0 INFO-VAX	Thu, 16 Jan 2003	Volume 2003 : Issue 31      Contents: Re: block site Re: blocksize of system  Re: Boot Contest change  Re: Control-P on al alpha 4100 Re: CSWS PHP 1.1 php.ini bug? * Re: decnet  - mailbox programming question DECnet-Plus vs DECnet IV Re: DECnet-Plus vs DECnet IV Re: DECnet-Plus vs DECnet IV Re: DECnet-Plus vs DECnet IV Re: DECnet-Plus vs DECnet IV Re: Disk drive too big?  Re: EVA and READL/WRITEL' Re: F$GETDVI ("AVL") in an IF statement B Re: Fibre channel:  Is the INIT/ERASE function built into the HBA?2 RE: Getting started with an AlphaStation 200 4/2332 RE: Getting started with an AlphaStation 200 4/2332 Re: Getting started with an AlphaStation 200 4/2332 RE: Getting started with an AlphaStation 200 4/2332 Re: Getting started with an AlphaStation 200 4/2332 Re: Getting started with an AlphaStation 200 4/2332 Re: Getting started with an AlphaStation 200 4/233 HardwareJ Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?- Re: HP Sets the Stage for Alpha's Last Hurrah  Incoming !  Ds20e Systems 4 Re: incomprehensible mountVerifyTimeout in a cluster4 Re: incomprehensible mountVerifyTimeout in a cluster4 Re: incomprehensible mountVerifyTimeout in a cluster1 It's 'open source', so it must be good.....right? ' Re: Looks like AMD has deep pockets now  Re: Microvax Floating point " Re: New Marvel machines? New news?" Re: New Marvel machines? New news?, Re: Newbie to OpenVMS (contrasting to Linux)0 Re: No mention of Digital in official HP history0 Re: No mention of Digital in official HP history0 RE: No mention of Digital in official HP history& Re: OpenVMS for AlphaStation 200 4/233& Re: OpenVMS for AlphaStation 200 4/233P OpenVMS Hobbyist licence (was: Getting started with an AlphaStation 200 4/233) 2 Re: OpenVMS VAX Hobbyist Kits  out-of-band ast  Re: out-of-band ast  Re: out-of-band ast  Re: Product kit question Re: Reading file in DCL = RE: searching for seemingly obscure disk activity light cable / Re: ShadowSets, MiniCopy, and SHUTDOWN - Take 2  Re: Suppressing SMTP headersP Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofP Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofP Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofP Re: The manual says "Do not use QIO for reading from a file" has anyone heard of Re: THREADCP problem" Re: Traceback and installed images% Re: VMS on Itanic boot report sighted  Re: What does RWCLU mean? ) Re: where are diskeeper logicals defined?  ZMENU - Menus for DCL   F ----------------------------------------------------------------------  % Date: Wed, 15 Jan 2003 21:57:34 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net> Subject: Re: block site 6 Message-ID: <3E262DAE.E3B4F975@earthlink.spamfree.net>   Sandeep Yelwatkar wrote: >  > Hi > 8 > I want to know how to find the block size of savesets. > ! > Any suggesstion is most welcome   H If the saveset file has not been corrupted by file transfers, etc., then" DIRECTORY/FULL will indicate this:  1 DJAS01::DDACHTERA$ back user$img:*.* exes.sav/sav " DJAS01::DDACHTERA$ dir/fu exes.sav   Directory DKA0:[DDACHTERA]  > EXES.SAV;1                    File ID:  (8942,3,0)            . Size:         2772/2772       Owner:    [30,1]" Created:   15-JAN-2003 21:51:38.48& Revised:   15-JAN-2003 21:51:47.37 (1) Expires:   <None specified>  Backup:    <No backup recorded>  Effective: <None specified>  Recording: <None specified>  File organization:  Sequential Shelved state:      Online  G File attributes:    Allocation: 2772, Extend: 0, Global buffer count: 0 $                     No version limitE Record format:      Fixed length 32256 byte records  <=== /BLOCKSIZE=  Record attributes:  None RMS attributes:     None Journaling enabled: None= File protection:    System:RWED, Owner:RWED, Group:RE, World:  Access Cntrl List:  None  " Total of 1 file, 2772/2772 blocks.