Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ethernet)

SGI hardware problems, solutions, tips, hacks, etc.
Forum rules
Any posts concerning pirated software or offering to buy/sell/trade commercial software are subject to removal.
User avatar
Gerhard.Lenerz
Posts: 330
Joined: Thu Jan 15, 2004 1:20 am
Location: Frankfurt/Main, Germany
Contact:

Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ethernet)

Unread postby Gerhard.Lenerz » Sun Dec 17, 2006 5:42 am

I'm trying to further expand my Onyx. I've installed a 013-1616-004 assembly (C8-100BT) into the first available VME slot on the machne (next to the IO4/VCAM).

Running autoconfig fails like this:

Code: Select all

probe spaced if_vfe adapter 0
if_vfe, adapter 0 probe space failed


The machine doesn't show any obvious errors during POST and I've also done 'update' on PROM level just to make sure I didn't miss anything. Does anyone have a similar adapter up and running? I know the machine I've stripped had only one installed but still the two bulkhead panels are labeled VFE2 and VFE3 (handwritten). So I wonder if there is some jumper magic to be done.

Just for the record, here is a log of the POST as well as a hinv:

Code: Select all

IP25 SCC(E) SGI Version 6  built 10:11:49 AM May  8, 1996
R10000 2.6 194MHz BE (4-2-2/9) 1MB

Initializing hardware inventory...              ...done.
    CPU 02/00 is bootmaster
Testing Secondary Cache...                      ...passed.
Testing and clearing bus tags...                ...passed.
Configuring memory...
    Using standard interleave algorithm.
Running built-in memory test... 01
                                                ...passed.
Writing cfginfo to memory
Initializing MPCONF blocks
Checking slave processor diag results..........
    Enabled 1024 Megabytes of main memory
    Enabled 4 processors
Downloading PROM header information...
Downloading PROM code...
Jumping into IO4 PROM.

PROM Segment Loader (R10000 IP25) SGI Version 2.1 Rev A MIPS3,   Sep  3, 1996
Loading and executing R10000 boot prom image...

IO4 PROM Monitor SGI Version 4.21 Rev A IP25,   Sep  3, 1996 (BE64)
Sizing caches...
Initializing exception vectors.
Initializing IO4 subsystems.
Fixing vpids...
Initializing environment
Piggyback reads enabled.
Initializing software and devices.
All initialization and diagnostics completed.
Bootmaster processor already started.
Starting processor #1
Starting processor #2
Starting processor #3
Checking hardware inventory...


                           Starting up the system...

               To perform system maintenance instead, press <Esc>

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.
>> update
>> hinv
                   System: IP25
                Processor: 194 Mhz R10000, 1M secondary cache
                Processor: 194 Mhz R10000, 1M secondary cache, (cpu 1)
                Processor: 194 Mhz R10000, 1M secondary cache, (cpu 2)
                Processor: 194 Mhz R10000, 1M secondary cache, (cpu 3)
              Memory size: 1024 Mbytes
                SCSI Tape: scsi(0)tape(3)
               SCSI CDROM: scsi(0)cdrom(7)
                SCSI Disk: scsi(1)disk(1)
                SCSI Disk: scsi(1)disk(8)
                SCSI Disk: scsi(1)disk(9)
                 Graphics: Reality-Engine 2 12 GEs  1 RM
>> auto


                           Starting up the system...

Loading dksc(1,1,8)/sash: 896+111372+16725+3848 entry: 0xa80000001a64791c
3739868+595724 entry: 0xa8000000000075b0
IRIX Release 6.5 IP25 Version 01100601 System V - 64 Bit
Copyright 1987-2003 Silicon Graphics, Inc.
All Rights Reserved.

WD95A SCSI controller 0 - single ended internal, rev 0, min xfer period 100ns
WD95A SCSI controller 1 - differential internal, rev 0, min xfer period 100ns
Configuring EPC in IO4 slot 3 padap 1 as et0
The system is coming up.


