1 INFO-VAX	Tue, 11 Oct 2005	Volume 2005 : Issue 568       Contents: Re: can't dismount ? Re: can't dismount ?* Re: Epson printers are electronic theieves* Re: Epson printers are electronic theievesC Re: fork() (Was: Re: Will "COM for OpenVMS" be ported to Itanium ?) C Re: fork() (Was: Re: Will "COM for OpenVMS" be ported to Itanium ?) ? Re: fork() (Was: Will "COM for OpenVMS" be ported to Itanium ?) ? Re: fork() (Was: Will "COM for OpenVMS" be ported to Itanium ?) / Re: HP : Massive strike and protest march today / Re: HP : Massive strike and protest march today / Re: HP : Massive strike and protest march today / Re: HP : Massive strike and protest march today / Re: HP : Massive strike and protest march today " Re: Migrating from Multinet to UCX" Re: Migrating from Multinet to UCX" Re: Migrating from Multinet to UCX) Re: Missed opportunity for VMS marketing. F Re: ODS-5 directory and header definitions, where? Was: [OT] - dealingF Re: ODS-5 directory and header definitions, where? Was: [OT] - dealingP Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing  with ;32P Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing with ;327! Re: Resetting Compaq SANSwitch 16 ! Re: Resetting Compaq SANSwitch 16  Samba-3 for OpenVMS & Re: TCPware starts up, then shuts down1 Re: VMS compatible terminal emulator for Mac OS X 6 Re: Why did VMS users go along with the itanium farce?6 Re: Why did VMS users go along with the itanium farce?6 Re: Why did VMS users go along with the itanium farce?6 Re: Why did VMS users go along with the itanium farce?6 Re: Why did VMS users go along with the itanium farce?1 Re: Will "COM for OpenVMS" be ported to Itanium ?  __PROBER(),  Re: __PROBER(),   F ----------------------------------------------------------------------  % Date: Tue, 11 Oct 2005 09:33:26 +0100 0 From: Chris Sharman <chris.sharman@sorry.nospam> Subject: Re: can't dismount ? 3 Message-ID: <diftcm$q2$1$830fa7a5@news.demon.co.uk>     Peter 'EPLAN' LANGSTOEGER wrote:o > In article <10OCT05.11031638@thuria.waisman.wisc.edu>, karcher@thuria.waisman.wisc.edu (Carl Karcher) writes:  > J >>In a previous article, Chris Sharman <chris.sharman@sorry.nospam> wrote: >>N >>->I tried "SDA> sho proc/chan all" - that didn't list anything on that disk. >>->   >>J >>Hmm. Then there are no files open on the disk (if you believe SDA). I've? >>had HP working on a similar problem a while back but on 7.3-2 E >>(3206154671). They chased vcbs and fcbs and found no reason why the I >>transaction count didn't match sda and "show dev/files". The conclusion E >>was something (kernel software) wasn't decrementing the transaction E >>count (sho dev). I'm guessing your trans cnt here is 8 (add one for    No. My trans cnt is 1.  G >>indexf.sys). We all suspected pathworks (you don't run that do you?). I >>Perhaps the common thread is OSU, but there's no kernel software there. ( >>that. Do you use Decnet Phase IV or V?  7 No pathworks on that machine ... it's just a webserver. H OSU on that disk, a test copy of apache, mysql, php etc on another disk. DECnet Phase IV.  N > I still think that the INSTALL is the problem. He wrote, that he deleted theK > images before he actually did deINSTALL them (INSTALL> REMOVE vs DELETE).   H Yes - I realised as soon as I'd done it that it would have been cleaner I the other way about. They all deinstalled cleanly though, despite having   been deleted first.   G >>HP provided an image that will set the transaction count to what ever I >>you want (i.e. 1) and then the disk can be dismounted. If you reference - >>that case number perhaps you can obtain it.    Scary.  F Perhaps I'll go for a reboot in the small hours after all. Not what I > wanted to do, but it beats a reboot in the not so small hours.   Thanks Chris    ------------------------------   Date: 11 OCT 2005 09:53:58 GMT4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher) Subject: Re: can't dismount ? 6 Message-ID: <11OCT05.09535854@thuria.waisman.wisc.edu>  H In a previous article, Chris Sharman <chris.sharman@sorry.nospam> wrote:   ->No. My trans cnt is 1.  G Not the same problem then. I always thought deleting an installed image D just removed it from the .dir and marked it for delete. Once all theG process referencing those sections go away the file should get deleted. E Have you tried anal/disk/rep/confirm? It should show you files marked  for delete.    --G -- Carl Karcher, Waisman Computing Services, Waisman Center, UW-Madison ; --                   karcher.notmorespam@waisman.wisc.edu      ------------------------------  % Date: Tue, 11 Oct 2005 09:23:05 -0400 * From: "Syltrem" <syltremzulu@videotron.ca>3 Subject: Re: Epson printers are electronic theieves 0 Message-ID: <11knf5qkivhg05a@corp.supernews.com>  H "Dr. Dweeb" <NOSPAM_5msg0h202@sneakemail.com> a crit dans le message de7 news:4349939e$0$78284$157c6196@dreader1.cybercity.dk...  > Anna Daptor wrote:> > > "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message* > > news:43497EE5.425D1A68@teksavvy.com... > >> Tony wrote:@ > >>> Have you tried using the driver software to change all the > >>> cartridges again but9 > >>> using the *same* cartrdiges? I have seen this work.  > >> > >>G > >> EPSON cartridges can only be inserted into te printer once. If you H > >> take a brand new cartridge, put it in printer, then take it out and? > >> put it back it, it will indicate 0 even though it is full.  > > H > > No it won't. Epson carts have a chip that remembers the ink level soI > > you *can* take them out of the printer and it'll know how much ink is $ > > in the cart when you replace it. > >  > I > I am not certain who is correct here, though I have seen others than JF # > mention this somewhere else IIRC.  > I > The printer "decided" that the cartridge was empty rather arbitrarily -  the J > final page printed flawlessly.  The % full for the colours seems to diveG > pretty rapidly, even when the only colour printed was some trivial 10  pixel  > logo.  > K > I think it is a giant scheme of deceit and lies and I am really mad about  > it.  > J > Sadly, like almost everything else in the computer world, no one gives a FF. K > and there is no where to lodge a complaint.  Well maybe there is - I will E > try during the week to harass support here and see what happens.  I  predict  > abject failure.  >  > Dr. Dweeb  >  > >> You need toJ > >> electronically reset the cartridge so it thinks it was never inserted > >> into a printer before.  > > , > > That's true if you've refilled the cart. >  >     I have an Epson printer at home.L Agreed that their cartridges are way too pricey, and mine cannot be replacedL by another brand (no one else makes them for some obscure technical reason). They cannot be refilled either.   E Anyway, they can be pulled out of the printer and reinserted, and the $ printer will remember the ink level.J I did that with Epson support 2 weeks ago when my printer suddenly refused to print colors (only black).   E And YMMV but even though my warranty was finished for awhile, after a L miserable result (colors were printing but not clean) support said "Ship theL printer to us and we will repair it under warranty". Shipping was free, too.K The printer was finally replaced for a brand new one of the newer model, in  3 days.   I They do have business practices I do not agree with but their support, at % least in this instance, was superior.   I Call support (as it seems you have not done so yet) and they may help you 
