From owner-freebsd-hackers  Thu May 27  9:32:12 1999
Delivered-To: freebsd-hackers@freebsd.org
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.26.10.9])
	by hub.freebsd.org (Postfix) with ESMTP
	id 2EDE914C0C; Thu, 27 May 1999 09:32:03 -0700 (PDT)
	(envelope-from bde@godzilla.zeta.org.au)
Received: (from bde@localhost)
	by godzilla.zeta.org.au (8.8.7/8.8.7) id CAA15122;
	Fri, 28 May 1999 02:31:48 +1000
Date: Fri, 28 May 1999 02:31:48 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199905271631.CAA15122@godzilla.zeta.org.au>
To: bde@zeta.org.au, bob@anet.donetsk.ua,
	freebsd-hardware@FreeBSD.ORG, mboers@datacompusa.com
Subject: Re: Multiproc kernel 3.1-Release & Cyclades Cyclom 8Yep PCI
Cc: freebsd-hackers@FreeBSD.ORG
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

>>There seems to be a problem with nested locks.  What was the panic
>>message?
>
>I have the same problem with my Cyclom Ye cards (both the isa & pci 
>variety)
>
>The kernel panics with:
>
>panic messages:
>---
>panic: rslock: cpu: 0, addr: 0xf026a15c, lock: 0x00000001
>mp_lock = 00000001; cpuid = 0; lapic.id = 00000000

That's the problem with nested locks.

>after crash debugging shows that the value of com is 0x3 at the time of 
>crash.  Since com is a pointer, I think we know that it is wrong.  I have 
>added a printf statement to a varitation of cy.c that shows that just 
>before the call to commctl, the value of com is correct (not 3).

I don't think that's the problem.

>Any help towards getting the cy driver up and running on the smp kernel 
>would be appreciated.

See hints in my previous mail.  The main problem is that disable_intr()
isn't doesn't nest properly and I only avoided this problem for !SMP
case.  The SMP disable_intr() is an evil macro that calls a locking
function which crashes if the lock is already held.

Bruce


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message