Regression for trunk on athlon_mp-pc-linux-gnu

Jan Hubicka jh@suse.cz
Sat Sep 6 21:30:00 GMT 2003


> One or more new FAIL's have appeared for athlon_mp-pc-linux-gnu:
>     g++.sum g++.dg/ext/typeof2.C
>     g++.sum g++.dg/opt/inline4.C
> +2003-09-02  Geoffrey Keating  <geoffk@apple.com>
> +
> +	PR 12161
> +	* decl2.c (mark_used): Use ggc_push_context/ggc_pop_context.
> +	* tree.c (cp_cannot_inline_tree_fn): Likewise.

Geoff,
at least the inline4.c failure appears to be related to your change
(we get very deep recursion on mark_used and get out of the maximal
context depth).  I am not sure what you were shooting for and how to
avoid this,  so would be possible for you to look into this, please?

The typeof2.C was mine, I will wait before looking at the others.

Honza
> +
> +2003-09-04  Nathan Sidwell  <nathan@codesourcery.com>
> +
> +	* cp-tree.h (finish_sizeof, finish_alignof): Remove.
> +	(expr_sizeof): Replace with ...
> +	(cxx_sizeof_or_alignof_expr): ... here.
> +	(cxx_sizeof_or_alignof_type): Make complain parameter a bool.
> +	* parser.c (cp_parser_unary_expression): Commonize alignof and
> +	sizeof handling.
> +	* pt.c (tsubst_copy_and_build): Adjust alignof and sizeof
> +	substitution.
> +	* semantics.c (finish_sizeof, finish_alignof): Remove.
> +	* typeck.c (cxx_sizeof_or_alignof_type): Complain parameter
> +	becomes bool. Set TREE_READONLY.
> +	(expr_sizeof): Replace with ...
> +	(cxx_sizeof_or_alignof_expr): ... here. Clear TREE_SIDE_EFFECTS.
> +
>  2003-09-04  Mark Mitchell  <mark@codesourcery.com>
> diff -u1r previous/CL/gcc/testsuite/ChangeLog current/CL/gcc/testsuite/ChangeLog
> --- previous/CL/gcc/testsuite/ChangeLog	2003-09-04 04:30:10.000000000 -0400
> +++ current/CL/gcc/testsuite/ChangeLog	2003-09-05 04:30:16.000000000 -0400
> @@ -1 +1,7 @@
> +2003-09-04  Matt Austern  <austern@apple.com>
> +
> +	* g++.dg/ext/fnname1.C: New test. (__func__ for C++.)
> +	* g++.dg/ext/fnname2.C: Likewise.
> +	* g++.dg/ext/fnname3.C: Likewise.
> +	
>  2003-09-04  Mark Mitchell  <mark@codesourcery.com>
> 



More information about the Gcc-regression mailing list