INFO-VAX Sat, 27 Jan 2007 Volume 2007 : Issue 54 Contents: Re: FPGA VAX Fun with GNV - Watch out for MNT. Re: Nonstop UNIX takes another loss Re: PPPD, TCPIP Services and incomplete negotiations Re: SHOW TERMINAL glitch (PPP) Re: VMS in the HP hierarchy ---------------------------------------------------------------------- Date: Sat, 27 Jan 2007 12:57:58 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: FPGA VAX Message-ID: In article , David Mathog wrote: >Tim Sneddon wrote: > >> FPGA VAX >> >> RT Logic is re-implementing VAX computer systems in FPGA for Hewlett >> Packard. By leveraging HP's VAX intellectual property, RT Logic will >> provide VAX single-board computers and peripherals to replace legacy >> systems critical to the maintenance of NATO's Air Defense System. > >How is this easier than dusting off the last VAX chip masks and doing a >run or two at a foundry? Given that 90 nm is pretty much standard these >days, and that's probably 10 times the resolution of the last VAX chip >that was made, cranking out a few actual chips shouldn't be that hard. Where did you get that idea? Have you priced out the set-up costs for a "run or two" at a foundry? How about cost of building (or perhaps refurbishing) the test equipment? Do you know of a foundry that exists that could fabricate using the right materials and processes for the last generation of VAX chips? You seem to think the only relevant property of a foundry is the minimum feature size it can produce (90nm). That's FAR from reality. Complex chips like a VAX are designed in conjuction with the foundry that will produce them, and the test equipment that will test each chip. There's no reason to think that the ancient tools and software that were used for VAX would work easily or cheaply with today's foundry equipment. For an application that probably needs less than 1000 chips, FPGA is likely MUCH cheaper. >That is, even if the VAX chip required cutting edge chip technology then >it's not anywhere near that now. > >Or during all the mergers did Dec/Compaq/HP manage to lose the masks, They were probably gone before the Digital/Compaq merger. But not lost; deliberatly tossed to save the considerable storage costs. They could perhaps be re-generated from CAD files (if those could be found and the data interpreted by today's software), but that might cost as much as the FPGA project. >so that they have no choice but to do it with an FPGA? For that matter, >why not just run a VAX emulator on an off the shelf CPU and skip the >custom chip entirely? A VAX emulator does meet the needs of most VAX customers. That's why the demand for new VAX hardware is so tiny. An FPGA design has the advantage that the set-up costs for the hardware are aleady paid, and the chip comes already electronically tested. The firmware that implements the VAX can be tested using conventional, and still actively used, VAX verification tools. ------------------------------ Date: Sat, 27 Jan 2007 10:06:59 -0500 From: "Peter Weaver" Subject: Fun with GNV - Watch out for MNT. Message-ID: <3e0701c74224$cc1ac1f0$2802a8c0@CHARONVAX> I decided to put GNV 2.1 on my Alpha/IA64 cluster the other day. My cluster has two system disks, one for the Alpha and one for the IA64. By default GNV creates a SYS$SYSDEVICE:[MNT] directory that contains pointers to each disk. It also creates a alias for SYS$COMMON:[GNV] pointing to SYS$SYSDEVICE:[000000]. So on my system the Alpha's SYS$SYSDEVICE:[MNT] has a link to the IA64's system disk, the IA64 system disk has a [MNT] directory with a pointer to the Alpha's system disk, the Alpha's system disk has a [MNT] directory with a pointer to the IA64's system disk, the IA64 system disk has a [MNT] directory with a pointer to the Alpha's system disk, the Alpha's system disk has a [MNT] directory with a pointer to the IA64's system disk, the IA64 system disk has a [MNT] directory with a pointer to the Alpha's system disk, the Alpha's system disk has a [MNT] directory with a pointer to the IA64's system disk, the IA64 system disk has a [MNT] directory with a pointer to the Alpha's system disk, the Alpha's system disk has a [MNT] directory with a pointer to the IA64's system disk. (I'll stop here, but DIR SYS$SYSDEVICE:[000000...] hasn't stopped yet.) But even with one system disk the SYS$COMMON:[GNV] points back to SYS$SYSDEVICE:[000000] which contains VMS$COMMON, which contains GNV which points back to [000000] which contains VMS$COMMON. I don't have the time to play right now, but I think installing GNV on a single system disk system will cause DIR SYS$SYSDEVICE:[000000...] to loop too. If anyone else out there has tried it on a single system disk system then let me know if DIR loops. I see in the readme notes there is a comment about editing PSX$UP_STARTUP.COM so I will have to see if I can fix this mess that way. If anyone else is bothered by this and have a way to report this then please do. Peter Weaver www.weaverconsulting.ca CHARON-VAX CHARON-AXP DataStream Reflection PreciseMail HP Commercial Hardware ------------------------------ Date: 27 Jan 2007 10:49:20 -0800 From: jreederhi@nac.net Subject: Re: Nonstop UNIX takes another loss Message-ID: <1169923760.858151.218500@a34g2000cwb.googlegroups.com> To clear up some confusion, HP NonStop servers are the old Tandem systems acquired when Compaq acquired Tandem and then HP acquired Compaq. NonStop servers achieve their fault-tolerance by packaging up to 16 processors in a node. Critical software processes are backed up with processes running in another processor. Their state is kept synchronized via checkpointing so that the backup process can take over instantly. NonStop's recent versions included Himalaya and now the NonStop Integrity (Itanium). Now to the question about what is meant by NonStop Unix. NonStop comes with a Unix-like POSIX personality called OSS (Open System Software). JVM and other open source software run on it. This is what was referred to as "NonStop Unix" at the Toronto Stock Exchange. The Exchange is replacing its NonStop trading system with a Linux system call Leapfrog. Similar to replacing a tractor with 10,000 chickens. www.availabilitydigest.com On Jan 17, 8:58 am, "n.ri...