Enabling gdb-interface Z packet by default

Enabling gdb-interface Z packet by default

[Frank Ch. Eigler]

Hi -

I would like to commit the following patch to sid/bsp/configrun-sid.in,
intended to by default support the gdb remote protocol "Z" packet.  The
purpose of this change is to add some resilience against unilateral 
changes in gdb land to designated breakpoint instructions.  

Instead of overwriting instruction memory, gdb uses the Z packet to
give us a PC value, which in turn is translated to a sid triggerpoint
on the PC register.  This has the drawback that it slows down the
simulation, and requires gdb to supply a proper PC value (compensated
for word-based addressing, harvard architectures, etc.).  Still, methinks
being able to function despite some gdb changes is worthwhile.

Any comments?

- FChE


2001-04-02  Frank Ch. Eigler  <fche@redhat.com>

	* configrun-sid.in (enable_z_packet): Change default to "true".

 
Index: configrun-sid.in
===================================================================
RCS file: /cvs/cvsfiles/devo/sid/bsp/configrun-sid.in,v
retrieving revision 1.55
diff -u -r1.55 configrun-sid.in
--- configrun-sid.in	2001/03/27 16:40:12	1.55
+++ configrun-sid.in	2001/04/02 19:09:36
@@ -402,7 +402,7 @@
       &sidconf_new("sw-profile-gprof", "gprof") . "\n";
   }
 
-$enable_z_packet = "false";
+$enable_z_packet = "true";
 $second_section = "# second section
 # settings
 set cpu step-insn-count $cpu_insn_count

-- 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE6yNWGVZbdDOm/ZT0RAg7WAJ9bF+ywdpE5FkMimqoek9HMuMeoXgCghrbN
Ipb7hG65gLxZaJRnfcDfsgM=
=iPtD
-----END PGP SIGNATURE-----

Re: Enabling gdb-interface Z packet by default

[Ben Elliston]

>>>>> "Frank" == Frank Ch Eigler <fche@redhat.com> writes:

  Frank> I would like to commit the following patch to sid/bsp/configrun-sid.in,
  Frank> intended to by default support the gdb remote protocol "Z" packet.  The
  Frank> purpose of this change is to add some resilience against unilateral 
  Frank> changes in gdb land to designated breakpoint instructions.  

Can you elaborate on why this change is necessary?  Are changes
occuring in GDB land w.r.t. the remote protocol?

Ben