linux-kernel - Re: [PATCH 1/3] futex: do not pagefault_disable in futex_atomic_cmpxchg

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]

Date:	Mon, 14 Mar 2011 10:13:17 +0100
From:	Peter Zijlstra <peterz@...radead.org>
To:	Darren Hart <darren@...art.com>
Cc:	Linus Torvalds <torvalds@...ux-foundation.org>,
	Michel Lespinasse <walken@...gle.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...e.hu>, Matt Turner <mattst88@...il.com>,
	Russell King <linux@....linux.org.uk>,
	David Howells <dhowells@...hat.com>,
	Tony Luck <tony.luck@...el.com>,
	Michal Simek <monstr@...str.eu>,
	Ralf Baechle <ralf@...ux-mips.org>,
	"James E.J. Bottomley" <jejb@...isc-linux.org>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Martin Schwidefsky <schwidefsky@...ibm.com>,
	Paul Mundt <lethal@...ux-sh.org>,
	"David S. Miller" <davem@...emloft.net>,
	Chris Metcalf <cmetcalf@...era.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/3] futex: do not pagefault_disable in
 futex_atomic_cmpxchg_inatomic()

On Sun, 2011-03-13 at 17:55 -0700, Darren Hart wrote:
> I wondered about the preempt vs pagefault disable 

In current mainline they happen to be the same, but for -rt they're not.

Its also conceptually different, with preempt disable we dis-allow any
scheduling, which thus also precludes most fault handlers.

Disabling pagefault can be done without also disabling preemption (as is
done in -rt) and simply means that we'll never try to handle the fault
and fallback to the exception table or give up and segfault the kernel.

Its not dis-similar to the whole rcu_read_lock() vs preempt_disable()
stuff, for a while one implied the other.
--
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/