> I put out this patch a week ago, I don't know if it will so much as compile,
> but you can hack your work above with the __int64 stuff and I think we should
> be all rounded out :-)

Whoops, I must have missed that post.
  
Okay, this works on Solaris, FreeBSD, and Linux.  

I think we are getting closer.  Comments?  

I also tried to add support for the INT64_C macro.  Ideally, INT64_C (if 
present) is always right - so the preference is to use INT64_C.  But, I guess 
it is possible for APR to make a different decision.

So, assume stdint.h says:

#define INT64_C(c) c ## LL

This would use a long long, but let's also assume that sizeof(int) == 8 - 
this patch would generate somewhere: 

typedef int apr_int64_t;
#define APR_INT64_C(c) INT64_C(c)

rather than the slightly more intuitive:

typedef int apr_int64_t;
#define APR_INT64_C(c) c

That may not be ideal.  But, this isn't a big deal and can be tweaked.

Wouldn't shock me if I get the HAVE v. HAS thing wrong...  -- justin