Date: Sun, 13 Aug 2000 09:05:40 -0700 From: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca> To: sos@FreeBSD.org Cc: Cy.Schubert@uumail.gov.bc.ca, freebsd-bugs@FreeBSD.org Subject: Re: kern/20573: ATA MFC Breaks -STABL Message-ID: <200008131606.e7DG6K900795@cwsys.cwsent.com> In-Reply-To: Your message of "Sun, 13 Aug 2000 04:32:56 PDT." <200008131132.EAA20386@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Based on my testing done so far, the previous behaviour has been
restored. The bug has been fixed. The messages produced are:
Aug 13 08:41:16 cwsys /kernel: ata1-master: timeout waiting to give
command=c8 s=d0 e=00
Aug 13 08:41:16 cwsys /kernel: ad2: error executing command - resetting
Aug 13 08:41:16 cwsys /kernel: ata1: resetting devices .. done
I'm not exactly sure whether the original cause of the problem, the
timeout itself, is a FreeBSD bug (PR 17592) or a drive problem. This
drive has suffered timeouts under FreeBSD using DMA mode ever since it
was new about 5 years ago, yet the Western Digital diagnostics see no
problem, nor does PIO mode have any problem.
This specific problem has been fixed and this PR can be closed. I've
updated PR 17592 to document that my Western Digital drive has the same
timeouts.
Regards, Phone: (250)387-8437
Cy Schubert Fax: (250)387-5766
Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca
Open Systems Group, ITSD, ISTA
Province of BC
In message <200008131132.EAA20386@freefall.freebsd.org>,
sos@FreeBSD.org writes
:
> Synopsis: ATA MFC Breaks -STABL
>
> State-Changed-From-To: open->analyzed
> State-Changed-By: sos
> State-Changed-When: Sun Aug 13 04:29:57 PDT 2000
> State-Changed-Why:
>
> I'm the ATA guy :)
>
> Please try the following patch it should help:
>
> Index: ata-disk.c
> ===================================================================
> RCS file: /home/ncvs/src/sys/dev/ata/ata-disk.c,v
> retrieving revision 1.73
> diff -u -r1.73 ata-disk.c
> --- ata-disk.c 2000/08/06 18:01:47 1.73
> +++ ata-disk.c 2000/08/13 11:25:21
> @@ -391,7 +391,7 @@
>
> if (ata_command(adp->controller, adp->unit, cmd,
> cylinder, head, sector, count, 0, ATA_IMMEDIATE)) {
> - printf("ad%d: error executing command\n", adp->lun);
> + printf("ad%d: error executing command", adp->lun);
> goto transfer_failed;
> }
>
> @@ -429,12 +429,21 @@
>
> transfer_failed:
> untimeout((timeout_t *)ad_timeout, request, request->timeout_handle);
> - request->bp->bio_error = EIO;
> - request->bp->bio_flags |= BIO_ERROR;
> - request->bp->bio_resid = request->bytecount;
> - devstat_end_transaction_bio(&adp->stats, request->bp);
> - biodone(request->bp);
> - free(request, M_AD);
> + printf(" - resetting\n");
> +
> + /* if retries still permit, reinject this request */
> + if (request->retries++ < AD_MAX_RETRIES)
> + TAILQ_INSERT_HEAD(&adp->controller->ata_queue, request, chain);
> + else {
> + /* retries all used up, return error */
> + request->bp->bio_error = EIO;
> + request->bp->bio_flags |= BIO_ERROR;
> + request->bp->bio_resid = request->bytecount;
> + devstat_end_transaction_bio(&adp->stats, request->bp);
> + biodone(request->bp);
> + free(request, M_AD);
> + }
> + ata_reinit(adp->controller);
> }
>
> int32_t
>
>
>
> Responsible-Changed-From-To: freebsd-bugs->sos
> Responsible-Changed-By: sos
> Responsible-Changed-When: Sun Aug 13 04:29:57 PDT 2000
> Responsible-Changed-Why:
>
> http://www.freebsd.org/cgi/query-pr.cgi?pr=20573
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200008131606.e7DG6K900795>