Kernel freeze on killall kissattach

[Roland Schwarz <roland.schwarz@blackspace.at>]

[-- Attachment #1.1: Type: text/plain, Size: 1414 bytes --]

Dear All,

please correct me if I am wrong, but as I understand it should not be 
possible to trigger a kernel freeze by just killing user space processes.

While trying to understand a certain behavior of the ax25 driver I came 
up with a small test case which I ran in a virtual machine (KVM):

First I tried with a kernel from Linus tree:

uname -r
5.17.0-rc6+

cat /etc/ax25/axports
ax0     OE1RSA-0        115200  255     7       Test A
ax1     OE1RSA-1        115200  255     7       Test B

sudo kissattach /dev/ptmx ax0
AX.25 port ax0 bound to device ax0
Awaiting client connects on
/dev/pts/1

sudo kissattach /dev/pts/1 ax1
AX.25 port ax1 bound to device ax1

axcall ax1 OE1RSA-0

Now in a second terminal:
sudo killall kissattach

First terminal:
GW4PTS AX.25 Connect v1.11
Trying...
*** Connected to OE1RSA-0
*** Cleared

==> But: Kernel freezes

I can see this behaviour also in the debian kernel:

uname -r
5.10.0-11-amd64

This kernel also freezes. I suspect this to be a bug because if I try to 
reboot the machine instead of the killall command the machine does not 
shut off and reboot which is a problem with a remotely installed system.

Any ideas?

73 de Roland oe1rsa

-- 
__________________________________________
   _  _  | Roland Schwarz
  |_)(_  |
  | \__) | mailto:roland.schwarz@blackspace.at
________| http://www.blackspace.at

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 840 bytes --]