From: Arthur Marsh Subject: Re: [Bugfix 3/3] eata: Enhance eata driver to support PCI device hot-removal Date: Sat, 19 Sep 2015 00:38:51 +0930 Message-ID: <55FC2903.6080102@internode.on.net> References: <1442200140-30808-1-git-send-email-jiang.liu@linux.intel.com> <1442200140-30808-4-git-send-email-jiang.liu@linux.intel.com> <55F6837A.5020807@suse.de> <20150916134211.GA21535@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20150916134211.GA21535@infradead.org> Sender: linux-kernel-owner@vger.kernel.org To: Christoph Hellwig , Hannes Reinecke Cc: Jiang Liu , Thomas Gleixner , Bjorn Helgaas , Dario Ballabio , "James E.J. Bottomley" , linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org, linux-scsi@vger.kernel.org, x86@kernel.org List-Id: linux-scsi@vger.kernel.org Christoph Hellwig wrote on 16/09/15 23:12: > Jiang, you also need to convert the driver to > scsi_add_host/scsi_remove_host from the legacy scsi_register interface, > otherwise the SCSI layer will be very unhappy. > > Take a look at commit 0d31f8759109cbc1e6fc196d08e6b0e8a9e93b3f for > example, the change should be straight forward. > I am pleased to note that when I tried a Linus git head kernel from the last 24 hours, the IRQ routing for my DPT2044W SCSI card using eata module worked again, although the shut-down/kexec issue remains. Arthur.