To: Gregory Haskins <ghaskins@novell.com>
Cc: mingo@elte.hu, rostedt@goodmis.org, tglx@linutronix.de,
       linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org,
       gregory.haskins@gmail.com
Subject: Re: [PATCH] seqlock: serialize against writers
From: Andi Kleen <andi@firstfloor.org>
References: <20080829154237.1196.66825.stgit@dev.haskins.net>
Date: Fri, 29 Aug 2008 18:09:16 +0200
In-Reply-To: <20080829154237.1196.66825.stgit@dev.haskins.net> (Gregory Haskins's message of "Fri, 29 Aug 2008 11:44:58 -0400")
Message-ID: <87abevpzv7.fsf@basil.nowhere.org>
User-Agent: Gnus/5.1008 (Gnus v5.10.8) Emacs/21.3 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 882
Lines: 24

Gregory Haskins <ghaskins@novell.com> writes:

> *Patch submitted for inclusion in PREEMPT_RT 26-rt4.  Applies to 2.6.26.3-rt3*
>
> Hi Ingo, Steven, Thomas,
>   Please consider for -rt4.  This fixes a nasty deadlock on my systems under
>   heavy load.

Does this even work under x86-64? x86-64 uses seqlocks in user space
in its vsyscalls. And read_lock() definitely doesn't work there because 
it writes.

You would need at least to disable vsyscall gettimeofday(), making
it much much slower.

Perhaps you tested on one of the systems where the vsyscalls need
to fallback for other reasons? (e.g. one using pmtimer for timing).

-Andi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/