Dan Carpenter <dan.carpenter@xxxxxxxxxx> writes: > Ah... Here is the first patch which adds eth_hw_addr_inherit() > > http://patchwork.ozlabs.org/patch/269365/ > > I don't think we've actually set dst->addr_len yet at this point so > it doesn't do the memcpy(). This doesn't work. Ouch. Yes, I see. The net_device is allocated using kzalloc just a few lines earlier and there is no ether_setup or similar. Actually, it doesn't look like it ever sets addr_len. Is that right? Does it work even before this patch? Anyway, thanks for spotting this. I'll drop this patch for now as I clearly don't understand how this driver works. I do note that the otherwise very similar code in the vt6655 driver uses alloc_etherdev, and therefore has a fair chance of working both before and after my change. Bjørn _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