 very well.   --   Syltrem   H http://pages.infinit.net/syltrem (OpenVMS related web site, en franais)   ------------------------------  # Date: Tue, 11 Oct 2005 16:10:22 GMT  From: pgx@pgrahams.com3 Subject: Re: Epson printers are electronic theieves 8 Message-ID: <pronk1h0vjvjfoueb9t9b2sahr84pkren9@4ax.com>  4 "Dr. Dweeb" <NOSPAM_5msg0h202@sneakemail.com> wrote:   |Gentlemen,  | J |Today, the reading on my black ink cartridge hit zero, while the other 5  |read between 23-28%G |I replaced the black cartridge with a 3rd party product from Gold-Ink.    |Failure - and worse.  | N |1: "The ink cartridge is different from Epson blah blah blah" message.  This 5 |will now appear EVERY time I try to print !!!!!!!!!!   C I would suggest that Gold-Ink is also guilty.  If their ink doesn't B work, they should not sell it (unless there is something "Special" about your particular printer);    Phil   ------------------------------    Date: 11 Oct 2005 07:54:08 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) L Subject: Re: fork() (Was: Re: Will "COM for OpenVMS" be ported to Itanium ?)3 Message-ID: <uKby+ZqGhu8R@eisner.encompasserve.org>   \ In article <434ADE22.8E7812B9@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > G > Under what circumstances would one want to copy your process to a new J > identical one which can then diverge and write stuff to its own memory ? > I > Or is this Unix' answer to global sections where the forked process can D > continue to read memory structures of its parent process while theC > parent continues to modify the memory ? (and as long as the child I > doesn't write to it, it maintains the shared copy "owned" by the parent  > ?   E    It's used in the same manner one might use threads.  In most cases B    it can be replaced by using the POSIX thread package, portable     between UNIX and VMS.  A    But a lot of existing code uses it.  For example, I started to F    modify the CVS server to use threads instead of fork(), so it couldD    be ported to VMS.  But then I got real work to do.  Basically theD    implementer of the CVS server just wanted a multi-threaded server"    to handle multiple connections.   ------------------------------  % Date: Tue, 11 Oct 2005 09:50:06 -0400 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>L Subject: Re: fork() (Was: Re: Will "COM for OpenVMS" be ported to Itanium ?). Message-ID: <434B8ACE.28696.2827C9F@localhost>  ( On 10 Oct 2005 at 17:33, JF Mezei wrote:E > Or is this Unix' answer to global sections where the forked process H > can continue to read memory structures of its parent process while theC > parent continues to modify the memory ? (and as long as the child B > doesn't write to it, it maintains the shared copy "owned" by the
 > parent ?  E The parent and child share the "mmap" memory structures (aka, global  @ section).  The regular stack/heap/code sections are copied, not " shared, to create the new process.  E In order for it to *really* work, the mmap data would have to be set  E up so there's an automatic memory barrier (on Alpha, memory fence on  ? Itanium), or the updates between processes aren't visible in a  8 deterministic fashion.  Wonder how *nixes handle that...      
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------  % Date: Tue, 11 Oct 2005 09:11:37 +0200 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>H Subject: Re: fork() (Was: Will "COM for OpenVMS" be ported to Itanium ?)< Message-ID: <434b65a3$0$90412$edfadb0f@dread16.news.tele.dk>   John Malmberg wrote: > Bill Gunshannon wrote:C >> I was under the impression that VMS didn't have a true fork() at C >> all, thus the big problem porting most OpenSource Unix software. C >> In the Windows world, fork(0 has been around dince the MSDOS/MSC  >> days. > 8 > Not in any of the documentation that I have ever seen. > H > The only fork() I have seen for the Microsoft world is in third party  > libraries.   If so then the argument:  0 Mono works on Windows => Mono should work on VMS   still is valid.    Arne   ------------------------------   Date: 11 Oct 2005 12:59:33 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)H Subject: Re: fork() (Was: Will "COM for OpenVMS" be ported to Itanium ?)+ Message-ID: <3r1r9lFhec53U1@individual.net>   4 In article <mvC2f.14282$R85.13851@news.cpqcorp.net>,: 	John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> writes: > Bill Gunshannon wrote: >>  C >> I was under the impression that VMS didn't have a true fork() at C >> all, thus the big problem porting most OpenSource Unix software. C >> In the Windows world, fork(0 has been around dince the MSDOS/MSC  >> days. > 8 > Not in any of the documentation that I have ever seen. > H > The only fork() I have seen for the Microsoft world is in third party  > libraries. >   H Mea culpa.  I remembered wrong.  I thought MSC had fork, but it actuallyG has spawn() (which makes more sense in the DOS world.)  I don't have my H MSDN Library online at the moment, so I can't say if current Windows nowL support fork(),  I beleive Turbo-C supported a limited (read broken) fork().   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Tue, 11 Oct 2005 20:37:56 +1300 $ From: "Lurker" <nowhere@nothing.com>8 Subject: Re: HP : Massive strike and protest march today5 Message-ID: <zXJ2f.17198$iM2.1405607@news.xtra.co.nz>   > "Dr. Dweeb" <NOSPAM_5msg0h202@sneakemail.com> wrote in message news:4348d214$0$78285   I > OK.  I give up.  I notice that I have several Moulinex appliances in my  > kitchen though.  > # > Just what has become of Moulinex?   + Well, nothing much - it just went bankrupt.    ------------------------------  % Date: Tue, 11 Oct 2005 00:46:57 -0700  From: Z <Z@no.spam> 8 Subject: Re: HP : Massive strike and protest march today+ Message-ID: <R5K2f.17346$dO5.3407@fe07.lga>   
