1 INFO-VAX	Sun, 02 Sep 2001	Volume 2001 : Issue 487       Contents:( RE: 100 million sale for Compaq to Sabre Re: 16 VLCs for sale Re: A very sad moment.!  Re: alpha - ia64 Re: alpha - ia64 Re: alpha - ia64 AlphaServer DS20's - Cheap  DCL command file to requeue jobs7 Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP. 7 Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP. 7 Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP.  Re: I hate Compaq  Re: I hate Compaq  Re: I hate Compaq  Re: I hate Compaq  Intel Press Releases RMS index file performance Re: Tunneling DECnet over IP Re: Tunneling DECnet over IP Re: Tunneling DECnet over IP Re: Tunneling DECnet over IP VMS 7.3 compatibility issues  Re: VMS 7.3 compatibility issues  F ----------------------------------------------------------------------  $ Date: Sat, 1 Sep 2001 14:46:01 -0400+ From: "Main, Kerry" <Kerry.Main@compaq.com> 1 Subject: RE: 100 million sale for Compaq to Sabre R Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4D49600@kaoexc01.americas.cpqcorp.net>   Jeff,   J >>> If someone shifts from a Ford Crown Victoria to a Lexus does that thanL mean that a Toyota Camry isn't good enough to replace a Ford Crown Victoria? <<  K Your comment and this thread's discussion reminds me of a "dog story" story J mentioned in Greg Pfisters cluster book. It dealt with Lou Gerstner's (IBMJ CEO) view of the press ie. no matter what IBM did about its mainframes, it would be reported negatively ..   J Story goes that Lou could throw a stick into the middle of a lake and haveL his new trained dog walk on top of the water and retrieve the stick. He thenJ stated that the headlines in the press the next day would be "IBM Buys Dog That Can't Swim !"   :-)    Regards,  
 Kerry Main Senior Consultant  Compaq Canada Corp.  Professional Services  Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----+ From: Jeff Killeen [mailto:Jeff@IDM-IO.com]  Sent: August 29, 2001 11:27 PM To: Info-VAX@Mvb.Saic.Com 1 Subject: Re: 100 million sale for Compaq to Sabre       : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3B8D9EFA.2A07256D@videotron.ca... > Jeff Killeen wrote: J > > Unlikely since they wanted five 9's reliability. To my knowledge there is no F > > hardware currently being manufactured for either VMS or Tru64 that providesI > > that level of reliability.  You can build high availability Wildfires  (thereH > > will be a session or BOF at CETS) but you can't build fault tolerant > > Wildfires... > K > Well, if the predecessor was IBM mainframes with MVS, and VMS wasn't good I > enough to replace the IBM mainframe, and they had to go to Tandem, then  what > does it say for VMS ?   I Nothing - period - How do you know that VMS wasn't good enough to replace J MVS?  All you know is they choose NSK.  They could have chosen NSK becauseJ of they wanted something better. It is very false logic what was suggestedF above.  It could be that VMS is better than MVS but neither MVS or VMS supplied what they wanted.  K If someone shifts from a Ford Crown Victoria to a Lexus does that than mean G that a Toyota Camry isn't good enough to replace a Ford Crown Victoria?   L It amazes me the few in this newsgroup who see everything as a proving theirC negative viewpoint of Compaq's handling of VMS.  It rained today in G Houston - that proves the point of view Compaq is trying to kill VMS...    ------------------------------  % Date: Sat, 01 Sep 2001 20:26:58 +0200 2 From: martin@radiogaga.harz.de (Martin Vorlaender) Subject: Re: 16 VLCs for sale ; Message-ID: <3b912872.524144494f47414741@radiogaga.harz.de>   , Chris Doran (chris_doran@my-deja.com) wrote:H > For those who aren't aware of it: you don't have to use them for theirE > designed purpose as workstations coupled to a bigger system -- just @ > load VMS and you have quite a nice baby VAX which can use a VT > terminal.   < I really shouldn't tell, but one of our main internet system= (the one handling the primary DNS and all mail) is a VLC with = attached StorageWorks disks. Never had it fail in - hmmm... - ' I think it's more than three years now.    cu,    Martin --  J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.de J One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Sat, 01 Sep 2001 13:57:00 -0400 ' From: Howard S Shubs <howard@shubs.net>   Subject: Re: A very sad moment.!< Message-ID: <howard-FA1188.13570001092001@enews.newsguy.com>  : In article <1ez18jk.83dgyffuif9qN%andekl_no@saaf_spam.se>,-  andekl_no@saaf_spam.se (Anders Eklf) wrote:   * > Howard S Shubs <howard@shubs.net> wrote: > G > > > I had no response to this - I just felt too old, and I'm only 36!  > >  > > Kids these days... > @ > Yeah - just a little setback, and they give up... Must be MTV.  3 Not what I meant.  I was sympathizing.  I'm 36 too.  --   Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  # Date: Sat, 01 Sep 2001 18:24:37 GMT 2 From: "Yousuf Khan" <bbbl67@nospam.yahoo.com.spam> Subject: Re: alpha - ia64 = Message-ID: <FJ9k7.71260$n75.17819613@news4.rdc1.on.home.com>   1 "Alex Johnson" <spectre@jhu.edu> wrote in message ! news:3B901219.194A55FB@jhu.edu...  > Yousuf Khan wrote: > > K > > One will note that when the Hammer line comes out from AMD, it's 64-bit  modeK > > will default to 64-bit registers for addressing operands, but to 32-bit I > > registers for data operands. Just proves that 64-bit is only good (so  far)) > > for addressing and nothing much more.  > I > Well, except when you keep your data as 32 bit and your addresses as 64 J > bit, then some bad programmer startes manipulating addresses as data andE > you end up forced into 32 bit addresses again.  Never trust the end F > user.  Make all integers 64 bit by default.  It isn't much slower toJ > compute 64 bit adds than 32 bit adds and space is not an issue any more.  B It's not possible to manipulate addresses as 32-bit unless you useK address-size-overrides, or code the whole program in the legacy 32-bit mode 1 in which case it will be run in a 32-bit segment.        Yousuf Khan    ------------------------------  % Date: Sat, 01 Sep 2001 16:08:19 -0700 & From: name99@mac.com (Maynard Handley) Subject: Re: alpha - ia64 7 Message-ID: <name99-0109011608190001@handma2.apple.com>   < In article <3B8DE7E0.6815693F@hda.hydro.com>, Terje Mathisen% <terje.mathisen@hda.hydro.com> wrote:    > Nick Maclaren wrote: > > A > > In article <B4kj7.54329$n75.13608013@news4.rdc1.on.home.com>, 5 > > Yousuf Khan <bbbl67@nospam.yahoo.com.spam> wrote: < > > >"Sarah Warner" <sarahamiee7@yahoo.com> wrote in message? > > >news:5vgj7.162589$Xr6.910102@news-server.bigpond.net.au... & > > >> EPIC is better than RISC, mate. > > >> > > >> Do a bit of research. > > > P > > >You must be the same person as Smitty, right? Both from Australia, what areG > > >the chances that there are two complete morons out of the millions 	 that come  > > >from country? > > C > > The UK is full of civilised, witty and educated Australians who G > > left because they couldn't stand the density of the morons in their  > > native country ....  >  > Ouch.  > G > I assume you've seen the news about the norwegian ship that picked up  > 483 boat refugees? > F > The _australian_ coast guard asked them to assist an overloaded boatD > that was about to sink, and the norwegian vessel of course did so. > = > At this point australia turned around and said (against all H > international maritime laws) that the ship could not go to the nearestJ > harbour (Australian Christmas Island) and offload the same refugees they > had been asked to pick up. > 7 > It is an election year in Australia, but even so! :-(   L And Norway allows in what fraction of its population every year in refugees?J The CIA world factbook lists Norway as having 2.13 immigrants per year perG 1000 population as opposed to 4.26 for Australia. Additionally the only C ethnic minority Norway appears to have is 20 000 Sami (from a total @ population of 4.5 million) as opposed to 7% Asian in Australia. J It's easy to be high and mighty about principles when you're not next door4 to a slow-motion disaster like Indonesia, isn't it?    Maynard    ------------------------------  # Date: Sun, 02 Sep 2001 00:02:39 GMT   From: "eddie" <NullVoid@att.net> Subject: Re: alpha - ia64 F Message-ID: <zGek7.6777$151.493073@bgtnsc05-news.ops.worldnet.att.net>  3 "Maynard Handley" <name99@mac.com> wrote in message 1 news:name99-0109011608190001@handma2.apple.com... > > In article <3B8DE7E0.6815693F@hda.hydro.com>, Terje Mathisen' > <terje.mathisen@hda.hydro.com> wrote:  >  > > Nick Maclaren wrote: > > > C > > > In article <B4kj7.54329$n75.13608013@news4.rdc1.on.home.com>, 7 > > > Yousuf Khan <bbbl67@nospam.yahoo.com.spam> wrote: > > > > >"Sarah Warner" <sarahamiee7@yahoo.com> wrote in messageA > > > >news:5vgj7.162589$Xr6.910102@news-server.bigpond.net.au... ( > > > >> EPIC is better than RISC, mate. .  .  . ? > > At this point australia turned around and said (against all J > > international maritime laws) that the ship could not go to the nearestL > > harbour (Australian Christmas Island) and offload the same refugees they > > had been asked to pick up.  ; I agree with Australia's decision to keep the refugees out.   F I would like to help everyone in the world but it just isn't possible.  A I understand why they want to get out of their country, but, they < are basically stealing resources from others for themselves.  E No country has the resources to help all of the world's unfortunates.    > > 9 > > It is an election year in Australia, but even so! :-(  >   . .  .  > 	 > Maynard    ------------------------------  $ Date: Sat, 1 Sep 2001 18:21:28 -04009 From: "D.B. Turner, islandco.com" <dbturner@islandco.com> # Subject: AlphaServer DS20's - Cheap / Message-ID: <tp2nd4rplq6v8a@news.supernews.com>   	 For Sale:   / DS20 with VMS EIP Licenses Refurbished (as new) $ Single CPU 500Mhz (Dual CPU capable) 1GB DEC Memory (4 x 256)% 3DLabs Oxygen VX1 32MB PCi Video Card  32x CD-ROM SCSI  4 x 18GB 7200RPM Storageworks '  KZPAC-CB 8MB 3 Channel RAID Controller  (Split bus kit installed)  DE500-BA 10/100 Ethernet Card  Keyboard Mouse  17" Monitor (NEW)  All this for $8995!!!!!!  % Limited Qty available at this price ! % 1 Year warranty from Island Computers    We sell Alpha's & Alpha Parts  http://www.islandco.com  sales@islandco.com Island Computers US Corp.  2700 Gregory Street  Savannah GA 31404  Tel: 912 447 6622  Fax: 912 201 0096    ------------------------------  % Date: Fri, 31 Aug 2001 20:54:13 -0400 ! From: "Airnews" <Kuff@Tessco.Com> ) Subject: DCL command file to requeue jobs O Message-ID: <DAC100152F603D79.3FAB745C8999546F.87C96A7F18847D20@lp.airnews.net>   K Thought I saw a command file that would re-create in a DCL script all of he 4 jobs queued in a system... that way one could changeJ "_$1$Dua10:[Foo]Test.Dat;37" to _$1$Dga1276:[Foo]Test.Dat;"  and re-submitJ them after zapping the queue file... thinking an OpenVMS upgrade that usesL this for cases where jobs are queued and they need to completely restructure the queue file...    ------------------------------  # Date: Sat, 01 Sep 2001 21:50:03 GMT + From: Jeff Campbell <jcampbell@ins-msi.com> @ Subject: Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP.+ Message-ID: <3B91509A.C2D2FB85@ins-msi.com>    sword7@speakeasy.org wrote:  > = > In comp.os.vms Jeff Campbell <jcampbell@ins-msi.com> wrote: G > > The Q22-Bus IOPAGE addresses are directly addressed in the physical / > > address scheme at 0x20000000 to 0x20001fff.  > C > > The console ROM code must map enough local memory space to boot G > > the system. The Q22-Bus devices will be doing DMA into the physical B > > Q22-Bus address space. The mapping array will cause the DMA to1 > > be to physical MicroVAX memory, page by page.  >  > Jeff:  > L > Thank you for information about Q-bus specs.  However, I have one questionH > for you about IOPAGE (20000000 to 20001FFF).  I noticed that in tracedG > execution log file and it tried to access 20001F40, etc.  Does anyone H > have information about IOPAGE area?  I assume that IOPAGE registers isK > for Q-bus registers.  Q-bus memory space is for drive registers, ethernet  > registers, etc...   H The Qbus has a signal wire called BB7SL: Bus Bank Seven Select Low. ThisG signal is asserted by the host CPU (MicroVAX, PDP11) to access the 8 KB H device IO space window. Effectively, BB7SL is another address line. In aF PDP11 the 16 bit physically addressable memory space is broken up intoC eight 8 KB segments. With no MMU or the MMU not enabled the CPU can # address 64 KB: (addresses in octal)       0000000  bank 0    0020000  bank 1    0040000  bank 2    0060000  bank 3    0100000  bank 4    0120000  bank 5    0140000  bank 6    0160000  bank 7  E When an address in the range 0160000 to 0177777 is asserted the BB7SL H is also asserted, indicating an IOPAGE access. Memory cards are requiredA to monitor this signal and not respond with BRPLYL when an IOPAGE  address cycle is in progress.   D When the MMU is enabled, BB7SL is asserted when a virtual address inC the IOPAGE range is asserted and the contents of the bank 7 mapping ? register (PAR 7) point to the last physically addressable bank.   B In the MicroVAX II the physical memory space 0x20000000-0x20001fff> corresponds to bank 7 addressing, generating the BB7SL signal.  D All IO device registers occur in the 8 KB bank 7 address space. Some> devices might also have memory that is visible on the Qbus. An@ ethernet interface might dual port it's memory pool so that bothC the host CPU (via the Qbus) and the onboard ethernet chip can share C the same memory. The MicroVAX would address this memory through the  0x30000000 space.   = The whole 22 bit Qbus memory space is addressable through the * physical addresses starting at 0x30000000.  < I forgot to mention the 0x20001f40 address before. This is aD register in the Qbus IOPAGE but implemented by the MicroVAX II chip.F It's called the Interprocessor Communications Register, IPCR. The IPCRG allows upto 4 CPUs to share the same Qbus! The IPCR defines these bits:   C    <15>     DMA QPE - DMA Qbus Address Space Parity Error. MSER <4> ,                       must be set to enable.       <14:09>  unused, read as zero  B    <08>     AUX HLT - Auxiliary Halt. Forces AUX CPU to enter HALT                       ROM code.        <07>     unused, read as zero  0    <06>     DBI IE  - Doorbell Interrupt Enable.  G    <05>     LM  EAE - Local Memory External Access Enable. Setting this E                       bit enables external access to local memory, by D                       means of the Q22-Bus Memory Mapping Registers.       <04:01>  unused, read as zero  D    <00>     DBI RQ  - Doorbell Interrupt Request. Enabled by DBI IE.  1   >>> I noticed that in traced execution log file +   >>> and it tried to access 20001F40, etc.   E The LM EAE bit (0x20) would be set after the ROM code has initialized  the B mapping register array to allow the bootstrap driver(s) to have IOD buffer(s) the booted device interface can R/W. I think (do not know)E the ROM code maps the whole Qbus address space into the first 4 MB of 0 main memory page for page. Would make sense. 8-)   >  > Thank you! >  > -- Tim Stark >  > --4 > Timothy Stark   <><     Inet: sword7@speakeasy.orgL > --------------------------------------------------------------------------G > "For God so loved the world, that he gave his only begotten Son, that J > whosoever believeth in him should not perish, but have everlasting life.0 > Amen." -- John 3:16 (King James Version Bible)  
 Jeff Campbell  n8wxs@arrl.net   ------------------------------  % Date: Sat, 01 Sep 2001 23:52:20 -0000  From: sword7@speakeasy.org@ Subject: Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP./ Message-ID: <tp2t5k72hno564@corp.supernews.com>    Jeff:   E Ok, thank you for information about IPCR (0x20001F40).  Also, I foundSF complete QD21 specs (same as UDA50/RXQDx).  I already had DEQNA specs 5 before.  Ok, I will finish my MicroVAX emulator soon.   
 Thank you!   -- Tim Stark   --  , Timothy Stark	<><	Inet: sword7@speakeasy.orgJ --------------------------------------------------------------------------F "For God so loved the world, that he gave his only begotten Son, that H whosoever believeth in him should not perish, but have everlasting life.. Amen." -- John 3:16 (King James Version Bible)   ------------------------------  % Date: Sun, 02 Sep 2001 00:01:37 -0000r From: sword7@speakeasy.org@ Subject: Re: Good news! MicroVAX II - SAVPSL, SAVPC, and SAVISP./ Message-ID: <tp2tn1a7sgd6d9@corp.supernews.com>t  9 In comp.os.vms antonio.carlini <arcarlini@iee.org> wrote:-( > If you get stuck with the MicroVAX II,' > you could pick up the tech manual for0 > the MicroVAX 2000 from3 >  http://208.190.133.201/decimages/moremanuals.htm0! > and have a go with that - there  > might be fewer unknowns.  J Thank you for info.  I found some userful information.  I downloaded some.J Also, I found QD21 specs (UDA50/RXQDx) for disk/tape drives that MicroVAX K II requires.  Also, I found KMC-11 and DUP-11 specs that I was looking for rE more than a year for my working KS10 emulator!  That is designed for   DECnet/ANF-10 network.  ) > The 78032 user guide is badly bound and') > falling apart, so I'll probably scan itr > quite soon anyway.     Go ahead and let me know.r  I >> Does anyone have information about restart code?  Only I know that 0x3 H >> is power-up state or so.  I had seen 0x5, 0x3, and 0x2 in diassembled >> VMB code.   > The documented codes are:i  ! > 02 - HALT (the signal) assertede) > 03 - initial power on or RESET asserted , > 04 - int. stack not valid during exception3 > 05 - machine check during machine check or kernal6& >       stack invalid during exception, > 06 - HALT instruction while in kernel mode > 07 - SCB vector <1:0> = 11 > 08 - SCB vector <1:0> = 10 > 09 - not listed !1, > 0A - CHMx execute while on interrupt stack& > 10 - ACV or TNV during machine check9 > 11 - ACV or TNV during kernel stack not valid exceptionm  G Thank you for information again.  03 is initial power on.  My guess was E right.  I already tried 03 and KA630 console program finally perform  I normally.  Whew.  I will implement that codes for 'halt' command, etc on h TS10 prompt.   -- Tim Stark   -- b, Timothy Stark	<><	Inet: sword7@speakeasy.orgJ --------------------------------------------------------------------------F "For God so loved the world, that he gave his only begotten Son, that H whosoever believeth in him should not perish, but have everlasting life.. Amen." -- John 3:16 (King James Version Bible)   ------------------------------   Date: 1 Sep 2001 18:09:34 -0500M- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: I hate Compaq3 Message-ID: <NBmtmjnzC45E@eisner.encompasserve.org>t  [ In article <9mqpca$hlj$1@pegasus.csx.cam.ac.uk>, nmm1@cus.cam.ac.uk (Nick Maclaren) writes:t  < > I do not know of a single Unix vendor that provides eitherB > reliable and functional signal and (hardware) exception handling? > (clearly necessary for truly robust programs), nor a debuggerr? > that will handle genuine problems.  And that is simply in theeA > context of a single-threaded application - the SMP situation isn > much, much worse.t  B But by restricting the search space to Unix, you are starting withB a candidate set of vendors who have similar restrictions, and also< have a common set of customer expectations.  