NetBSD Problem Report #14811

Received: (qmail 22135 invoked from network); 2 Dec 2001 17:47:02 -0000
Message-Id: <20011202174706.6134B1110F@www.netbsd.org>
Date: Sun,  2 Dec 2001 09:47:06 -0800 (PST)
From: gavan@coolfactor.org
Sender: nobody@netbsd.org
Reply-To: gavan@coolfactor.org
To: gnats-bugs@gnats.netbsd.org
Subject: eh and other drivers that use podloader don't work
X-Send-Pr-Version: www-1.0

>Number:         14811
>Category:       port-acorn26
>Synopsis:       eh and other drivers that use podloader don't work
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    bjh21
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Dec 02 17:48:00 +0000 2001
>Closed-Date:    Mon Dec 03 20:56:54 +0000 2001
>Last-Modified:  Mon Dec 03 20:56:54 +0000 2001
>Originator:     Gavan Fantom
>Release:        -current
>Organization:
>Environment:
Not booted multi-user yet, but based on -current sources from 2 days ago
>Description:
Drivers using podloader don't work, leading to a panic in podloader_call. Every byte read from the podule is returned as 'ef', an obviously incorrect value.

The problem lies in the code that wraps the call to the podule loader, which inadvertantly overwrites the return value from the podule loader immediately after its return.
>How-To-Repeat:
Install an i-cubed Etherlan card in an Archimedes and boot an arm26 kernel with the eh driver compiled in. The kernel will find eh0 and then hang for a while. A panic follows in podloader_call.
>Fix:
RCS file: /cvsroot/syssrc/sys/arch/arm26/podulebus/podloader_asm.S,v
retrieving revision 1.2
diff -r1.2 podloader_asm.S
62a63
>       mov     r2, r0
66a68
>       mov     r0, r2
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->analyzed 
State-Changed-By: bjh21 
State-Changed-When: Mon Dec 3 07:45:53 PST 2001 
State-Changed-Why:  
Gavan's diagnosis is correct. 


Responsible-Changed-From-To: port-arm26-maintainer->bjh21 
Responsible-Changed-By: bjh21 
Responsible-Changed-When: Mon Dec 3 07:45:53 PST 2001 
Responsible-Changed-Why:  
I've got a patch that should fix the problem rather more solidly. 
State-Changed-From-To: analyzed->closed 
State-Changed-By: bjh21 
State-Changed-When: Mon Dec 3 12:54:59 PST 2001 
State-Changed-Why:  
Gavan has confirmed that the patch posted to port-arm26 fixes this, and I've 
comitted that patch. 
[ podloader_asm.S 1.3, podulebus.c 1.16 ] 
>Unformatted:

Home	PR Database Search