Re: WARN... Device 'cpu1' does not have a release() function, it is broken and must be fixed. when doing 'xl vcpu-set <guest_id> 1'

[Greg KH <gregkh@suse.de>]

On Mon, Jan 23, 2012 at 01:06:01PM -0500, Konrad Rzeszutek Wilk wrote:
> When I bring a CPU down in a guest (which should be the same as bringing a
> CPU down using the ACPI framework), I get this:
> 
> [   14.484206] SMP alternatives: switching to UP code
> [   14.514287] ------------[ cut here ]------------
> [   14.514318] WARNING: at /home/konrad/linux-linus/drivers/base/core.c:194 device_release+0x82/0x90()
> [   14.514354] Device 'cpu1' does not have a release() function, it is broken and must be fixed.
> [   14.514386] Modules linked in: radeon fbcon tileblit font ttm bitblit softcursor drm_kms_helper xen_blkfront xen_netfront xen_fbfront fb_sys_fops sysimgblt sysfillrect syscopyarea xen_kbdfront xenfs xen_privcmd
> [   14.514557] Pid: 22, comm: xenwatch Not tainted 3.3.0-rc1 #1
> [   14.514586] Call Trace:
> [   14.515094]  [<ffffffff810897fa>] warn_slowpath_common+0x7a/0xb0
> [   14.515094]  [<ffffffff810898d1>] warn_slowpath_fmt+0x41/0x50
> [   14.515094]  [<ffffffff813dea02>] device_release+0x82/0x90
> [   14.515094]  [<ffffffff812dfb85>] kobject_release+0x45/0x90
> [   14.515094]  [<ffffffff812dfa4c>] kobject_put+0x2c/0x60
> [   14.515094]  [<ffffffff813de6a2>] put_device+0x12/0x20
> [   14.515094]  [<ffffffff813df409>] device_unregister+0x19/0x20
> [   14.515094]  [<ffffffff813e4edf>] unregister_cpu+0x4f/0x80
> [   14.515094]  [<ffffffff81052c7c>] arch_unregister_cpu+0x1c/0x20
> [   14.515094]  [<ffffffff8137f787>] handle_vcpu_hotplug_event+0xc7/0xd0
> [   14.515094]  [<ffffffff8137b900>] xenwatch_thread+0xb0/0x180
> [   14.515094]  [<ffffffff810ad3f0>] ? wake_up_bit+0x40/0x40
> [   14.515094]  [<ffffffff8137b850>] ? split+0xf0/0xf0
> [   14.515094]  [<ffffffff810acd16>] kthread+0x96/0xa0
> [   14.515094]  [<ffffffff816151e4>] kernel_thread_helper+0x4/0x10
> [   14.515094]  [<ffffffff8160cc80>] ? retint_restore_args+0x5/0x6
> [   14.515094]  [<ffffffff816151e0>] ? gs_change+0x13/0x13
> [   14.515094] ---[ end trace 8f70af51a2e2611f ]---
> 
> Looking at "commit e032d80774315869aa2285b217fdbbfed86c0b49
> Author: Greg Kroah-Hartman <gregkh@suse.de>
> Date:   Mon Jan 16 14:40:28 2012 -0800
> 
>     mce: fix warning messages about static struct mce_device
> "
> 
> it looks like the corret fix is to make the 'cpu_devices' in
> arch/x86/kernel/topology.c to be changed to be more dynamic
> (or perhaps have an empty release function)?
> 
> Is anybody else hitting this with ACPI CPU hot-unplug? Or do I have
> the privilige of being the first? Oh, I hadn't done a full bisection
> but v3.2 does not have this.

Kay, this is a mess.

This cpu system device is is interconnected with the different arches
and their cpu-specific structures.  Some arches have a static array,
some allocate a huge structure (struct arch_cpu * NUM_CPUS), and others
try to do the right thing with DECLARE_PER_CPU() but don't quite get it
right, making that a static array per cpu.

To unwind all of this, is much beyond 3.3 material, as I'm sure I'll get
it wrong, and have a bunch of non-x86-64 build problems along the way.

Any objection to me just doing the "hack" of the empty release function
at the moment to get rid of this warning, and then clean it all up
properly for 3.4?

thanks,

greg k-h