A different OS,3 about which I have no ax to grind, might be OS/400.D  D > And, while the debugger situation is a little better, I don't knowC > of ANY modern system where it is possible to debug stack trashingoA > in an application signal handler, to take one relatively simplec > (sic) example.  @ This example seems to be rather C-centric, and lack of automatic? buffer limit checking in C is probably the major cause of blowno stacks.n  A So yes, a more robust error detectiong environment (lots of guardaC pages, etc) might be desired by those who choose C.  But apparently.; not by enough of them to make a it profitable for a vendor.t  D > As I have posted before, I have proof positive that the insanitiesA > of the C and POSIX standards (which are better than Microsoft'sdB > conventions) cause occasional, unpredictable application failureC > on a couple of Unix systems, and very good evidence that the samerA > is true for virtually every one I have used in the past decade.:    If it hurts when you do that,...   ------------------------------   Date: 1 Sep 2001 23:26:16 GMTh( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: I hate Compaq0 Message-ID: <9mrqqo$h4o$1@pegasus.csx.cam.ac.uk>  3 In article <NBmtmjnzC45E@eisner.encompasserve.org>,s. Larry Kilgallen <Kilgallen@SpamCop.net> wrote:\ >In article <9mqpca$hlj$1@pegasus.csx.cam.ac.uk>, nmm1@cus.cam.ac.uk (Nick Maclaren) writes: >h= >> I do not know of a single Unix vendor that provides eitheroC >> reliable and functional signal and (hardware) exception handlingu@ >> (clearly necessary for truly robust programs), nor a debugger@ >> that will handle genuine problems.  And that is simply in theB >> context of a single-threaded application - the SMP situation is >> much, much worse. > C >But by restricting the search space to Unix, you are starting withAC >a candidate set of vendors who have similar restrictions, and also5= >have a common set of customer expectations.  A different OS,h4 >about which I have no ax to grind, might be OS/400.  > No, I am not restricting the search space to Unix, but it is a= BETTER search space than most others :-(  I could explain whyv> OS/400 isn't relevant (and both MVS and VMS are), but it would< be a bit tedious.  This is a change over time, far more than between systems.  ? In particular, the two systems that the IA-64 has to support ifa> it is to be a commercial success are Windows XXX and Unix.  If? it can't support both of those, it is in dead trouble.  And the ? interrupt handling on Unix is in much better shape than that onh, Windows XXX.  Which is why I referred to it.  E >> And, while the debugger situation is a little better, I don't knoweD >> of ANY modern system where it is possible to debug stack trashingB >> in an application signal handler, to take one relatively simple >> (sic) example.- >-A >This example seems to be rather C-centric, and lack of automaticr@ >buffer limit checking in C is probably the major cause of blown >stacks. >nB >So yes, a more robust error detectiong environment (lots of guardD >pages, etc) might be desired by those who choose C.  But apparently< >not by enough of them to make a it profitable for a vendor.  C No, that's not the point.  That may be the CAUSE of trashed stacks,@@ but my point was the much lower quality of modern debuggers than@ those of yore.  Let us say that I am an implementor of a secure,@ high-level language - how am I expected to debug a problem in my own run-time system?  ? The point is that a debugger shouldn't fall over, no matter howy; knotted the program it is debugging has got.  To say that a @ debugger need handle only errors that are trapped by the program> it is debugging begs the question of why the thing is provided in the first place.n  : When I started computing, there weren't any such things as= debuggers, and life became a LOT easier when I started to use A systems which had them.  What really disgusts me is that I am nowd@ having to use pre-debugger techniques because the damn debuggers? don't work.  And very, very few of the modern generation can don that :-(     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QG, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679d   ------------------------------   Date: 1 Sep 2001 19:27:09 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen)e Subject: Re: I hate Compaq3 Message-ID: <oUSeMIrZmepE@eisner.encompasserve.org>r  [ In article <9mrqqo$h4o$1@pegasus.csx.cam.ac.uk>, nmm1@cus.cam.ac.uk (Nick Maclaren) writes:l5 > In article <NBmtmjnzC45E@eisner.encompasserve.org>,e0 > Larry Kilgallen <Kilgallen@SpamCop.net> wrote:  B >>This example seems to be rather C-centric, and lack of automaticA >>buffer limit checking in C is probably the major cause of blowni	 >>stacks.o >>C >>So yes, a more robust error detectiong environment (lots of guard E >>pages, etc) might be desired by those who choose C.  But apparently0= >>not by enough of them to make a it profitable for a vendor.p > E > No, that's not the point.  