NetBSD Problem Report #88

From gnats  Fri Jan 21 07:15:36 1994
Received: from cray-ymp.acm.stuorg.vt.edu (root@cray-ymp.acm.stuorg.vt.edu [128.173.43.251]) by sun-lamp.cs.berkeley.edu (8.6.4/8.6.4) with ESMTP id HAA14950 for <gnats-bugs@sun-lamp.cs.berkeley.edu>; Fri, 21 Jan 1994 07:14:31 -0800
Message-Id: <199401211514.KAA11038@cray-ymp.acm.stuorg.vt.edu>
Date: Fri, 21 Jan 1994 10:14:12 -0500
From: briggs@cray-ymp.acm.stuorg.vt.edu
Reply-To: briggs@cray-ymp.acm.stuorg.vt.edu
To: gnats-bugs@sun-lamp.cs.berkeley.edu
Subject: Problem with latest kernel sources from sun-lamp.
X-Send-Pr-Version: 3.01.6

>Number:         88
>Category:       port-mac68k
>Synopsis:       kernel fails
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    gnats-admin
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Jan 21 07:20:02 +0000 1994
>Closed-Date:    Sun Jan 23 15:53:49 +0000 1994
>Last-Modified:  Thu Jan 02 18:14:48 +0000 2014
>Originator:     Allen Briggs
>Release:        current
>Organization:
none
>Environment:
	Kernel.  SE/30 8MB RAM.
System: NetBSD kitten.async.vt.edu 0.9a GENERIC#1 mac68k


>Description:
When I try to build the latest kernel from sun-lamp on my SE/30,
the kernel fails to exec init.  It looks like the arguments are
hosed when they are get to execve.  Has something changed in the
copyin/copyout semantics, or in locore?

>How-To-Repeat:
	Build kernel.  Boot.
>Fix:
	Unknown.  Work in progress.
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed 
State-Changed-By: briggs 
State-Changed-When: Sun Jan 23 07:53:49 PST 1994 
State-Changed-Why:  
The problem was caused by the fact that copyinstr and friends were 
limited to a 64K buffer and execve was running into that limit. 
The fix was actually to split the copy functions from locore.s 
to a general m68k/copy.s and fix copyinstr.  The other functions 
should be fixed, too. 

>Unformatted:

Home	PR Database Search