[prev in list] [next in list] [prev in thread] [next in thread] 

List:       xen-ppc-devel
Subject:    [XenPPC] GDB Summary: Was: warning: gpproxy does not work with
From:       Jimi Xenidis <jimix () watson ! ibm ! com>
Date:       2006-06-27 9:46:39
Message-ID: EF8F030C-16AE-4C2A-8965-B79E86FB4251 () watson ! ibm ! com
[Download RAW message or body]

On Jun 26, 2006, at 7:07 PM, Jimi Xenidis wrote:

First off, I had a really old GDB, one I built from 2003-08-12-cvs,  
but I upgraded my gcc (gcc (GCC) 4.1.2 20060613 ) and the result was  
not debuggable, (auto variables were KAKA).
So I grabbed the gdb CVS head (GNU gdb 6.5.50.20060626-cvs) and gave  
it a try.

> they seem to have changed the protocol quite a bit and I don't have  
> the python-kung-fu to make it happy.

Thankfully, my gdb-kung-fu _is_ up to snuff :) and it is possible to  
tell gdb what and what not to use.  To continue to use gpproxy stick  
these into your .gdbinit:

   set remote X-packet 0
   set remote read-aux-vector-packet 0
   set remote symbol-lookup-packet 0
   set remote verbose-resume-packet 0
   set remote Z-packet 1
   set remote get-thread-local-storage-address-packet 0
   set remote supported-packets-packet 0
   set remotetimeout 5

It is way, way, WAY faster now, gdb is more intelligent about the  
remote commands that it uses and "next" and "step" command which were  
painfully slow work at a much more reasonable speed.

Oh yeah, my debugging makes sense and the KAKA has gone away.

-JX

_______________________________________________
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel
[prev in list] [next in list] [prev in thread] [next in thread]