linux-kernel - Re: [ RT PATCH] seqlock: serialize against writers

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:	Fri, 29 Aug 2008 13:02:58 -0400 (EDT)
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Stephen Hemminger <shemminger@...tta.com>
cc:	Gregory Haskins <ghaskins@...ell.com>, mingo@...e.hu,
	tglx@...utronix.de, linux-kernel@...r.kernel.org,
	linux-rt-users@...r.kernel.org, gregory.haskins@...il.com
Subject: Re: [ RT PATCH] seqlock: serialize against writers


/me adds "RT" to subject.

On Fri, 29 Aug 2008, Stephen Hemminger wrote:
> 
> I have mixed feelings about this. The point of this primitive was for
> cases where write contention was rare and writers only did small updates.
> So what you did was fix the primitive for cases where it is being misused.
> Do you have a real example where this is a problem? If so then the
> user of seqlock should be fixed, rather than fixing seqlock.
> 


I think there's some confusion here, because of the changes to raw. The 
raw_seqlock_t and mainline seqlock_t has no issue, and should not be 
touched.

What Gregory is solving is an issue in -rt where we let write seqlocks be 
preempted.  Thus this causes issues with determinism. This is a special 
case for -rt, and -rt only. It should not affect mainline in anyway.

Sorry for the noise.

-- Steve

--
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/