John Klos | 9 Nov 00:52 2002

Binary packages for m68k NetBSD-1.6

Hello,

The first 1,390 binary packages for m68k NetBSD-1.6 have been uploaded to
ftp.netbsd.org. There are many, many more packages to go, but the build
machine suffered from disk problems and had to be restarted, and this
substantially delayed the building.

Soon a permanent m68k build machine will be colocated and will build
pkgsrc continuously. This machine is lilith.sixgirls.org, and information,
if you're interested, can be found at:
http://www.sixgirls.org/lilith/

More binary packages to come!

John Klos
Sixgirls Computing Labs

kunal agarwal | 10 Nov 07:33 2002
Picon

Virtual Memory Exhausted

Hello,
While running a prog on sun-solaris using gcc, its saying:
Virtual Memory Exhausted
and after a couple of times trying its saying:
ld:fatal: file /dev/zero:cannot mmap file:resource temporarily unavailable
collect2: ld returned 1 exit status

Any help would be appreciated.

Thanks,
Kunal Agarwal

Manuel Bouyer | 10 Nov 20:43 2002

Re: Virtual Memory Exhausted

On Sat, Nov 09, 2002 at 10:33:54PM -0800, kunal agarwal wrote:
> Hello,
> While running a prog on sun-solaris using gcc, its saying:
> Virtual Memory Exhausted
> and after a couple of times trying its saying:
> ld:fatal: file /dev/zero:cannot mmap file:resource temporarily unavailable
> collect2: ld returned 1 exit status
> 
> Any help would be appreciated.

Maybe try to raise per-process limits ?
'unlimit' in (t)csh

--

-- 
Manuel Bouyer <bouyer <at> antioche.eu.org>
     NetBSD: 23 ans d'experience feront toujours la difference
--

Adam Kropelin | 19 Nov 03:14 2002
Picon

1.6 vm_fault on ie packet rx

Trying to install 1.6 on my 3/280 I get the below vm_fault soon after I
ifconfig ie0 (onboard). It appears to happen on packet rx. The quoted fault
is from the netbsd.RAMDISK kernel (netbooted) but it happens with the
miniroot as well. 1.5.3 is happy so this seems to be a new issue with 1.6.

I apologize for being something of a NetBSD newbie...in the Linux world we
have a tool to 'symbolize' dumps like this. If I need to do that to make it
useful for you, hit me with a clue-stick and I'll get to it...

--Adam

vm_fault(0xe0c9598, 0x0, 0x2, 0) -> 0xd
trap type=0x8, code=0x105, v=0x0
pid = -1, pc = 0E0A4166, ps = 2300, sfc = 1, dfc = 1
Registers:
             0        1        2        3        4        5        6
7
dreg: 0FA5BE74 0000FFFF 0000F9C0 00000001 00000000 000000F9 000000F9
000000F9
areg: 0FA5BE74 0FA5BE82 0E176434 0E176000 0E0EB620 00000000 0FA5BE84
0DFFFFFC

Kernel stack (0FA5BD4C):
A5BD4C: 0E094A8C 0FA5BD9C 00000080 0000F9C0 00000001 00000000 000000F9
000000F9
A5BD6C: 000000F9 0E176434 0E176000 0E0EB620 00000000 00000000 00000000
0FA5BE84
A5BD8C: 0E0040EC 00000008 00000105 00000000 0FA5BE74 0000FFFF 0000F9C0
00000001
A5BDAC: 00000000 000000F9 000000F9 000000F9 0FA5BE74 0FA5BE82 0E176434
(Continue reading)

WILEY, MR. JAMES | 19 Nov 15:39 2002

RE: 1.6 vm_fault on ie packet rx


I see the same thing on a 3/110 with the miniroot and the install kernels.  If I put in the Generic kernel the
problems goes away, but the machine is unusably slow (only 4 MB - anyone have a memory expansion that will
work in a 3/110?).  I am working on trying to compile a custom kernel, but am having disk error problems as
well as lack of time problems...

James L. Wiley
Mgr. Client/Server Computing
Information Resources
Tarleton State University

Every imaginable disclaimer applies.

-----Original Message-----
From: port-sun3-owner <at> netbsd.org [mailto:port-sun3-owner <at> netbsd.org]On
Behalf Of Adam Kropelin
Sent: Monday, November 18, 2002 8:15 PM
To: port-sun3 <at> netbsd.org
Subject: 1.6 vm_fault on ie packet rx

Trying to install 1.6 on my 3/280 I get the below vm_fault soon after I
ifconfig ie0 (onboard). It appears to happen on packet rx. The quoted fault
is from the netbsd.RAMDISK kernel (netbooted) but it happens with the
miniroot as well. 1.5.3 is happy so this seems to be a new issue with 1.6.

