From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: [PATCHSET 08/11] add hotplug support, take 2 Date: Fri, 12 May 2006 00:32:06 +0900 Message-ID: <11473615261297-git-send-email-htejun@gmail.com> Reply-To: Tejun Heo Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Return-path: Received: from wr-out-0506.google.com ([64.233.184.227]:60996 "EHLO wr-out-0506.google.com") by vger.kernel.org with ESMTP id S1030267AbWEKPcQ (ORCPT ); Thu, 11 May 2006 11:32:16 -0400 Received: by wr-out-0506.google.com with SMTP id 55so355916wri for ; Thu, 11 May 2006 08:32:13 -0700 (PDT) In-Reply-To: Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: jgarzik@pobox.com, alan@lxorguk.ukuu.org.uk, axboe@suse.de, albertcc@tw.ibm.com, forrest.zhao@intel.com, efalk@google.com, linux-ide@vger.kernel.org, htejun@gmail.com Hi. This is part of patchset series described in [T]. This is the second take of add-hotplug-support patchset. Changes from the last take[L] are * No ap->hotplug_flags and no atomic bitops helpers. All are handled via dev->dflags and ap->eh_info. Things are much simpler now. * Tighter event handling. Unless a new device is plugged in the middle of reset (there's no generic and reliable way to handle this), no hotplug event is lost. * Stronger SCSI synchronization. There still is a race condition during SCSI scan after hotplugging, but the window is really tiny and there doesn't seem to be a waterproof way to fix it. I don't think the race will ever get triggered and even if it does the worst that can happen is not attaching a SCSI device for a newly detected ATA device. Other than this race condition, everything should be watertight. * Disabled devices are not automatically rescanned on user scan request unless the user explicitly requests scan of the device by specifying the device ID or using wild card. * Probing during driver loading and detaching while unloading are all handled by hotplug. There is only one probing/detaching path now. * Unused ->probe_init, ->probe_reset and all related functions killed. This patchset is against upstream (acc696d93dcf993dec123d69d599979e1456ffec) + [1] prep-for-new-EH patchset + [2] new-EH-framework patchset, take 3 + [3] new-EH-implementation patchset, take 3 + [4] merge-irq-pio patchset + [5] add-NCQ-support patchset, take 3 + [6] prep-for-hotplug support, take 2 + [7] prep-LLDDs-for-hotplug-support, take 1 -- tejun [T] http://article.gmane.org/gmane.linux.ide/9957 [L] http://article.gmane.org/gmane.linux.ide/9587 [1] http://article.gmane.org/gmane.linux.ide/9959 [2] http://article.gmane.org/gmane.linux.ide/9984 [3] http://article.gmane.org/gmane.linux.ide/9995 [4] http://article.gmane.org/gmane.linux.ide/10005 [5] http://article.gmane.org/gmane.linux.ide/10011 [6] http://article.gmane.org/gmane.linux.ide/10028 [7] http://article.gmane.org/gmane.linux.ide/10063