From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030855Ab2HHT6D (ORCPT ); Wed, 8 Aug 2012 15:58:03 -0400 Received: from mx1.redhat.com ([209.132.183.28]:29112 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1030829Ab2HHT5q (ORCPT ); Wed, 8 Aug 2012 15:57:46 -0400 Date: Wed, 8 Aug 2012 15:57:30 -0400 From: Vivek Goyal To: Herton Ronaldo Krzesinski Cc: Jiri Kosina , Andrew Morton , Jens Axboe , Tejun Heo , linux-kernel@vger.kernel.org, Ben Hutchings Subject: Re: [PATCH 2/2] floppy: error handling fixes on do_floppy_init Message-ID: <20120808195730.GG15623@redhat.com> References: <1344450293-5557-1-git-send-email-herton.krzesinski@canonical.com> <1344450293-5557-3-git-send-email-herton.krzesinski@canonical.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1344450293-5557-3-git-send-email-herton.krzesinski@canonical.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Aug 08, 2012 at 03:24:53PM -0300, Herton Ronaldo Krzesinski wrote: > While looking at commit 3f9a5aa ("floppy: Cleanup disk->queue before > caling put_disk() if add_disk() was never called") I noticed some > problems with the error handling and cleanup: > > * missing cleanup (put_disk) if blk_init_queue fails, dr is decremented > first in the error handling loop > * if something fails in the add_disk loop, there is no cleanup of > previous iterations in the error handling. > * "if (disks[dr]->queue)" check is bogus, when reaching there for each > dr should exist an queue allocated, and it doesn't take into account > iterations where add_disk wasn't done, if failure happens in add_disk > loop. > * floppy_module_exit doesn't reset queue pointer if add_disk wasn't > done. Hey, these seem to be multiple cleanups. Can you break these down into individual patches. Review becomes easy. [..] > + blk_cleanup_queue(disks[dr]->queue); > + /* > + * put_disk() may not be paired with add_disk() and > + * will put queue reference one extra time. fix it. > + */ > + if (dr > registered || !(allowed_drive_mask & (1 << dr)) || > + fdc_state[FDC(dr)].version == FDC_NONE) > disks[dr]->queue = NULL; I think checking for FDC_NONE and allowed_drive_mask() in multiple places is becoming unreadable now. Can we just maintain a separate array to keep track of disks on which we have called add_disk() and do the cleanup accordingly. static unsigned short disk_registered[N_DRIVE]; /* do add_disk */ disk_registered[drive] = 1; out_put_disk: while(dr--) { if (disks[dr]->queue && !disk_registered[dr]) { blk_cleanup_queue() disks[dr]->queue = NULL; } } Same disk_registered[] can be used for your other loop of remove drives. Also it can be used in cleaning up code in floppy_module_exit(). I think this will make code much more readable. Right now this error handling loop is just getting too complicated. Thanks Vivek