linux-kernel - RFC: remove or modify check_for_bios

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>] [day] [month] [year] [list]

Message-ID: <4C75AD28.8060101@zytor.com>
Date:	Wed, 25 Aug 2010 16:54:16 -0700
From:	"H. Peter Anvin" <hpa@...or.com>
To:	LKML <linux-kernel@...r.kernel.org>
CC:	Ingo Molnar <mingo@...e.hu>, Thomas Gleixner <tglx@...utronix.de>,
	Arjan van de Ven <arjan@...radead.org>,
	Jeremy Fitzhardinge <jeremy@...p.org>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: RFC: remove or modify check_for_bios_corruption()

check_for_bios_corruption() was added to discover how many systems have
BIOSes with problems corrupting low memory.  The answer really is very
simple... almost all.  In fact, I have seen claims that Windows 7
doesn't use any memory below the 1 MiB point at all for this reason.

As such, I have queued up a patch in -tip to reserve the low 64K for all
BIOSes, with an option (suggested by Andrew) to adjust the reservation
further.

Since checking for corruption makes the memory unavailable anyway and
since it is now established the memory is too unreliable to use, I am
suggesting removing the check code completely.

The other alternative is to have it optionally scan from the reserved
threshold to some higher point, and only kick in if that gap is nonzero.
 That way it doesn't continually run for most users, but it might be
useful to confirm corruption if we explicitly suspect it on some system.

Thoughts?

	-hpa
--
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/