This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Bug in C++ tree inlining or in PPC backend?
Am Don, 13 Jan 2000 schrieb Mark Mitchell:
>Franz --
>
> Thanks for working further on this bug.
>
> I'm not sure your patch is correct, though; basically, if I read it
>right, it's making sure that a temporary whose address is taken inside
>an RTL-expression is preserved outside that RTL expression. But, it's
>like that should not be happenning.
Well, I see this happening (don't nail me down on the terms I use here):
1. a string 'object' #1 is created in stackslot X
2. more string objects are created, which eventually reuse stackslot X
3. the objects get used
4. object #1 gets deleted/destroyed. If stackslot X got used in 2., that causes
the corruption.
So I would say it's either the problem I tried to fix, or some kind of
'scheduling' problem with object creation/deletion?
> I think we still need to track this down further to figure out where
>exactly the bug lies.
Sure. So I'll continue to wade thru code I don't understand :-). I'll be away
til Sunday though.
Franz.