[PATCH] use osmocom auth API instead of direct calls

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/OpenBSC@lists.osmocom.org/.

Max.Suraev at fairwaves.co
Tue Jul 22 12:48:10 UTC 2014


Curiously I do not see this patch in patchwork anymore. Comments are inline.

22.07.2014 12:01, Holger Hans Peter Freyther пишет:
> On Wed, Jun 04, 2014 at 06:21:54PM +0200, Max Suraev wrote:
> 
> Hi!
> 
> sorry for the late reply (and I also intend to reply for reviving
> our wednesday meetings. I could talk about TCAP/MAP..)
> 

Neat.


>> @@ -76,6 +41,11 @@ int auth_get_tuple_for_subscr(struct gsm_auth_tuple *atuple,
>>  {
>>  	struct gsm_auth_info ainfo;
>>  	int i, rc;
>> +	static struct osmo_sub_auth_data auth = {
>> +	    .type = OSMO_AUTH_TYPE_GSM
>> +	};
> 
> Why is that static?
> 

Probably crawled from static functions I've replaced :)
It's not really needed here.

>>  
>> +	memcpy(auth.u.gsm.ki, ainfo.a3a8_ki, sizeof(auth.u.gsm.ki));
>> +
>> +	if (osmo_auth_gen_vec(vec, &auth, atuple->rand) < 0)
>> +		return -1;
>> +
>> +	memcpy(atuple->sres, vec->sres, 4);
>> +	memcpy(atuple->kc, vec->kc, 8);
> 
> in terms of API, how hard would be an in-situ operation?
> 

Ideally we should use osmocom's auth_vec directly instead of our own atuple. But this
change would be more intrusive and I recall Sylvain been working on something like
that so I've decided to go for smaller patch which does the job of using proper API.

-- 
best regards,
Max, http://fairwaves.co




More information about the OpenBSC mailing list