Yoshihiro-san
Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx> writes:
>> >> Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx> writes:
>> >> > Sadly, to change the role ("host" and "peripheral") of USB3.0 DRD
>> >> > controller on R-Car Gen3, software has to set the DRD_CON register
>> >> > which is included in a usb3.0 peripheral controler's register.
>> >> > To simply implementation, the previous code always set peripheral mode
>> >> > as hardcoded. However, to support usb role swap in the future,
>> >> > the hardcoded is not good. So, this patch adds sysfs "role" to set
>> >> > the mode by a user.
>> >> > After applied this patch, since the DRD controller will act as host
>> >> > mode after probed, a user needs to change the mode via the sysfs.
>> >> >
>> >> > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx>
>> >>
>> >> in patch 3, you add extcon to monitor ID and VBUS pins. Do you really
>> >> need this sysfs file at all?
>> >
>> > Yes. This is because this "role" sysfs file cannot monitor the ID and VBUS pins.
>> > The "role" only monitors the mode which is "host" or "peripheral".
>>
>> Right, I understand it can't monitor the state of ID/VBUS, but do you
>> have a usecase for that file that's not covered by the other patches in
>> the series? :-)
>
> Thank you for the comment. Since my English skill is not good,
> I don't understand your question.
oh, that's alright. I'll try to explain:
I'm asking if you have any situation where you need userspace to change
the role manually. I understand it's a very nice feature to have during
development because it frees us from searching for "the right cable";
but if the only application of this sysfs file is durinb development,
then perhaps it should be exposed using debugfs instead ;-)
Hopefully that's easier to understand now.
--
balbi
Attachment:
signature.asc
Description: PGP signature
