[RFC] dwarf2_read_address(): sign extend as appropriate

[Kevin Buettner <kevinb at redhat dot com>]

Any comments regarding the patch below?

I wrote it last summer, so my memory about it is rather hazy aside
from the fact that it fixed a bunch of C++ frame base problems on
mips64 targets.

It shouldn't affect any architecture other than mips because mips is
the only one which defines an integer_to_address method.  It also
addresses Andrew's FIXME comment from nearly four years ago...

	* dwarf2expr.c (unsigned_address_type): Add forward declaration.
	(dwarf2_read_address): Sign extend return address as required by
	target architecture.

Index: dwarf2expr.c
===================================================================
RCS file: /cvs/src/src/gdb/dwarf2expr.c,v
retrieving revision 1.19
diff -u -p -r1.19 dwarf2expr.c
--- dwarf2expr.c	9 Jan 2007 17:58:50 -0000	1.19
+++ dwarf2expr.c	20 Apr 2007 23:23:50 -0000
@@ -33,6 +33,7 @@
 
 static void execute_stack_op (struct dwarf_expr_context *,
 			      gdb_byte *, gdb_byte *);
+static struct type * unsigned_address_type (void);
 
 /* Create a new context for the expression evaluator.  */
 
@@ -205,9 +206,24 @@ dwarf2_read_address (gdb_byte *buf, gdb_
     error (_("dwarf2_read_address: Corrupted DWARF expression."));
 
   *bytes_read = TARGET_ADDR_BIT / TARGET_CHAR_BIT;
-  /* NOTE: cagney/2003-05-22: This extract is assuming that a DWARF 2
-     address is always unsigned.  That may or may not be true.  */
-  result = extract_unsigned_integer (buf, TARGET_ADDR_BIT / TARGET_CHAR_BIT);
+
+  /* Some architectures (such as MIPS) use signed addresses.  Those that
+     do will have registered a gdbarch_integer_to_address method.  Use
+     that method if it exists, otherwise fall back to extracting an
+     unsigned integer as the address.
+     
+     Note:  kevinb/2006-07-10:  The use of `unsigned_address_type' in
+     the gdbarch_integer_to_address() call below refers to the type of
+     `buf' and has no bearing on the signedness of the address being
+     returned.  In all of the integer-to-address conversion methods
+     being used by GDB at the time that this comment was written, this
+     type was used only to determine the size of `buf'.  */
+  if (gdbarch_integer_to_address_p (current_gdbarch))
+    result = gdbarch_integer_to_address (current_gdbarch,
+                                         unsigned_address_type(), buf);
+  else
+    result = extract_unsigned_integer (buf,
+                                       TARGET_ADDR_BIT / TARGET_CHAR_BIT);
   return result;
 }