mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andre Hedrick <andre@aslab.com>
To: Mark Hahn <hahn@physics.mcmaster.ca>
Cc: "Christian Bornträger" <linux-kernel@borntraeger.net>,
	linux-kernel@vger.kernel.org,
	"Alan Cox" <alan@lxorguk.ukuu.org.uk>
Subject: Re: RFC (patch below) Re: ide drive problem?
Date: Sat, 29 Sep 2001 13:37:03 -0700 (PDT)	[thread overview]
Message-ID: <Pine.LNX.4.31.0109291333300.7545-100000@postbox.aslab.com> (raw)
In-Reply-To: <Pine.LNX.4.10.10109291627300.9176-100000@coffee.psychology.mcmaster.ca>


Mark, you have a valid point but if they see this level of noise the
auto-dma-downgrade feature set exclusive to Linux will ramp down the max.
performance from init to a reasonable threshold.  The object is to
stablize the transfer rates first and then maintain performance.

You know the song and dance my friend, if anyone does.  Very few could
keep me in check, but you have managed quite well and I am impressed!

Cheers,

Andre Hedrick
CTO ASL, Inc.
Linux ATA Development
-----------------------------------------------------------------------------
ASL, Inc.                                    Tel: (510) 857-0055 x103
38875 Cherry Street                          Fax: (510) 857-0010
Newark, CA 94560                             Web: www.aslab.com

On Sat, 29 Sep 2001, Mark Hahn wrote:

> > This is an error that I am considering removing form the user's view.
> > For the very fact/reason you are pointing out; however, it becomse
> > more painful when performing error sorting.
>
> yes, most of the time, the warning scares people unnecessarily.
> but if the machine is seeing lots of CRC failures, there should
> definitely be some prominent messages.  perhaps something simple
> like producing a warning if more than a few of recent IOs
> have had CRC problems:
>
> int crcState = 0;
>
> on successful IO:
> 	crcState >>= 1;
>
> on CRC failure:
> 	if (crcState)
> 		printk("dang, CRC failed on hda, see http://whatever");
> 	crcState = 1 << 10;
>
> so if >10 IOs succeed between CRC failures, there's no warning.
> (uh, I guess that would actually be 9, since presumably the retry
> would succeed...)  keeping a global count of CRC failures would be
> kind of nice, too.
>
> regards, mark hahn.
>


      reply	other threads:[~2001-09-29 20:53 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-09-28  4:02 Steven Joerger
2001-09-28  5:44 ` David Grant
2001-09-28  5:55   ` Steven Joerger
2001-09-28 11:52   ` clemens
2001-09-28 12:54     ` Vojtech Pavlik
2001-09-28 15:25       ` 2 GB file limitation Bobby Hitt
2001-09-28 16:21         ` David Lang
2001-09-29  0:18           ` Jeff Chua
2001-09-29 11:21             ` Nicholas Knight
2001-09-29 13:17               ` Jeff Chua
2001-09-29 13:51             ` Luigi Genoni
2001-09-30  8:23               ` Gábor Lénárt
2001-09-30  8:59                 ` safemode
     [not found]                 ` <20010930085955.76BE016EBB@vega.digitel2002.hu>
2001-09-30  9:21                   ` Gábor Lénárt
2001-09-30 15:59                     ` Luigi Genoni
2001-09-30 17:38                 ` GCC 2.95, 2.96 and 3.0 on linear algebra (was RE: 2 GB file limitation) M. Edward Borasky
2001-09-30 18:11                   ` Luigi Genoni
2001-09-30 20:35                     ` M. Edward Borasky
2001-09-28 13:09     ` ide drive problem? Gábor Lénárt
2001-09-28 14:05       ` Alan Cox
2001-09-28 15:48         ` ide drive problem? RFC Christian Bornträger
2001-09-29 15:21         ` RFC (patch below) Re: ide drive problem? Christian Bornträger
2001-09-29 20:16           ` Andre Hedrick
2001-09-29 20:33             ` Dave Jones
2001-09-29 20:26               ` Andre Hedrick
2001-09-29 21:15                 ` Anders Eriksson
2001-09-29 22:21                   ` Andre Hedrick
2001-09-30  6:01                     ` Anders Eriksson
2001-09-29 20:34             ` Nicholas Knight
2001-09-29 20:48             ` Mark Hahn
2001-09-29 20:37               ` Andre Hedrick [this message]

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=Pine.LNX.4.31.0109291333300.7545-100000@postbox.aslab.com \
    --to=andre@aslab.com \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=hahn@physics.mcmaster.ca \
    --cc=linux-kernel@borntraeger.net \
    --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®