That may be the CAUSE of trashed stacks,tB > but my point was the much lower quality of modern debuggers thanB > those of yore.  Let us say that I am an implementor of a secure,B > high-level language - how am I expected to debug a problem in my > own run-time system? > A > The point is that a debugger shouldn't fall over, no matter how0= > knotted the program it is debugging has got.  To say that ayB > debugger need handle only errors that are trapped by the program@ > it is debugging begs the question of why the thing is provided > in the first place.   @ Actually, I think things have gotten better now that the bulk of' the debugger is in a different process.s  @ Since modern debuggers have just a small execution engine in theA same process context as the program being debugged, there is lessaE to be trashed.  To eliminate that remaining possibility would requireh@ running in inner mode, so the program with the buffer overrun or@ bad pointer arithmetic (another flaw from the past) cannot trash" data that belongs to the debugger.  9 Of course one of the things a debugger does is show you ao< symbolic trace of the stacks for various tasks (threads, for< C folks).  But if the application under test has trashed the8 stack badly enough, there is not much there to show you.  7 A debugger _could_ set up an intercept to keep a shadow ; copy of the stack in inner mode memory, solving that issue.c  = Personally I have no desire to pay higher prices so operating < systems will include a debugger that robust, since those are not problems I encounter.e   ------------------------------  % Date: Sat, 01 Sep 2001 21:09:56 -0400e+ From: Tim Shoppa <shoppa@trailing-edge.com>d Subject: Re: I hate Compaq1 Message-ID: <3B914EA4.27DF0BE8@trailing-edge.com>n   Nick Maclaren wrote: > 5 > In article <NBmtmjnzC45E@eisner.encompasserve.org>,e0 > Larry Kilgallen <Kilgallen@SpamCop.net> wrote:D > >So yes, a more robust error detectiong environment (lots of guardF > >pages, etc) might be desired by those who choose C.  But apparently> > >not by enough of them to make a it profitable for a vendor. > E > No, that's not the point.  That may be the CAUSE of trashed stacks, B > but my point was the much lower quality of modern debuggers than > those of yore.  ; It's not everything to all people (and it certainly doesn'te: compare to a real debugger on a real high-level language),; but Rational's Purify is remarkable in its ability to catchtB many of the scores of memory corruption errors in every C program.  D I will refrain from slamming C here because I've done it plentifully in other threads :-)   Tim.   ------------------------------  $ Date: Sat, 1 Sep 2001 15:18:38 -0400+ From: "Main, Kerry" <Kerry.Main@compaq.com>d Subject: Intel Press ReleasessR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4D49601@kaoexc01.americas.cpqcorp.net>  G Apologies if this was previously posted, but I thought the readers of =R thisH list might be interested in the following press releases from Intel -=20  B http://www.intel.com/pressroom/archive/releases/20010830corp_a.htmE "Intel Appoints Four New Fellows, Names New Vice President - Former =A CompaqB Employees Join Intel as Part of Itanium(tm) Processor Agreement=20H SANTA CLARA, Calif., Aug. 30, 2001 - Intel Corporation today announced = thatG four individuals have been appointed to the company's highest technicalMG position, Intel Fellow. The new Intel Fellows are Joel S. Emer, TryggveaG Fossum, William J. Grundmann and P. Geoffrey Lowney. In addition, Intel B today announced that Daniel J. Casaletto has been appointed vice =
 president, Intel Architecture Group.OG Today's appointments are the result of the recent agreement with CompaqiG Computer Corporation to transfer key enterprise processor technology tonC Intel and consolidate Compaq's entire 64-bit server family on the =e Intel=AE$ Itanium(tm) microprocessor family. " [see url for rest of release]c: http://www.intel.com/pressroom/kits/bios/dcasaletto.htm=20H "Daniel J. Casaletto is Vice President of the Intel Architecture Group = andcD General Manager of the Massachusetts Microprocessor Design Center in Shrewsbruy, Mass. B Most recently, Daniel J. Casaletto was Vice President of the AlphaF Development Group at Compaq Computer Corporation. He was responsible = for the @ development and delivery of Alpha microprocessors and supportingC technologies for entry-level server systems. His organization was =0	 primarilykF made up of hardware and software engineers developing microprocessors,F related chipsets, motherboards, firmware, IC package design and CAD. = His H group also oversaw Alpha microprocessor production, and worked closely = with the semiconductor foundries. " [see url for rest of release]  Regards,  
