Date: Mon, 13 Apr 2015 12:38:16 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: George Spelvin <linux@horizon.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Two other ways to do latched seqcounts
Message-ID: <20150413103816.GE5029@twins.programming.kicks-ass.net>
References: <20150413073510.GC5029@twins.programming.kicks-ass.net>
 <20150413092605.2469.qmail@ns.horizon.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150413092605.2469.qmail@ns.horizon.com>
User-Agent: Mutt/1.5.21 (2012-12-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 874
Lines: 19

On Mon, Apr 13, 2015 at 05:26:05AM -0400, George Spelvin wrote:
> > I'm assuming you're writing to me because of the latched rb-tree;
> > because that's the most recent related thing I posted ;-)
> 
> Basically yes, although it was the documentation you added to the
> latched seqlock code in particular.
> 
> I haven't checked the users of your rb-tree code to see how large and
> frequently read the trees are, but if a read is expensive, then avoiding
> retries by incrementing the seqlock twice per update starts to become
> interesting.

Right, so I use it for modules, and updates are near non existent
under normal usage.
--
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/