Hi Jonas,
On 03/31/2017 05:37 AM, Jonas Bonn wrote:
On 03/31/2017 10:35 AM, Lukasz Nowak wrote:
Hi Jonas,
On 30/03/17 21:53, Jonas Bonn wrote:
The rule might match for some device but setup_telitqmi does not attempt
to do anything for interfaces that aren't driven by the qmi_wwan driver
so it's pointless.
As I described before, this rule is not pointless, it prevents this rule:
{ "telit", "option", "1bc7" },
from taking over the device, and using setup_telit instead of
setup_telitqmi.
Lukasz
Yes, but in the context of the transition to hwdb it should not be
necessary anymore. That said, I see the problem and I think it still
persists in the hwdb model. The problem is the Telit catch-all rule; we
really need to get rid of it but that risks rendering a bunch of modems
that work today non-functional. This needs some thought.
Why don't we unify setup_telit & setup_telitqmi and let the unified
function figure out whether the particular telit modem is QMI based or
not. Similar to what setup_sierra does.
Regards,
-Denis
_______________________________________________
ofono mailing list
[email protected]
https://lists.ofono.org/mailman/listinfo/ofono
