Thank you so much Mark and Matt for the update.
I downloaded your two kits and I tested them (one at a time) with SIMH
V3.8-1 under OpenVMS V8.4 Update600 environment. I confirm you that with
these new two kits the CPUSPINWAIT crash seems to be gone.
Unfortunally under the SIMH environment the network still doesn't work
(protocols used : SCS, Decnet, LAT and TCP/IP - all correctly configured
on XQA: device). Basically, the same problems observed three weeks ago, when
I started my tests on this environment.
It's very strange because with his rx2660, Mark is able to work with the
emulated network adapter without error. What am I doing wrong ?
From Integrity side :
SYSTEM_YODA > run SYS$SYSDEVICE:[SIMH]VAX-I64.EXE
sim> show xq eth
ETH devices:
0 LL0 (VMS Device: _LLA0:)
1 IE1 (VMS Device: _EIB0:) <-- I reserve this network adapter for SIMH
emulator !!!
sim> set cpu 512m
sim> set cpu conhalt
sim> load -r ka655x.bin
sim> attach NVR ka655.nvr
sim> set telnet 5305
sim> set TQ enable
sim> set TS disable
sim> set LPT enable
sim> set RQB disable
sim> set RQC disable
sim> set RQD disable
sim> set RL disable
sim> set RY disable
sim> set rq2 disable
sim> set rq3 enable
sim> set tq1 enable
sim> set tq2 disable
sim> set tq3 disable
sim> set RL disable
sim> set LPT disable
sim> set TQ TK70
sim> set rq0 ra92
sim> attach rq0 ovmsvaxsys.dsk
sim> set rq1 ra92
sim> attach rq1 private.dsk
sim> set xq type=delqa
sim> attach xq eth1
sim> boot cpu
From SIMH side :
LAN Data Structures
-------------------
LSB address = A1C07580
Device state = 0013 Inited,Run,Timer
-- XQA Unit Summary 24-MAR-2012 18:50:23 --
UCB UCB Addr Fmt Value Client State
--- -------- --- ----- ------ -----------
XQA0 A1BAFF00
XQA16 A1D3CE40 Eth 60-03 DECNET 0017 Strtn,Len,Uniq,Strtd
XQA17 A1BB0180 Eth 60-01 MOPDL 001F
Strtn,Len,Uniq,Share,Strtd
XQA18 A1D37DC0 Eth 90-00 LOOP 001D Strtn,Uniq,Share,Strtd
XQA19 A1C7D880 Eth 60-04 LAT 0015 Strtn,Uniq,Strtd
XQA20 A1C80300 Eth 08-00 IP 0015 Strtn,Uniq,Strtd
XQA21 A1C7F900 Eth 08-06 ARP 0015 Strtn,Uniq,Strtd
XQA22 A1C7EF00 Eth 86-DD 0015 Strtn,Uniq,Strtd
XQA23 A1BBDF00 Eth 00-00 0000
XQA24 A1C80D00 Eth 00-00 0000
SYSTEM_SIMH> ifconfig -a
LO0: flags=100c89<UP,LOOPBACK,NOARP,MULTICAST,SIMPLEX,NOCHECKSUM>
inet 127.0.0.1 netmask ff000000 ipmtu 4096
QE0: flags=c43<UP,BROADCAST,RUNNING,MULTICAST,SIMPLEX>
inet 192.168.1.8 netmask ffffffc0 broadcast 192.168.1.63 ipmtu 1500
SYSTEM_SIMH> ping 192.168.1.1 ! (the router)
PING 192.168.1.1 (192.168.1.1): 56 data bytes
----192.168.1.1 PING Statistics----
4 packets transmitted, 0 packets received, 100% packet loss
%SYSTEM-F-TIMEOUT, device timeout
SYSTEM_SIMH> ping 192.168.1.5 ! (the Integrity Server)
PING 192.168.1.5 (192.168.1.5): 56 data bytes
----192.168.1.5 PING Statistics----
4 packets transmitted, 0 packets received, 100% packet loss
%SYSTEM-F-TIMEOUT, device timeout
From Integrity side :
LAN Data Structures
-------------------
-- EIB Device Summary 24-MAR-2012 18:51:45 --
LSB address = 9E81A000
Device state = 00003103 RUN,INITED,RING_AVAIL,FP_CPUAFF,AUTO_ENA
UCB UCB Addr Fmt Value Client State
--- -------- --- ----- ------ -----------
EIB0 8E266D00
EIB6 8ECFF640 Eth 00-00 0000
EIB12 8E924B80 Eth 60-06 USER 0015 STRTN,UNIQ,STRTD
SYSTEM_YODA >ifconfig -a
LE0: flags=c43<UP,BROADCAST,RUNNING,MULTICAST,SIMPLEX>
*inet 192.168.1.5 netmask ffffffc0 broadcast 192.168.1.63 ipmtu 1500
LO0: flags=100c89<UP,LOOPBACK,NOARP,MULTICAST,SIMPLEX,NOCHECKSUM>
inet 127.0.0.1 netmask ff000000 ipmtu 4096
SYSTEM_YODA >ping 192.168.1.8 ! (the simh Ip address)
PING 192.168.1.8 (192.168.1.8): 56 data bytes
----192.168.1.8 PING Statistics----
4 packets transmitted, 0 packets received, 100% packet loss
%SYSTEM-F-TIMEOUT, device timeout
It is very interesting to observe that the LAT protocol behaves differently
(Incoming and
Outcoming connection are enabled on both nodes - rx1600 and SIMH ) :
SYSTEM_SIMH> mc latcp sho service
Service Name Status Identification
---------------- ----------- -------------------------------------------------
SIMH Available @SYS$MANAGER:ANNOUNCE.COM
VAX_SIMULATOR Available SIMH-Vax running on rx1620 & OpenVMS V8.4
SYSTEM_YODA >mc latcp sho service
Service Name Status Identification
---------------- ----------- -------------------------------------------------
JEDY Available OpenVMS Alias Name Cluster
SIMH Available @SYS$MANAGER:ANNOUNCE.COM
VAX_SIMULATOR Available SIMH-Vax running on rx1620 & OpenVMS V8.4
YODA Available @SYS$MANAGER:ANNOUNCE.COM
SYSTEM_YODA >set host/lat VAX_SIMULATOR
%LAT-F-PROTERR, communication protocol error
-LAT-I-END, control returned to node YODA
SYSTEM_YODA >set host/lat simh
%LAT-F-PROTERR, communication protocol error
-LAT-I-END, control returned to node YODA
It seems that LAT protocol under SIMH is able to transmit but it is not
able to receive :
EIB12 8E9279C0 Eth 60-06 USER 0015 STRTN,UNIQ,STRTD
(Integrity Server)
XQA11 A1D50080 Eth 60-04 LAT 0015 Strtn,Uniq,Strtd (Simh)
Any hint will be appreciated.
Best regards,
/Maurizio
----- Original Message -----
From: "Matt" <[email protected]>
To: "Mark Pizzolato - Info Comm" <[email protected]>
Cc: "Maurizio De Tommaso (Gmail)" <[email protected]>; "Hittner,
David T (IS)" <[email protected]>; "Simh Trailing-Edge Mailing List"
<[email protected]>
Sent: Thursday, March 22, 2012 10:37 PM
Subject: Re: [Simh] Simh and OpenVMS Integrity V8.4 Network Support issues
On 22/03/2012 20:42, Mark Pizzolato - Info Comm wrote:
Well, I looked into this. I've got an rx2600 recently acquired cheap on
eBay. I updated the firmware on this system to the latest before
installing VMS and doing any testing.
I started with the vms-pcap.zip downloaded from:
https://github.com/markpizz/simh/downloads
Everything worked as expected. No crashes and full Ethernet behavior.
I then took the vms-pcap.zip from http://www.9track.net/simh/pcap
I expected these to be the same since I worked with Matt last year when I
picked up his fixes and merged them into the code on github.
They were not the same. The code from http://www.9track.net/simh/pcap
produced the crashes which Maurizio described.
Looking for what was different, I compared the source. The code on
github.com had a version of pcapvcm.c which had comments describing it as
Version X-3. The code from 9track.net had comments describing it as
Version X-2. There was also a difference in snprintf.c which cleaned up
compiler warnings on some platforms. Additionally, there were an extra
file in the zip file from github.com called 0readme_pcap-VMS.txt which
describes the relative directory structure the pcap-vms code should have
with respect to the simh source which references it (via simh's
descrip.mms file).
I'm not sure why Matt's website has the older code in its zip file since
I got the code on github from Matt. The right answer seems to use the
code from github.
Meanwhile, I also looked at the libpcap provided with OpenVMS 8.4 TCP/IP
5.7 ECO3. This libpcap implementation does NOT have pcap_sendpacket, so
it won't be useful for simh networking.
- Mark Pizzolato
I know what's happened here. I recently had to move to a new server, due
to disk problems and the website content was restored from the copy on
my local system, which should have been up to date but clearly wasn't. I
think I've uploaded the correct version now (it identifies itself as
X-3). Sorry for any inconvenience this has caused.
Maurizio, does this fix the problems you were seeing?
Matt
_______________________________________________
Simh mailing list
[email protected]
http://mailman.trailing-edge.com/mailman/listinfo/simh
