On Mon, Aug 17, 2015 at 4:40 PM, Nick Podolak <[email protected]> wrote:
> > On Sat, Aug 15, 2015 at 3:25 AM, Hans Dedecker <[email protected]> wrote: > >> It would be better to make this a device setting like promiscuous >>> mode,macaddr, etc ... so it can be configured via UCI for every device and >>> thus not only restricted for GRE >>> >> >> Hans >> > > You make an excellent point and I agree, but I'm unsure of some of the > details. > > Is promisc mode controllable from UCI? I didn't think it was but I can't > tell from looking at these parts of netifd. > Promiscuous mode is configurable from UCI; see device.c as all dev_attrs entries can be set via UCI > > Help me understand the extent of this implementation. Looks like each of > these would need modification. Did i miss anything? > device.c > static const struct blobmsg_policy dev_attrs[__DEV_ATTR_MAX] = { > device_merge_settings(struct device *dev, struct device_settings *n) > device_init_settings(struct device *dev, struct blob_attr **tb) > device_dump_status(struct blob_buf *b, struct device *dev) > system-linux.c > system_if_get_settings(struct device *dev, struct device_settings *s) > system_if_apply_settings(struct device *dev, struct device_settings *s, > unsigned int apply_mask) > > UCI is all passthrough, right? nothing there would have to change? > > Your analysis regarding UCI and the impacted functions is correct; the promiscuous mode implementation can be used as guidance for the multicast device support. Hans > -Nick >
_______________________________________________ openwrt-devel mailing list [email protected] https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
