When the phy is forced in host mode, only the first hot plug and hot remove works. That is actually because the driver execute the OTG workaround, whereas it is not applicable in host or device mode. Indeed, to work correctly, the VBUS sense and session end comparator must be enabled, what is only possible when the phy is in OTG mode. Only execute the workaround if the phy is in OTG mode. Signed-off-by: Alexandre Bailon <abailon@xxxxxxxxxxxx> --- drivers/usb/musb/da8xx.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c index ec93602..be10c51 100644 --- a/drivers/usb/musb/da8xx.c +++ b/drivers/usb/musb/da8xx.c @@ -145,6 +145,17 @@ static void otg_timer(unsigned long _musb) unsigned long flags; /* + * We should only execute the OTG workaround when the phy is in OTG + * mode. The workaround require the VBUS sense and the session end + * comparator to be enabled, what is only possible if the phy is in + * OTG mode. As the workaround is only required to detect if the + * controller must act as host or device, we can safely exit OTG is + * not in use. + */ + if (musb->port_mode != MUSB_PORT_MODE_DUAL_ROLE) + return; + + /* * We poll because DaVinci's won't expose several OTG-critical * status change events (from the transceiver) otherwise. */ -- 2.7.3 -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
