From mboxrd@z Thu Jan 1 00:00:00 1970 From: Josh Durgin Subject: Re: C++ api for librados 0.25 Date: Thu, 24 Feb 2011 10:38:33 -0800 Message-ID: <4D66A5A9.4050107@dreamhost.com> References: <4D669D48.6060202@dreamhost.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mail.hq.newdream.net ([66.33.206.127]:51140 "EHLO mail.hq.newdream.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755601Ab1BXSid (ORCPT ); Thu, 24 Feb 2011 13:38:33 -0500 In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Colin McCabe Cc: ceph-devel@vger.kernel.org On 02/24/2011 10:13 AM, Colin McCabe wrote: > On Thu, Feb 24, 2011 at 10:02 AM, Josh Durgin wrote: >> - aio_create_completion could be moved into Pool since it's only >> used within the context of a pool. (librbd would need this unless >> Pool had a get_rados method). > > Hmm. I'm not seeing the pool dependency for aio_create_completion. It > looks like this has nothing to do with any particular pool. It's just > a callback? My thinking was that the only reason for creating one is for doing aio, which is done on a pool. Looking at the implementation, however, there's no dependency on the RadosClient at all. These could just be constructors of AioCompletion (in the interface and implementation). -Josh Durgin