This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [RFA/RFC] dwarf2-frame read_reg
- From: Daniel Jacobowitz <drow at false dot org>
- To: Michael Snyder <msnyder at redhat dot com>
- Cc: gdb-patches at sourceware dot org, Jim Blandy <jimb at red-bean dot com>
- Date: Fri, 5 May 2006 15:44:29 -0400
- Subject: Re: [RFA/RFC] dwarf2-frame read_reg
- References: <443C7529.7020206@redhat.com>
On Tue, Apr 11, 2006 at 08:34:01PM -0700, Michael Snyder wrote:
> I want you guys to vett this change. I was getting wrong results
> on a target where sizeof (SP) != sizeof (void *). The local func
> read_reg was calling extract_unsigned_integer with the wrong size.
> 2006-04-11 Michael Snyder <msnyder@redhat.com>
>
> * dwarf2-frame.c (read_reg): Use register type instead of
> builtin_data_pointer_type to extract register's value.
I think the thread consensus was to do the attached. I've tested it on
x86_64, and I'm fairly confident in it. Any concerns? (Concerns that
come with test results for some other platform are the best kind!)
--
Daniel Jacobowitz
CodeSourcery
2006-05-05 Daniel Jacobowitz <dan@codesourcery.com>
* dwarf2-frame.c: Include "value.h".
(read_reg): Use unpack_long and register_type.
* Makefile.in (dwarf2-frame.o): Update.
Index: Makefile.in
===================================================================
RCS file: /cvs/src/src/gdb/Makefile.in,v
retrieving revision 1.811
diff -u -p -r1.811 Makefile.in
--- Makefile.in 23 Apr 2006 14:15:01 -0000 1.811
+++ Makefile.in 5 May 2006 19:41:34 -0000
@@ -1907,7 +1907,8 @@ dwarf2expr.o: dwarf2expr.c $(defs_h) $(s
dwarf2-frame.o: dwarf2-frame.c $(defs_h) $(dwarf2expr_h) $(elf_dwarf2_h) \
$(frame_h) $(frame_base_h) $(frame_unwind_h) $(gdbcore_h) \
$(gdbtypes_h) $(symtab_h) $(objfiles_h) $(regcache_h) \
- $(gdb_assert_h) $(gdb_string_h) $(complaints_h) $(dwarf2_frame_h)
+ $(gdb_assert_h) $(gdb_string_h) $(complaints_h) $(dwarf2_frame_h) \
+ $(value_h)
dwarf2loc.o: dwarf2loc.c $(defs_h) $(ui_out_h) $(value_h) $(frame_h) \
$(gdbcore_h) $(target_h) $(inferior_h) $(ax_h) $(ax_gdb_h) \
$(regcache_h) $(objfiles_h) $(exceptions_h) $(elf_dwarf2_h) \
Index: dwarf2-frame.c
===================================================================
RCS file: /cvs/src/src/gdb/dwarf2-frame.c,v
retrieving revision 1.59
diff -u -p -r1.59 dwarf2-frame.c
--- dwarf2-frame.c 5 Apr 2006 20:01:19 -0000 1.59
+++ dwarf2-frame.c 5 May 2006 19:41:35 -0000
@@ -32,6 +32,7 @@
#include "symtab.h"
#include "objfiles.h"
#include "regcache.h"
+#include "value.h"
#include "gdb_assert.h"
#include "gdb_string.h"
@@ -214,7 +215,13 @@ read_reg (void *baton, int reg)
buf = alloca (register_size (gdbarch, regnum));
frame_unwind_register (next_frame, regnum, buf);
- return extract_typed_address (buf, builtin_type_void_data_ptr);
+
+ /* Convert the register to an integer. This returns a LONGEST
+ rather than a CORE_ADDR, but unpack_pointer does the same thing
+ under the covers, and this makes more sense for non-pointer
+ registers. Maybe read_reg and the associated interfaces should
+ deal with "struct value" instead of CORE_ADDR. */
+ return unpack_long (register_type (gdbarch, regnum), buf);
}
static void