Re: Kernel freeze under heavy (transmit) load


From: Jouni Malinen (jkmaline_at_cc.hut.fi)
Date: 2002-06-18 07:46:51 UTC



On Fri, Jun 14, 2002 at 10:24:03PM +0100, Devaney, Matthew wrote:

> Looking back through this thread it seems that most of the people who have
> reported this problem are running SMP - coincidence?
>
> Jouni - if you're reading this =).. Any thoughts on what might be causing
> these freezes or any hints on what we can do to get more information about
> what's causing it?

Sure I'm reading, but this thread was delayed till today ;-).

My main development host happens to use SMP (with kernel.org 2.4.19-pre10) and I have been unable to freeze it under haevy load.. For some reason I cannot use PCI Prism2.5 cards in this host (they are not found at all), but I have both Compaq WL200 using external pcmcia-cs with hostap_cs and PLX-based card with hostap_plx in this setup.

Since I cannot duplicate the problem, I would need to get some data about the crashes. Do you get any messages from the kernel (when running in text console at the moment of freeze)? Does magic SysRQ work after the freeze? Have you tested disabling PRISM2_USE_CMD_COMPL_INTERRUPT in prism2_config.h?

-- 
Jouni Malinen                                            PGP id EFC895FA


This archive was generated by hypermail 2.1.4.