From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from buildserver.ru.mvista.com (unknown [213.79.90.228]) by ozlabs.org (Postfix) with ESMTP id D5718B7BB3 for ; Fri, 6 Nov 2009 07:30:02 +1100 (EST) Date: Thu, 5 Nov 2009 23:30:01 +0300 From: Anton Vorontsov To: Scott Wood Subject: Re: [PATCH 1/3] powerpc/83xx/suspend: Clear deep_sleeping after devices resume Message-ID: <20091105203001.GA629@oksana.dev.rtsoft.ru> References: <20090923190041.GA18944@oksana.dev.rtsoft.ru> <20090923190113.GA19932@oksana.dev.rtsoft.ru> <043A102D-9F70-4287-B34E-17FF9AE90EE1@kernel.crashing.org> <4AF30406.3080101@freescale.com> <2F8E2E66-8398-4ACD-9697-6C2AB6E39CD3@kernel.crashing.org> <4AF32C0B.4010109@freescale.com> <7A8DBFC0-69D9-406B-B1AC-3EAC8EE26C58@kernel.crashing.org> <4AF32F82.8040801@freescale.com> <20091105200931.GA30577@oksana.dev.rtsoft.ru> <4AF334AB.1030401@freescale.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 In-Reply-To: <4AF334AB.1030401@freescale.com> Cc: linux-ppc list Reply-To: avorontsov@ru.mvista.com List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Thu, Nov 05, 2009 at 02:25:15PM -0600, Scott Wood wrote: > Anton Vorontsov wrote: > >On Thu, Nov 05, 2009 at 02:03:14PM -0600, Scott Wood wrote: > >>Kumar Gala wrote: > >>>On Nov 5, 2009, at 1:48 PM, Scott Wood wrote: > >>>>As for patch 3, Ben objected to the sleep-nexus stuff on IRC. > >>>Is sleep-nexus new? I thought we've had that for a bit. > >>It's been around in a few dts files, but as was noted, nothing uses > >>this stuff yet. > > > >So I should just drop the sleep-nexus changes? I can also > >prepare a patch that removes sleep-nexus from 8313rdb.dts. > >But how should we handle the sleep = <> properties then? > > We could still have some sort of nexus node that is off to the side > (pointed to with sleep-parent) and not inserting itself into the > hierarchy. > > Or, we could allow multiple nodes to refer to the same sleep ID, and > use the ID rather than a node to tie things together. If we do > that, we'll probably want a simple index rather than a set of bits > in the sleep property, so it can correspond to some kernel object > that has some bookeeping info. Perhaps we could tie into the clock > bindings that were discussed on devtree-discuss in August. Yeah, reusing the clk api would be the best. Anyway, since we don't use the sleep-nexus stuff, I'd rather just add the power management controller nodes. -- Anton Vorontsov email: cbouatmailru@gmail.com irc://irc.freenode.net/bd2