From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org ([63.228.1.57]:41853 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751670Ab3GEXFZ (ORCPT ); Fri, 5 Jul 2013 19:05:25 -0400 Message-ID: <1373065511.4446.11.camel@pasglop> Subject: Re: [PATCH 1/8] PCI: Add pcibios_stop_dev() From: Benjamin Herrenschmidt To: Bjorn Helgaas Cc: Gavin Shan , linuxppc-dev , "linux-pci@vger.kernel.org" Date: Sat, 06 Jul 2013 09:05:11 +1000 In-Reply-To: References: <1372993054-25730-1-git-send-email-shangw@linux.vnet.ibm.com> <1372993054-25730-2-git-send-email-shangw@linux.vnet.ibm.com> <1373063807.4446.10.camel@pasglop> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-pci-owner@vger.kernel.org List-ID: On Fri, 2013-07-05 at 16:49 -0600, Bjorn Helgaas wrote: > I know we have paths where we save config space, do a reset, and > restore config space. I don't like those because some quirks aren't > applied and I don't believe restoring config space is sufficient to > make the device safe to use. There could be internal state that we > aren't restoring. It's also possible that a firmware update means the > device is different than it was before the reset, e.g., BARs could be > different types or sizes. True. Ok, I'll stick to the actual hotplug path then, you make good points. Cheers, Ben.