Oliver Neukum <oliver@xxxxxxxxxx> writes:
> Am Montag, 27. Februar 2012, 12:58:27 schrieb Bjørn Mork:
>> Oliver Neukum <oliver@xxxxxxxxxx> writes:
>>
>> > Now, I am confused. I thought I had acked everything outstanding.
>> > What did I miss?
>>
>> Could just as well be me having missed it, but I cannot remember seeing
>> a definite ack the the patch set in the subject. I.e. the
>> "[PATCH v2 0/5] usb: cdc-wdm: subdriver support"
>> set.
>>
>> You sort of pre-approved the concept and provided valuable feedback on
>> patch 3/5, based on which I sent an updated version of that single
>> patch. But I didn't dare consider any of those a final ack.
>>
>
> +/* using a counter to merge subdriver requests with our own into a combined state */
> static int qmi_wwan_manage_power(struct usbnet *dev, int on)
> {
> - dev->intf->needs_remote_wakeup = on;
> - return 0;
> + atomic_t *pmcount = (void *)&dev->data[1];
>
> Where is this initialized?
struct usbnet is allocated as a private part of the netdev by
usbnet_probe => alloc_etherdev => alloc_netdev_mqs => kzalloc
So the whole data[5] array should already be properly initialized to 0.
Do you think that the counter should be explicitly initialized anyway to
avoid any confusion? I have also wondered if it would be a good idea to
overlay a private struct on top of the data[5] array, to better document
the driver's usage of the array. That would probably make even more
sense if we do explicit initialization. Any thoughts on that?
Bjørn
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
