kgdb support for gdb
Amit S. Kale
amitkale@linsyssoft.com
Tue Oct 19 12:55:00 GMT 2004
On Monday 04 Oct 2004 8:44 pm, Andrew Cagney wrote:
> > Index: src/gdb/remote.c
> > ===================================================================
> > --- src.orig/gdb/remote.c 2004-09-25 01:21:18.000000000 +0530
> > +++ src/gdb/remote.c 2004-10-01 12:58:49.000000000 +0530
> > @@ -271,12 +271,13 @@
> >
> > /* Default maximum number of characters in a packet body. Many
> > remote stubs have a hardwired buffer size of 400 bytes
> > - (c.f. BUFMAX in m68k-stub.c and i386-stub.c). BUFMAX-1 is used
> > + (c.f. BUFMAX in m68k-stub.c and i386-stub.c).
> > + BUFMAX is defined as 1024 in most kgdb stubs. BUFMAX-1 is used
> > as the maximum packet-size to ensure that the packet and an extra
> > NUL character can always fit in the buffer. This stops GDB
> > trashing stubs that try to squeeze an extra NUL into what is
> > already a full buffer (As of 1999-12-04 that was most stubs. */
> > - rs->remote_packet_size = 400 - 1;
> > + rs->remote_packet_size = 1024 - 1;
> >
> > /* Should rs->sizeof_g_packet needs more space than the
> > default, adjust the size accordingly. Remember that each byte is
>
> As the comment explains, the change is wrong. Can I suggest starting a
> new thread on gdb@ explaining the underlying problem that you're trying
> to fix here (however, can you also delay it a bit, gdb@'s currently
> dominated by MI questions).
This problem went away after changing from "target i386-lk" so "osabi
Linux-kernel". i386-linux target defines all the registers etc. correctly, so
the problem doesn't appear.
-Amit
More information about the Gdb-patches
mailing list