From mboxrd@z Thu Jan  1 00:00:00 1970
From: will.deacon@arm.com (Will Deacon)
Date: Thu, 29 Apr 2010 17:04:04 +0100
Subject: [PATCH 0/6] ARM: pmu: provide a registration mechanism for	IRQs
	[v3]
In-Reply-To: <20100428213619.GA16525@n2100.arm.linux.org.uk>
References: <1269530198-19572-1-git-send-email-will.deacon@arm.com>
	<20100428213619.GA16525@n2100.arm.linux.org.uk>
Message-ID: <000501cae7b5$97a76fb0$c6f64f10$@deacon@arm.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi Russell,

> Please can we have an ordered set which doesn't break compilation.
> What I think should happen is the first patch should introduce the
> new definitions required for the platform data, and the rest of the
> first patch should come at the very end.

Sorry about this. I've split up the first patch so that the first
half is now:


ARM: pmu: add enum describing PMU types
    
This patch adds an enum describing the potential PMU device types in
preparation for PMU device registration via platform devices.

diff --git a/arch/arm/include/asm/pmu.h b/arch/arm/include/asm/pmu.h
index 2829b9f..44bec1f 100644
--- a/arch/arm/include/asm/pmu.h
+++ b/arch/arm/include/asm/pmu.h
@@ -12,6 +12,11 @@
 #ifndef __ARM_PMU_H__
 #define __ARM_PMU_H__
 
+enum arm_pmu_type {
+       ARM_PMU_DEVICE_CPU      = 0,
+       ARM_NUM_PMU_DEVICES,
+};
+
 #ifdef CONFIG_CPU_HAS_PMU
 
 struct pmu_irqs {


which gets rid of the compilation failures. I can either supersede
the existing patch with two more or resubmit the whole series. Which
would you prefer?

Thanks,

Will