Partially fix hppa64-hpux bootstrap, 2/2

[John David Anglin]

> Thanks for picking up the ball here.  I was going to say that an
> unmodified tree now gets as far as _absvsi2 and then crashes, but I
> see that that's addressed by the patch you cite.  Your patch seems a
> plausible fix, but I'd like Jeff's opinion - it may be that a more
> appropriate fix is to change the varray to allocate using xmalloc.


Agreed.  I think it depends on how important it is to free the memory
allocated by the function.  Possibly, using NOGC and VARRAY_FREE would
also work.

Dave
-- 
J. David Anglin                                  dave.anglin@nrc-cnrc.gc.ca
National Research Council of Canada              (613) 990-0752 (FAX: 952-6602)