From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Hansen Subject: Re: possible_cpus broken in linux-next Date: Wed, 29 May 2013 15:14:12 -0700 Message-ID: <51A67DB4.9090404@sr71.net> References: <51A6367E.1060400@intel.com> <1369852557.16819.54.camel@misato.fc.hp.com> <1369853414.16819.56.camel@misato.fc.hp.com> <51A650BC.7030604@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <51A650BC.7030604@intel.com> Sender: linux-kernel-owner@vger.kernel.org To: "Rafael J. Wysocki" Cc: Toshi Kani , Dave Hansen , Stephen Rothwell , Greg Kroah-Hartman , LKML , linux-acpi@vger.kernel.org List-Id: linux-acpi@vger.kernel.org On 05/29/2013 12:02 PM, Rafael J. Wysocki wrote: >>> diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c >>> index 130ba0b..b9f0eec 100644 >>> --- a/drivers/base/cpu.c >>> +++ b/drivers/base/cpu.c >>> @@ -281,6 +281,7 @@ int __cpuinit register_cpu(struct cpu *cpu, int num) >>> cpu->dev.bus = &cpu_subsys; >>> cpu->dev.release = cpu_device_release; >>> cpu->dev.offline_disabled = !cpu->hotpluggable; >>> + cpu->dev.offline = !cpu_online(num); >>> #ifdef CONFIG_ARCH_HAS_CPU_AUTOPROBE >>> cpu->dev.bus->uevent = arch_cpu_uevent; >>> #endif This gets things working for me again. Thanks for the quick response! Tested-by: Dave Hansen