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   Date: 16 Jan 2003 01:08:33 GMT# From: hoff@hp.nospam (Hoff Hoffman)   Subject: Re: blocksize of system* Message-ID: <b050mh$nnt$5@web1.cup.hp.com>  g In article <v2b0i6iv5cfgeb@corp.supernews.com>, "Sandeep Yelwatkar" <Sandeep_Yelwatkar@bmc.com> writes:   L :I am using system script call spkitbld.com to build the kit. In this scriptI :/block is set equal to 9000. I want to know if  changing the /block=8192 8 :affect the savesets the block size of savesets created.E :Is there any other way to change the block size of savesets created.   H   SPKITBLD.COM is intended to package the older VMSINSTAL-format productE   installation kits, and particularly using a consistent and reliable ;   mechanism for the preparation and the kit-creation steps.   B   Various more recent product installation kits will use the newer   PCSI installation mechanisms.   C   I would encourage you not to alter the options that are chosen by    SPKITBLD.COM.   C   Terminology: Files containing OpenVMS DCL commands are refered to B   as "command procedures" or as "procedures", and not generally as   "scripts".  F   Suggestion: please explain why you want to change the BACKUP savesetE   block size.  An error?  If so, what commands, specific symptoms and    error messages are seen?    F   You are asking terse and pointed questions, and these might or mightD   not get you the best answer -- you might well get a direct answer,C   but potentially not the best one.  For some suggestions on how to E   ask questions -- it is tougher than it looks -- and for some of the D   information that you will want to provide to the folks that might E   seek to answer your questions, please see the introductory sections    of the OpenVMS FAQ.   N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------    Date: 16 Jan 2003 01:15:52 +0800, From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: Boot Contest change- Message-ID: <87ptqynxkn.fsf@prep.synonet.com>   2 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:   > John Smith wrote:   D > > Just curious...how are they going to protect against boot sector > > viruses on a FAT partition?    3 > How would the virus get to those blocks of disk ?   @ Just install lozedoze on you SAN without partitioning the drives, so it can't improve the rest of the drives.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  + Date: Wed, 15 Jan 2003 23:35:53 +0000 (UTC) " From: "Yi Yi" <Yi_Yi@vodafone.net>' Subject: Re: Control-P on al alpha 4100 1 Message-ID: <b04r8o$74k$1@knossos.btinternet.com>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:gNdx$3cBtFlE@eisner.encompasserve.org... ? > In article <1f83b2aa.0301150703.1e96e681@posting.google.com>, . millerk@hushmail.com (Kevin T. Miller) writes:
 > > Hello. > > @ > > Would a Control-P ever cause major file corruption to disks?H > > Is it ever a good idea to perform a control-P when the machine is upF > > and running? Would this cause disk controllers to go bad also? Any( > > help greatly appreciated. Thank you. > C >    Control-P does different things on different systems.  On many A >    systems today it halts the processor.  If you don't continue H >    the processor and do a clean shutdown your file system's free blockI >    count will be stale.  If you do this on a system running a DBMS such * >    as Oracle you can hose the data base. >   E If you want to bring the box down quickly,  and do not have that much  memory, then do    $ mcr opccrash  H This forces a crash, but at least flushes out unwritten I/O, so the data should still be "clean".  K I have Control-P many a box, and as far as VMS is concerned the only issues L I have had are with the queue database and rights list.  But these have beenD few and far between.  In general you get away with it, but its not a% recommend practise for a live system.   I Control-P is a good way of getting a box down quickly.  This is useful if D for example its a DNS server and it has got its knickers in a twist.K Shutting it down normally you have to Control-y the shutdown so many times, K saves the hassle.  If you have a run away process or two and performance is K like a dog.  