Re: [patch,fortran] Fix corner case where consecutive calls to date_and_time can return times out of sequence (PR30015)

[Brooks Moses <brooks dot moses at codesourcery dot com>]

Paul Thomas wrote:
> Tobias,
> > > The change by Mat Cross is therefore:
> > > - Moving the  gettimeofday() block up in the file
> > > - Using the result of gettimeofday() for lt and value[7], instead of
> > > using time() for lt and gettimeofday() for value[7]
> > >  date_and_time.c |  
> I am sorry but I have stared at the patch and the explanation and am 
> none the wiser :-)
>
> There is obviously something that I am missing in the mechanics of 
> time() and gettimeofday().  I need an explanation as to where this time 
> shifting comes from, I think.  Is it simply that time() and 
> gettimeofday() are different?  If so why?

I believe that the problem comes across because time() and 
gettimeofday() are separate function calls, made at slightly different 
times.  When the returned result is calculated by assembling pieces of 
the results from those two calls, it may produce a value which is not 
within the range of either if the wrong time-counter happens to roll 
over in the tiny span of time between the two function calls.

At least, that's my impression from the conversation; I haven't looked 
at the patch in any detail.

- Brooks