@sympatico.ca" wrote: > NonstopUNIXtakes another loss as the Toronto stock exchange moves to > LINUX on AMD Opteron or Intel Xeon > > http://www.itbusiness.ca/it/client/en/home/News.asp?id=41862 > > Neil Rieck > Waterloo, Ontario, Canada. ------------------------------ Date: Sat, 27 Jan 2007 05:31:15 -0500 From: JF Mezei Subject: Re: PPPD, TCPIP Services and incomplete negotiations Message-ID: <45bb2a50$0$5969$c3e8da3@news.astraweb.com> Forrest Kenney wrote: > There is an undocumented tracing capability built into PPPD. To=20 > enable it use the command set ttxx/trace/full. The /full is optional > it just tracks more packet types. =20 Sorry for delay in followups, but have been quite busy checking everythin= g,=20 including resoldering serial connectors to ensure full modem control. (la= ck=20 of carrier detect caused the VMS terminla driver to complain). Now however, I cannot get any PPP connection going. Here is a trace at the serial level (between a MAC and the Alpha. This=20 starts just after the PPPD CONNECT command was manually issued. The short story, the MAC sends configuration requests, but no data comes = back from the alpha after it has issued the "CONNECTTERM" message. R denotes characters sent from the Mac. L denotes characters sent by the Alpha. followed by minutes,seconds,hundreds of seconds followed by the dash, followed by number of characters in that buffer. R:56:48:92- 1 =B0 0 D L:56:48:99- 96 =B0=B0=B0%PPPD-I-CONNECTTERM, converting connection on de= vice _VTA 00025554242444444554542266676776662666666766626626676662= 5554 DAD50004D9D3FEE534452DC03FE65249E703FEE5349FE0FE04569350= F641 139: to a Point-to-Point connection=B0 333327626256667276256667266666676660 139A04F0100F9E4D4FD0F9E403FEE5349FED R:56:51:35- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:56:53:26- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:56:55:18- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:56:57:09- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:56:59:00- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:57:00:92- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:57:02:83- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE R:57:04:74- 45 ~=B0}#=C0!}!%} }4}"}&} } } } }%}&=C7=C0=FC=CE}'}"}(}"=A5}= <~ 7F72C272272737272727272727272CCFC72727272A737 EFD301D15D0D4D2D6D0D0D0D0D5D670CED7D2D8D25DCE RE: PPPD SHOW TTA0: This correctly picks up the speed from the terminal device. But it defaul= st=20 to XON/XOFF flow control. But even if I get the /FLOW=3DHARD in the PPPD,= it=20 doesn't seem to make a difference. Note that I can succesfully login=20 manually and issue the PPPD command interactively and there are no issues= =20 with flow control at that time. RE: PPPD SET TRACE /FULL If I do SET TRACE=3Dtrace.log /FULL, I get a nasty insufficient privilege= =20 error after it has converted the line. No trace log is created. If I don't specify a file, it seems to go to MBA163: But which process=20 consumes the mailbox data ? Must I manually start the logger utility from another terminal to capture= =20 the data ? Also, I realise that PPPD doesn't support authentication. Will it fail if= a=20 client tries to authenticate ? Or will it just respond with a success=20 status no matter what credentials are presented ? ------------------------------ Date: Sat, 27 Jan 2007 13:54:16 +0000 (UTC) From: moroney@world.std.spaamtrap.com (Michael Moroney) Subject: Re: SHOW TERMINAL glitch (PPP) Message-ID: JF Mezei writes: >$ show term tta0 >Terminal: _ASN10: Device_Type: Unknown Owner: No Owner >Physical terminal: _TTA0: >Terminal Characteristics: ... >%SYSTEM-F-BADPARAM, bad parameter value >Why the error message at the bottom ? I played with PPP long (6+ years) ago and remember seeing this. I looked into it enough to see $ show terminal issued 2 QIOs against the terminal port, one (must have been a SENSEMODE) worked, the other failed with the BADPARAM status. I don't remember what the other function was or what information it was trying to retrieve. ------------------------------ Date: Sat, 27 Jan 2007 09:00:11 -0800 From: "Tom Linden" Subject: Re: VMS in the HP hierarchy Message-ID: On Sat, 27 Jan 2007 05:09:28 -0800, Robert Deininger wrote: >> Since VMS was included in VMS, I have noticed Stallard acquiring >> responsabilities and growing his empire. All I know of the guy is his >> May >> 7th memo. > In actual fact, after he wrote that memo, which was intentionally > designed > to drive away as much VMS business as possible, he has done absoultely > NOTHING except cash his paychecks. Even with your fox-hole impaired > tunnel vision, you haven't missed anything that Scott has done. The margins on the VMS business exceed that of all other lines of business, save printers, so why would you want to drive it away? -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ End of INFO-VAX 2007.054 ************************