Some one unauthorized has gained access to your box and you do  not want them to do any more.   C The disk controllers will not care the box has gone down, after all H uncompleted I/O is in the cache. I cannot for see a scenario when a disk* controller, HSZ, HSJ, etc would get upset.   Yi Yi    ------------------------------  % Date: Wed, 15 Jan 2003 14:28:03 -0500 5 From: Matthew Doremus <Matthew.Doremus@hp.com.NoSpam> & Subject: Re: CSWS PHP 1.1 php.ini bug?, Message-ID: <3E25B643.8000002@hp.com.NoSpam>  G This is a problem in the CSWS_PHP 1.0 & 1.1 command line component.  I  A am putting together a fix for this and was wondering if you'd be  I interested in testing it.  I'll contact you offline in regard to getting  ' the images to you if you're interested.    Mark Daniel wrote:   > Mark Daniel wrote: > D >> I am finding it difficult to change configuration parameters via  >> PHP.INI.  >>J >> To demonstrate (this uses the PHP.EXE that is delivered with CSWS PHP): >>% >>   $ PHP = "$PHP_ROOT:[BIN]PHP.EXE" - >>   $ PIPE PHP -i | SEARCH SYS$INPUT: EXPOSE  >>I >> Note that the displayed value (amongst all the HTML markup) is 'on'.   B >> Now EDIT the file PHP_ROOT:[000000]PHP.INI and change the line E >> "php_expose = on" to php_expose = off".  Do the above pipe|search  D >> again and see if the reported value of php_expose has changed to ; >> "off".  I'll bet it hasn't, and neither can most (all?)  5 >> non-'extension' directives in the file be changed.  >>F >> I am not using this with CSWS and MOD_PHP and so cannot comment on F >> whether PHP.INI can be used in that configuration.  It should work  >> using the above in any case.  >>2 >> Can anyone confirm or dispute this observation? >> > , > OK.  It doesn't work with CSWS 1.3 either! > # >   $ type APACHE$ROOT:[PHP]PHP.INI  >   ; D >   ; Uncomment any of the extension below to have them automaticly  > loaded by PHP  >   ;  >   expose_php=off >   extension=php_bcmath.exe >   extension=php_calendar.exe >   extension=php_ctype.exe  >   8< snip 8< >  > php_info() ... >  >   8< snip 8<6 >   error_reporting    no value               no value0 >   expose_php         On                     OnB >   extension_dir      /php_root/extensions   /php_root/extensions >   8< snip 8<@ >   _SERVER["SERVER_SOFTWARE"] Apache/1.3.26 (OpenVMS) PHP/4.1.1 >   8< snip 8< >  > Log a software call? > H > +--------------------------------------------------------------------+F >  Mark Daniel                         http://wasd.vsm.com.au/adelaideG >  mailto:Mark.Daniel@wasd.vsm.com.au (Mark.Daniel@dsto.defence.gov.au) H > +--------------------------------------------------------------------+ >    ------------------------------   Date: 15 Jan 2003 23:15:23 GMT# From: hoff@hp.nospam (Hoff Hoffman) 3 Subject: Re: decnet  - mailbox programming question * Message-ID: <b04q2b$nnt$2@web1.cup.hp.com>  c In article <fbc2930d.0301140726.7a2b86ff@posting.google.com>, jiannisf@yahoo.com (Jiannisf) writes: F :I want to create a process in OVMS, which will communicate via Decnet7 :at one side and from an AST mailbox at the other side.   C   You've various semi-related and potentially unrelated terminology B   in that sentence, and I am not entirely certain what you intend.A   I will assume this is an application process which communicates *   with both DECnet and mailbox mechanisms.  F   DECnet task-to-task communications -- and particularly what is knownF   as non-transparent task-to-task -- has a construct confusingly knownB   as a network mailbox.  This network mailbox is used by a processF   acting as a server, and is how the process can receive DECnet statusC   and connection events.  It is NOT used as a data link for network C   communications with the remote host; user data does not generally C   pass through this mailbox.  (OK, I'm ignoring the small amount of B   application data that can transit the DECnet connect handshake.)  C   There are various DECnet code examples available in SYS$EXAMPLES: D   and on the OpenVMS Freeware, among other sources.  The OpenVMS FAQ;   has pointers to these and other OpenVMS coding resources.   E   A basic DCL-based transparent task-to-task example is posted at the    Ask The Wizard area.  3 :How can I create an AST for the Network messages ?  : G :Since now I'm using an infinite loop which reads network (sys$waitfr ( E :sys$get(RAB) ), but that approach doesnt work with AST mailboxes (?)   E   I really dislike sys$waitfr and event flags in general, personally. D   These certainly have their place in simple applications, but theseC   event flags do not scale particularly well.  I much prefer to use F   ASTs -- though that said, ASTs are probably not something that a newD   OpenVMS programmer is quite ready to work with.  Simple is better,*   at least for the first few applications.  C   You probably don't want to use RMS and the file system to access  A   mailboxes, nor do you probably want to use RMS to access remote @   tasks using DECnet task-to-task communications.  This is oftenD   possible, but the RMS record access semantics can have undesirable1   side-effects during application communications.   A   http://www.openvms.compaq.com/freeware/freeware50/srh_examples/ '   has nice.c, a DECnet-related example.   B   http://www.openvms.compaq.com/wizard/ topic (159) has the DECnetB   task-to-task DCL coding example.  Other topics of interest thereD   are (1910) and (7359).  For general discussions of programming andE   programming bugs and for pointers to many other programming-related B   topics, please see topic (1661).  Mailbox topics include (1893),6   (1937), (5045), (6004), (6497), (6905), (8015), etc.  @   The AskQ site referenced in the FAQ will also provide you withA   various programming examples and related information.  (This is %   one of many pointers in the FAQ...)   N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Wed, 15 Jan 2003 18:52:16 -0600 . From: Lyndon Bartels <lbartels@pressenter.com>! Subject: DECnet-Plus vs DECnet IV . Message-ID: <3E25ADDF.56C069BC@pressenter.com>   Hello,  H I've been using DECnet Phase IV on all my machines. But it seems that to> accomplish some of my goals, I need to move to DECnet Plus....  C I just finished an install of 7.3-1 and 'Plus' on my workstation to F start testing things out. So I haven't dug into any documentation yet.     But....   B Is there any documentation talking about the comparisons? Just theF rudimentery stuff... If I wanted to define a node on IV. What would be the analaous stuff on Plus.     E Any pointers would be great. I hate calling up CSC for questions like < that. I'll be reading docs for a long time before I do that.       Thanks in advance,   Lyndon     --  G My opinions are mine and mine alone. They seldom align with those of my 	 employer.     H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Wed, 15 Jan 2003 20:23:08 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>% Subject: Re: DECnet-Plus vs DECnet IV / Message-ID: <3E25FB6B.C8D875DC@vl.videotron.ca>    Lyndon Bartels wrote: D > Is there any documentation talking about the comparisons? Just theH > rudimentery stuff... If I wanted to define a node on IV. What would be > the analaous stuff on Plus.   L It,s been a while, but go into MC NCP and use the help there. I *think* thatM there is a way to get the fake NCP to translate the real DECNET commands into  the Decnet-plus stuff.  L When I installed DECNET-5, configuring it to talk to a decnet-4 node was not> too difficult. I was able to do set host and copy operations.   G But you'll need the manuals to try to figure out which files need to be M modified to add the commands you need because therte is a whole hiearchy that I gets executed ina  certain order, and if you enter a command in the wrong L file, chances are it will bet overrules by a contradicting command in a file executed later.    ------------------------------  % Date: Wed, 15 Jan 2003 21:22:15 -0500 3 From: "Homer J. Simpson" <hsimpson@burnsenergy.com> % Subject: Re: DECnet-Plus vs DECnet IV 6 Message-ID: <TAoV9.29791$Jm2.28384@news.bellsouth.net>  , Not sure if this will help, but on eBay. . .