Messages in this thread |  | | Date | Sun, 15 Jan 2023 09:25:55 -0800 | Subject | Re: [PATCH v13 1/3] usb: typec: tcpm: Add callbacks to mitigate wakeups due to contaminant | From | Guenter Roeck <> |
| |
On 1/14/23 01:32, Badhri Jagan Sridharan wrote: > On some of the TCPC implementations, when the Type-C port is exposed > to contaminants, such as water, TCPC stops toggling while reporting OPEN > either by the time TCPM reads CC pin status or during CC debounce > window. This causes TCPM to be stuck in TOGGLING state. If TCPM is made > to restart toggling, the behavior recurs causing redundant CPU wakeups > till the USB-C port is free of contaminant. > > [206199.287817] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [206199.640337] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [206199.985789] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > > (or) > > [ 7853.867577] Start toggling > [ 7853.889921] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [ 7855.698765] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected] > [ 7855.698790] state change TOGGLING -> SNK_ATTACH_WAIT [rev3 NONE_AMS] > [ 7855.698826] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev3 NONE_AMS] > [ 7855.703559] CC1: 0 -> 0, CC2: 5 -> 5 [state SNK_ATTACH_WAIT, polarity 0, connected] > [ 7855.856555] CC1: 0 -> 0, CC2: 5 -> 0 [state SNK_ATTACH_WAIT, polarity 0, disconnected] > [ 7855.856581] state change SNK_ATTACH_WAIT -> SNK_ATTACH_WAIT [rev3 NONE_AMS] > [ 7855.856613] pending state change SNK_ATTACH_WAIT -> SNK_UNATTACHED @ 170 ms [rev3 NONE_AMS] > [ 7856.027744] state change SNK_ATTACH_WAIT -> SNK_UNATTACHED [delayed 170 ms] > [ 7856.181949] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [ 7856.187896] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [ 7857.645630] CC1: 0 -> 0, CC2: 0 -> 0 [state TOGGLING, polarity 0, disconnected] > [ 7857.647291] CC1: 0 -> 0, CC2: 0 -> 5 [state TOGGLING, polarity 0, connected] > [ 7857.647298] state change TOGGLING -> SNK_ATTACH_WAIT [rev3 NONE_AMS] > [ 7857.647310] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @ 170 ms [rev3 NONE_AMS] > [ 7857.808106] CC1: 0 -> 0, CC2: 5 -> 0 [state SNK_ATTACH_WAIT, polarity 0, disconnected] > [ 7857.808123] state change SNK_ATTACH_WAIT -> SNK_ATTACH_WAIT [rev3 NONE_AMS] > [ 7857.808150] pending state change SNK_ATTACH_WAIT -> SNK_UNATTACHED @ 170 ms [rev3 NONE_AMS] > [ 7857.978727] state change SNK_ATTACH_WAIT -> SNK_UNATTACHED [delayed 170 ms] > > To mitigate redundant TCPM wakeups, TCPCs which do have the needed hardware > can implement the check_contaminant callback which is invoked by TCPM > to evaluate for presence of contaminant. Lower level TCPC driver can > restart toggling through TCPM_PORT_CLEAN event when the driver detects > that USB-C port is free of contaminant. check_contaminant callback also > passes the disconnect_while_debounce flag which when true denotes that > the CC pins transitioned to OPEN state during the CC debounce window. > > Signed-off-by: Badhri Jagan Sridharan <badhri@google.com> > --- > Changes since v12: > * None. > Changes since v11: > * None.
Why did you drop the Reviewed-by: tags if there was no change ?
Same question applied to the second patch of the series.
Guenter
|  |