*BSD News Article 11167


Return to BSD News archive

Received: by minnie.vk1xwt.ampr.org with NNTP
	id AA1336 ; Tue, 23 Feb 93 14:35:49 EST
Newsgroups: comp.unix.bsd
Path: sserve!manuel.anu.edu.au!munnari.oz.au!uunet!UB.com!quack!dfox
From: dfox@quack.sac.ca.us (David Fox)
Subject: Re: [386bsd] npx-0.1 causes lockup in npxprobe() - now what? (
Message-ID: <fZiQfbX@quack.sac.ca.us>
Organization: The Duck Pond public unix: +1 408 249 9630, log in as 'guest'.
References: <9302121437.AA09426@styx.iqm.unicamp.br>
Date: 14 Feb 1993 09:20:58 UTC
Lines: 49

You bellowed:

>I applied npx-0.3-diff with the same results of David Fox, the system
>hangs on boot. Then I comented out the fp_divideby0() as sugested and 
>the new kernel booted ok but issued the follow message:
>
>npx0<Error reporting broken (no IRQ13)> at 0xf0 on isa

And, I got the same results as you.  I haven't yet gone in to uncomment
the fp_divideby0 part, though, but I might.  The code in npx-0.3-diff 
already has a portion of the irq13 stuff essentially commented out with
an #if 0, #endif, as you might have noticed.

I imagine if I comment out the fp_divideby0() again that it'll provide
the same results that I have now.  The other part of this quandary is that
I've recompiled the kernel with gcc-2.3.3, but my working kernel is still
compiled with gcc-1.39.  The first time I tried the new gcc-2.3.3 kernel,
the machine went into a reboot loop.  I got some patches, and they helped
that situation, but the thing still locks up hard in npxprobe.

>With this kernel I started my torture test: I put an molecular eletronic
>structure abinitio program to run in background, then started a rayshade
>and a raypaint of the same scene description file and checked if  the 

That's pretty tortuous.  I had two rayshade processes going on at the same
time myself.

>Then I called xv to display a gif image and the system hang :-(

Yes, and I've gotten some hangs in little things (od -f file is one, and
the fpetest supplied with the patch is another).  It seems if it has to
do any status-related stuff it'll hang, but in normal NDP operations it
works well.  And it's quite possible the error signal handling is broken.
This is as near as I can tell.  And as far as I can tell, the irq13 needs
to be latched, and the code doesn't do this unless it can run properly
through the irq13 part.

>Pedro


-- 
------------------------------------------------------------------------
in real life: david fox            email: dfox@quack.sac.ca.us
5479 Castle Manor Drive,                  
San Jose, CA 95129                 "Thanks for letting me change some
USA                                 magnetic patterns on your HD" -me
408-253-7992                        
"You've just won an absolutely FREE trip to the Isles of Langerhans" -me
------------------------------------------------------------------------