From mboxrd@z Thu Jan 1 00:00:00 1970 From: Linus Torvalds Subject: Re: [RFC] Asynchronous suspend/resume - test results Date: Mon, 21 Dec 2009 15:00:56 -0800 (PST) Message-ID: References: <200912212136.12665.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: Received: from smtp1.linux-foundation.org ([140.211.169.13]:53418 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755997AbZLUXBD (ORCPT ); Mon, 21 Dec 2009 18:01:03 -0500 In-Reply-To: <200912212136.12665.rjw@sisk.pl> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Rafael J. Wysocki" Cc: Alan Stern , pm list , ACPI Devel Maling List , LKML , Dmitry Torokhov On Mon, 21 Dec 2009, Rafael J. Wysocki wrote: > > Do you think it's a good idea to mark all SCSI devices as async? Well, for > testing it probably is, but I doubt in general. As long as things aren't multi-path, or multi-lun, it's probably fine. With multi-path (ie same disk reachable multiple ways), it's probably fine too, but let's face it, who the hell knows? But it's not going to be an issue for laptops and normal desktops. For multi-lun, I suspect the luns are related, and there might be some ordering issues. But it's _likely_ all fine there too. The nice thing about systems with wires (ie USB, SCSI, SATA) is that it's really hard to make different devices have odd dependencies on each other. The wires tend to _be_ the topology. Linus