http://bugzilla.kernel.org/show_bug.cgi?id=9772
[EMAIL PROTECTED] changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
------- Comment #2 from [EMAIL PROTECTED] 2008-01-23 23:58 -------
Hi, Daniel,
In order to hot remove a CPU,
Please echo 0 > /sys/devices/system/cpu/cpuX/online first before echo 1 >
/sys/bus/acpi/drivers/processor/ACPI0007:0X/eject.
Is this work for you?
Alex,
(In reply to comment #1)
> Ugh, it also seems that acpi_processor_handle_eject() calls
> arch_unregister_cpu(), which removes the /sys/devices/system/cpu/cpuX/online
> attribute, so a udev script can't do the right thing.
IMO, for cpu hot removal, it should work like this:
users prepare to hot removal a CPU, notification ACPI_NOTIFY_EJECT_REQUEST is
generated. acpi_processor_hotplug_notify send a KOBJ_OFFLINE uevent.
Userspace should echo 0 > /sys/devices/system/.../cpuX/online to offline the
processor, and then echo 1 > /sys/bus/acpi/.../ACPI0007:0x/eject to remove the
processor from Linux device tree, including the acpi dev and sysdev.
> One possible solution would be for acpi_processor_hotplug_notify() to poke a
> finger into the base cpu driver and call cpu_down() when it receives an eject
> request, but that would negate the whole point of emitting uevents and
> letting userspace take care of it.
IMO, it works, but you're right that user space should take care of this.
> The other possibility is to move the call to arch_unregister_cpu into the
> cpu_down path.
Then what about the acpi device node for processor? maybe we need a platform
hook here?
--
Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
acpi-bugzilla mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