[.....SNIP.....]


aragorn 2# hinv
4 194 MHZ IP25 Processors
CPU: MIPS R10000 Processor Chip Revision: 2.6
FPU: MIPS R10010 Floating Point Chip Revision: 2.6
Main memory size: 1024 Mbytes, 1-way interleaved
Instruction cache size: 32 Kbytes
Data cache size: 32 Kbytes
Secondary unified instruction/data cache size: 1 Mbyte
Integral SCSI controller 0: Version WD33C95A, single ended, revision 0
  Tape drive: unit 3 on SCSI controller 0: 8mm(8500) cartridge
  CDROM: unit 7 on SCSI controller 0
Integral SCSI controller 1: Version WD33C95A, differential, revision 0
  Disk drive: unit 1 on SCSI controller 1
  Disk drive: unit 8 on SCSI controller 1
  Disk drive: unit 9 on SCSI controller 1
ASO 6-port Serial board 0: revision 2.1.26, Ebus slot 3, IO Adapter 5
Integral EPC serial ports: 4
Integral EPC parallel port: Ebus slot 3
RealityEngineII Graphics Pipe 0 at IO Slot 3 Physical Adapter 2 (Fchip rev 2)
Integral Ethernet controller: et0, Ebus slot 3
Iris Audio Processor: version A2 revision 1.1.0 unit 0, Ebus slot 3 adapter 5
I/O board, Ebus slot 3: IO4 revision 1
VME bus: adapter 13
VME bus: adapter 0 mapped to adapter 13
EPC external interrupts
aragorn 3#


Thanks for any hint.


Gerhard

User avatar
landy
Posts: 40
Joined: Wed Oct 04, 2006 8:43 am
Location: Germany

Unread postby landy » Sun Dec 17, 2006 6:36 am

I believe I have the same board (labled 013-1616-004) like you in my Challenge L .

When I run
autoconfig -fv

I get

Code: Select all

if_vfe, adapter 0 probe space failed
probe spaced if_vfe adapter 0
TOOLROOT=/var/sysgen/root /var/sysgen/root/usr/bin/cc  -c -I -DVENICE -DIP19 -DIP19 -D_K64PROM32 -xansi -D_K64PROM32 -D_PAGESZ=16384 -D_KERNEL -DSTATIC=static -D_MIPS3_ADDRSPACE -DR4000_DADDIU_WAR -DFAST_LOCORE_TFAULT -DPTE_64BIT -DEVEREST -DBADVA_WAR -DMP_R4000_BADVA_WAR -DECC_RECOVER -DCLOCK_CTIME_IS_ABSOLUTE -DULI -DCKPT -DDELAY_CEL_WRITES -DNO_MFC_NOPS -DIDLESTACK_FASTINT -DSW_FAST_CACHE_SYNCH -DUSE_PTHREAD_RSA -DMC3_CFG_READ_WAR -DIO4_CFG_READ_WAR -D_SHAREII -D_MP_NETLOCKS -DMP_STREAMS=1 -DMP -DIP19 -DR4000 -g3 -O2 -G 8 -TENV:kernel -OPT:space=on -DEBUG:optimize_space=on -OPT:Olimit=0 -CG:unique_exit=on -TENV:strings_not_gprelative=on -TENV:X=1 -OPT:IEEE_arithmetic=1 -OPT:roundoff=0 -OPT:wrap_around_unsafe_opt=off -fullwarn -diag_error 1116,1196,1551,1413,1498,1009,1110,1171,1047,1140,1164,1412,1069,1035,1183 -LANG:vla=off -non_shared -mips3 -64 -woff 1685,515,608,658,799,803,852,1048,1233,1499,1501 -DDEFAULTSEMAMETER=1 -I.  -I/usr/include -I/var/sysgen/boot /var/sysgen/master.c -o /tmp/master1047.o
TOOLROOT=/var/sysgen/root /var/sysgen/root/usr/bin/ld  -non_shared -e start -N -woff 133 -G 8 -woff 84,47,39,17 -mips3 -64 -nostdlib -T a800000000004000 -wall  kernel.o /tmp/master1047.o -o /unix.install
... blabla...
lboot: WARNING: there are files in /var/sysgen with mtime in the future
lboot: WARNING: setting /unix.install mtime to Sat Jan 14 05:52:14 2006
challenge 6#


and booting the system

Code: Select all


                           Starting up the system...

IRIX Release 6.5 IP19 Version 07091541 System V - 64 Bit
Copyright 1987-2001 Silicon Graphics, Inc.
All Rights Reserved.

WD95A SCSI controller 0 - single ended internal, rev 0, min xfer period 100ns
WD95A SCSI controller 1 - differential internal, rev 0, min xfer period 100ns
[b]vfe board0: IP-6200 100BaseTX EtherNet Ver 2.1 (B.17) - A16[0x6000], A32[0x19500
000][/b]
Configuring EPC in IO4 slot 5 padap 1 as et0
The system is coming up.

boot: WARNING: kernel seems current but has a modification time in the future
[b]vfe0: 100 Mb/s halfduplex[/B]


You know this link: http://192.48.170.165/library/tpl/cgi-b ... lnotes/vfe

I will take out the board if you want.
Michael

User avatar
shel
Posts: 304
Joined: Fri Jan 13, 2006 11:25 am
Location: Uzes, France or Seattle, WA, USA

Re: Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ether

Unread postby shel » Sun Dec 17, 2006 4:37 pm

Gerhard.Lenerz wrote:I'm trying to further expand my Onyx. I've installed a 013-1616-004 assembly (C8-100BT) into the first available VME slot on the machne (next to the IO4/VCAM).

Running autoconfig fails like this:

Code: Select all

probe spaced if_vfe adapter 0
if_vfe, adapter 0 probe space failed


I have a dual-VFE board in my Challenge L. It works, but not very well.

I installed it long enough ago that I forget exectly what I did, but I think I had to install some software from the IRIX 6.5 disk set in order to get the OS to recognize the board. Once I did that, and fiddled around a bit with various settings, I was able to get it to work.

However, I've never been able to get it to run as fast as the normal et0 10base network port. My understanding is that the VFE board is locked at 100mbs half-duplex, and doesn't announce that. Since most switches default to full-duplex, all sorts of collisions occur. However, even using a HD-FD converter, I wasn't able to get anything like 100mbps, or even 10mbps, out of the VFE board.

So, I'l be reading this thread with interest. I've heard that these boards, and their drivers, aren't very good, but someone must know how to make them sing and dance.

-Shel

User avatar
Gerhard.Lenerz
Posts: 330
Joined: Thu Jan 15, 2004 1:20 am
Location: Frankfurt/Main, Germany
Contact:

Unread postby Gerhard.Lenerz » Sun Dec 17, 2006 11:41 pm

Michael, if you could take a snapshot of the jumpers of your board that would be great and very much appreciated. :)

Shel, I was also thinking about the "driver" missing but from IRIX 6.5 it is included along with IRIX. Looking at the relevant directories I do have for example the vfe headers and object files for the kernel. But still, I might be missing something else.

I read that it doesn't perform good and isn't really stable but I hoped to get at least any 10+MBit performance out of it because im running almost anything remote (including X applications). We'll see how good or bad it works. Configuring ports on the network side (outside the Onyx) is probably going to be trouble as it is connected to a cheap LAN/DSL/WLAN router.



Gerhard

User avatar
shel
Posts: 304
Joined: Fri Jan 13, 2006 11:25 am
Location: Uzes, France or Seattle, WA, USA

Unread postby shel » Mon Dec 18, 2006 8:29 am

Gerhard.Lenerz wrote:Shel, I was also thinking about the "driver" missing but from IRIX 6.5 it is included along with IRIX. Looking at the relevant directories I do have for example the vfe headers and object files for the kernel. But still, I might be missing something else.

Looking for files whose names contain "vfe" shows me ...

