linux-kernel - Re: "modularized" 2.4.34.4 -> ide-core "unresolved symbols"

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: <20070508223922.GB31591@1wt.eu>
Date:	Wed, 9 May 2007 00:39:22 +0200
From:	Willy Tarreau <w@....eu>
To:	Zbigniew Baniewski <zb@...id.com.pl>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: "modularized" 2.4.34.4  ->  ide-core "unresolved symbols"

On Wed, May 09, 2007 at 12:14:25AM +0200, Zbigniew Baniewski wrote:
> On Tue, May 08, 2007 at 11:16:46PM +0200, Willy Tarreau wrote:
> 
> > depmod:         init_cmd640_vlb
> 
> There wasn't such message.

No, because it seems that your debian scripts hide it (or log it somewhere
else). But if you do "make modules_install" from the kernel tree (as I did),
you get it.

> I made another kernel (config below) - unfortunately, this time it gives up
> completely, saying:
> 
> -------------------------------
> ...ordinary boot messages, ending with "hda:... sectors... w/Cache..."
> /dev/ide/host0/bus0/target0/lun0:<1>Unable to handle kernel paging request at virtual address ffffff00
> printing eip:
> 
> ...several verses of sanscrit-like language...

This "sanscrit-like" language is an oops (in fact a kernel panic, but the
dump format is called an oops). It is what developers use to find bugs.

I see that you're using devfs. It is unfortunate for you that you're using
all the parts which are not reputed for their reliability :-/ I'm pretty
sure that initialization order matters here but you cannot interact much
with it, except by forcing some drivers to "=y".

Are you aware of any kernel version on which this config worked ? I'd
like to help you get the minimal working config, ie: the one with the
lower number of hard-linked drivers.

> <0>Kernel panic: Attempted to kill the idle task!
> In idle task - not syncing.
> --------------------------------
> 
> Moved the disk to quite different machine - with the same result. So, I'm
> afraid, it's not hardware-related problem.

Do not worry about the hardware here. The problem lies in the
initialization path of some sensible drivers. Did you enable DEVFS for a
particular reason, or just because you enabled almost everything ? If the
later, then you should retry with it disabled. It's absolutely not certain
that it will work, but it may.

If you're pretty certain that you need DEVFS, then for a test, please
set a max of IDE options to YES instead of MODULE. At least all the ones
you need to boot on this system.

As I said, it's a tricky area, and it's for this reason that distros
enable IDE to YES as well as some other entries I don't have in mind.

Regards,
Willy

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/