This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [RFA/stabs] Fix gdb/280 infinite loop in userdef.exp
- From: Jim Blandy <jimb at zwingli dot cygnus dot com>
- To: Daniel Jacobowitz <drow at mvista dot com>
- Cc: gdb-patches at sources dot redhat dot com
- Date: 03 Feb 2002 17:25:30 -0500
- Subject: Re: [RFA/stabs] Fix gdb/280 infinite loop in userdef.exp
- References: <20020201185437.A26371@nevyn.them.org>
Approved. Could you also put a comment in the stabs reader saying
that this isn't really a kosher way to construct types?
Daniel Jacobowitz <drow@mvista.com> writes:
> Another problem exposed by the way the stabs reader handles types. It does
> '*type = *xtype' at one point, which dies horribly for qualified types (and
> messes up pointer types, too...). I don't immediately see a way to get rid
> of that line, so I made an interface to do the copy correctly.
>
> For reference, that appears to be the only place in GDB this is actually
> done. Other readers must handle forward-referenced types somehow...
>
> Without this patch we enter an infinite loop because the cv-chain is
> corrupted.
>
> OK to commit?
>
> --
> Daniel Jacobowitz Carnegie Mellon University
> MontaVista Software Debian GNU/Linux Developer
>
> 2002-02-01 Daniel Jacobowitz <drow@mvista.com>
>
> PR gdb/280
> * gdbtypes.c (replace_type): New function.
> * gdbtypes.h (replace_type): Add prototype.
> * stabsread.c (read_type): Use replace_type.
>
> Index: gdbtypes.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/gdbtypes.c,v
> retrieving revision 1.38
> diff -u -p -r1.38 gdbtypes.c
> --- gdbtypes.c 2002/01/22 19:57:40 1.38
> +++ gdbtypes.c 2002/02/01 23:48:45
> @@ -519,6 +519,32 @@ finish_cv_type (struct type *type)
> }
> }
>
> +/* Replace the contents of ntype with the type *type.
> +
> + This function should not be necessary, but is due to quirks in the stabs
> + reader. This should go away. It does not handle the replacement type
> + being cv-qualified; it could be easily fixed to, but it should go away,
> + remember? */
> +void
> +replace_type (struct type *ntype, struct type *type)
> +{
> + struct type *cv_chain, *as_chain, *ptr, *ref;
> +
> + cv_chain = TYPE_CV_TYPE (ntype);
> + as_chain = TYPE_AS_TYPE (ntype);
> + ptr = TYPE_POINTER_TYPE (ntype);
> + ref = TYPE_REFERENCE_TYPE (ntype);
> +
> + *ntype = *type;
> +
> + TYPE_POINTER_TYPE (ntype) = ptr;
> + TYPE_REFERENCE_TYPE (ntype) = ref;
> + TYPE_CV_TYPE (ntype) = cv_chain;
> + TYPE_AS_TYPE (ntype) = as_chain;
> +
> + finish_cv_type (ntype);
> +}
> +
> /* Implement direct support for MEMBER_TYPE in GNU C++.
> May need to construct such a type if this is the first use.
> The TYPE is the type of the member. The DOMAIN is the type
> Index: gdbtypes.h
> ===================================================================
> RCS file: /cvs/src/src/gdb/gdbtypes.h,v
> retrieving revision 1.25
> diff -u -p -r1.25 gdbtypes.h
> --- gdbtypes.h 2002/01/20 19:42:04 1.25
> +++ gdbtypes.h 2002/02/01 23:48:45
> @@ -1062,6 +1062,8 @@ extern struct type *make_cv_type (int, i
>
> extern void finish_cv_type (struct type *);
>
> +extern void replace_type (struct type *, struct type *);
> +
> extern int address_space_name_to_int (char *);
>
> extern char *address_space_int_to_name (int);
> Index: stabsread.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/stabsread.c,v
> retrieving revision 1.23
> diff -u -p -r1.23 stabsread.c
> --- stabsread.c 2002/01/20 19:12:23 1.23
> +++ stabsread.c 2002/02/01 23:48:46
> @@ -2531,7 +2531,7 @@ again:
> }
> else if (type_size >= 0 || is_string)
> {
> - *type = *xtype;
> + replace_type (type, xtype);
> TYPE_NAME (type) = NULL;
> TYPE_TAG_NAME (type) = NULL;
> }