Code: Select all

/ # find / | grep -i vfe | more
/var/sysgen/master.d/if_vfe
/usr/local/lib/perl5/site_perl/5.6.1/sys/if_vfe.ph
/usr/include/sys/if_vfe.h
/usr/share/lib/perl5/site_perl/irix-n32/sys/if_vfe.ph
/usr/cpu/sysgen/IP25boot/if_vfe.o
/hw/node/vmebus/21/vfe
/hw/node/vmebus/21/vfe/0
/hw/net/vfe0

Got all those?

I wish I could remember what I did to make it work the first time. I remember having to install the software after I installed the board, since I got the VFE board long after I got the machine, and the software wasn't installed by default.

-Shel

User avatar
Gerhard.Lenerz
Posts: 330
Joined: Thu Jan 15, 2004 1:20 am
Location: Frankfurt/Main, Germany
Contact:

Unread postby Gerhard.Lenerz » Mon Dec 18, 2006 11:28 am

Let me see... i've got:

Code: Select all

/usr/cpu/sysgen/IP25boot/if_vfe.o
/usr/share/lib/perl5/site_perl/irix-n32/sys/if_vfe.ph
/usr/include/sys/if_vfe.h
/var/sysgen/master.d/if_vfe


The files in the /hw directory will probably turn up only if the machine knows about the hardware.

Speaking of that I did look at the /hw directory:

Code: Select all

aragorn 17# ls -la
total 0
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 .
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 ..
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 13
aragorn 18# pwd
/hw/node/vmebus


Good or bad? I don't know. #13 is apparently the VME host.


Gerhard

User avatar
landy
Posts: 40
Joined: Wed Oct 04, 2006 8:43 am
Location: Germany

Unread postby landy » Mon Dec 18, 2006 12:39 pm

Michael, if you could take a snapshot of the jumpers of your board that would be great and very much appreciated. :)


Gerhard, I'll be back at home on friday. If you have still problems on friday I 'll take the board out and make some pictures.

User avatar
shel
Posts: 304
Joined: Fri Jan 13, 2006 11:25 am
Location: Uzes, France or Seattle, WA, USA

Unread postby shel » Mon Dec 18, 2006 6:47 pm

Gerhard.Lenerz wrote:Let me see... i've got:

Code: Select all

/usr/cpu/sysgen/IP25boot/if_vfe.o
/usr/share/lib/perl5/site_perl/irix-n32/sys/if_vfe.ph
/usr/include/sys/if_vfe.h
/var/sysgen/master.d/if_vfe


The files in the /hw directory will probably turn up only if the machine knows about the hardware.

Yeah, but you're missing a couple of the others, I think.

Gerhard.Lenerz wrote:Speaking of that I did look at the /hw directory:

Code: Select all

aragorn 17# ls -la
total 0
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 .
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 ..
drwxr-xr-x    2 root     sys            0 Dec 18 11:19 13
aragorn 18# pwd
/hw/node/vmebus


Good or bad? I don't know. #13 is apparently the VME host.

I don't know what difference it makes, but ...

Code: Select all

/ # cd /hw/node/vmebus
/hw/node/vmebus # ls -la
total 0
drwxr-xr-x    2 root     sys            0 Dec 18 18:37 .
drwxr-xr-x    2 root     sys            0 Dec 18 18:37 ..
drwxr-xr-x    2 root     sys            0 Dec 18 18:37 21
/hw/node/vmebus # ls -la 21
total 0
drwxr-xr-x    2 root     sys            0 Dec 18 18:38 .
drwxr-xr-x    2 root     sys            0 Dec 18 18:38 ..
drwxr-xr-x    2 root     sys            0 Dec 18 18:38 vfe
/hw/node/vmebus # ls -la 21/vfe
total 0
drwxr-xr-x    2 root     sys             0 Dec 18 18:38 .
drwxr-xr-x    2 root     sys             0 Dec 18 18:38 ..
crw-------    1 root     sys         0,158 Dec 18 18:38 0

Keep in mind that mine is installed, and works (sort of) but isn't activated at the moment. I only need one Ethernet connection, and the et0 is faster, so I don't use the vfe.

