Re: possible_cpus broken in linux-next

From: Toshi Kani
Date: Wed May 29 2013 - 14:36:16 EST


On Wed, 2013-05-29 at 10:10 -0700, Dave Hansen wrote:
> If I boot with: maxcpus=2 possible_cpus=4, I get
>
> # grep . /sys/devices/system/cpu/cpu[0-9]*/online'
> /sys/devices/system/cpu/cpu1/online:1
> /sys/devices/system/cpu/cpu2/online:1
> /sys/devices/system/cpu/cpu3/online:1
>
> on bad kernels, and this on working ones:
>
> /sys/devices/system/cpu/cpu1/online:1
> /sys/devices/system/cpu/cpu2/online:0
> /sys/devices/system/cpu/cpu3/online:0
>
>
> I also get -EINVAL if I try to re-offline them in this state. 2 and 3
> don't show up in /proc/cpuinfo, so sysfs just looks broken here. This
> happens in a KVM guest, so it should be dirt-simple for anyone to reproduce.
>
> I bisected it down to:
>
> > commit 0902a9044fa5b7a0456ea4daacec2c2b3189ba8c
> > Author: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > Date: Fri May 3 00:25:49 2013 +0200
> >
> > Driver core: Use generic offline/online for CPU offline/online
> >
> > Rework the CPU hotplug code in drivers/base/cpu.c to use the
> > generic offline/online support introduced previously instead of
> > its own CPU-specific code.
> >
> > For this purpose, modify cpu_subsys to provide offline and online
> > callbacks for CONFIG_HOTPLUG_CPU set and remove the code handling
> > the CPU-specific 'online' sysfs attribute.
> >
> > This modification is not supposed to change the user-observable
> > behavior of the kernel (i.e. the 'online' attribute will be present
> > in exactly the same place in sysfs and should trigger exactly the
> > same actions as before).
> >
> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > Acked-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> > Reviewed-by: Toshi Kani <toshi.kani@xxxxxx>


Thanks for the report. The following patch fixes the problem.
-Toshi

====
From: Toshi Kani <toshi.kani@xxxxxx>
Subject: [PATCH] ACPI: Fix sysfs cpu/online of offlined cpus

As reported by Dave Hansen, sysfs cpu/online shows 1 for
offlined cpus at boot.
https://lkml.org/lkml/2013/5/29/403

Fix this problem by initializing dev.offline with cpu_online()
when registering a cpu.

Reported-by: Dave Hansen <dave.hansen@xxxxxxxxx>
Signed-off-by: Toshi Kani <toshi.kani@xxxxxx>
---
drivers/base/cpu.c | 1 +
1 file changed, 1 insertion(+)

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
--
1.8.1.4



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/