NetBSD Problem Report #52415

From www@NetBSD.org  Mon Jul 17 13:23:43 2017
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id AA7267A181
	for <gnats-bugs@gnats.NetBSD.org>; Mon, 17 Jul 2017 13:23:43 +0000 (UTC)
Message-Id: <20170717132333.6FB4F7A20E@mollari.NetBSD.org>
Date: Mon, 17 Jul 2017 13:23:33 +0000 (UTC)
From: charlie@amigaspirit.hu
Reply-To: charlie@amigaspirit.hu
To: gnats-bugs@NetBSD.org
Subject: lost interrupt and device timeout messages in dmesg
X-Send-Pr-Version: www-1.0

>Number:         52415
>Category:       port-amiga
>Synopsis:       lost interrupt and device timeout messages in dmesg
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    jdolecek
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Jul 17 13:25:00 +0000 2017
>Closed-Date:    Mon Oct 22 21:45:58 +0000 2018
>Last-Modified:  Mon Oct 22 21:45:58 +0000 2018
>Originator:     Karoly Balogh
>Release:        7.1
>Organization:
amigaspirit.hu
>Environment:
NetBSD bsdmiga 7.1 NetBSD 7.1 (GENERIC.201703111743Z) amiga

Boot dmesg: http://dmesgd.nycbug.org/index.cgi?do=view&id=3222
>Description:
I have NetBSD 7.1 installed on my Amiga 1200 with a Blizzard 1260 CPU card. Stock GENERIC kernel.

Under heavy I/O load (example, compiling from pkgsrc) lines like this appear in dmesg:

wdc0:0:0: lost interrupt
        type: ata tc_bcount: 512 tc_skip: 15872
wd0a: device timeout writing fsbn 3748895 of 3748864-3748895 (wd0 bn 6370955; cn 25281 tn 2 sn 17), retrying
wd0: soft error (corrected)
wdc0:0:0: lost interrupt
        type: ata tc_bcount: 512 tc_skip: 5632
wd0a: device timeout writing fsbn 890667 of 890656-890667 (wd0 bn 3512727; cn 13939 tn 1 sn 36), retrying
wd0: soft error (corrected)
wdc0:0:0: lost interrupt
        type: ata tc_bcount: 512 tc_skip: 5632
wd0a: device timeout writing fsbn 897867 of 897856-897867 (wd0 bn 3519927; cn 13967 tn 3 sn 54), retrying
wd0: soft error (corrected)
wdc0:0:0: lost interrupt
        type: ata tc_bcount: 512 tc_skip: 15872
wd0a: device timeout writing fsbn 1128671 of 1128640-1128671 (wd0 bn 3750731; cn 14883 tn 3 sn 26), retrying
wd0: soft error (corrected)

The disk in question is the single disk connected to the internal IDE port of the A1200, it's an SD card on an SD2IDE44pin adapter, with a 8GB class 10 card. Using a different SD card doesn't fix the problem, the hardware doesn't report/experiences any problems with AmigaOS 3.1 at least. The Blizzard 1260 is not overclocked, runs at 50Mhz.

The machine is very stable otherwise, runs with week+ uptimes.
>How-To-Repeat:
I can provide remote SSH access to the box if necessary for debugging.
>Fix:

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: port-amiga-maintainer->jdolecek
Responsible-Changed-By: jdolecek@NetBSD.org
Responsible-Changed-When: Mon, 17 Jul 2017 19:10:58 +0000
Responsible-Changed-Why:
I see these under QEMU sometimes, and have some error handling changes
on jdolecek-ncq branch in works which might address this. It's not amiga
specific. Stay tuned.


State-Changed-From-To: open->feedback
State-Changed-By: jdolecek@NetBSD.org
State-Changed-When: Sun, 22 Oct 2017 13:29:02 +0000
State-Changed-Why:
Can you please try -current sources? I've seen similar errors on QEMU
for i386 host, and they don't happen after the NCQ branch was merged,
perhaps it would help your amiga too.


State-Changed-From-To: feedback->closed
State-Changed-By: jdolecek@NetBSD.org
State-Changed-When: Mon, 22 Oct 2018 21:45:58 +0000
State-Changed-Why:
No feedback provided in year, closing.


>Unformatted:

Home	PR Database Search