NetBSD Problem Report #7272

Received: (qmail 13555 invoked from network); 28 Mar 1999 23:46:43 -0000
Message-Id: <199903282338.AAA00873@black-star.demon.co.uk>
Date: Mon, 29 Mar 1999 00:38:08 +0100 (BST)
From: mpumford@black-star.demon.co.uk
Reply-To: mpumford@black-star.demon.co.uk
To: gnats-bugs@gnats.netbsd.org
Subject: Hangs at startup after printing real memory=xxx
X-Send-Pr-Version: 3.95

>Number:         7272
>Category:       port-arm32
>Synopsis:       Hangs at startup after printing real memory=xxx
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    port-arm32-maintainer
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Mar 28 15:50:01 +0000 1999
>Closed-Date:    Tue Apr 06 06:46:10 +0000 1999
>Last-Modified:  Tue Apr 06 07:01:15 +0000 1999
>Originator:     Mike Pumford
>Release:        current 28/03/1999
>Organization:
>Environment:
System: NetBSD black-star.demon.co.uk 1.3K NetBSD 1.3K (BLACK-STAR) #40: Thu Mar 25 01:18:39 GMT 1999 mpumford@black-star.demon.co.uk:/usr/src/netbsd/sys/arch/arm32/compile/BLACK-STAR arm32

Acorn RiscPC 200MHz SA110. 48MB DRAM, 2MB VRAM.
Simtec IDE interface with 6.4GB UDMA Seagate attached
Acorn SCSI interface with Pioneer DRM602X attached
Acorn Etherlan 600A Netslot ethernet card.


>Description:
	Kernel hangs at bootup after printing real memory=xxx line. I added
	printf's to machdep.c and discoved the hang is in pmap_enter while 
	allocating the 289'th buffer. If machdep is forced to allocate only
	288 buffers the kernel hangs after the printing:
	avail mem=xxx
	using 288 buffers containing xx bytes of memory.
>How-To-Repeat:
	Boot kernel on RiscPC.
>Fix:
	No idea.
>Release-Note:
>Audit-Trail:

From: Mike Pumford <mpumford@black-star.demon.co.uk>
To: gnats-bugs@gnats.netbsd.org, port-arm32-maintainer@netbsd.org
Cc:  Subject: Re: port-arm32/7272: Hangs at startup after printing real memory=xxx 
Date: Sat, 03 Apr 1999 23:37:50 +0100

 I have tracked this problem further. It appears to be related to the number of 
 SIMM's in the machine.

 The machine has 1 32MB SIMM and 1 16MB SIMM. At boot this is described in the 
 bootconfig structure as 3DRAM blocks. The system also has 2MB of VRAM.

 When the 16MB SIMM is removed the machine boots but panics as follows just 
 after the 'root file system type: ffs'

 The panic is as follows:

 [u]vm_fault(0xf355e0b0, b018000,1,0) -> 1
 Unhandled trap (frame = 0xf3569d74)
 Data abort: 'Translation fault (section)' status=005 address=0b018f04 PC=f012ef5
 8
 Stopped in sh at _pmap_enter_pv+0x64: ldr r3,[r4,#0x0004]

 _pmap_enter_pv(_pmap_enter_pv+x010)
 _pmap_enter(_pmap_enter+0x10)
 _uvm_fault(_uvm_fault+0x10)
 _data_abort_handler(_data_abort_handler+0x10)

 This is with 04/02/1999 supped current.






From: Mark Brinicombe <mark@causality.com>
To: mpumford@black-star.demon.co.uk
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: port-arm32/7272: Hangs at startup after printing real memory=xxx
Date: Sun, 4 Apr 1999 21:16:50 -0700 (PDT)

 This problem has been identified and a solution is being created. This
 lossage is a side effect of the recent pmap changes.

 		Mark


State-Changed-From-To: open->closed 
State-Changed-By: mark 
State-Changed-When: Mon Apr 5 23:46:10 PDT 1999 
State-Changed-Why:  
arm32/riscpc/rpc_machdep 1.26 contains a fix for this problem. 
Due to the recent changes in the arm32 pmap the virtual address space used 
for the SA110 fast cache clean area was incorrect and this resulted 
in corruption of some of the pmap bootpvent pages when the bootstrap 
code mapped the fast cache clean area. A rework of the way the space is 
reserved for the cache clean area resolves the problem though the solution 
used is really a hack and should be better addressed in the future. 
>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.