--0-566846030-1043167343=:23613
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit


clnp_raw.c is part of the kernel, and I need the kernel to coredump at a specific point in its execution.  Basically I want to edit clnp_raw.c to cause a coredump and then recompile the kernel with the modified clnp_raw.c
of if someone has code that just grabs the whole of memory and writes it to a file...
I can't find debugger in the man pages.
I need some more detailed help...  I saw the panic() command but couldn't get it to work, I'm just too stupid to figure it out.
Philip
 Jachym Holecek <freza@psi.cz> wrote:


If you want to force userland coredump, use abort(3). To
force a kernel crashdump, I would invoke Debugger(); and
use its sync command.

HTH,
-- Jachym Holecek

On Tue, Jan 21, 2003 at 04:06:57PM +0000, Philip Christian wrote:
> I have been trying to sendto() using a CLNP RAW_SOCK
> and failing.
> 
> So I started messing with the OSI stack and
> recompiling to see what is going on.
> 
> Basically when sendto() is invoked it is calling
> clnp_raw.c but does not seem to pass the OSI address
> specified in sendto() onto clnp_raw.c properly.
> 
> If I manually put an address into clnp_raw.c and
> recompile then a CLNP packet is then actually sent.
> 
> As the address structure is pretty complicated, with
> structures of pointers pointing to structures of
> pointers etc I think that the only way that I am going
> to figure out what is going on is to do invoke a
> memory dump inside clnp_raw.c so that I can see where
> the bytes have gone...
> 
> Can someone give me a fragment of code that will cause
> a coredump ?




---------------------------------
With Yahoo! Mail you can get a bigger mailbox -- choose a size that fits your needs

--0-566846030-1043167343=:23613
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

<P>clnp_raw.c is part of the kernel, and I need the kernel to coredump at a specific point in its execution.&nbsp; Basically I want to edit clnp_raw.c to cause a coredump and then recompile the kernel with the modified clnp_raw.c
<P>of if someone has code that just grabs the whole of memory and writes it to a file...
<P>I can't find debugger in the man pages.
<P>I need some more detailed help...&nbsp; I saw the panic() command but couldn't get it to work, I'm just too stupid to figure it out.
<P>Philip
<P>&nbsp;<B><I>Jachym Holecek &lt;freza@psi.cz&gt;</I></B> wrote:
<BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">
<P><BR><BR>If you want to force userland coredump, use abort(3). To<BR>force a kernel crashdump, I would invoke Debugger(); and<BR>use its sync command.<BR><BR>HTH,<BR>-- Jachym Holecek</P>
<P>On Tue, Jan 21, 2003 at 04:06:57PM +0000, Philip Christian wrote:<BR>&gt; I have been trying to sendto() using a CLNP RAW_SOCK<BR>&gt; and failing.<BR>&gt; <BR>&gt; So I started messing with the OSI stack and<BR>&gt; recompiling to see what is going on.<BR>&gt; <BR>&gt; Basically when sendto() is invoked it is calling<BR>&gt; clnp_raw.c but does not seem to pass the OSI address<BR>&gt; specified in sendto() onto clnp_raw.c properly.<BR>&gt; <BR>&gt; If I manually put an address into clnp_raw.c and<BR>&gt; recompile then a CLNP packet is then actually sent.<BR>&gt; <BR>&gt; As the address structure is pretty complicated, with<BR>&gt; structures of pointers pointing to structures of<BR>&gt; pointers etc I think that the only way that I am going<BR>&gt; to figure out what is going on is to do invoke a<BR>&gt; memory dump inside clnp_raw.c so that I can see where<BR>&gt; the bytes have gone...<BR>&gt; <BR>&gt; Can someone give me a fragment of code that will cause<BR>&gt; a co
 redump ?</P></BLOCKQUOTE><p><p><br><hr size=1><a href="http://uk.yahoo.com/mail/tagline_xtra/?http://uk.docs.yahoo.com/mail_storage.html"><b><font face="Arial" size="2">With Yahoo! Mail you can get a bigger mailbox -- choose a size that fits your needs</font></b></a><br>
--0-566846030-1043167343=:23613--