[PATCH 2/8] w1: introduce helper to generate mac address from 1-wire id

[Jean-Christophe PLAGNIOL-VILLARD]

On 09:33 Mon 29 Oct     , Sascha Hauer wrote:
> On Sun, Oct 28, 2012 at 02:24:06AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj at jcrosoft.com>
> > ---
> >  include/w1_mac_address.h |   64 ++++++++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 64 insertions(+)
> >  create mode 100644 include/w1_mac_address.h
> > 
> > diff --git a/include/w1_mac_address.h b/include/w1_mac_address.h
> > new file mode 100644
> > index 0000000..ed93e0e
> > --- /dev/null
> > +++ b/include/w1_mac_address.h
> > @@ -0,0 +1,64 @@
> > +/*
> > + * Copyright (C) 2012 Jean-Christophe PLAGNIOL-VILLARD <plagnio at jcrosoft.com>
> > + *
> > + * Under GPLv2 only
> > + */
> > +
> > +#ifndef __W1_MAC_ADDRESS_H__
> > +#define __W1_MAC_ADDRESS_H__
> > +
> > +static inline int w1_mac_address_register(int eth_num, bool local,
> > +					 char * oui, char *w1_dev)
> 
> This function really deserves a comment what it does. From the name I
> assume that it turns something w1 related into a MAC address, but I have
> no clue what.
> 
> Also I think the 'local' parameter should not exist and be unconditional
> true. I don't think any vendor with a real ieee registered oui will
> match their mac addresses with ids from a w1 eeprom.
as you can request specific id format for the serial number 1-wire (smem)

you can use it for real OUI

with this way you can prouve than you manage your OUI correctly if you need to
request an other one

Best Regards,
J.
> 
> > +{
> > +	char buf[6];
> > +	const char *val;
> > +	u64 id;
> > +	int nb_oui = 0;
> > +	int i, shift;
> > +	char *tmp = NULL;
> > +	int ret = 0;
> > +
> > +	if (!local && !oui)
> > +		return -EINVAL;
> > +
> > +	if (oui) {
> > +		nb_oui = strlen(oui);
> 
> A oui always has three bytes. You can't use strlen here as for example
> Freescale has the oui 00-04-9f
> 
> > +		if (!local && nb_oui != 3)
> > +			return -EINVAL;
> > +
> > +		nb_oui = min(nb_oui, 3);
> > +
> > +		for (i = 0; i < nb_oui; i++)
> > +			buf[i] = oui[i];
> > +	}
> > +
> > +	tmp = asprintf("%s.id", w1_dev);
> > +	if (!tmp)
> > +		return -ENOMEM;
> > +
> > +	val = getenv("w1-1-0.id");
> 
> tmp is never used. I assume it should be used here, right?
> 
> > +	if (!val) {
> > +		ret = -EINVAL;
> > +		goto err;
> > +	}
> > +
> > +	id = simple_strtoull(val, NULL, 16);
> > +	if (!id) {
> > +		ret = -EINVAL;
> > +		goto err;
> > +	}
> > +
> > +	for (i = nb_oui, shift = 40; i < 6; i++, shift -= 8)
> > +		buf[i] = (id >> shift) & 0xff;
> > +
> > +	if (local)
> > +		buf[0] |= 0x2;
> > +
> > +	eth_register_ethaddr(eth_num, buf);
> > +
> > +err:
> > +	free(tmp);
> > +	return ret;
> > +}
> > +
> > +#endif /* __W1_MAC_ADDRESS_H__ */
> > -- 
> > 1.7.10.4
> > 
> > 
> > _______________________________________________
> > barebox mailing list
> > barebox at lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/barebox
> > 
> 
> -- 
> Pengutronix e.K.                           |                             |
> Industrial Linux Solutions                 | http://www.pengutronix.de/  |
> Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
> Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |