Subject: 20030904 snapshot test results (and bootable CD image available)
To: None <>
From: sgimips NetBSD list <>
List: port-sgimips
Date: 09/07/2003 22:28:58

I've just performed extensive testing of Chris Sekiya's 20030904 snapshot.
Tested were 9 differing SGI Mips machines.  With the exception of two,
NetBSD runs well on all.  The only problems encountered so far are with
the sysinst installation tool.


	Working platforms:     O2 R10K, O2 R5K, Indy R5K, Indy R4600, 
			       Indy R4400 (200 and 175mhz), Indigo 2 R4000
	Non-working platforms: Indigo 2 Impact R10000, Challenge S R5000
There are various sysinst(8) problems to report:

1.	Can't set a timezone during the installation process -- segfault
2.	Can't set a timezone via Utility Menu -- No timezones listed
3.	During an install from an ISO CD containing packages, progress(1)
	does nothing.
4.	If more than one hard drive is attached, sysinst segfaults during
5.	Odd (intermittent at best) panics on an INDY R4600 (could be
	attributed to intermittent hardware problems).

I'll look in to these, but this is the preliminary "supported" list from
my own testing.

I have made available a bootable CD image at
along with some rudimentary installation instructions.   This is the
bootable CD I used for all the tests here.

Thanks to Chris and others for putting all their time into the R10K
stuff and for putting this snapshot together.


Tested Platforms:
Machine Type		CPU	Mhz	Boot	Install	Network	cvs+bld	Notes
=======================	=======	=======	=======	=======	=======	=======	=======
Indigo 2 Impact 10000	R10000	195	Fail	N/A	N/A	N/A	(1) ELF-64
O2			R10000	175	OK	OK	fxp0	OK
O2			R5000	180	OK	OK	fxp0	OK
Challenge S		R5000	180	OK	OK	sq0	N/A
Indy 			R5000	150	OK	OK	sq0	N/A
Indy 			R4400	200	OK	OK	sq0	OK
Indy			R4400	175	OK	OK	sq0	N/A
Indy			R4600	100	OK	OK	sq0	N/A
Indigo 2		R4000	100	OK	OK	sq0	N/A	ecoff

"Network" indicates networking worked after multi-user boot and which
interface was used.  "cvs+bld" indicates a full cvs download and build
of the full NetBSD sources was completed. 

(1) The Indigo 2 Impact R10000 machine is not yet supported due to
    ELF-64 kernel requirements (as opposed to the ELF-32 binaries used
    by all the other platforms)
(2) My Challenge S won't boot NetBSD/sgimips with this snapshot.  It boots
    into IRIX 6.5.7 fine.   See the boot log below.

All tests were performed using the following CD boot command line:

boot -f scsi(0)disk(6)rdisk(0)partition(8)/boot.elf scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP32

Indigo 2 (ecoff-based boot)
boot -f scsi(1)disk(6)rdisk(0)partition(8)/boot scsi(1)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22

INDYs and Challenge S:
boot -f scsi(0)disk(6)rdisk(0)partition(8)/boot.elf scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22



     Please choose the timezone that fits you best from the list below.  Press
     RETURN to select an entry.
     Press 'x' followed by RETURN to quit the timezone selection.
     Default:        UTC
     Selected:       CST6CDT
     Local time:     Sun Sep 7 16:23:34 2003 CDT
     uid 0, pid 140, command sysinst, on /: file system full

    /: write failed, file system is full
    [1]   Segmentation fault      

To avoid this problem, do not select a timezone but instead
select "x" and press ENTER to bypass timezone selection.  You can
set the timezone later by issuing a commands similar to the following
after you have booted the installed OS:

	# rm /etc/timezone
	# ln -s /usr/share/zoneinfo/US/Pacific /etc/timezone


When installing from an ISO CD image, progress(1) hangs.  Works
fine when installing from an FFS CD filesystem.

I do not know a workaround for this except to install using
the following methods:

A.	Use an FFS filesystem on the CD instead of an ISO/CD9660 one :)

B.	Use the following method to hack through an install:

	1.	Use sysinst to partition and mkfs the drive.
	2.	^C out of sysinst before proceeding with the
		install media selection.
	3.	mount the ISO cd:   # mount -t cd9660 -o ro /dev/cd0a /mnt
	4.	Manually extract the packages:
		# cd /targetroot
		# for i in /mnt/binary/sets/*.tgz; do
		> echo Installing package $i
		> gzip -dc $i | tar -xepf -
		> done
	5.	Make all device nodes:
		# cd /etc
		# ./MAKEDEV all
	6.	You will need to manually set a root password, set the
		timezone, and other such initial set-ups.


If more than one drive is installed disk partitioning fails
	    /: write failed, file system is full
	    [1]   Segmentation fault
When the screen would be displayed with your selected partitions
and sizes.

I do not know a workaround for this except to disconnect all but
the drive you intend to install to.   You can reconnect the other
drives after the installation has completed.

PANIC WHILE MAKING DEVICE NODES	- single transient occurence

 Making device nodes ...
 IP22 bus error: cpu_stat 00000400 addr 0fb20030, gio_stat 00000000 addr 1fbc4003
panic: TLB out of universe: ksp 0xc61bbd58 epc 0x8829078c vaddr 0xc015f000
Stopped in pid 696.1 (sh) at    0x88285ac4:     jr      ra
                bdslot: nop
db> ps
 PID           PPID     PGRP        UID S   FLAGS LWPS          COMMAND    WAIT
>696            108      108          0 2     0x2    1               sh
 108             15      108          0 2  0x4002    1               sh  piperd
 15               8       15          0 2  0x4002    1          sysinst  select
 8                1        8          0 2  0x4002    1               sh    wait
 7                0        0          0 2 0x20200    1         aiodoned aiodone
 6                0        0          0 2 0x20200    1          ioflush  syncer
 5                0        0          0 2 0x20200    1           reaper  reaper
 4                0        0          0 2 0x20200    1       pagedaemon pgdaemo
 3                0        0          0 2 0x20200    1       lfs_writer lfswrit
 2                0        0          0 2 0x20200    1         scsibus0  sccomp
 1                0        1          0 2  0x4000    1             init    wait
 0               -1        0          0 2 0x20200    1          swapper schedul
db> trace
88285ac0+4 (8bfff000,32,0,0) ra 881fd8c4 sz 0
881fd79c+128 (882df40c,32,0,0) ra 882906e0 sz 40
88290670+70 (882df40c,32,0,0) ra 88288190 sz 48
88287cc4+4cc (882df40c,32,0,0) ra 882710f4 sz 80
8827087c+878 (88ce8340,558000,0,1) ra 88289c30 sz 296
trap+58c (ff13,558000,0,1) ra 88283cc8 sz 72
mips3_UserGenException+d0 (ff13,558000,0,558b70) ra 0 sz 0
User-level: pid 696.1

This occured on an R4600 INDY.    I have no idea why.  Wonky hardware perhaps?
I booted the installed drive, re-made /dev, fsck'd, and all seemed fine.

On a second attempt to install from scratch, this panic did not occur.

ALSO on the same machine a few install tests later:

>> boot -f scsi(0)disk(6)rdisk(0)partition(8)/boot.elf scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22

Cannot load scsi(0)disk(6)rdisk(0)partition(8)/boot.elf.
Illegal f_magic number 0x1, expected MIPSELMAGIC or MIPSEBMAGIC.

