On Tue, Oct 15, 2024 at 08:11:02PM +0200, Patrice Dumas wrote:
> > diff --git a/info/util.c b/info/util.c
> > index 7ecb90f06b..a511471b9f 100644
> > --- a/info/util.c
> > +++ b/info/util.c
> > @@ -34,9 +34,12 @@ xvasprintf (char **ptr, const char *template, va_list ap)
> > int
> > xasprintf (char **ptr, const char *template, ...)
> > {
> > + int ret;
> > va_list v;
> > va_start (v, template);
> > - return xvasprintf (ptr, template, v);
> > + ret = xvasprintf (ptr, template, v);
> > + va_end (v);
> > + return ret;
> > }
>
> If the above change is relevant (looks like so to me), we should
> probably do it in texi2any C code that is very similar too.
The question is do we actually need to use va_end?
>From the libc Info manual:
In the GNU C Library, ‘va_end’ does nothing, and you need not ever
use it except for reasons of portability.
and
(In practice, with most C compilers, calling ‘va_end’ does nothing.
This is always true in the GNU C compiler. But you might as well
call ‘va_end’ just in case your program is someday compiled with a
peculiar compiler.)
So I suppose we should use va_end throughout, if only to avoid anyone
worrying about it in the future.