 Lurker wrote: # >>Just what has become of Moulinex?   - > Well, nothing much - it just went bankrupt.   0 Delphi is about to join them, thanks to the UAW.  1 Another evil corporation slayed by a labor union!    ------------------------------  % Date: Tue, 11 Oct 2005 21:05:44 +1300 $ From: "Lurker" <nowhere@nothing.com>8 Subject: Re: HP : Massive strike and protest march today5 Message-ID: <MlK2f.17204$iM2.1406261@news.xtra.co.nz>   F "Z" <Z@no.spam> wrote in message news:R5K2f.17346$dO5.3407@fe07.lga... > Lurker wrote: % > >>Just what has become of Moulinex?  > / > > Well, nothing much - it just went bankrupt.  > 2 > Delphi is about to join them, thanks to the UAW. > 3 > Another evil corporation slayed by a labor union!    And I would add to that:  4 The local bus driver's union recently went on strike3 for a couple of days. Not surprisingly it seriously 2 disrupted transportation throughout the city. Many8 people couldn't get to work at all if they haven't heard2 about the strike beforehand. Others had to take to9 the cars with the net result of an additional congestion. 7 Where were the greens btw? Surely all that extra petrol / being burned up should have concerned them? And 2 in any case, why are the unions allowed to disrupt other people's lives?    ------------------------------  % Date: Tue, 11 Oct 2005 10:36:08 +0200  From: S <soterroatyahoodotcom>8 Subject: Re: HP : Massive strike and protest march today& Message-ID: <434b796b$1@news1.ethz.ch>  
 Lurker wrote: 4 > in any case, why are the unions allowed to disrupt > other people's lives?   C Now this one is quite simple: because the bus drivers are not your  H slaves. They are people as well which seem to need more than just being " fed and watered like house plants.   S    ------------------------------  % Date: Tue, 11 Oct 2005 04:37:08 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 8 Subject: Re: HP : Massive strike and protest march today, Message-ID: <434B7998.77A5A270@teksavvy.com>   Z wrote:2 > Delphi is about to join them, thanks to the UAW. > 3 > Another evil corporation slayed by a labor union!     G This one is a bit different. Delphi used to be GM. GM spun it off a few F years ago (I think 1999). It has struggled since because GM puts priceA pressure on the products it buys from Delphi, yet Delphi is still C straddled with the burdens that GM had agreed to back when it owned  Delphi. (eg: benefits etc).   H Car companies and Wall Mart have long hung suppliers by the neck, tryingD to extract every penny out of suppliers who are not very profitable.  H In many cases, the drive to be competitive results in large corporationsB focusing hardships on a fairly focused area of the corporation andG letting the rest keep their comfortable budgets. Often, it is the wrong H areas that are asked to make sacrifices and that is simpky because headsF of the "good" areas manage to convince the CEO that they should not beF the target of cuts and that the CEO should instead cut budgets of such and such departments.    ------------------------------    Date: 11 Oct 2005 11:09:48 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)+ Subject: Re: Migrating from Multinet to UCX , Message-ID: <434b9d7c$1@news.langstoeger.at>  ` In article <434B0F82.A815221E@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: >JF Mezei wrote: >> David J Dachtera wrote:L >> > Our site is migrating from Multinet to TCP/IP Services for OpenVMS V5.42 >> > (I call it simply "UCX" for obvious reasons). >>  I >> Since this is generally regarded as a downgrade, could you provide the , >> reasons for this move ? Is it financial ? > ' >I agree, it is a FUNCTIONAL downgrade.   < Also agreed. But no longer the big difference it used to be.E Slightly differences are still there however (like in TCPware FTP you F have to quote mixedcase Filenames while in TCPIP you don't need quotesI or in TCPware the 2nd and 3rd parameters to FTP are username and password J while in TCPIP you have to use qualifiers and not parameters and so on ;-)  F >Management has made it clear that the question is not open to debate.  H In case there is a shopstopper, I would like to see them debating again.E It does hurt seeing them heading into the wall without discussions...   E >That said, I do hope the performance we hope to gain will offset the ? >increased operating costs we expect as a result of the switch.   F Performance ? You mean the new scalable TCPIP kernel ? I hope you alsoI tested that your application does run on this new kernel (V5.4 has both). I Ours unfortunately still does not (and no fix in sight) and so we are now = bound to TCPIP V5.4 and therefor also to OpenVMS Alpha V7.3-2   M OTOH, switching from TCPware to UCX removed finger pointing of support folks.    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 11 Oct 2005 11:26:38 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)+ Subject: Re: Migrating from Multinet to UCX , Message-ID: <434ba16e$1@news.langstoeger.at>  m In article <1128999596.b157899a6833250e0542ce23fa3ba065@teranews>, Wayne Sewell <wayne@tachysoft.com> writes: M >One way to limit the effects of the downgrade is to disable the ftp and smtp 4 >services in the stack and use hgftp and mx instead.  K ...and use DCPS for the printer queues (DCPS$SMB instead of TCPWARE_LPRSMB, D TCPWARE_VMSLPRSMB, TCPWARE_TSSYM, TCPIP$LPD_SMB and TCPIP$TELNETSYM)  A And MX would be an upgrade for both TCPIP and TCPware/Multinet...   G And also PSC's SSH would also be work better on TCPIP than TCPIP's SSH.   K Too bad that MX/SSH brings fingerpointing again and freeware is disallowed.    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 11 Oct 2005 09:59:04 -0700 From: bob@instantwhip.com + Subject: Re: Migrating from Multinet to UCX C Message-ID: <1129049944.741066.140080@g47g2000cwa.googlegroups.com>   4 TCPWARE wins performance hands down as it still uses the vms kernel ...   ------------------------------    Date: 11 Oct 2005 07:57:20 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 2 Subject: Re: Missed opportunity for VMS marketing.3 Message-ID: <2DWha0S2u7O6@eisner.encompasserve.org>   a In article <434adf52$1@usenet01.boi.hp.com>, "Paul A. Jacobi" <Paul.Jacobi@nospam.hp.com> writes:  > O > OpenVMS is included in the ad for the RX4640 on page 33 in the September 26,  0 > 2005 issue of EETIMES.  Is that recent enough? >   F    I was reading some literature on 1553B bus support from SBS.  I wasB    surprised to see both OpenVMS and VMS listed (I guess they want1    your business no matter how you pronounce it).   C    This means there's still at least one vendor who still thinks of <    VMS as a viable real-time platform.  Too bad it isn't HP.   ------------------------------    Date: 11 Oct 2005 07:49:08 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) O Subject: Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing 3 Message-ID: <X$Yf0goth1uh@eisner.encompasserve.org>   p In article <1128995203.65afe672fd4d768562a2b0f37db87a63@teranews>, Tim E Sneddon <tesneddon@bigpond.com> writes: > B > RSTS had this syntax too. I'm pretty sure TOPS-10/20 had similar< > syntax, but you could have sub-directories inside the UFD.  