On Sun, Nov 25, 2012 at 12:20:47AM +0800, Wen Congyang wrote:
> At 2012/11/24 1:50, Vasilis Liaskovitis Wrote:
> > Consider the following sequence of operations for a hotplugged memory device:
> >
> > 1. echo "PNP0C80:XX"> /sys/bus/acpi/drivers/acpi_memhotplug/unbind
> > 2. echo "PNP0C80:XX"> /sys/bus/acpi/drivers/acpi_memhotplug/bind
> > 3. echo 1>/sys/bus/pci/devices/PNP0C80:XX/eject
> >
> > The driver is successfully re-bound to the device in step 2. However step 3 will
> > not attempt to remove the memory. This is because the acpi_memory_info enabled
> > bit for the newly bound driver has not been set to 1. This bit needs to be set
> > in the case where the memory is already used by the kernel (add_memory returns
> > -EEXIST)
>
> Hmm, I think the reason is that we don't offline/remove memory when
> unbinding it
> from the driver. I have sent a patch to fix this problem, and this patch
> is in
> pm tree now. With this patch, we will offline/remove memory when
> unbinding it from
> the drriver.
ok. Which patch is this? Does it require driver-core changes?
>
> Consider the following sequence of operations for a hotplugged memory
> device:
>
> 1. echo "PNP0C80:XX" > /sys/bus/acpi/drivers/acpi_memhotplug/unbind
> 2. echo 1 >/sys/bus/pci/devices/PNP0C80:XX/eject
>
> If we don't offline/remove the memory, we have no chance to do it in
> step 2. After
> step2, the memory is used by the kernel, but we have powered off it. It
> is very
> dangerous.
How does power-off happen after unbind? acpi_eject_store checks for existing
driver before taking any action:
#ifndef FORCE_EJECT
if (acpi_device->driver == NULL) {
ret = -ENODEV;
goto err;
}
#endif
FORCE_EJECT is not defined afaict, so the function returns without scheduling
acpi_bus_hot_remove_device. Is there another code path that calls power-off?
thanks,
- Vasilis
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
