Re: libltdl & 64-bit lint

[Gary V. Vaughan]

On Sun, Sep 30, 2001 at 11:05:52PM -0500, address@hidden wrote:
> 2001-09-30  Albert Chin-A-Young <address@hidden>
> 
>         * ltdl.c: change some types to size_t from int because
>         strlen() returns size_t. argz_len changed to size_t in
>         foreach_dirinpath() because argzize_path() takes 3rd
>         argument as size_t, not int. Based on lint run from
>         Bob Friesenhahn <address@hidden>.
> 
> On Sun, Sep 30, 2001 at 11:04:35AM -0500, Bob Friesenhahn wrote:
> > Here is the output from Sun's 64-bit lint tool when run on libltdl
> > current as of September 29. You will notice that there are a number of
> > type down-conversions going on when the code is 64-bit.  In order to
> > be safe, these types should jive for both the ILP32 and LP64 data type
> > models.
> > 
> > (390) warning: suspicious comparison of unsigned with 0: op "<="
> 
> static lt_ptr
> realloc (ptr, size)
>      lt_ptr ptr;
>      size_t size;
> {
>   if (size <= 0)
> 
> Is size_t always unsigned?

Yup.  This must be a result of careless code tweaking on my part.  I've
made this into `==' in addition to the changes in your patch.

I wonder, should we also test for a system definition of size_t at
configure time?  And presumably substitute for `unsigned long' if the
system definition is missing?

> > (1994) warning: cast from 64-bit integer to 32-bit integer
> 
>   sprintf (filename, "%.*s/%s", (int) dirname_len, dirname, dlname);
> 
> According to sprintf(3) on a Linux box, when using %*, the field width
> must be of type int.

Agreed.

> > (2643) warning: comparing 32-bit integer with 64-bit integer
> 
>       for (i = 0; i < ext - base_name; ++i)
> 
> ext and base_name are char *. Should we make i a size_t?

Seems like the right thing to do from here.  I'm also adding this change
to your patch.

> > (2718) warning: passing 64-bit integer arg, expecting 32-bit integer: 
> > fgets(arg 2)
> 
>           if (!fgets (line, line_len, file))
> 
> line_len is size_t. fgets expects second argument to be int.

Hmmm.  That sucks.  I'll cast it down to an int for now.  But if someone
has the time to implement an rpl_fgets in the spirit of the other
replacement functions near the top of ltdl.c, I'd be very grateful...

Appying to HEAD and branch-1-4.

Cheers,
        Gary.
-- 
  ())_. Gary V. Vaughan     gary@(oranda.demon.co.uk|gnu.org)
  ( '/  Research Scientist  http://www.oranda.demon.co.uk       ,_())____
  / )=  GNU Hacker          http://www.gnu.org/software/libtool  \'      `&
`(_~)_  Tech' Author        http://sources.redhat.com/autobook   =`---d__/