 Kerry Main Senior Consultantw Compaq Canada Corp.  Professional Services  Voice: 613-592-4660e Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com   ------------------------------  $ Date: Sun, 2 Sep 2001 02:20:26 +0800" From: "Kenneth" <best@hotmail.com># Subject: RMS index file performancee0 Message-ID: <9mr8q0$ja45@imsp212.netvigator.com>  J I want to investigate the performance of the RMS index file that have beenJ used for a long time. According to the "File Application" manual I have toJ set the file statistic on to the selected file and capature the statistic.G How do I used with these statistic data? Is there any threshold value ?r   ------------------------------  $ Date: Sat, 1 Sep 2001 19:07:17 -0400  From: John Santos <JOHN@egh.com>% Subject: Re: Tunneling DECnet over IP 6 Message-ID: <1010901185422.61567A-100000@Ives.egh.com>  + On Sat, 1 Sep 2001, Robert Deininger wrote:e   [...]M  L > With DECnet-plus I don't think something like VTUN is needed.  It might beL > fine for DECnet-IV tunnelling, though routing via a DECnet-plus node might > be easier.  I One issue that HECnet is trying to address is connecting various PDP-11's I over the Internet.   Most PDP-11's (RSTS/E, various RSX variants, RT-11?)8I support DECnet (Phase IV, but not Phase V), but I don't think any support H TCP/IP.  (Well, there was a version of TCPWare for RSX, but I don't know if it still exists.)  D Routing through a Phase V system (i.e. a VMS system) is an option ifE you have one, but lots of people have old PC's laying around that canc run Linux or BSD, etc.  I > Since TCPIP can do IP over serial links via PPP, and DECnet-plus can donG > decnet-over tcpip, it's likely possible to do DECnet-over PPP.  But In > haven't tried it yet.   G Yes, that should work fine.  I used to have two DECservers (a 90M and a2F 700) connected via a pair of dialup 28.8 modems running PPP.  Before IC got dialout working at the local end, I set the local server's port B up as a dedicated outbound service, defined a LTA: device on a VMSE system to point to it, and ran PPP (TCPWare variant) on it to connectiB to the remote site.  It ran in this mode for a couple of weeks andE worked fine for both TCP/IP and DECnet-plus DECnet-over-tcpip.  (This @ was long since replaced by a leased 56K line and a pair of CiscoB routers, so I can't supply reliable details of configuration, etc.	 anymore.),   -- e John Santosy Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Sat, 1 Sep 2001 19:21:30 -0400  From: John Santos <JOHN@egh.com>% Subject: Re: Tunneling DECnet over IP-6 Message-ID: <1010901191546.61567B-100000@Ives.egh.com>  + On Sat, 1 Sep 2001, Robert Deininger wrote:   D > In article <1010831225453.61567D-100000@Ives.egh.com>, John Santos > <JOHN@egh.com> wrote:  >  > @ > > PhaseV VAX).  (BTW, I am getting frequent errors of the form > > N > > %DIRECT-E-OPENIN, error opening REMOTEPHASEV"user password"::*.*; as input1 > > -RMS-F-SYS, QIO system service request failed . > > -SYSTEM-F-LINKEXIT, network partner exited > > E > > when I try to use the VAX 4200 as the intermediate system, though-C > > sometimes it works.  Maybe it is just too slow and something iseH > > timing out??  Seems to be solid using either Alpha as the PM router. > H > I remember getting a lot of ethernet errors (not just DECnet) on a VAXF > 4600, when I was doing a lot of tests with DECRAM disks.  Whenever IG > INITed a RAM disk, the connection to other cluster members was up and L > down.  So it does seem possible to choke the ethernet on these systems.  IH > never tried the same on another VAX, for lack of extra memory.  Alphas! > never showed the same behavior.o  G We used to see lots of that and the cluster losing and regaining quorum0A when doing heavy ethernet traffic (not just from the cluster, for > example a PC server doing backups of other PC's), but using anC ethernet switch seemed to cure that.  However there may have been a1D internode backup going on at the time.  (Alpha backing up one of the, VAX'es disks to the Alpha's 8mm tape drive.)  A P.S.  You could also be flooding the VAX'es QBUS if it is using ahB DELQA.  (I don't know what the through-put of the onboard EthernetE chip is on some (all?) 4000's, but I think a QBUS is limited to aboutu
 1 MByte/sec.).   -- o John Santost Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sun, 02 Sep 2001 01:24:20 +0100e' From: Elliott Roper <elliott@yrl.co.uk>B% Subject: Re: Tunneling DECnet over IP 2 Message-ID: <020920010124201927%elliott@yrl.co.uk>  B In article <1010901191546.61567B-100000@Ives.egh.com>, John Santos <JOHN@egh.com> wrote:e    C > P.S.  You could also be flooding the VAX'es QBUS if it is using anD > DELQA.  (I don't know what the through-put of the onboard EthernetG > chip is on some (all?) 4000's, but I think a QBUS is limited to aboutx > 1 MByte/sec.)t  D You could get about 3MB/sec on QBus. 6MB/sec was possible with burstF mode DMA controllers. The earlier Unibus was about 2MB/sec despite not$ multiplexing address and data lines.  