From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH 1/4] scsi: Allow hosts to be flagged as hotpluggable Date: Thu, 16 Jul 2009 01:12:28 +0000 Message-ID: <1247706748.8632.3.camel@mulgrave.site> References: <1247701438-18266-1-git-send-email-mjg@redhat.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from bedivere.hansenpartnership.com ([66.63.167.143]:58625 "EHLO bedivere.hansenpartnership.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756909AbZGPBMe (ORCPT ); Wed, 15 Jul 2009 21:12:34 -0400 In-Reply-To: <1247701438-18266-1-git-send-email-mjg@redhat.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Garrett Cc: linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org, linux-ide@vger.kernel.org On Thu, 2009-07-16 at 00:43 +0100, Matthew Garrett wrote: > Userspace may wish to make policy decisions based on whether a host > supports device hotplug or not - for example, AHCI link power management > disables hotplug, so may only be desirable on hotplug ports. Add > support for marking hosts as hotpluggable in order to allow userspace to > treat them appropriately. OK, so I don't really understand what the hotplug flag means. You seem to be setting it unconditionally on most sata HBAs. If it just means "bus is hotpluggable", it should be set to 1 at initialisation and the few non hot plug busses (like SPI) get to reset it. However, by definition SATA (like SAS) is a hotplug bus ... why isn't it set for some SATA controllers ... is it because the HBA itself does something wrong when a hotplug event comes in? James