make check failure on mainline CVS

[Andreas Jaeger]

Graham Stott <graham.stott@btinternet.com> writes:

> Andreas,
>
> Investigating further I've found that the problem is that the GC collector
> kicks in when the entries on the pending_attributes tree-list as returned
> by get_pending_attributes are not accessable via any gc roots hence they
> are subject to collection by GC.
>
> A quick hack would be to push and pop the GC context at the begining and
> end of the  build_unit_elab routine. Such a hack does indeed fix the
> failure.

But is this a proper solution?  Can you ask on the main list about
this?

> Graham
> ps. the GC collector got invoked during a call to expand_expr_stmt within
> build_uint_elab.

Thanks for investigating this,

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.inka.de
    http://www.suse.de/~aj