Hi Paul, > Also, at the top of dwc2_driver_probe() there is a call to > dwc2_set_all_params(), which is in a different module and thus won't > link. Ah, good point. I guess I should have added a EXPORT_SYMBOL_GPL(dwc2_set_all_params) as well when adding the declaration to hcd.h in the previous patch. > I think that shouldn't be there anyway, since it is already called in > dwc2_hcd_init(). It is there on purpose. There are two instances of core_params: The one allocated in platform.c are the "requested" values. These are fixed to -1 in this patch, but get populated with DT data in the next patch. There is a second instance, allocated and initialized to -1 by hcd_init, which stores the actual values used. The various dwc2_set_param_* functions validate the values from the first instance and store them in the second. We can probably merge these into one instance if that is desirable, but that one instance should then be allocated inside platform.c, since it must be writable and needs data from the dt loaded into it (so it cannot be a global const struct shared between all devices, like in pci.c). Gr. Matthijs -- 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
