From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steve Lord Subject: Re: barriers vs. reads Date: Thu, 24 Jun 2004 11:39:11 -0500 Sender: linux-fsdevel-owner@vger.kernel.org Message-ID: <40DB03AF.7090706@xfs.org> References: <20040623214845.A21586@almesberger.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: linux-fsdevel@vger.kernel.org Return-path: Received: from [63.81.117.10] ([63.81.117.10]:34367 "EHLO mail00hq.adic.com") by vger.kernel.org with ESMTP id S266153AbUFXQkN (ORCPT ); Thu, 24 Jun 2004 12:40:13 -0400 To: Werner Almesberger In-Reply-To: <20040623214845.A21586@almesberger.net> List-Id: linux-fsdevel.vger.kernel.org Werner Almesberger wrote: > BTW, regarding overlapping requests, I wonder if there's a data > structure that gives O(log requests) or such lookups for ranges. > The best I could spontaneously think of would be > O(new_request_size*log(requests*avg_request_size)) > which isn't pretty. > > BTW2, is O_DIRECT actually a Linux-only thing, or is there some > ancestor whose semantics we may want to preserve ? I've had a > quick look at POSIX, but they don't seem to have direct IO. Irix has O_DIRECT, Solaris has something too, but it is not in the posix specs. Cray Unicos is the oldest implementation I came across. Irix explicitly lets multiple readers and writers into a file at once with O_DIRECT. The assumption being that the application which does this is doing its own coordination and will not shoot itself in the foot. Steve Lord