Exception: <vector=Normal>
Status register: 0x30004803<CU1,CU0,IM7,IM4,IPL=???,MODE=KERNEL,EXL,IE>
Cause register: 0x8010<CE=0,IP8,EXC=RADE>
Exception PC: 0x9fc43cb0, Exception RA: 0x9fc43f40
Read address error exception, bad address: 0xfffffffd
Local I/O interrupt register 1: 0x80 <VR/GIO2>
  Saved user regs in hex (&gpda 0xa8740e48, &_regs 0xa8741048):
  arg: 1 1 a8748c70 1
  tmp: 256 a874a150 a87fe6d8 1 a8747560 0 a8747560 1
  sve: 0 a874a150 f a87fe8d8 a87fe848 c000000 a87fe990 9fc563d4
  t8 a8747560 t9 0 at ffffffff v0 42 v1 42 k1 bad11bad
  gp 883034e0 fp 9fc4bec8 sp a87fe748 ra 9fc43f40

PANIC: Unexpected exception

[Press reset or ENTER to restart.]

A second attempt to boot succeeded.  Perhaps there was bad data
read from the CD-ROM?

Perhaps my R4600 is unwell since both of the above went away on
second attempts.


When you enter the "Utility Menu" and select "Set Timezone," there
are no timezones listed to choose from.


System Maintenance Menu

1) Start System
2) Install System Software
3) Run Diagnostics
4) Recover System
5) Enter Command Monitor

Option? 5
Command Monitor.  Type "exit" to return to the menu.
>> hinv -v -p -n
Usage: inventory:        hinv [-v] [-t [-p]]
>> hinv -v -t -p
system ARC SGI-IP22 key 0
  processor CPU MIPS-R5000 key 0
    processor FPU MIPS-R5000FPC key 0
    cache primary icache 32 Kbytes (block 2 lines, line 32 bytes)
    cache primary dcache 32 Kbytes (block 2 lines, line 32 bytes)
    cache secondary cache 512 Kbytes (block 1 lines, line 32 bytes)
    memory main 192 Mbytes
  controller network ec0 key 0
  v: net SEEQ NQ8003 - type: Ethernet MTU: 1500 HW: 8:0:69:a:b8:72
  p: network(0)
    peripheral network key 0 ( Input Output )
    p: network(0)network(0)
  adapter SCSI WD33C93B key 0
  p: scsi(0)
    controller disk SGI IBM DORS-32160 key 2
    p: scsi(0)disk(2)
      peripheral disk unit 0 ( Input Output )
      v: scsidisk - - - blocksize: 512 maxblocks: 4197404
      p: scsi(0)disk(2)rdisk(0)
    controller CDROM TOSHIBA CD-ROM XM-5701TA key 6
    p: scsi(0)cdrom(6)
      peripheral disk unit 0 ( ReadOnly Removable Input )
      p: scsi(0)cdrom(6)rdisk(0)
  controller serial IP22 tty key 0
  p: serial(0)
    peripheral line key 0 ( ConsoleIn ConsoleOut Input Output )
    p: serial(0)line(0)
  controller serial IP22 tty key 1
  p: serial(1)
    peripheral line key 0 ( ConsoleIn ConsoleOut Input Output )
    p: serial(1)line(0)
  controller keyboard pckbd key 0
  p: key(0)
  controller pointer pcms key 0
  p: pointer(0)
>> boot -f scsi(0)disk(6)rdisk(0)partition(8)/boot.elf scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22

NetBSD/sgimips 1.6X Bootstrap, Revision 1.2
(wileyc@izu, Thu Sep  4 15:13:46 JST 2003)

Mem block: type 0 base 0x0 size 0x1000
Mem block: type 1 base 0x1000 size 0x1000
Mem block: type 5 base 0x8002000 size 0xc000
Mem block: type 3 base 0x800e000 size 0x732000
Mem block: type 6 base 0x8740000 size 0xc0000
Mem block: type 3 base 0x8800000 size 0xb800000
Local storage a87ff168
argv[0]: scsi(0)disk(6)rdisk(0)partition(8)/boot.elf
argv[1]: scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22
argv[2]: ConsoleIn=serial(0)
argv[3]: ConsoleOut=serial(0)
argv[4]: SystemPartition=scsi(0)disk(2)rdisk(0)partition(8)
argv[5]: OSLoader=sash
argv[6]: OSLoadPartition=scsi(0)disk(2)rdisk(0)partition(0)
argv[7]: OSLoadFilename=/unix
argv[8]: OSLoadOptions=INST
Boot: scsi(0)disk(6)rdisk(0)partition(0)/netbsd-INSTALL_IP22
devopen: scsi(0)disk(6)rdisk(0)partition(0) type scsi file /netbsd-INSTALL_IP22
Starting at 0x88069000

