Return-Path: linux-nfs-owner@vger.kernel.org Received: from bombadil.infradead.org ([198.137.202.9]:38915 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753099Ab3JANTn (ORCPT ); Tue, 1 Oct 2013 09:19:43 -0400 Date: Tue, 1 Oct 2013 06:19:40 -0700 From: Christoph Hellwig To: Benny Halevy Cc: Christoph Hellwig , "J. Bruce Fields" , linux-nfs@vger.kernel.org Subject: Re: [PATCH RFC v0 05/49] pnfsd: introduce pnfsd header files Message-ID: <20131001131940.GA32431@infradead.org> References: <52447EA0.7070004@primarydata.com> <1380220810-12909-1-git-send-email-bhalevy@primarydata.com> <20130929114327.GB25750@infradead.org> <52481939.7060405@primarydata.com> <20130929121345.GA21083@infradead.org> <52481B11.2080407@primarydata.com> <20130929122130.GI21083@infradead.org> <20130929123553.GA7510@infradead.org> <5249977E.5060106@primarydata.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <5249977E.5060106@primarydata.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Mon, Sep 30, 2013 at 06:23:42PM +0300, Benny Halevy wrote: > This makes sense for blocks for its use of the generic block allocation and mapping > calls (and it needs a new call for committing uninitialized extents) > But for objects there are no such calls and the integration with exofs > is pretty intimate. That's just because there is no proper split between exofs and the pnfsd-objects layer. The split between the two doesn't seem too hard and would dramatically improve the interface. With that and moving the recall handling on truncate to generic code where it belongs almost nothin pnfs-specific will be left in exofs.