= http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=2301544234   = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message ) news:3E25FB6B.C8D875DC@vl.videotron.ca...  > Lyndon Bartels wrote: F > > Is there any documentation talking about the comparisons? Just theJ > > rudimentery stuff... If I wanted to define a node on IV. What would be > > the analaous stuff on Plus.  > I > It,s been a while, but go into MC NCP and use the help there. I *think*  thatJ > there is a way to get the fake NCP to translate the real DECNET commands into > the Decnet-plus stuff. > J > When I installed DECNET-5, configuring it to talk to a decnet-4 node was not ? > too difficult. I was able to do set host and copy operations.  > I > But you'll need the manuals to try to figure out which files need to be J > modified to add the commands you need because therte is a whole hiearchy thatK > gets executed ina  certain order, and if you enter a command in the wrong I > file, chances are it will bet overrules by a contradicting command in a  file > executed later.    ------------------------------  % Date: Wed, 15 Jan 2003 22:25:17 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)% Subject: Re: DECnet-Plus vs DECnet IV L Message-ID: <rdeininger-1501032225170001@user-2ive3r6.dialup.mindspring.com>  = In article <3E25ADDF.56C069BC@pressenter.com>, Lyndon Bartels   <lbartels@pressenter.com> wrote:   >Hello,  > I >I've been using DECnet Phase IV on all my machines. But it seems that to ? >accomplish some of my goals, I need to move to DECnet Plus....  > D >I just finished an install of 7.3-1 and 'Plus' on my workstation toG >start testing things out. So I haven't dug into any documentation yet.   ( You need to look at the documentation...  3 http://www.openvms.compaq.com/doc/decnetplus73.html     
 I suggest:J 1.  Look at the table of contents for all of the manuals to get an idea of what is in each of them.  C 2.  Don't skip the planning guide.  You can likely skip most of the @ sophisticated options, but pay some attention to the stuff about namespaces.   H 3.  The "local" namespace is analogous to the functionality of DECnet-IVI node name concepts.  The configuration will typically convert a DECnet-IV I name database to a DECnet-plus local namespace.  There is a separate tool @ for managing the local namespace, decnet_register.  For example: $ netreg :== $decnet_register  $ netreg  G This tool uses a character-cell menu interface by default, but there is 3 also a command interface which is easier to script.   E 4.  The DECdns namespace was designed for very large installations.   D Small installations will typically not use it.  But even for a dozenH systems, it lets you centralize name management instead of keeping it on
 each node.  F 5.  If you configure TCPIP, the domain namespace connects DECnet names4 with TCPIP names, and lets you do DECnet-over-tcpip.  G 6.  Except for the namespaces, the rest of DECnet-plus is controlled by J the configuration tool (@sys$manager:netconfigure) and by the NCL utility,I which is weird compared to almost everything else in VMS.  NCL takes some I getting used to.  Look at the Installation and Basic Configuration manual J to get started. There is also a DECwindows graphical interface that can do$ some monitoring and configuration.   $ run sys$system:net$mgmt   I This is not a complete tool.  IMHO, they never finished implementing it.  J One useful feature, if you hunt in the menus a bit: the graphical tool canJ echo the NCL commands it uses to accomplish its tasks.  This is a nice was to learn some useful NCL.   J 7.  There is also an NCP emulator tool, which is again not comprehensive. < NCL and NCP just don't map into one another in a simple way.    J When I first configured DECnet-plus, I found it fairly bewildering.  UntilH I spent a few hours in the docs.  After that, almost everything seems at@ least as sensible as DECnet-IV.  Different, but no more bizarre.   ------------------------------  % Date: Wed, 15 Jan 2003 22:01:54 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>% Subject: Re: DECnet-Plus vs DECnet IV 6 Message-ID: <3E262EB2.1CBB82DA@earthlink.spamfree.net>   Lyndon Bartels wrote:  >  > Hello, > J > I've been using DECnet Phase IV on all my machines. But it seems that to@ > accomplish some of my goals, I need to move to DECnet Plus....  
