Shouldn't we favour reads over writes? Or is it just that
> the elevator is not doing its job right, so reads suffer?
You are probably just seeing starvation due to the very long queues.
--
Jens Axboe
--
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/
next prev parent reply other threads:[~2001-10-26 14:39 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-10-24 10:42 xmm2 - monitor Linux MM active/inactive lists graphically Zlatko Calusic
2001-10-24 10:42 ` Zlatko Calusic
2001-10-24 14:26 ` Marcelo Tosatti
2001-10-24 14:26 ` Marcelo Tosatti
2001-10-25 0:25 ` Zlatko Calusic
2001-10-25 0:25 ` Zlatko Calusic
2001-10-25 1:50 ` Simon Kirby
2001-10-25 4:19 ` Linus Torvalds
2001-10-25 4:19 ` Linus Torvalds
2001-10-25 4:57 ` Linus Torvalds
2001-10-25 4:57 ` Linus Torvalds
2001-10-25 12:48 ` Zlatko Calusic
2001-10-25 12:48 ` Zlatko Calusic
2001-10-25 16:31 ` Linus Torvalds
2001-10-25 16:31 ` Linus Torvalds
2001-10-25 17:33 ` Jens Axboe
2001-10-25 17:33 ` Jens Axboe
2001-10-26 9:45 ` Zlatko Calusic
2001-10-26 9:45 ` Zlatko Calusic
2001-10-26 10:08 ` Zlatko Calusic
2001-10-26 10:08 ` Zlatko Calusic
2001-10-26 14:39 ` Jens Axboe [this message]
2001-10-26 14:39 ` Jens Axboe
2001-10-26 14:57 ` Zlatko Calusic
2001-10-26 14:57 ` Zlatko Calusic
2001-10-26 15:01 ` Jens Axboe
2001-10-26 15:01 ` Jens Axboe
2001-10-26 16:04 ` Linus Torvalds
2001-10-26 16:04 ` Linus Torvalds
2001-10-26 16:57 ` Linus Torvalds
2001-10-26 16:57 ` Linus Torvalds
2001-10-26 17:19 ` Linus Torvalds
2001-10-26 17:19 ` Linus Torvalds
2001-10-28 17:30 ` Zlatko Calusic
2001-10-28 17:30 ` Zlatko Calusic
2001-10-28 17:34 ` Linus Torvalds
2001-10-28 17:34 ` Linus Torvalds
2001-10-28 17:48 ` Alan Cox
2001-10-28 17:48 ` Alan Cox
2001-10-28 17:59 ` Linus Torvalds
2001-10-28 17:59 ` Linus Torvalds
2001-10-28 18:22 ` Alan Cox
2001-10-28 18:22 ` Alan Cox
2001-10-28 18:46 ` Linus Torvalds
2001-10-28 18:46 ` Linus Torvalds
2001-10-28 19:29 ` Alan Cox
2001-10-28 19:29 ` Alan Cox
2001-10-28 22:04 ` Linus Torvalds
2001-10-28 18:56 ` Andrew Morton
2001-10-28 18:56 ` Andrew Morton
2001-10-30 8:56 ` Jens Axboe
2001-10-30 8:56 ` Jens Axboe
2001-10-30 9:26 ` Zlatko Calusic
2001-10-30 19:07 ` Josh McKinney
2001-10-28 19:13 ` Barry K. Nathan
2001-10-28 19:13 ` Barry K. Nathan
2001-10-28 21:42 ` Jonathan Morton
2001-10-28 21:42 ` Jonathan Morton
2001-11-02 5:52 ` Zlatko's I/O slowdown status Andrea Arcangeli
2001-11-02 5:52 ` Andrea Arcangeli
2001-11-02 20:14 ` Zlatko Calusic
2001-11-02 20:14 ` Zlatko Calusic
2001-11-02 20:16 ` Jeffrey W. Baker
2001-11-02 20:36 ` John Alvord
2001-11-02 21:16 ` Zlatko Calusic
2001-11-02 20:26 ` Rik van Riel
2001-11-02 21:22 ` Zlatko Calusic
2001-11-02 20:57 ` Andrea Arcangeli
2001-11-02 20:57 ` Andrea Arcangeli
2001-11-02 23:23 ` Simon Kirby
2001-11-02 23:23 ` Simon Kirby
2001-11-02 23:37 ` Miquel van Smoorenburg
2001-10-27 13:14 ` xmm2 - monitor Linux MM active/inactive lists graphically Giuliano Pochini
2001-10-27 13:14 ` Giuliano Pochini
2001-10-28 5:05 ` Mike Fedyk
2001-10-28 5:05 ` Mike Fedyk
2001-10-25 9:07 ` Zlatko Calusic
2001-10-25 9:07 ` Zlatko Calusic
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=20011026163958.C3324@suse.de \
--to=axboe@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=marcelo@conectiva.com.br \
--cc=torvalds@transmeta.com \
--cc=zlatko.calusic@iskon.hr \
/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.