hi harald,
sorry for not getting back earlier. I honestly did not realize that
there
was a second attachment to your mail. We typically use inline
patches, and one
patch per mail on this list. Once again, sorry for the delay.
next time i will send the patches inline...
I'm now looking at the code in detail, I shall be getting back to you
at some
point during the next days.
in order to remove md5 code, you can use a simple XOR to mix the input data. i think it is good enough to not select the same random number as the base station.
regards,
andreas
Hi Andreas,
On Fri, Oct 30, 2009 at 01:29:36PM +0100, Andreas.Eversberg wrote:
in order to remove md5 code, you can use a simple XOR to mix the input data. i think it is good enough to not select the same random number as the base station.
I think we could use the random() / srandom() functions, which are standard POSIX API. We could seed it with something like the time it takes us to receive the first location update request after boot, or something unpredictable like that.
In the end, we have at least three parts in our system that need a 32bit random value: The TMSI, P-TMSI and the RTP source ID
Regards, Harald
-
Andreas.Eversberg -
Harald Welte