From: Mike Anderson <andmike@us.ibm.com>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: dougg@torque.net, Mark Haverkamp <markh@osdl.org>,
Christoffer Hall-Frederiksen <hall@jiffies.dk>,
linux-scsi@vger.kernel.org,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux aacraid devel <linux-aacraid-devel@dell.com>
Subject: Re: Problem with aacraid driver in 2.5.63-bk-latest
Date: Wed, 12 Mar 2003 16:50:46 -0800 [thread overview]
Message-ID: <20030313005046.GB14373@beaverton.ibm.com> (raw)
In-Reply-To: <1047517604.23902.39.camel@irongate.swansea.linux.org.uk>
Alan Cox [alan@lxorguk.ukuu.org.uk] wrote:
> On Wed, 2003-03-12 at 23:55, Douglas Gilbert wrote:
> > /*
> > * Limit max queue depth on a single lun to 256 for now. Remember,
> > * we allocate a struct scsi_command for each of these and keep it
> > * around forever. Too deep of a depth just wastes memory.
> > */
> > if(tags > 256)
> > return;
> > ....
>
> I can see the memory consideration. However the thing can really handle big
> queues well. Possibly we should be setting the queue to 512 / somefunction(volumes)
> though to avoid the worst case overcommit here
I agree with Doug that the previous comment is out of date and we could
raise the value, but we also should never leave the function with the
possibility that the queue_depth is 0.
The patch below is something Patrick and I where discussing though I
believe he indicated that I should print out the value we where setting
the queue_depth to. It was only compiled and not tested on any devices.
-andmike
--
Michael Anderson
andmike@us.ibm.com
scsi.c | 32 +++++++++++++++++++++++---------
1 files changed, 23 insertions(+), 9 deletions(-)
------
--- 1.96/drivers/scsi/scsi.c Fri Feb 21 13:46:58 2003
+++ edited/drivers/scsi/scsi.c Wed Mar 12 16:05:42 2003
@@ -926,15 +926,28 @@
/*
* refuse to set tagged depth to an unworkable size
*/
- if(tags <= 0)
- return;
+ if(tags <= 0) {
+ printk(KERN_WARNING "(scsi%d:%d:%d:%d) "
+ "%s, tag value to small\n"
+ "disabled\n", SDpnt->host->host_no,
+ SDpnt->channel, SDpnt->id, SDpnt->lun,
+ __FUNCTION__);
+
+ SDpnt->queue_depth = 1;
+ }
/*
- * Limit max queue depth on a single lun to 256 for now. Remember,
- * we allocate a struct scsi_command for each of these and keep it
- * around forever. Too deep of a depth just wastes memory.
+ * Limit max queue depth on a single lun to 256 for now.
+ * Too deep of a depth just wastes memory.
*/
- if(tags > 256)
- return;
+ if(tags > 256) {
+ printk(KERN_WARNING "(scsi%d:%d:%d:%d) "
+ "%s, tag value to big\n"
+ "disabled\n", SDpnt->host->host_no,
+ SDpnt->channel, SDpnt->id, SDpnt->lun,
+ __FUNCTION__);
+
+ SDpnt->queue_depth = 256;
+ }
spin_lock_irqsave(&device_request_lock, flags);
SDpnt->queue_depth = tags;
@@ -949,9 +962,10 @@
break;
default:
printk(KERN_WARNING "(scsi%d:%d:%d:%d) "
- "scsi_adjust_queue_depth, bad queue type, "
+ "%s, bad queue type, "
"disabled\n", SDpnt->host->host_no,
- SDpnt->channel, SDpnt->id, SDpnt->lun);
+ SDpnt->channel, SDpnt->id, SDpnt->lun,
+ __FUNCTION__);
case 0:
SDpnt->ordered_tags = SDpnt->simple_tags = 0;
SDpnt->queue_depth = tags;
next prev parent reply other threads:[~2003-03-13 0:39 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20030228133037.GB7473@jiffies.dk>
2003-03-12 23:06 ` Mark Haverkamp
2003-03-13 0:18 ` Alan Cox
2003-03-12 23:55 ` Douglas Gilbert
2003-03-13 1:06 ` Alan Cox
2003-03-13 0:50 ` Mike Anderson [this message]
2003-03-13 23:13 ` Mark Haverkamp
2003-03-14 7:05 ` Denis Vlasenko
2003-03-14 10:18 ` Mike Anderson
2003-03-13 1:31 ` Douglas Gilbert
2003-03-13 23:27 ` Doug Ledford
2003-03-13 15:42 ` Mark Haverkamp
2003-03-13 23:17 ` Doug Ledford
2003-03-13 23:22 ` Mark Haverkamp
2003-03-13 23:25 ` Doug Ledford
2003-03-14 14:57 ` Alan Cox
2003-03-14 15:34 ` Mark Haverkamp
2003-03-14 16:48 ` Alan Cox
2003-03-03 10:05 Christoffer Hall-Frederiksen
2003-03-03 13:40 ` Alan Cox
2003-03-03 14:46 ` Christoffer Hall-Frederiksen
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=20030313005046.GB14373@beaverton.ibm.com \
--to=andmike@us.ibm.com \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=dougg@torque.net \
--cc=hall@jiffies.dk \
--cc=linux-aacraid-devel@dell.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=markh@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 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®