[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: AX.25 Debian bug report



Hi Ralf,

Damn, I thought the source code reading times where over ;-)

Back to business, I am not sure what is happening in glibc. I only
know one thing, the <netax5/ax25.h>, <netrom/netrom.h> and
<netrose/rose.h> are not the same as the ones from the 2.6 kernel in
<linux/ax25.h>, <linux/netrom.h> and <linux/rose.h>. A quick scan of
the Makefile concluded me that these files in glibc where linked to
<bits/syscall.h> or another one.

Perhaps I am naive and a purist, but the files are incorrect. This
concluded for me that the files in glibc need to updated. As a
consequence the libax25 and its dependencies, needs also an update.

At this moment my system is running stable for 4 days, with a 2.4
kernel. With 2.6 kernel it crashed after more or less 16 hours.

Returning on the source code reading! Random call checking in libax25
and the kernel concluded me that most things are in order, accept
netromd and nrattach.

$ netromd uses obsolete (PF_INET,SOCK_PACKET)

The biggest change I found between 2.4 and 2.6 are in the mkiss and
bqpether module.

So I think, I have to file a new bug report that the mkiss module is
not stable in my situation. Where it is used as a loop device through
internet.

I am running an etch system, with all the ham related packages. The
only thing I changed is that I installed the last 2.4 kernel-image
from sarge, with its dependencies from etch.

Greetings, Arjan

On Thu, May 1, 2008 at 3:30 PM, Ralf Baechle <ralf@linux-mips.org> wrote:
> Hi Arjan,
>
>  regarding http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=478693 - can
>  you elaborate on the problem?  What I read on the Debian bugtracker
>  doesn't make very much sense to me, I'm afraid.  What you're writing
>  reads like backward compatibility breakage - but we haven't changed
>  any interfaces.
>
>  In particular the ax25. headers are not involved in generation of syscall.h
>  (I assume you actually mean <bits/syscall.h>; <syscall.h> is not
>  generated and trivial) and anyway AX.25, NETROM or ROSE do not have their
>  own syscalls.
>
>  73 de DL5RB op Ralf
>
>  ---
>  DL5RB / M0LRB / AB1IN
>  Loc.    JN47BS / CQ 14 / ITU 28 / DOK A21
>  Packet: DL5RB@DB0FHN.#BAY.DEU.EU
>


Reply to: