WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] [PATCH][RFC] Gdbserver-xen support for HVM Guests... getting

To: <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] [PATCH][RFC] Gdbserver-xen support for HVM Guests... getting further.
From: "Zulauf, John" <john.zulauf@xxxxxxxxx>
Date: Fri, 30 Nov 2007 16:50:23 -0800
Delivery-date: Fri, 30 Nov 2007 16:51:04 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: AcgztCep6XAJy0OTTqySYR4Ir/K1JA==
Thread-topic: [PATCH][RFC] Gdbserver-xen support for HVM Guests... getting further.
All,

below is a minimal patch that allows me to attach, stepi, x,
disassemble, break with a 64bit HVM guest (2MB pages).  The RFC is for
the 1st fix below, also, others testing/experience trying this patch
would be welcome.  

1st fix: Diasable the gpfn to mfn mapping of to_ma.  This seems to be
the right thing to do (at least as the map_domain_va_* functions are
using it), as xc_map_foreign_range takes gpfn's for HVM domains.  Yet,
the comments in the file indicate that the mapping in to_ma is important
and specifically so for page tables.  Using a small HVM domain with
known page tables, I can verify that the correct pages are *only* loaded
with the to_ma mapping disabled.  Can anyone explain?  Is there some
*other* conditional case in which this would be correct?
 
2nd fix:  2M page calculation -- this is a simple misplacement of
parentheses that was causing the page directory entry to be shifted left
12.

Testing -- able to attach to 64bit HVM domains and info registers,
stepi, break, disassemble, x etc.
Issues Remaining
        cannot ^C to stop a domain  (wait for domain never returns)
        
        must explicitly "set architecture i386:x86_64" in the gdb
session to connect to 64 bit remote targets
        
        attempting to print variable contents or addresses "n" through
hvmloader (built -g) 
        resulted in gdb error messages: 
                dwarf2_read_address: Corrupted DWARF expression


Patch Signed-off by John Zulauf john.zulauf@xxxxxxxxx


Attachment: ptrace_fix.diff
Description: ptrace_fix.diff

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] [PATCH][RFC] Gdbserver-xen support for HVM Guests... getting further., Zulauf, John <=