B    I don't recall any numerical directories on my TOPS-20 systems.  H    Under TOPS-10, you had to logon using your UIC instead of a username.   ------------------------------  # Date: Tue, 11 Oct 2005 13:18:44 GMT ( From: Alan Greig <greigaln@netscape.net>O Subject: Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing = Message-ID: <UYO2f.100695$RW.95686@fe2.news.blueyonder.co.uk>    Bob Koehler wrote:r > In article <1128995203.65afe672fd4d768562a2b0f37db87a63@teranews>, Tim E Sneddon <tesneddon@bigpond.com> writes: > B >>RSTS had this syntax too. I'm pretty sure TOPS-10/20 had similar< >>syntax, but you could have sub-directories inside the UFD. >  > D >    I don't recall any numerical directories on my TOPS-20 systems.  C It certainly wasn't the native TOPS-20 form but TOPS-10 compatible  B utilities would accept a PPN format. The TOPS-20 EXEC (DCL equiv) E TRANSLATE command would translate back and forward between directory  D name and simulated PPN (based on the TOPS-20 directory number). The C jsyses to do this were, IIRC, stppn and ppnst. These were the only  H native TOPS-20 system routines with knowledge of PPNs. All the rest was H done by the PA1050 TOPS-10 compatibility package. But you could fire up ) PIP, for example, and use TOPS-10 format.   J >    Under TOPS-10, you had to logon using your UIC instead of a username. >    --  
 Alan Greig   ------------------------------    Date: 11 Oct 2005 02:27:42 -04003 From: Rich Alderson <news@alderson.users.panix.com> Y Subject: Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing  with ;32 . Message-ID: <mddpsqc600x.fsf@panix5.panix.com>  - Tim E Sneddon <tesneddon@bigpond.com> writes:    > Bob Koehler wrote:  J >> In article <1128968851.313884.122060@g43g2000cwa.googlegroups.com>, "Ed% >> Wilts" <ewilts@ewilts.org> writes:   H >>> What I still don't understand, though, is why dev:[0,0] is valid butH >>> only when dev is not rooted to a subdirectory.  Sure it's goofy, but >>> why shouldn't it work?  F >>    For compatability with RSX and ODS-1, where [g,m] refered to theI >>    three digit octal group and member number of a directory who's full E >>    name is gggddd.DIR;1 in [0,0].  Early versions of VMS had [1,3] $ >>    instead of [SYSLOST] and such.  B > RSTS had this syntax too. I'm pretty sure TOPS-10/20 had similar< > syntax, but you could have sub-directories inside the UFD.  N Tops-10 started out with a project-programmer number of the [1,1] type; later,N subdirectories were added, up to 5 deep, with SIXBIT names, like [10,7,FORTRA]' for the FORTRAN installation directory.   M Tops-20 inherited from TENEX the alphanumeric directory designation <FOO> and K added hierarchical subdirectories <FOO.BAR>.  No limit on number of levels, + only a limit on length of filespec strings.   I Both operating systems are still in use, like VMS.  See, for example, the  PDPplanet.com website.   --  L Rich Alderson                                       | /"\ ASCII ribbon     |L news@alderson.users.panix.com                       | \ / campaign against |L "You get what anybody gets. You get a lifetime."    |  x  HTML mail and    |L                          --Death, of the Endless    | / \ postings         |   ------------------------------    Date: 11 Oct 2005 10:05:13 -0700$ From: "Ed Wilts" <ewilts@ewilts.org>Y Subject: Re: ODS-5 directory and header definitions, where? Was: [OT] - dealing with ;327 B Message-ID: <1129050313.461461.52920@g14g2000cwa.googlegroups.com>   Bob Koehler wrote:l > In article <1128968851.313884.122060@g43g2000cwa.googlegroups.com>, "Ed Wilts" <ewilts@ewilts.org> writes: > > H > > What I still don't understand, though, is why dev:[0,0] is valid butH > > only when dev is not rooted to a subdirectory.  Sure it's goofy, but > > why shouldn't it work? > E >    For compatability with RSX and ODS-1, where [g,m] refered to the H >    three digit octal group and member number of a directory who's fullD >    name is gggddd.DIR;1 in [0,0].  Early versions of VMS had [1,3]# >    instead of [SYSLOST] and such.   A It is weird and my background is RSX (1980 or so) before VMS so I C remember it well.  What works is $1$dga100:[0,0] and user1:[0,0] if D user1: points to the physical device name so some level of concealedC directory specs work.  It doesn't go to the next level though.  Not @ that it should - it's ancient and should be obsolete but it doesE confuse users in that the device can sometimes work and sometimes not C work (and by not working, the user gets a no such directory error).   E >    You can set up a 000000.DIR;1 in a rooted directory, pointing to C >    itself, so that it works with [0,0], but I wouldn't do it if I  >    were you.  E You can't do that with standard tools.  If your rooted directory (say B ewsub) is user1:[ewilts.], ewsub:[000000] implicitely exists - youB can't create/dir ewsub:[000000].  Typing [0,0] is sure easier thanC [000000] though.  Perhaps an alias for [0] pointing to [000000] :-)   F Some days us old guys just need to drive the engineers nuts with weirdA backwards compatibility issues with 25-year old apps :-).  Yes, I  remember PIP for VMS...   	    .../Ed    ------------------------------  # Date: Tue, 11 Oct 2005 07:50:23 GMT A From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> * Subject: Re: Resetting Compaq SANSwitch 16? Message-ID: <39K2f.126584$G8.111356@text.news.blueyonder.co.uk>   I Doesn't look good. I've not seen that, nor have I run that old a firmware I version. Have you a support contract under which you can get the firmware < updated? Have you asked in the ITRC? Have you tried Brocade?   --     Hope this helps, Colin. ) colin DOT butcher AT xdelta DOT co DOT uk E It's not mine, but I like this definition: Legacy = stuff that works.    ------------------------------  % Date: Tue, 11 Oct 2005 08:12:59 -0500 " From: "Schroeder, AJ" <aj1@qg.com>* Subject: Re: Resetting Compaq SANSwitch 16) Message-ID: <digdos$q98$1@sxnews1.qg.com>   M "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> wrote in message  9 news:39K2f.126584$G8.111356@text.news.blueyonder.co.uk... K > Doesn't look good. I've not seen that, nor have I run that old a firmware K > version. Have you a support contract under which you can get the firmware > > updated? Have you asked in the ITRC? Have you tried Brocade? >   K <sigh> I do not have a support contract on this equipment since it is only   used for testing purposes.  L I'll try the ITRC and see if anyone has any other ideas, if not, then maybe : I can contact Brocade. Something doesn't make sense there.  + Thank you for the help/suggestions on this.   