 Such as ... ?   < That is, what do you need to do that DECnet-IV can't handle?   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 15 Jan 2003 16:42:31 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)   Subject: Re: Disk drive too big?< Message-ID: <cf15391e.0301151642.a544359@posting.google.com>  p cdandrea@nd.edu (Christopher D'Andrea) wrote in message news:<70aee1c6.0301091149.cbc2d89@posting.google.com>...H > I have tried to install a new (18 GB) disk drive in my VAX machine butG > get an "aritmetic trap divide by zero" error when I try to initialize B > it. I suspect that the drive may be too large for the machine to> > recognize.  Is there some way to partition the drive or even+ > initialize it at a slightly smaller size?   E As I recall, in OpenVMS VAX versions prior to V6.0, there was a limit B in the XQP (IIRC, the Relative Volume Number was stored in the topA byte of a longword when handling retrieval pointers, so retrieval F pointer LBNs were limited to a field width of 3 bytes (24 bits), whichC limited Files-11 volume size to 8 gigabytes (in 1K x 1K x 1K terms, B where 1K = 1024, or 8.5 gigabytes in disk-marketing terms, where aB gigabyte is a billion (US) bytes).  If a disk is mounted /FOREIGN, there's no problem.   F One workaround folks used in those days was to use the host-based RAIDA software to divide a single disk up into pieces of 8 gigabytes or E less, and then mount each of them.  See my DECUS presentation on Disk  Partitioning on OpenVMS atJ http://www.geocities.com/keithparris/decus_presentations/s99_partition.ppt   ------------------------------    Date: 15 Jan 2003 20:49:43 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) ! Subject: Re: EVA and READL/WRITEL < Message-ID: <cf15391e.0301152049.f16ced2@posting.google.com>  f young_r@encompasserve.org (Rob Young) wrote in message news:<BXbgXq5csEI8@eisner.encompasserve.org>...5 > Does Enterprise Virtual Array support READL/WRITEL?   B I can't imagine that it wouldn't, considering that all the HSZ/HSGD models have supported READL/WRITEL.  I don't have ready access to anE EVA at the moment, but if someone else does, it is relatively easy to @ check.  There's a discussion of this in the Volume Shadowing forC OpenVMS Manual; section 4.9.4.1 in the 7.3 version shows how to use ? SDA to check if a SCSI disk supports the simulated Forced-Error D operations or not (it will have "nofe" set in the "Characteristics:"D section of an SDA> SHOW DEVICE DKAxx: display if it does NOT support READL/WRITEL).   For example:  E $1$DGA41 [ARGON$DGA41]                 EMC SYMMETRIX    UCB: 80DF7040   < Device status:   08020810 online,valid,lcl_valid,exfunc_supp Characteristics: 1C4D5008 0 dir,mpdev_member,fod,shr,avl,mnt,elg,idv,odv,rnd7                  25050201 clu,nnm,nlt,shd,scsi,nofe,dtn 3                                                ^^^^    ------------------------------    Date: 16 Jan 2003 00:32:54 +0800, From: Paul Repacholi <prep@prep.synonet.com>0 Subject: Re: F$GETDVI ("AVL") in an IF statement- Message-ID: <87y95mnzk9.fsf@prep.synonet.com>   ( David M Smith <dsmit115@csc.com> writes:  N > On 09 Jan 2003 07:41:40 +0800, Paul Repacholi <prep@prep.synonet.com> wrote: > 3 > >spamsink2001@yahoo.com (Alan E. Feldman) writes:   ; > >> The lexical function F$GETDVI assumes the latter view.   : > >It uses the definitions lated out in the MSCP spec. The< > >UDA-50 manaul has a good section on 'EXISTS', 'AVAILABLE'1 > >and 'ONLINE' and the differences between them.   B > Can you tell me where I might find the "MSCP spec"? Is it easily) > available online, or in paper doc only?   