From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: Re: [linux-pm] [RFC][PATCH] PM: Introduce new top level =?iso-8859-1?q?suspend=09and_hibernation_callbacks?= (rev. 2) Date: Wed, 26 Mar 2008 21:46:38 +0100 Message-ID: <200803262146.39899.rjw@sisk.pl> References: Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org To: Alan Stern Cc: Oliver Neukum , ACPI Devel Maling List , linux-pm@lists.linux-foundation.org, Alexey Starikovskiy , Johannes Berg , LKML List-Id: linux-acpi@vger.kernel.org On Wednesday, 26 of March 2008, Alan Stern wrote: > On Wed, 26 Mar 2008, Oliver Neukum wrote: >=20 > > Am Mittwoch, 26. M=E4rz 2008 15:40:27 schrieb Alan Stern: > > > Remember, the whole purpose of this is to let drivers know when t= he=20 > > > system is going to sleep or waking up. =A0Proper handling of devi= ces is=20 > > > up to the drivers, not up to the core. > >=20 > > Then declare these methods void. We cannot introduce methods that d= eliberately > > ignore errors. Reporting is also better done in the drivers. >=20 > That decision is up to Rafael. Changing the methods to return void i= s=20 > okay with me. But that's not what they currently do, either. If I change the methods to void and it turns out in the future that it'= s better if they return error codes, it will be rather difficult to go ba= ck and change everything. For this reason, I'd prefer to retain the returning= of error codes. What exactly do you whink would be wrong with using the error codes to = avoid resuming the children of devices that failed to resume? Rafael