From mboxrd@z Thu Jan 1 00:00:00 1970 From: Josh Durgin Subject: Re: [PATCH 0/4] libceph: kill the "trail" portion of message data Date: Mon, 11 Mar 2013 15:44:07 -0700 Message-ID: <513E5E37.9010503@inktank.com> References: <513CEE83.4040900@inktank.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-pb0-f52.google.com ([209.85.160.52]:59504 "EHLO mail-pb0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753506Ab3CKWot (ORCPT ); Mon, 11 Mar 2013 18:44:49 -0400 Received: by mail-pb0-f52.google.com with SMTP id ma3so4262409pbc.11 for ; Mon, 11 Mar 2013 15:44:48 -0700 (PDT) In-Reply-To: <513CEE83.4040900@inktank.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Alex Elder Cc: ceph-devel@vger.kernel.org On 03/10/2013 01:35 PM, Alex Elder wrote: > The trail portion of message data was added to support > two distinct sets of data for an osd request--one a > pagelist for providing parameters to object method > calls; and a second a page array for receiving data > back from the result of such a call. > > It's always been a bit of a weird thing bolted onto > a message though, and with the rework of the messenger > code it can now be removed. > > This series eliminates the trail by allowing the osd > client to record a (non-trail) pagelist for data, and > using the fact that we now distinguish incoming from > outgoing data to allow that to be specified distinct > from the page array used for the incoming response. > > Having done this, we can eliminate the trail from the > ceph message structure, and then that allows some > code to be simplified. > > These patches are available in the "review/wip-kill-trail" > branch of the ceph-client git repository. That branch > is based on branch "review/wip-cursor". > > -Alex > > [PATCH 1/4] libceph: have osd requests support pagelist data > [PATCH 2/4] libceph: kill osd request r_trail > [PATCH 3/4] libceph: kill message trail > [PATCH 4/4] libceph: more cleanup of write_partial_msg_pages() These look good. Reviewed-by: Josh Durgin