From mboxrd@z Thu Jan 1 00:00:00 1970 From: Len Brown Subject: Re: [RFC] ACPI interrupt statistics Date: Sat, 24 Nov 2007 00:34:24 -0500 Message-ID: <200711240034.24747.lenb@kernel.org> References: <200711030040.53126.lenb@kernel.org> <200711191942.35575.carlos@strangeworlds.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from hera.kernel.org ([140.211.167.34]:43462 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750999AbXKXFem (ORCPT ); Sat, 24 Nov 2007 00:34:42 -0500 In-Reply-To: <200711191942.35575.carlos@strangeworlds.co.uk> Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Carlos Corbacho Cc: Linux-acpi@vger.kernel.org On Monday 19 November 2007 14:42, Carlos Corbacho wrote: > On Saturday 03 November 2007 04:40:53 Len Brown wrote: > > For your review and testing... > > > > The patch below exports interrupt counters from within ACPI > > so that developers can better understand the cause of ACPI interrupts. > > > > My intent is to have something like this built in by default > > starting in 2.6.25. The version below applies to 2.6.21 > > and is originally based on a patch by Luming Yu. > > There have been a fair few changes since then, and this patch will probably > need a fair bit of work. > > 1) Back in 2.6.21, acpi_subsys was defined as a 'struct subsystem' in > acpi_bus.h - it's now a 'struct kset' yep, that is what broke it. > 2) Greg KH has an RFC patch lined up that does even more work in this area - > http://lkml.org/lkml/2007/11/2 yeah, i noticed that too -- we already ran amuck of them with some other patches in the acpi tree... akpm patched it up in mm. > 3) From my limited understanding of the area and intent, this patch seems to > be a combination of kset work + the interrupt support - could it be split > into two functionally separate patches to cover these instead? the patch I sent to the list is actually a roll-up of a patch series, so more patches will appear in the tree when it hits 2.6.25. > (Basically, the only bit I'm really interested in is the > EXPORT_SYMBOL(acpi_subsys), as acpi_subsys can currently only be used from > built-in drivers, which is preventing me from converting wmi.c to a module). hmm, okay, i didn't know that. but feel free to pre-pend a non-wmi patch to your wmi series if you need it. thanks, -Len