I apologize for being something of a NetBSD newbie...in the Linux world we
have a tool to 'symbolize' dumps like this. If I need to do that to make it
useful for you, hit me with a clue-stick and I'll get to it...

--Adam
(Continue reading)

David Brownlee | 19 Nov 14:45 2002
Picon

RE: 1.6 vm_fault on ie packet rx

On Tue, 19 Nov 2002, WILEY, MR. JAMES wrote:

>
> I see the same thing on a 3/110 with the miniroot and the install
> kernels.  If I put in the Generic kernel the problems goes away, but the
> machine is unusably slow (only 4 MB - anyone have a memory expansion
> that will work in a 3/110?).  I am working on trying to compile a custom
> kernel, but am having disk error problems as well as lack of time
> problems...

	If you have any other unix box available you can cross compile
	a NetBSD kernel using that:
	http://www.netbsd.org/Documentation/cross/

--

-- 
		David/absolute          -- www.netbsd.org: No hype required --

Chuck Silvers | 19 Nov 16:47 2002

Re: 1.6 vm_fault on ie packet rx

hi,

yea, the register dump isn't so useful.  try building a kernel with
"options DDB" turned on and with the attached patch applied.
that should give a stack trace with function names when it crashes.

-Chuck

On Mon, Nov 18, 2002 at 09:14:42PM -0500, Adam Kropelin wrote:
> Trying to install 1.6 on my 3/280 I get the below vm_fault soon after I
> ifconfig ie0 (onboard). It appears to happen on packet rx. The quoted fault
> is from the netbsd.RAMDISK kernel (netbooted) but it happens with the
> miniroot as well. 1.5.3 is happy so this seems to be a new issue with 1.6.
> 
> I apologize for being something of a NetBSD newbie...in the Linux world we
> have a tool to 'symbolize' dumps like this. If I need to do that to make it
> useful for you, hit me with a clue-stick and I'll get to it...
> 
> --Adam
Index: arch/sun3/sun3/trap.c
===================================================================
RCS file: /cvsroot/syssrc/sys/arch/sun3/sun3/trap.c,v
retrieving revision 1.106
diff -u -r1.106 trap.c
--- arch/sun3/sun3/trap.c	2002/10/20 02:37:37	1.106
+++ arch/sun3/sun3/trap.c	2002/11/19 15:45:18
 <at>  <at>  -78,6 +78,7  <at>  <at> 
 #ifdef DDB
(Continue reading)

Adam Kropelin | 19 Nov 23:41 2002
Picon

Re: 1.6 vm_fault on ie packet rx

----- Original Message ----- 
From: "Chuck Silvers" <chuq <at> chuq.com>
> hi,
> 
> yea, the register dump isn't so useful.  try building a kernel with
> "options DDB" turned on and with the attached patch applied.
> that should give a stack trace with function names when it crashes.

Will do...

--Adam

Adam Kropelin | 22 Nov 05:29 2002
Picon

Re: 1.6 vm_fault on ie packet rx

----- Original Message -----
From: "Chuck Silvers" <chuq <at> chuq.com>
> hi,
>
> yea, the register dump isn't so useful.  try building a kernel with
> "options DDB" turned on and with the attached patch applied.
> that should give a stack trace with function names when it crashes.
>
> -Chuck

I built a RAMDISK kernel with the sun3trap patch and DDB enabled. (Well,
first I build a 1.6 compat toolchain, *then* 24 hours of compiling later I
built a 1.6 kernel ;)

Unfortunately it can't seem to mount the ramdisk root filesystem. Only
change from the stock RAMDISK config was to enable DDB.

Do I need, perhaps, to build a ramdisk filesystem image as a second step?
Any hints? Boot messages follow...

--Adam

starting program at 0x4000
console is ttya
[ using 134572 bytes of netbsd ELF symbol table ]
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002
    The NetBSD Foundation, Inc.  All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
    The Regents of the University of California.  All rights reserved.

(Continue reading)

Chuck Silvers | 23 Nov 02:31 2002

Re: 1.6 vm_fault on ie packet rx

On Thu, Nov 21, 2002 at 11:29:14PM -0500, Adam Kropelin wrote:
> I built a RAMDISK kernel with the sun3trap patch and DDB enabled. (Well,
> first I build a 1.6 compat toolchain, *then* 24 hours of compiling later I
> built a 1.6 kernel ;)
> 
> Unfortunately it can't seem to mount the ramdisk root filesystem. Only
> change from the stock RAMDISK config was to enable DDB.
> 
> Do I need, perhaps, to build a ramdisk filesystem image as a second step?
> Any hints? Boot messages follow...

er, yea, if you want to use the ramdisk stuff you need to load an image
into the kernel, I think mdsetimage is the tool that does that.

it would be easier to just use an NFS root rather than mess with a ramdisk,
if you can do that...

-Chuck


Gmane