linux-kernel - Re: [PATCH] unicore32 fix: remove arch-specific futex support

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: <alpine.LFD.2.00.1103301446110.22418@localhost6.localdomain6>
Date:	Wed, 30 Mar 2011 14:50:29 +0200 (CEST)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Arnd Bergmann <arnd@...db.de>
cc:	Guan Xuetao <gxt@...c.pku.edu.cn>,
	Paul Mundt <lethal@...ux-sh.org>, linux-kernel@...r.kernel.org,
	linux-arch@...r.kernel.org
Subject: Re: [PATCH] unicore32 fix: remove arch-specific futex support

On Wed, 30 Mar 2011, Arnd Bergmann wrote:

> On Wednesday 30 March 2011, Thomas Gleixner wrote:
> > The problem is that you need cmpxchg in user space as well.
> > 
> > ARM does this with some syscall magic for machines which do not have
> > cmpxchg.
> 
> Ah, right. I wonder how sh does it.
> 
> I also remember that the original unicore32 ABI had support for a cmpxchg
> system call, but that was removed with the move to the generic syscall
> table.
> 
> It is of course possible to add it back using the __NR_arch_specific_syscall
> entry point.

Either that or you take a trap. None of them is pretty. 

If the only purpose is futex support, then we can add an extra flag to
the futex ops and handle it there via asm-generic/futex.h. That would
avoid an extra syscall in the contended case.

Thanks,

	tglx
--
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/