linux-kernel - Re: HPA Resume patch

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 for Android: free password hash cracker in your pocket

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

Message-Id: <20060827141656.0e4d2c17.vsu@altlinux.ru>
Date:	Sun, 27 Aug 2006 14:16:56 +0400
From:	Sergey Vlasov <vsu@...linux.ru>
To:	Lee Trager <Lee@...turesInMotion.net>
Cc:	B.Zolnierkiewicz@...a.pw.edu.pl, linux-ide@...r.kernel.org,
	linux-kernel@...r.kernel.org, akpm@...l.org
Subject: Re: HPA Resume patch

On Sun, 27 Aug 2006 04:42:03 -0400 Lee Trager wrote:

> This patch fixes a problem with computers that have HPA on their hard
> drive and not being able to come out of resume from RAM or disk. I've
> tested this patch on 2.6.17.x and 2.6.18-rc4 and it works great on both
> of these. This patch also fixes the bug #6840. This is my first patch to
> the kernel and I was told to e-mail the above people to get my patch
> into the kernel. If I made a mistake please be gentle and correct me ;)

The patch adds a call from ide.c to a function inside ide-disk.c - this
won't work when IDE support is built as modules (it will cause a
circular dependency between ide-core and ide-disk modules).

The proper way to do such calls is to add a new method to ide_driver_t
and call it from generic_ide_resume().  Also, if the ide_do_drive_cmd()
call failed, it is probably unsafe to reset HPA, so you need to check
the result and call the resume method only if the low-level resume has
succeeded.

And please and "-p" to diff options.

Content of type "application/pgp-signature" skipped