 AJ Schroeder     ------------------------------    Date: 11 Oct 2005 07:37:25 -0700 From: "Joerg" <spi@equicon.de> Subject: Samba-3 for OpenVMSC Message-ID: <1129041445.129575.134970@g44g2000cwa.googlegroups.com>   	 Hi folks,   @ I will (re-)start using distributed Netbeans for OpenVMS and I'm@ looking for a binary for Samba-3. Does anybody knows if there is "installation kit" for it?   Joerg    ------------------------------  % Date: Tue, 11 Oct 2005 06:22:46 -0400 ' From: Rich Whalen <whalenr@process.com> / Subject: Re: TCPware starts up, then shuts down 8 Message-ID: <td4nk11u05hv8a1446jjmrpr85pmov7dd2@4ax.com>  B On Mon, 10 Oct 2005 12:34:58 -0400, "Chris" <an.other@not_here.ca> wrote:  J >I apologize for posting this here, but have had no answer from Process asI >yet, and vmsnet.networks.tcpware seems deserted.  Any al all suggestions  >gratefully received.  >  >Chris > 1 >Here is the NETCP.LOG that results after a boot:  > B >TCPware(R) for OpenVMS V5.6-2 Copyright (c) 2002 Process Software > / >**  6-OCT-2005 14:52:38 NETCP logging started.  > M >**  6-OCT-2005 14:52:41 Line LPB-0 (127.0.0.1) started by SYSTEM (20400205). F >**  6-OCT-2005 14:52:42 Line EWA-0 (172.22.120.204) started by SYSTEM >(20400205)  >. >   Line EWA-0 is using _EWA5:. H >**  6-OCT-2005 14:52:42 Updating multicast address list for line EWA-0:$ >   Adding address 01-00-5E-00-00-01D >**  6-OCT-2005 14:52:42 Line EWA-1 (172.22.128.2) started by SYSTEM >(20400205). >   Line EWA-1 is using _EWB4:. H >**  6-OCT-2005 14:52:42 Updating multicast address list for line EWA-1:$ >   Adding address 01-00-5E-00-00-01I >**  6-OCT-2005 14:52:42 TCP started (or restarted) by SYSTEM (20400205). I >**  6-OCT-2005 14:52:42 UDP started (or restarted) by SYSTEM (20400205). J >**  6-OCT-2005 14:52:42 INET started (or restarted) by SYSTEM (20400205).I >**  6-OCT-2005 14:52:42 UCX started (or restarted) by SYSTEM (20400205). 1 >%RPC-E-LISTENERR, error listening for connection I >-SYSTEM-F-THIRDPARTY, network logical link disconnected by a third party   B An exception producing error is happening when starting RPC.  ThisE causes the NETCP's exception handler to take over and shut everything D down.  Check your configuration of NFS, CNFS and RPC.  Try disabling; them one at a time and seeing which one clears the problem.    ------------------------------    Date: 11 Oct 2005 06:46:01 -0700' From: "Dick Piccard" <piccard@ohio.edu> : Subject: Re: VMS compatible terminal emulator for Mac OS XB Message-ID: <1129038361.167890.53370@g47g2000cwa.googlegroups.com>    stuie_norris@yahoo.com.au wrote: > Hi Group,  > G > I have just install a wireless network and I would like to be able to H > connect to my DS10L running VMS 7.3-2 from my powerbook G4 running Mac
 > OSX 10.3.x.  > E > I have tried terminal which is pretty poor.  Could not even get the F > ctrl-z to work correctly.  Also tried powerterm - but this is not asF > nice as I had hoped - plus pretty pricy.  Virtual keyboard and fonts > were pretty poor.  > G > Any other suggestions.  Does WRQ still market a terminal emulator for 
 > Mac OSX? > A > Main use is to run EVE (edit / tpu) for editoring source files.  >  > Thanks >  > Stuart  F I have been very happy using VersaTerm.  It runs in classic mode under	 Mac OS X.   ) See   http://www.synergy.com/versfeat.htm   F I have been using version 4.6.3 happily since 1991-92, and have had noD problems under Mac OS X, except that some of the keys F1 through F15G are hijacked by the operating system.  I don't know whether the current E version (5.07???) exhibits the same problem, or not.  VT220 emulation   appears to be correct otherwise.  C There is a VersaTerm Pro product, which as I understand it provides E emulations of more terminal models (e.g., ReGIS-compatible graphics), F but which does not provide any better an emulation of the VT220, which is all I have needed.    ------------------------------  % Date: Tue, 11 Oct 2005 13:31:51 +0200  From: S <soterroatyahoodotcom>? Subject: Re: Why did VMS users go along with the itanium farce? & Message-ID: <434ba2a7$1@news1.ethz.ch>   Larry Kilgallen wrote:E > We have had people come to this newsgroup with backgrounds in other J > operating systems who think drivers are best done as user mode programs.  B That being QNX :) But to get that properly done you need also the 5 specialized hardware to support it - rarely the case.    S    ------------------------------  + Date: Tue, 11 Oct 2005 11:13:04 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk? Subject: Re: Why did VMS users go along with the itanium farce? ) Message-ID: <dig6o0$2ed$1@news.mdx.ac.uk>   V In article <3qvqg9Fgmj4oU2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:* >In article <die9rf$ct4$1@news.mdx.ac.uk>," >	david20@alpha2.mdx.ac.uk writes:Y >> In article <3qvf10Fgk47hU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: , >>>In article <didv3t$93h$1@news.mdx.ac.uk>,$ >>>	david20@alpha2.mdx.ac.uk writes:h >>>> In article <NvwskviolK5n@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:O >>>>>In article <didppl$7jb$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes:  >>>>> U >>>>>> How different is the structure of a VMS device driver from a Unix/Linux device  >>>>>> driver ?  >>>>> 8 >>>>>What makes you think they would be at all similar ? >>>>> H >>>>>We have had people come to this newsgroup with backgrounds in otherM >>>>>operating systems who think drivers are best done as user mode programs.  >>>>  P >>>> Although I have 20 years experience of managing and programming VMS systemsO >>>> I've never had occasion to investigate Device Drivers hence the reason for M >>>> enquiring as to whether the NDISwrapper route to getting Windows drivers 8 >>>> working on Linux could be adapted for use with VMS. >>>>   >>> F >>>Well, considering that the NDIS Drivers are all x86 and VMS doesn't) >>>run on x86, my guess would be no.  :-)  >>>  >>  K >> Which is why I mentioned that I wasn't sure whether this only worked on   >> Linux running on x86/x86-64.  >>  O >> However if that is the "only" problem then it suggests that a port of VMS to O >> x86-64 might have benefits from the possibilty of utilising windows drivers   >> for certain applications. >>   >>  K >> However I'm still not sure that this restriction is actually true since  B >> there seem to be rpms for ndiswrapper for sparc, alpha, ppc etc >>   >> see for instance  >>  ^ >> http://rpmfind.net/linux/RPM/cooker/cooker/sparc/media/main/ndiswrapper-1.2-2mdk.sparc.html >>   >> and >>  b >> http://rpmfind.net/linux/RPM/cooker/cooker/sparc64/media/main/ndiswrapper-1.2-2mdk.sparc64.html >>   >> and >>  _ >> http://rpmfind.net/linux/RPM/cooker/cooker/alpha/media/main/ndiswrapper-0.12-1mdk.alpha.html  >>   >> and >>  Z >> http://rpmfind.net/linux/RPM/cooker/cooker/ppc/media/main/ndiswrapper-1.2-3mdk.ppc.html >>   > C >From The NDIS Developer's Reference located at http:/www.ndis.com: B >     "Current NDIS versions used by Windows For Workgroups (WFW),B >      Windows 9X Windows NT, Windows 2000, Windows XP and Windows= >      Server 2003 are Microsoft proprietary specifications."  > F >Still think there are NDIS drivers for anything other than x86?  JustD >because the wrappers compile doesn't mean they have any real value. > 8 >And then we have this in the installation instructions:F >   "you need to download appropriate Windows XP driver for your card" > ; >Any idea where I would download a Sparc Windows XP driver?  > ; >You can't believe everything you see on the INTERNET.  :-)  >   P What would be the point in Mandrake spending time building and providing access 9 for downloading of rpms for a program that wouldn't work.   K My understanding is that NDIS is an API between the protocol stack and the  M actual physical device. It was developed so that a network card could support D many protocols such as IPX, TCP/IP, NetBui, Banyan etc concurrently.L With NDIS wrapper an environment is provided so that calls to the HAL and NT6 kernel are translated into Linux compatible functions.G So the fact that the NDIS drivers are for windows on x86 is pretty much  irrelevent.   J You don't download a Sparc Windows XP driver. You just download the normalO Windows XP driver. That's the whole point - you are using the normal Windows XP  drivers under Linux.      
 David Webb Security team leader CCSS Middlesex University       >bill  >  >-- K >Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves E >bill@cs.scranton.edu     |  and a sheep voting on what's for dinner.  >University of Scranton   | B >Scranton, Pennsylvania   |         #include <std.disclaimer.h>      ------------------------------   Date: 11 Oct 2005 13:06:46 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)? Subject: Re: Why did VMS users go along with the itanium farce? + Message-ID: <3r1rn6Fhec53U2@individual.net>   ) In article <dig6o0$2ed$1@news.mdx.ac.uk>, ! 	david20@alpha2.mdx.ac.uk writes: X > In article <3qvqg9Fgmj4oU2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:+ >>In article <die9rf$ct4$1@news.mdx.ac.uk>, # >>	david20@alpha2.mdx.ac.uk writes: Z >>> In article <3qvf10Fgk47hU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:- >>>>In article <didv3t$93h$1@news.mdx.ac.uk>, % >>>>	david20@alpha2.mdx.ac.uk writes: i >>>>> In article <NvwskviolK5n@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: P >>>>>>In article <didppl$7jb$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes: >>>>>>V >>>>>>> How different is the structure of a VMS device driver from a Unix/Linux device >>>>>>> driver ? >>>>>>9 >>>>>>What makes you think they would be at all similar ?  >>>>>>I >>>>>>We have had people come to this newsgroup with backgrounds in other N >>>>>>operating systems who think drivers are best done as user mode programs. >>>>> Q >>>>> Although I have 20 years experience of managing and programming VMS systems P >>>>> I've never had occasion to investigate Device Drivers hence the reason forN >>>>> enquiring as to whether the NDISwrapper route to getting Windows drivers9 >>>>> working on Linux could be adapted for use with VMS.  >>>>>  >>>>G >>>>Well, considering that the NDIS Drivers are all x86 and VMS doesn't * >>>>run on x86, my guess would be no.  :-) >>>> >>> L >>> Which is why I mentioned that I wasn't sure whether this only worked on   >>> Linux running on x86/x86-64. >>> P >>> However if that is the "only" problem then it suggests that a port of VMS toP >>> x86-64 might have benefits from the possibilty of utilising windows drivers  >>> for certain applications.  >>>  >>> L >>> However I'm still not sure that this restriction is actually true since C >>> there seem to be rpms for ndiswrapper for sparc, alpha, ppc etc  >>>  >>> see for instance >>> _ >>> http://rpmfind.net/linux/RPM/cooker/cooker/sparc/media/main/ndiswrapper-1.2-2mdk.sparc.html  >>>  >>> and  >>> c >>> http://rpmfind.net/linux/RPM/cooker/cooker/sparc64/media/main/ndiswrapper-1.2-2mdk.sparc64.html  >>>  >>> and  >>> ` >>> http://rpmfind.net/linux/RPM/cooker/cooker/alpha/media/main/ndiswrapper-0.12-1mdk.alpha.html >>>  >>> and  >>> [ >>> http://rpmfind.net/linux/RPM/cooker/cooker/ppc/media/main/ndiswrapper-1.2-3mdk.ppc.html  >>>  >>D >>From The NDIS Developer's Reference located at http:/www.ndis.com:C >>     "Current NDIS versions used by Windows For Workgroups (WFW), C >>      Windows 9X Windows NT, Windows 2000, Windows XP and Windows > >>      Server 2003 are Microsoft proprietary specifications." >>G >>Still think there are NDIS drivers for anything other than x86?  Just E >>because the wrappers compile doesn't mean they have any real value.  >>9 >>And then we have this in the installation instructions: G >>   "you need to download appropriate Windows XP driver for your card"  >>< >>Any idea where I would download a Sparc Windows XP driver? >>< >>You can't believe everything you see on the INTERNET.  :-) >> > R > What would be the point in Mandrake spending time building and providing access ; > for downloading of rpms for a program that wouldn't work.  > M > My understanding is that NDIS is an API between the protocol stack and the  O > actual physical device. It was developed so that a network card could support F > many protocols such as IPX, TCP/IP, NetBui, Banyan etc concurrently.N > With NDIS wrapper an environment is provided so that calls to the HAL and NT8 > kernel are translated into Linux compatible functions.I > So the fact that the NDIS drivers are for windows on x86 is pretty much 
