From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nigel Cunningham Subject: Re: [linux-pm] [PATCH 0/6] [-mm]: ACPI: duplicate ACPI procfs functions in sysfs Date: Thu, 25 Jan 2007 13:52:28 +1100 Message-ID: <1169693548.9381.1.camel@nigel.suspend2.net> References: <1168083306.5619.34.camel@localhost.localdomain> <20070107111509.GA4792@ucw.cz> <200701242103.57153.lenb@kernel.org> Reply-To: nigel@nigel.suspend2.net Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <200701242103.57153.lenb@kernel.org> Sender: linux-acpi-owner@vger.kernel.org To: Len Brown Cc: Pavel Machek , "linux-acpi@vger" , linux-pm@osdl.org List-Id: linux-pm@vger.kernel.org Hi. On Wed, 2007-01-24 at 21:03 -0500, Len Brown wrote: > > > Patch 03-05: > > > add ACPI sleep attributes in sysfs. > > > /proc/acpi/sleep is already deprecated by /sys/power/state. > > > > Does that mean we drop standby (S1) capability on PCs? > > I think we need to make /sys/power/state handle S1. > > There are two cases > > 1. Platform supports S1, but does not support S3. > > This is more common. You see this a lot on server-class machines. > > We could make "mem" simply mean S1 here b/c it is effectively > the closest thing to S3. > > 2. Platform supports both S1 and S3. > > This is pretty rare -- at least on the systems I've got. > I'd like the generic interface be able to describe and handle this case. > > I'm open to suggestions on what to call S1 if it isn't called "mem". 'standby'? That's what occurs to me from previous M$ usage. Regards, Nigel