Re: [PATCH v3] usb: typec: tcpm: Fix sink caps op current check

[William McVicker <willmcvicker@google.com>]

On 10/31/2023, Guenter Roeck wrote:
> On 10/31/23 18:28, Badhri Jagan Sridharan wrote:
> > TCPM checks for sink caps operational current even when PD is disabled.
> > This incorrectly sets tcpm_set_charge() when PD is disabled.
> > Check for sink caps only when PD is enabled.
> > 
> > [   97.572342] Start toggling
> > [   97.578949] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected]
> > [   99.571648] CC1: 0 -> 0, CC2: 0 -> 4 [state TOGGLING, polarity 0, connected]
> > [   99.571658] state change TOGGLING -> SNK_ATTACH_WAIT [rev3 NONE_AMS]
> > [   99.571673] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev3 NONE_AMS]
> > [   99.741778] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed 170 ms]
> > [   99.789283] CC1: 0 -> 0, CC2: 4 -> 5 [state SNK_DEBOUNCED, polarity 0, connected]
> > [   99.789306] state change SNK_DEBOUNCED -> SNK_DEBOUNCED [rev3 NONE_AMS]
> > [   99.903584] VBUS on
> > [   99.903591] state change SNK_DEBOUNCED -> SNK_ATTACHED [rev3 NONE_AMS]
> > [   99.903600] polarity 1
> > [   99.910155] enable vbus discharge ret:0
> > [   99.910160] Requesting mux state 1, usb-role 2, orientation 2
> > [   99.946791] state change SNK_ATTACHED -> SNK_STARTUP [rev3 NONE_AMS]
> > [   99.946798] state change SNK_STARTUP -> SNK_DISCOVERY [rev3 NONE_AMS]
> > [   99.946800] Setting voltage/current limit 5000 mV 500 mA
> > [   99.946803] vbus=0 charge:=1
> > [  100.027139] state change SNK_DISCOVERY -> SNK_READY [rev3 NONE_AMS]
> > [  100.027145] Setting voltage/current limit 5000 mV 3000 mA
> > [  100.466830] VBUS on
> > 
> > Cc: stable@vger.kernel.org
> > Fixes: 803b1c8a0cea ("usb: typec: tcpm: not sink vbus if operational current is 0mA")
> > Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
> 
> Reviewed-by: Guenter Roeck <linux@roeck-us.net>

Thanks for catching this Badhri! I verified this fixes the charging issues seen
on Pixel 6 and 7 with the mainline and v6.1 kernel. Feel free to add

Tested-by: Will McVicker <willmcvicker@google.com>

Thanks,
Will
> 
> > ---
> > Changes since v2:
> > * Fix the "Fixes" tag
> > * Refactor code based on Guenter Roeck's suggestion.
> > 
> > Changes since v1:
> > * Fix commit title and description to address comments from Guenter Roeck
> > ---
> >   drivers/usb/typec/tcpm/tcpm.c | 3 ++-
> >   1 file changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
> > index 058d5b853b57..afc791ab6d4f 100644
> > --- a/drivers/usb/typec/tcpm/tcpm.c
> > +++ b/drivers/usb/typec/tcpm/tcpm.c
> > @@ -4273,7 +4273,8 @@ static void run_state_machine(struct tcpm_port *port)
> >   				current_lim = PD_P_SNK_STDBY_MW / 5;
> >   			tcpm_set_current_limit(port, current_lim, 5000);
> >   			/* Not sink vbus if operational current is 0mA */
> > -			tcpm_set_charge(port, !!pdo_max_current(port->snk_pdo[0]));
> > +			tcpm_set_charge(port, !port->pd_supported ||
> > +					pdo_max_current(port->snk_pdo[0]));
> >   			if (!port->pd_supported)
> >   				tcpm_set_state(port, SNK_READY, 0);
> > 
> > base-commit: c70793fb7632a153862ee9060e6d48131469a29c
>