D DELQA would get close to theoretical 10Mbit/sec of original ethernetA for short bursts on sunny days. In practice it would manage aboutm& 600KB/sec. Far below a maxed out QBus.   ------------------------------  # Date: Sun, 02 Sep 2001 00:45:37 GMT-L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")% Subject: Re: Tunneling DECnet over IPt8 Message-ID: <00A01696.5F81C750@SSRL04.SLAC.STANFORD.EDU>   In article <rdeininger-0109011000570001@user-2iveba2.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:1C >In article <1010831212112.61567B-100000@Ives.egh.com>, John Santosc ><JOHN@egh.com> wrote: >g/ >> On Fri, 31 Aug 2001, Robert Deininger wrote:  >aH >> I don't think DECnet-plus can route between a local Phase-IV end nodeF >> and a remote Phase V node, when the only connection between the twoI >> phase V nodes is IP.  IOW, if I have a LAN with some phase IV and some F >> phase V nodes, and a similar remote LAN with a similar mixture, theH >> Phase V nodes can talk to any local node and all the remote Ph V, butH >> the phase IV nodes can only talk to local nodes.  (If you have DECnetB >> circuits, then the Ph IV nodes can of course talk to any system, >> so connected via a Ph IV or Ph V router.) >> tI >> This is because the DECnet-over-IP service operates at the applicationtF >> link level (in Ph IV terminology), end-system to end-system, rather9 >> than as a virtual DECnet circuit at the routing level.  >> AB >> If I'm wrong about this, I would love to know how to do it!  It0 >> certainly doesn't "just work" out of the box. >yI >I probably got it wrong.  We don't have any phase IV nodes anymore, so Id >never had a chance to try it. >e, >How about poor man's routing?  Can NODE1 do >  $DIR NODE2::NODE3::WHATEVERF >if NODE1 is phase IV, and NODE2 and NODE3 are phase V with only an IP >connection?  G Yes.  (I tried explicitly this from a Phase IV 6.2 system to two 7.2-1 i Phase V systems.)t   Can't SET HOST, though.r   -- Alanr  O ===============================================================================t0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056dM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210pO ===============================================================================-   ------------------------------  $ Date: Sun, 2 Sep 2001 02:15:08 +0800" From: "Kenneth" <best@hotmail.com>% Subject: VMS 7.3 compatibility issuesc0 Message-ID: <9mr8g4$jag2@imsp212.netvigator.com>  G I am using EV5 processor and want to upgrade to EV6 (21264a) processor.qF Compaq says I need to recompile my C program with the C version 6.0 orK above. Does anyone out there has the problem in running your application inrI EV6 which is compile with C below 6.0? Or do you have encounter any thirdl0 party product which is not compatiable with EV6?   ------------------------------  $ Date: Sat, 1 Sep 2001 20:05:15 -0400  From: John Santos <JOHN@egh.com>) Subject: Re: VMS 7.3 compatibility issuesc6 Message-ID: <1010901193007.61567C-100000@Ives.egh.com>  " On Sun, 2 Sep 2001, Kenneth wrote:  I > I am using EV5 processor and want to upgrade to EV6 (21264a) processor.tH > Compaq says I need to recompile my C program with the C version 6.0 orM > above. Does anyone out there has the problem in running your application inbK > EV6 which is compile with C below 6.0? Or do you have encounter any thirdS2 > party product which is not compatiable with EV6?  C Normally, no you don't have to recompile.  However, there was a bug @ in some old compilers (actually in the GEM optimizer) that couldB generate incorrect code sequences.  These sequences worked fine onD EV5 and before, but could break on an EV6.  The newer compilers haveC this bug fixed, so recompiling guarantees it won't bite you.  TherehC is also a program you can run which will scan an image and tell you F if the invalid code sequences occur.  (It can get false positives fromG data that looks like code.)  I can't remember the name of this program,i0 but it should be in the V7.0-V7.2 release notes.  C If you are only intending to run your programs on your new EV6, you A can recompile with the right optimization switches and get betteroA performance.  (If you optimize for EV6 and run the resulting .EXEi> on EV5, it will work, but it might be slower than it would run# un-optimized or optimized for EV5.)   A Generally 3rd party products are compiled without optimization orhA with generic optimizations, to get performance that is acceptable A everywhere rather than great performance on some systems and poor ? performance on others.  Something that is specifically intendedl? for high performance might be shipped either as source (compileu@ on-site) or as multiple object libraries optimized for differentA systems, or as multiple .EXE's.  We usually ship object librariess for the most flexibility.a   -- l John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   End of INFO-VAX 2001.487 ************************