Re: XView bug/PPC machines available for developers?
> > > Huh? Your mileage WILL vary. That defeats the entire point of using
> > > __va_copy(). avlist1 is, I assume, a va_list; a va_list is an array
> > > type, and so *avlist1 means avlist1[0], which does not dereference an
> > > uninitialized pointer. This patch is wrong.
> >
> > avlist is _not_ a va_list. Please look up the definition of avlist in
> > lib/libxview/attr/attr.h yourself. avlist is a pointer to some u32 chunk
> > of data, either a pointer to a avlist attribute or value IIRC.
> > As such, __va_copy just so happens to work for other archs by accident.
>
> *sigh*
>
> Sorry, Michael, I'm too trigger happy. You're right; I bet I
> introduced this bug myself.
No sweat - and you didn't introduce the bug from what I've seen on the
BTS, it looks like Hartmut did when he did the glibc2.1 fixes for xview.
See #72387 ...
> > Feel free to point out a URL (searchable Debian ML archives would be nice
> > to have, sometimes).
>
> Sorry, don't have it any more :(
I'll dig around for it in the archive. I'm still looking for a trick to
make passing of va_lists to other functions work portably...
Michael
Reply to: