From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752336Ab2GURr4 (ORCPT ); Sat, 21 Jul 2012 13:47:56 -0400 Received: from ogre.sisk.pl ([193.178.161.156]:50835 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751401Ab2GURry (ORCPT ); Sat, 21 Jul 2012 13:47:54 -0400 From: "Rafael J. Wysocki" To: Linus Torvalds Subject: Re: [RFC] firmware load: defer request_firmware during early boot and resume Date: Sat, 21 Jul 2012 19:53:35 +0200 User-Agent: KMail/1.13.6 (Linux/3.5.0-rc5+; KDE/4.6.0; x86_64; ; ) Cc: Ming Lei , Linux Kernel Mailing List , "linux-usb" , Alan Stern , "Greg Kroah-Hartman" , Oliver Neukum , Linux PM list References: In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201207211953.35791.rjw@sisk.pl> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Saturday, July 21, 2012, Linus Torvalds wrote: > On Fri, Jul 20, 2012 at 5:33 AM, Ming Lei wrote: > > The RFC patch is just for discussing if the idea of deferring > > request_firmware is doable for addressing the issue of > > request_firmware in resume path, which is caused by driver > > unbind/rebind during resume. > > NAK. [...] > Seriously. People should load their firmware *before* the > suspend/resume cycle. And if that isn't possible, then the system > should ABSOLUTELY NOT silently say "whatever" and defer it until > later. We should have that big failure and the big noisy warning, and > drivers that need to defer need to do so themselves, so that we never > *ever* have that silent automatic defer situation. Well, thanks. I totally agree, but didn't know how to say this. :-) Rafael