10 GCC regressions, 6 new, with your patch on 2003-03-10T18:57:36Z.

[Geoff Keating]

"GCC regression checker" <geoffk@apple.com> writes:

> With your recent patch, GCC has some regression test failures, which
> used to pass.  There are 6 new failures, and 4
> failures that existed before and after that patch; 0 failures
> have been fixed.
> 
> The new failures are:
> native gcc.sum gcc.c-torture/compile/20001123-2.c,
> native gcc.sum gcc.c-torture/execute/930603-2.c
> native gcc.sum gcc.c-torture/execute/930628-1.c
> native gcc.sum gcc.c-torture/execute/divconst-2.c
> native gcc.sum gcc.c-torture/unsorted/loop-1.c,
> native gcc.sum gcc.c-torture/unsorted/xorn.c,

> +2003-03-10  Josef Zlomek  <zlomekj@suse.cz>
> +
> +	* cfgcleanup.c (outgoing_edges_match): Compare the jump tables. 
> +	(try_crossjump_to_edge): Replace refereces to one jump table by
> +	references to identical jump table.
> +	* loop.c (load_mems): Moved setting the JUMP_LABEL to replace_label.
> +	(replace_label): Moved to rtlanal.c.
> +	(struct rtx_pair): Moved to rtl.h.
> +	* rtl.h (struct rtx_pair): Moved from loop.c.  
> +	(replace_label): New extern function.
> +	(subrtx_p): New extern function.
> +	(tablejump_p): New extern function.
> +	* rtlanal.c (replace_label): Moved from loop.c.  
> +	(subrtx_p_1): New static function.
> +	(subrtx_p): New function.
> +	(tablejump_p): New function.
> +
> +Mon Mar 10 15:30:36 CET 2003  Jan Hubicka  <jh@suse.cz>
> +
> +	* cfgcleanup.c (merge_blocks): Return where to iterate next.
> +	(try_optimize_cfg): Use return value of merge_blocks

Josef, Jan, I'm sure this is one of your cfgcleanup.c patches.  The
symptom is a RTL-checking failure.  There doesn't seem to be a switch
statement involved, so I'm thinking Jan's patch.

-- 
- Geoffrey Keating <geoffk@geoffk.org>