 > irrelevent.  > L > You don't download a Sparc Windows XP driver. You just download the normalQ > Windows XP driver. That's the whole point - you are using the normal Windows XP  > drivers under Linux. >   M Yes, but if your running on a Sparc CPU how do you execute the x86 NDIS code?  The driver is executable code.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 11 Oct 2005 16:22:30 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)? Subject: Re: Why did VMS users go along with the itanium farce? + Message-ID: <3r2766Fhf5amU1@individual.net>   ) In article <digmpq$7g6$1@news.mdx.ac.uk>, ! 	david20@alpha2.mdx.ac.uk writes: X > In article <3r21gcFhapjpU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:+ >>In article <diggha$58a$1@news.mdx.ac.uk>, # >>	david20@alpha2.mdx.ac.uk writes: Z >>> In article <3r1rn6Fhec53U2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:- >>>>In article <dig6o0$2ed$1@news.mdx.ac.uk>, % >>>>	david20@alpha2.mdx.ac.uk writes: \ >>>>> In article <3qvqg9Fgmj4oU2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:/ >>>>>>In article <die9rf$ct4$1@news.mdx.ac.uk>, ' >>>>>>	david20@alpha2.mdx.ac.uk writes: ^ >>>>>>> In article <3qvf10Fgk47hU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:1 >>>>>>>>In article <didv3t$93h$1@news.mdx.ac.uk>, ) >>>>>>>>	david20@alpha2.mdx.ac.uk writes: m >>>>>>>>> In article <NvwskviolK5n@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: T >>>>>>>>>>In article <didppl$7jb$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes:
 >>>>>>>>>>Z >>>>>>>>>>> How different is the structure of a VMS device driver from a Unix/Linux device >>>>>>>>>>> driver ?
 >>>>>>>>>>= >>>>>>>>>>What makes you think they would be at all similar ? 
 >>>>>>>>>>M >>>>>>>>>>We have had people come to this newsgroup with backgrounds in other R >>>>>>>>>>operating systems who think drivers are best done as user mode programs.
 >>>>>>>>> U >>>>>>>>> Although I have 20 years experience of managing and programming VMS systems T >>>>>>>>> I've never had occasion to investigate Device Drivers hence the reason forR >>>>>>>>> enquiring as to whether the NDISwrapper route to getting Windows drivers= >>>>>>>>> working on Linux could be adapted for use with VMS. 
 >>>>>>>>>  >>>>>>>>K >>>>>>>>Well, considering that the NDIS Drivers are all x86 and VMS doesn't . >>>>>>>>run on x86, my guess would be no.  :-) >>>>>>>> >>>>>>> P >>>>>>> Which is why I mentioned that I wasn't sure whether this only worked on $ >>>>>>> Linux running on x86/x86-64. >>>>>>> T >>>>>>> However if that is the "only" problem then it suggests that a port of VMS toT >>>>>>> x86-64 might have benefits from the possibilty of utilising windows drivers ! >>>>>>> for certain applications.  >>>>>>>  >>>>>>> P >>>>>>> However I'm still not sure that this restriction is actually true since G >>>>>>> there seem to be rpms for ndiswrapper for sparc, alpha, ppc etc  >>>>>>>  >>>>>>> see for instance >>>>>>> c >>>>>>> http://rpmfind.net/linux/RPM/cooker/cooker/sparc/media/main/ndiswrapper-1.2-2mdk.sparc.html  >>>>>>>  >>>>>>> and  >>>>>>> g >>>>>>> http://rpmfind.net/linux/RPM/cooker/cooker/sparc64/media/main/ndiswrapper-1.2-2mdk.sparc64.html  >>>>>>>  >>>>>>> and  >>>>>>> d >>>>>>> http://rpmfind.net/linux/RPM/cooker/cooker/alpha/media/main/ndiswrapper-0.12-1mdk.alpha.html >>>>>>>  >>>>>>> and  >>>>>>> _ >>>>>>> http://rpmfind.net/linux/RPM/cooker/cooker/ppc/media/main/ndiswrapper-1.2-3mdk.ppc.html  >>>>>>>  >>>>>>H >>>>>>From The NDIS Developer's Reference located at http:/www.ndis.com:G >>>>>>     "Current NDIS versions used by Windows For Workgroups (WFW), G >>>>>>      Windows 9X Windows NT, Windows 2000, Windows XP and Windows B >>>>>>      Server 2003 are Microsoft proprietary specifications." >>>>>>K >>>>>>Still think there are NDIS drivers for anything other than x86?  Just I >>>>>>because the wrappers compile doesn't mean they have any real value.  >>>>>>= >>>>>>And then we have this in the installation instructions: K >>>>>>   "you need to download appropriate Windows XP driver for your card"  >>>>>>@ >>>>>>Any idea where I would download a Sparc Windows XP driver? >>>>>>@ >>>>>>You can't believe everything you see on the INTERNET.  :-) >>>>>> >>>>> V >>>>> What would be the point in Mandrake spending time building and providing access ? >>>>> for downloading of rpms for a program that wouldn't work.  >>>>> Q >>>>> My understanding is that NDIS is an API between the protocol stack and the  S >>>>> actual physical device. It was developed so that a network card could support J >>>>> many protocols such as IPX, TCP/IP, NetBui, Banyan etc concurrently.R >>>>> With NDIS wrapper an environment is provided so that calls to the HAL and NT< >>>>> kernel are translated into Linux compatible functions.M >>>>> So the fact that the NDIS drivers are for windows on x86 is pretty much  >>>>> irrelevent.  >>>>> P >>>>> You don't download a Sparc Windows XP driver. You just download the normalU >>>>> Windows XP driver. That's the whole point - you are using the normal Windows XP  >>>>> drivers under Linux. >>>>>  >>>>Q >>>>Yes, but if your running on a Sparc CPU how do you execute the x86 NDIS code? " >>>>The driver is executable code. >>>> >>> < >>> Inside the emulated environment provider by ndiswrapper.P >>> The driver is just bytes of data to be read and interpreted by the emulator. >>J >>Are you trying to say that ndiswrapper has a complete x86 emulator builtG >>into it?  Somehow I doubt that.  The NDIS driver is a program written ( >>for the x86 cpu.  It is not just data. >>B >>KTC120.SYS:   MS Windows PE 32-bit Intel 80386 native executable; >>ME102MAN.sys: MS-DOS executable (EXE), OS/2 or MS Windows  >>G >>Those are NDIS drivers.  