Re: [Patch, fortran] PR25289 Cannot handle record numbers large than huge(0_4)

[Jerry DeLisle <jvdelisle at verizon dot net>]

:REVIEWMAIL:

FX Coudert wrote:
> > In libgfortran you can see that gfc_offset is defined in terms of 
> > off_t and it is 64 bits there.  However, at the point of interest in 
> > trans-io.c, off_t is int4 and not int8.
> >
> > I found this out trying gfc_get_int_type (sizeof(off_t)) which fails 
> > miserably and in -fdump-tree-original you can see that rec is assigned 
> > (int4) n, with the int4 cast.
>
> [...]
>  > If I use gfc_get_int_type (sizeof(off64_t)) all appears to work fine.
>  > Is it safe to assume off64_t is defined for all targets of interest?
>
> You cannot do that kind of things anyhow, because the host off64_t 
> (which will be used when the front-end is compiled) might not have the 
> same size as the target off64_t (which will be used for the library). On 
> that topic, you can e.g. read the thread that lead to 
> http://gcc.gnu.org/ml/fortran/2005-12/msg00136.html According to Richard 
> Henderson, that problem is solvable 
> (http://gcc.gnu.org/ml/fortran/2005-12/msg00142.html), but I'm not sure 
> if he's in possession of a constructive proof of existence ;-)

I remember reading your post on the subject.  Now I understand first hand what 
the problem is.  :o)

> One possibility it to define a gfc_large_int to be integer(kind=8) if it 
> exists, and integer(kind=4) otherwise (and do the same in the library).

There are int32_t and int64_t defined.  Maybe use those.  I will think about 
this some more and see if there are any other comments.

Thanks,

Jerry