On Thu, Mar 11, 2021 at 9:08 PM Badhri Jagan Sridharan <badhri@xxxxxxxxxx> wrote: > > On Thu, Mar 11, 2021 at 5:33 AM Heikki Krogerus > <heikki.krogerus@xxxxxxxxxxxxxxx> wrote: > > > > Hi, > > > > On Thu, Mar 11, 2021 at 02:03:12AM -0800, Badhri Jagan Sridharan wrote: > > > This change allows the driver to configure input current/voltage > > > limit for the charging path. The driver sets current_max and voltage_max > > > values of the power supply identified through chg-psy-name. > > > > > > The change also exposes the data_role and the orientation as a extcon > > > interface for configuring the USB data controller. > > > > This looks wrong to me. Why wouldn't you just register your device as > > a separate psy that supplies your charger (which is also a psy, right)? > > Hi Heikki, > > Looks like that would pretty much make it reflect the same values as > "tcpm-source-psy-" exposed > by tcpm. So experimenting with making the charger power supply a supplicant. > However, noticed that the "tcpm-source-psy-" does not have calls to > power_supply_changed(). > So the notifiers are not getting invoked. > Trying to fix that to see if just "tcpm-source-psy-" helps the case > without me trying to create another > one which almost would reflect the same values. Let me know if you > think that might not work. Hi Heikki, With "[PATCH] usb: typec: tcpm: Invoke power_supply_changed for tcpm-source-psy-" which I just sent, "tcpm-source-psy-" seems to do the job. So the set/get_current_limit and pd_capable callbacks are not needed. Please do review "[PATCH] usb: typec: tcpm: Invoke power_supply_changed for tcpm-source-psy-". Thanks, Badhri > > For now, refactored the patches to only include changes related to > data path and sending > them in. Will follow up with patches for the charger path once I am > done with the above approach > and some validation. > > Thanks, > Badhri > > > > > > thanks, > > > > -- > > heikki