One for a Kingston TX120 and the other for a  >>Netgear ME102 Wireless card. >> > G > The drivers are WINDOWS executables in PE format. They are not LINUX  J > executables in ELF format. Hence some level of emulation has to be done " > even if both are running on x86. >   G I know that, but the claim was that the x86 NDIS drivers were usable on H architectures other than x86 and I don't see how that would be possible.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 11 Oct 2005 09:55:49 -0700 From: mhatch@ics.com? Subject: Re: Why did VMS users go along with the itanium farce? C Message-ID: <1129049749.060854.325130@g14g2000cwa.googlegroups.com>    JF Mezei wrote:  > David J Dachtera wrote: L > > In pondering this and considering all the past arguments quetsioning theF > > value of an office automation package on OpenVMS, I think I have a > > better picture in my head: > >  > > Back-end Tier: > > (GS-Class/Superdome) > > OpenVMS  > > DECwindows/MOTIF > > SWB  > > OpenOffice > > Oracle, RdB or mySQL > G > Until VMS management agree to put DECwindows/Motif on te VMS roadmap, + > should anyone be betting the bank on it ?  > F > Consider that on VMS, especially VAX, Motix is woefully out of date.D > Does Open Motif make use of widgets which are not present on VMS ? >   G Open Motif does include additional widgets (tab stack), etc (tooltips). C beyond what is offered OpenVMS Open Motif 2.3 (about to go to beta) F includes UTF8, png/jpg support on buttons and anti-alias font support.  D To be fair to OpenVMS, none of the UNIX platforms ship OpenMotif 2.x@ either because of the original license under which OpenMotif wasG released. The Open Group, I assume to protect their royalty stream from G Motif, released OpenMotif under a license that prevents it being ported ' to "non-open source operating systems."   4 This is probably more than you wanted to know... ;-)   Mark   ------------------------------  % Date: Tue, 11 Oct 2005 09:15:05 +0200 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>: Subject: Re: Will "COM for OpenVMS" be ported to Itanium ?< Message-ID: <434b6673$0$90419$edfadb0f@dread16.news.tele.dk>   healyzh@aracnet.com wrote:# > Arne Vaj <arne@vajhoej.dk> wrote: 2 >>It should actually not be that hard to port Mono	 >>to VMS.  > 5 >>If it runs on Windows, Linux and *BSD, then porting  >>to VMS should be possible. >  >>(and Mono includes C#) >  > J > I've not had time to look at what would be required to port Mono to VMS,N > however, I suspect it would be easier to try to port "DotGNU Portable .NET".J > http://www.dotgnu.org/pnet.html  I believe Mono has a fair amount of CPUI > specific code (though I might be wrong about this), while Portable .NET 
 > doesn't.  7 It is my impression that Mono is ahead both technically  and market wise over DotGNU.  > How do DoTGNU avoid CPU specific code in a JIT compiler ? Have8 they written most of the stuff in managed code (like the Jikes RVM for Java) ?   I > Of course the question might be, is anything like this needed.  I don't 8 > know, but I for one wouldn't mind having it available.  ? There are coming a lot of C# programmers out of school the next  5-10 years.    Arne   ------------------------------    Date: 11 Oct 2005 09:25:17 -0700$ From: "roger" <rogerntucker@msn.com> Subject: __PROBER(),C Message-ID: <1129047917.301067.281640@g47g2000cwa.googlegroups.com>   B I've been thinking about operating systems in general and OpenVMS,A and how data is moved from user-space to kernel space or from the  kernel space to user-space.   F After the jump to kernel mode, OpenVMS probes the user-space addressesG to check for read or write.  Often this is done at a different place in G the code than the actual read or write to user-space.  For instance, in F sys$getjpi(), getjpi() does a prober() on the item list before reading= it, and a probew() on the item list buffers to make sure it's B writable, but may do a lot of other instructions in between beforeC actually writing to user space.  Prober and probew date back to VAX A days when it was in hardware, they are implemented in pal code on C alpha, and sys_pal routines on itanium.  Obviously, the kernel code @ must be at or below ASTDEL or a user page fault would bug check.  E Now for my question - please think hard about it before you answer...   ? Now with multiple kernel threads - what prevents another thread F running on another processor from messing with the user memory betweenC the time of the probe but before the actual read or write.  I think = it's unlikely, but is that a hole, or am I missing something?   E To be clear, CPU A running thread 1 protects or deletes the user-mode D virtual address space that CPU B running thread 2 has already probed  but hasn't written yet.  Ouch...   Thanks,  Roger    ------------------------------    Date: 11 Oct 2005 10:25:48 -0700$ From: "roger" <rogerntucker@msn.com> Subject: Re: __PROBER(),C Message-ID: <1129051548.441390.250930@g49g2000cwa.googlegroups.com>    Posted for Fisher Burns...   Hi Roger...   C A very insightful question.  This is (one of) the reason(s) for the E inner mode semaphore which allows only one kernel thread to execute a E kernel (or exec) system service or AST at the same time.   I say "one F of" because there are a number of subtle assumptions in various placesD that if you are in mode x, nothing can execute in mode x+n until you	 complete.   B There are a few services (and ASTs), called thread-safe, which areF specially marked and do not take out the inner mode semaphore.  Sadly,G these are few and far between.  Also, on Alpha there is the notion of a C "tolerant" service.  That is, one which can run in combination with ? other tolerant services, but not with an intolerant (exclusive)  service.G For example, if you knew you depended only on not having another thread C unmap a virtual address out from under you, you could mark yourself ? tolerant (and the various mapping services would be exclusive). C Unfortunately, that feature was never used, and for the same reason F that there are few thread-safe services:  It is difficult to determineG exactly what subtle dependencies each service has grown over the years.    Burns   > P.S.  You are welcome to post this on comp.os.vms if you wish.   ------------------------------   End of INFO-VAX 2005.568 ************************