-Shel

User avatar
R-ten-K
Posts: 1889
Joined: Mon Nov 15, 2004 10:36 pm
Location: Nor Cal

Re: Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ether

Unread postby R-ten-K » Tue Dec 19, 2006 12:22 pm

shel wrote:
Gerhard.Lenerz wrote:I'm trying to further expand my Onyx. I've installed a 013-1616-004 assembly (C8-100BT) into the first available VME slot on the machne (next to the IO4/VCAM).

Running autoconfig fails like this:


I installed it long enough ago that I forget exectly what I did, but I think I had to install some software from the IRIX 6.5 disk set in order to get the OS to recognize the board. Once I did that, and fiddled around a bit with various settings, I was able to get it to work.


running autoconfig will not enable the driver, since it is not installed by default. I had an onyx with a VFE a loooong time ago, basically search your overlays for the vfe.sw.eoe package, and then run the autoconfig again. I used to get these packages out of the ftp gate server at sgi.com (to give you an idea how long ago this was :-)), dunno if they still have it up. But I think I located it in the 3rd CD of a early release 6.5.x (x<10), but don't quote me on this.


However, I've never been able to get it to run as fast as the normal et0 10base network port. My understanding is that the VFE board is locked at 100mbs half-duplex, and doesn't announce that. Since most switches default to full-duplex, all sorts of collisions occur. However, even using a HD-FD converter, I wasn't able to get anything like 100mbps, or even 10mbps, out of the VFE board.


Same experience here, I had to force my switch to do half duplex for the onyx's port. Same with the gfe fast etherne I had for my olde indy in the same network, both interfaces only negotiate speed not duplexing. And both interfaces are half duplex only and do not share that information with the other interface in the negotiation. A lot of hubs/switches time out if they do not hear about the duplexing, so they revert to their worst setting... and so does the SGI interface. I.e. major retardation going on. I was able to get OK performance by forcing the if_vfe and if_gfe config files to set their devices to 100Tx/HD and also forcing the same on the networ port they were connected to.

Anyhow, I also had an FDDI ring in the same network, and at least the onyx was much happier there. The vfe seemed to be a half assed product.
"Was it a dream where you see yourself standing in sort of sun-god robes on a
pyramid with thousand naked women screaming and throwing little pickles at you?"

User avatar
shel
Posts: 304
Joined: Fri Jan 13, 2006 11:25 am
Location: Uzes, France or Seattle, WA, USA

Re: Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ether

Unread postby shel » Tue Dec 19, 2006 2:44 pm

R-ten-K wrote:
shel wrote:However, I've never been able to get it to run as fast as the normal et0 10base network port. My understanding is that the VFE board is locked at 100mbs half-duplex, and doesn't announce that. Since most switches default to full-duplex, all sorts of collisions occur. However, even using a HD-FD converter, I wasn't able to get anything like 100mbps, or even 10mbps, out of the VFE board.

Same experience here, I had to force my switch to do half duplex for the onyx's port. Same with the gfe fast etherne I had for my olde indy in the same network, both interfaces only negotiate speed not duplexing. And both interfaces are half duplex only and do not share that information with the other interface in the negotiation. A lot of hubs/switches time out if they do not hear about the duplexing, so they revert to their worst setting... and so does the SGI interface. I.e. major retardation going on. I was able to get OK performance by forcing the if_vfe and if_gfe config files to set their devices to 100Tx/HD and also forcing the same on the networ port they were connected to.

I'll have to take a look at if_vfe and see if I can just lock it to 100/HD. Certainly, futzing with ifconfig settings doesn't do anything useful, which is why I resorted to the protocol converter ... which didn't do anything useful, either.
Anyhow, I also had an FDDI ring in the same network, and at least the onyx was much happier there.

I have some kind of VME optical networking card for my Challenge, as well, but no other optical networking stuff to hook it to.
The vfe seemed to be a half assed product.

Indeed. It really seems to be a "checkbox" item; something the marketroids can use in a "competitive evaluation" to make things look good.

