(2010/08/04 5:58), Rafael J. Wysocki wrote:
> On Tuesday, August 03, 2010, Rafael J. Wysocki wrote:
>> On Tuesday, August 03, 2010, Kenji Kaneshige wrote:
>>> (2010/08/03 6:53), Rafael J. Wysocki wrote:
>>>> From: Rafael J. Wysocki<rjw@xxxxxxx>
>>>
>>> <snip.>
>>>
>>>> + mutex_lock(&osc_lock);
>>>> + status = acpi_pci_query_osc(root, root->osc_support_set, ctrl_mask);
>>>> + mutex_unlock(&osc_lock);
>>>> +
>>>
>>> One more comment here.
>>>
>>> I think we can skip acpi_pci_query_osc() if all of queried controls are
>>> already granted to OS. Please see below
>>>
>>> mutex_lock(&osc_lock);
>>> if ((root->osc_control_set& *ctrl_mask) == *ctrl_mask) {
>>> *ctrl_mask = root->osc_control_set;
>>> goto out;
>>> }
>>> status = acpi_pci_query_osc(root, root->osc_support_set, ctrl_mask);
>>> mutex_unlock(&osc_lock);
>>> out:
>>
>> Well I guess you mean:
>>
>> mutex_lock(&osc_lock);
>> if ((root->osc_control_set& *ctrl_mask) != *ctrl_mask)
>> status = acpi_pci_query_osc(root, root->osc_support_set, ctrl_mask);
>> mutex_unlock(&osc_lock);
>>
>> Otherwise we would return with the mutex held. :-)
>
Oops... sorry...
> Updated patch is appended, please tell me what you think.
Looks good to me. The below (in your updated patch) was what I wanted to mean
> + mutex_lock(&osc_lock);
> + if ((*mask & root->osc_control_set) == *mask)
> + *mask = root->osc_control_set;
> + else
> + status = acpi_pci_query_osc(root, root->osc_support_set, mask);
> + mutex_unlock(&osc_lock);
Thanks,
Kenji Kaneshige
_______________________________________________
linux-pm mailing list
linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/linux-pm
