From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Elder Subject: rbd: wrap up of initialization rework Date: Tue, 30 Oct 2012 16:11:01 -0500 Message-ID: <50904265.2050603@inktank.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mail-ie0-f174.google.com ([209.85.223.174]:38230 "EHLO mail-ie0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934561Ab2J3VLF (ORCPT ); Tue, 30 Oct 2012 17:11:05 -0400 Received: by mail-ie0-f174.google.com with SMTP id k13so1110383iea.19 for ; Tue, 30 Oct 2012 14:11:03 -0700 (PDT) Sender: ceph-devel-owner@vger.kernel.org List-ID: To: ceph-devel@vger.kernel.org These four patches sort of finish out this bunch of patches that have been reworking how rbd initialization of devices is done. With this foundation in place, probing for parents of layered device fits in fairly neatly. [PATCH 1/4] rbd: don't pass rbd_dev to rbd_get_client() This makes getting a reference to a client not depend on a particular rbd_dev. [PATCH 2/4] rbd: consolidate rbd_dev init in rbd_add() This just moves some code around to make the next patch simpler. [PATCH 3/4] rbd: define rbd_dev_{create,destroy}() helpers These helpers isolate some initialization and make cleanup in error paths easier. [PATCH 4/4] rbd: encapsulate last part of probe This moves a big hunk of code into its own function, leaving rbd_add() much shorter and easier to follow. -Alex