Oh, well, the 10/HD et0 is actually fast enough for what I use the machine for. Given the SCSI bus speeds, it's never going to be a super-high-speed fileserver, anyway.

-Shel

User avatar
Gerhard.Lenerz
Posts: 330
Joined: Thu Jan 15, 2004 1:20 am
Location: Frankfurt/Main, Germany
Contact:

Unread postby Gerhard.Lenerz » Sat Jan 06, 2007 10:05 am

Little update: I didn't get any further but I have found some pictures of the board. The quality is poor but I think it's visible that only four jumpers are set (one black, three white).

Image

Image

Before I try random jumper positions (best guess would probably to remove the black one), can someone confirm any working settings?


Gerhard

User avatar
jan-jaap
Donor
Donor
Posts: 4881
Joined: Thu Jun 17, 2004 11:35 am
Location: Wijchen, The Netherlands
Contact:

Unread postby jan-jaap » Sat Jan 06, 2007 2:20 pm

Gerhard.Lenerz wrote:Little update: I didn't get any further but I have found some pictures of the board. The quality is poor but I think it's visible that only four jumpers are set (one black, three white).

Before I try random jumper positions (best guess would probably to remove the black one), can someone confirm any working settings?


Gerhard

check near the bottom of this thread:
viewtopic.php?t=9527
:PI: :Indigo: :Indigo: :Indy: :Indy: :Indy: :Indigo2: :Indigo2: :Indigo2IMP: :Octane: :Octane2: :O2: :O2+: Image :Fuel: :Tezro: :4D70G: :Skywriter: :PWRSeries: :Crimson: :ChallengeL: :Onyx: :O200: :Onyx2: :O3x02L:
To accentuate the special identity of the IRIS 4D/70, Silicon Graphics' designers selected a new color palette. The machine's coating blends dark grey, raspberry and beige colors into a pleasing harmony. (IRIS 4D/70 Superworkstation Technical Report)

User avatar
skywriter
Posts: 3290
Joined: Fri Mar 14, 2003 5:22 am
Location: Trump-proof bunker
Contact:

Re: Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ether

Unread postby skywriter » Sat Jan 06, 2007 4:03 pm

R-ten-K wrote:Anyhow, I also had an FDDI ring in the same network, and at least the onyx was much happier there. The vfe seemed to be a half assed product.


strooth. i still use FDDI for all my pre-O2 stuff with a cisco FDDI->100baset bridge.
:Skywriter:

DECUS Member 368596

User avatar
Gerhard.Lenerz
Posts: 330
Joined: Thu Jan 15, 2004 1:20 am
Location: Frankfurt/Main, Germany
Contact:

Unread postby Gerhard.Lenerz » Sat Mar 17, 2007 7:42 am

Thanks to Jan-Jaaps help I've now everything set up nicely and now I see this in hinv:

Code: Select all

VME 100BaseTX Fast Ethernet: vfe0


It is true, it isn't terribly fast. But for the time being it is still better than the builtin ethernet port.

I should probably also note for future reference the jumper configuration that did the trick: J8 is set to ON-OFF-OFF-OFF-ON-ON as seen from left to right when viewed from the upper card edge. Every other jumper is in the OFF/OPEN position.


Gerhard

User avatar
mapesdhs
Posts: 2495
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: Challenge/Onyx VFE (C8-100BT 2-port 100Base-T Fast Ethernet)

Unread postby mapesdhs » Thu Feb 16, 2017 1:52 am

Sorry for resurrecting an old thread!

Does anyone know exactly where to find the vfe.sw.eoe package? I've been checking my various CDs, but no luck so far.

Ian.
I'm working on a charitable PC build for the Learn Engineering YouTube channel. Please PM/email/call if you'd like to contribute!
Donations of any kind of item I can sell to provide funds are also most welcome.
mapesdhs@yahoo.com
+44 (0)7434 635 121


Return to “SGI: Hardware”

Who is online

Users browsing this forum: No registered users and 4 guests