From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin Herrenschmidt Subject: Re: 2.6.29-rc3: tg3 dead after resume Date: Sat, 07 Feb 2009 14:27:26 +1100 Message-ID: <1233977246.31963.64.camel@pasglop> References: <200901312346.35265.rjw@sisk.pl> <200902010111.08433.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: "Rafael J. Wysocki" , Parag Warudkar , Matt Carlson , "netdev@vger.kernel.org" , Linux Kernel Mailing List , "David S. Miller" , Andrew Morton To: Linus Torvalds Return-path: Received: from gate.crashing.org ([63.228.1.57]:54798 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751661AbZBGD1l (ORCPT ); Fri, 6 Feb 2009 22:27:41 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Sat, 2009-01-31 at 16:51 -0800, Linus Torvalds wrote: > > On Sat, 31 Jan 2009, Linus Torvalds wrote: > > > > For a USB host controller, for example, it really could make sense to do > > that - just leave all the core host controller stuff running, and the only > > thing the "suspend()" callback does is to send the commands to the actual > > devices, it doesn't necessarily touch the host controller itself at all. > > Same is quite likely true of things like video graphics adapters. Again, > for all the same reasons. Think about all those fbcon drivers. They will > use DMA for things. And again, there are very compelling debugging reasons > to not suspend them for real until suspend_late (if even then). Actually, there are things like service processors etc... that only know to communicate via DMA and little else via a little mailbox in some reserved area of system memory... Ben.