nsym 0x1 ssym 0x8873c55c esym 0x8873c770
IOC rev 1, machine Indy (Guiness), board rev 3
Timer calibration, got 900000 cycles (900000, 900000, 900000)
CPU clock speed = 180.00Mhz
zs channel 0 had address 0xbfbd9830
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003
    The NetBSD Foundation, Inc.  All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
    The Regents of the University of California.  All rights reserved.

NetBSD 1.6X (INSTALL_IP22) #0: Thu Sep  4 15:32:19 JST 2003
192 MB memory, 169 MB free, 768 KB for ARCS, 9932 KB in 2483 buffers
mainbus0 (root): SGI-IP22 [SGI, 690ab872], 1 processor
cpu0 at mainbus0: MIPS R5000 CPU (0x2310) Rev. 1.0 with built-in FPU Rev. 1.0
cpu0: 32KB/32B 2-way set-associative L1 Instruction cache, 48 TLB entries
cpu0: 32KB/32B 2-way set-associative write-back L1 Data cache
imc0 at mainbus0 addr 0x1fa00000
imc0: Revision 3
gio0 at imc0
hpc0 at gio0 addr 0x1fb80000: SGI HPC3
zsc0 at hpc0 offset 0x59830
zstty0 at zsc0 channel 1 (console i/o)
zstty1 at zsc0 channel 0
sq0 at hpc0 offset 0x54000: SGI Seeq 80c03
sq0: Ethernet address 08:00:69:0a:b8:72
wdsc0 at hpc0 offset 0x44000: WD33C93B SCSI, rev=0, target 7
scsibus0 at wdsc0: 8 targets, 8 luns per target
dsclock0 at hpc0 offset 0x60000
hpc1 at gio0 addr 0x1fb00000: SGI HPC3
zsc at hpc1 offset 0x59830 not configured
sq at hpc1 offset 0x54000 not configured
wdsc at hpc1 offset 0x44000 not configured
dsclock at hpc1 offset 0x60000 not configured
biomask 07 netmask 07 ttymask 0f clockmask bf
md0: internal 4096 KB image area
scsibus0: waiting 2 seconds for devices to settle...
MSGOUT Failed!
wd33c93_intr: LCI asr:c0 csr:1f
probe(wdsc0:0:2:0): wdsc0: timed out; asr=0x20 [acb 0x896f6000 (flags 0x1, dleft 20)], <state 5, nexus 0x896f6000, resid 20, msg(q 80,o 0)>probe(wdsc0:0:2:0): ABORT in timeout: csr=0xff, asr=0x20
probe(wdsc0:0:2:0): sending ABORT command
probe(wdsc0:0:2:0): Resetting bus
: WD33C93B SCSI, rev=0, target 7
probe(wdsc0:0:2:0): wdsc0: timed out; asr=0x00 [acb 0x896f6000 (flags 0x41, dleft 20)], <state 8, nexus 0x896f6000, resid 20, msg(q 80,o 0)>probe(wdsc0:0:2:0): ABORT in timeout: csr=0x01, asr=0x00
probe(wdsc0:0:2:0): sending ABORT command
probe(wdsc0:0:2:0): sending DISCONNECT to target
wd33c93_wait: TIMEO @913 with asr=x0 csr=x1
wd33c93_wait: TIMEO @913 with asr=x0 csr=x1
wd33c93_wait: TIMEO @913 with asr=x0 csr=x1
<... ad nauseum - 1 (about 40-50 repeats) until it hard resets...>