linux-kernel - Re: 2.6.29 regression: ATA bus errors on resume

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <49CAA4BC.1040102@gmail.com>
Date:	Wed, 25 Mar 2009 23:40:12 +0200
From:	Niel Lambrechts <niel.lambrechts@...il.com>
To:	Jeff Garzik <jeff@...zik.org>
CC:	"linux.kernel" <linux-kernel@...r.kernel.org>,
	Linux IDE mailing list <linux-ide@...r.kernel.org>,
	Arjan van de Ven <arjan@...radead.org>,
	"Rafael J. Wysocki" <rjw@...k.pl>,
	linux-scsi <linux-scsi@...r.kernel.org>
Subject: Re: 2.6.29 regression: ATA bus errors on resume

On 03/25/2009 08:06 AM, Jeff Garzik wrote:
>
> So from this, a few observations and open questions:
>
> Why didn't these SCSI commands get retried?
>
> Were they left over from prior to resume?
>
> Did SCSI error out the commands too soon?  The probing is not complete
> until AFTER the sector errors, I note.
>
> Did the system resume before ACPI resume and SCSI resume completed? 
> It sure looks that way, from the log.
>
> Does the asynchronous probing play a role here?
It looks like there are a few different scenarios in what could happen:

I had another go at hibernating on 2.6.29 running with and without X,
and this time round it resumed with a *clean* file-system. Without X, it
looked to be better off as there were not even any ATA errors logged,
but I'm not sure if this is just because of the system being almost
entirely idle.

Going on your comments of earlier, when hibernating in X it looks like
this time the SCSI initialization completes normally before any errors, 
being 'ATA bus error' messages for a while until a hard reset occurs.

I've attached the messages from both, hopefully this gives someone more
of a clue of what is happening.

cheers
Niel

View attachment "hibernate-x.txt" of type "text/plain" (17374 bytes)

View attachment "hibernate-console.txt" of type "text/plain" (12741 bytes)