linux-kernel - Re: [PATCH 0 of 2] x86: a couple of misc patches

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

Message-Id: <4787333F.76E4.0078.0@novell.com>
Date:	Fri, 11 Jan 2008 08:13:35 +0000
From:	"Jan Beulich" <jbeulich@...ell.com>
To:	"Jeremy Fitzhardinge" <jeremy@...p.org>
Cc:	"Ingo Molnar" <mingo@...e.hu>,
	"Glauber de Oliveira Costa" <glommer@...il.com>,
	"Andi Kleen" <ak@...e.de>, "LKML" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 0 of 2] x86: a couple of misc patches

>2. I noticed the _PAGE_PCD|_PAGE_PWT combination being used a lot, so
>I created _PAGE_NOCACHE to wrap them up.  asm-x86/fb.h uses plain
>_PAGE_PCD; should it be _PAGE_NOCACHE too?

Setting PCD but not PWT (or the other way around) is yielding not fully
defined behavior (model specific) as per the doc, so it doen't seem to be
too good an idea to use this. However, adding PWT here may not be what
is desirable either (as it'll make the referenced memory UC unconditionally,
whereas WC is generally more desirable here). It would seem to me,
though, that PWT alone would be the better choice in this situation
(despite this also yielding model specific behavior).

Jan

--
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/