Re: Random numbers (again) broken on 64-bit platforms

[Andy Wingo]

Hi,

On Sun 01 Aug 2010 20:44, Andreas Rottmann <address@hidden> writes:

> I've spotted an issue with your fix: the way 64-bit random numbers are
> generated in scm_random() is bogus;

Grr, indeed. Fix applied, thanks.

>> Would you be interested in [importing another rng]? We would need
>> some test suites too, I think, and possibly changes to the scm_t_rng
>> structure.
>>
> Sorry, I don't have the inclination to work on this ATM. Also, random
> number tests are kinda hard to write -- it's random stuff, after all
> :-).

Too bad, we could really use it, and now is the time for any needed ABI
breaks. Thanks anyway for the fix!

Andy
-- 
http://wingolog.org/