mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Patrick Mansfield <patmans@us.ibm.com>
To: Adam Kropelin <akropel1@rochester.rr.com>
Cc: Jens Axboe <axboe@suse.de>, linux-kernel@vger.kernel.org
Subject: Re: 2.5.46: ide-cd cdrecord (almost) success report
Date: Wed, 6 Nov 2002 15:52:35 -0800	[thread overview]
Message-ID: <20021106155235.A17479@eng2.beaverton.ibm.com> (raw)
In-Reply-To: <20021106233325.GA29940@www.kroptech.com>; from akropel1@rochester.rr.com on Wed, Nov 06, 2002 at 06:33:25PM -0500

On Wed, Nov 06, 2002 at 06:33:25PM -0500, Adam Kropelin wrote:
> On Wed, Nov 06, 2002 at 10:11:44AM -0800, Patrick Mansfield wrote:

> > What queue depth is the AIC setting?
> > 
> > SCSI in 2.5.x no longer copies the request, so if you have a queue
> > depth larger than the allocated requests there might not be
> > any free requests left for the blk layer to play with.
> > 
> > AIC default queue depth is 253 (with 2.5.46 queue depth can be set to 1
> 
> Are you talking tcq depth here? Best as I can tell, 2.5.46 defaults to
> 16. Lowering it to 2 doesn't seem to help.

Yes, but not the IDE code (I have know idea what IDE defaults to). If you
have sg its proc interface can dump them, but AIC should default to 253,
on my system with AIC adapter (not the aic7xxx_old adapter, it looks like
it defaults to 32) I typically see:

cat /proc/scsi/sg/device_hdr /proc/scsi/sg/devices
host    chan    id      lun     type    opens   qdepth  busy    online
0       0       0       0       0       2       253     0       1
0       0       1       0       0       1       253     0       1
0       0       15      0       3       0       2       0       1

The last device does not support tcq (id 15), so it gets a depth of 2.

The AIC driver also prints the depth on init/boot like so:

	scsi0:A:0:0: Tagged Queuing enabled.  Depth 253

My .config has (default cmd per device):

CONFIG_SCSI_AIC7XXX=y
CONFIG_AIC7XXX_CMDS_PER_DEVICE=253
CONFIG_AIC7XXX_RESET_DELAY_MS=15000

-- Patrick Mansfield

  reply	other threads:[~2002-11-06 23:46 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-11-06  4:13 Adam Kropelin
2002-11-06  7:22 ` Jens Axboe
2002-11-06  7:44   ` Jens Axboe
2002-11-06 23:25     ` Adam Kropelin
2002-11-07 14:33       ` Jens Axboe
2002-11-06 15:56   ` Adam Kropelin
2002-11-06 18:11     ` Patrick Mansfield
2002-11-06 23:33       ` Adam Kropelin
2002-11-06 23:52         ` Patrick Mansfield [this message]
2002-11-07  0:05           ` Adam Kropelin
2002-11-06 18:45     ` Jens Axboe
2002-11-06 23:23       ` Adam Kropelin

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=20021106155235.A17479@eng2.beaverton.ibm.com \
    --to=patmans@us.ibm.com \
    --cc=akropel1@rochester.rr.com \
    --cc=axboe@suse.de \
    --cc=linux-kernel@vger.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

all inboxes | Powered by JetHome®