F Good question. I got a copy of the UDA-50 programers kit before it wasE pulled. Long ago... I don't think the full MSCP spec has ever escaped  as yet.      --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 16 Jan 2003 02:25:03 +0800, From: Paul Repacholi <prep@prep.synonet.com>K Subject: Re: Fibre channel:  Is the INIT/ERASE function built into the HBA? - Message-ID: <87hecanudc.fsf@prep.synonet.com>   B "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net> writes:  F > > My question: Is the "ERASE" part of INIT/ERASE built into the HBA?F > > Does the OS tell the HBA to "Go erase that bad boy and let me know > > when you are done."?   D > Based on input from past postings, I'm inclined to believe that it@ > invokes a $QIO which causes some things to happen at the lower= > layers, tho I'm not quite sure what magic is afoot there...   B The driver is emulating the MSCP ERASE funtion, so a single ERASE @ for the entire device will work as it should. Note, that this isD not an `industry standard' funtion, so it may go away at any moment.     --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 15 Jan 2003 14:32:17 -0500 ! From: VAXVMS <bounce@notmail.com> ; Subject: RE: Getting started with an AlphaStation 200 4/233 K Message-ID: <BA52530E3149734A9BAABDBBFA808E4903027BBB@rlghncst964.usps.gov>    You may not, but you can???   " That statement contradicts itself.  3 I have two live hobbyist systems, one Alpha and one 	 MicroVAX.   3 I just looked at all four license files and none of / them have any such version restriction in them.   : Oddly enough, they all do refer to "your signature on this form" and "the VAX OpenVMS"   K $!      Upon your qualification for this license and your signature on this F $!     form, Compaq Computer Corporation ("COMPAQ") will grant you theC $!     right to use the VAX OpenVMS on a single computer ("Licensed  $!     Computer").    C I guess all the Alphas are illegal, then, and I'm illegal because I  didn't sign the emails. :^)   E Paging Doctor Kafka...Doctor Kafka, please pick up the white courtesy  phone...   WWWebb  G =======================================================================   C Sorry. I was not clear. I should have written "you may not, but you F can". The OpenVMS Hobbyist licence expressively (sp?) mention that youH are not allowed to use another VMS version than the one available on the Hobbyist CD.  C >>>Or does this license only work with the CDs in the hobbyist kit?  >> >>Yes, it does.  >>H > No the licenses should work with any VMS distribution you can get your hands  on.   G I was again talking about the VMS licence. For the Layered Products, he E should register his system on the Montagar site to get valid "normal" @ licences PAK to use "normal" software from the Software Library.   D.   ========================  William W. Webb / DSSC/RLM, USPS OpenVMS Support Services& 4924 Green Road Raleigh, NC 27616-2800: 919.874.3043 <FirstInitialDotLastNameAtEmailDotUSPSDotGov>   ------------------------------  % Date: Wed, 15 Jan 2003 11:32:49 -0800 $ From: Shane Smith <ssmith@icius.com>; Subject: RE: Getting started with an AlphaStation 200 4/233 0 Message-ID: <01C2BC89.D9113440@sulfer.icius.com>  D I'm fairly sure in the early days of the program you were limited toE certain releases and certain layered products. For example clustering G wasn't originally allowed, if I recall correctly. It was revamped later E based on customer feedback, and they opened it up to all versions and  pretty much all products.    Shane    -----Original Message-----( From: VAXVMS [mailto:bounce@notmail.com]* Sent: Wednesday, January 15, 2003 11:32 AM To: Info-VAX@Mvb.Saic.Com ; Subject: RE: Getting started with an AlphaStation 200 4/233      You may not, but you can???   " That statement contradicts itself.  3 I have two live hobbyist systems, one Alpha and one 	 MicroVAX.   3 I just looked at all four license files and none of / them have any such version restriction in them.   : Oddly enough, they all do refer to "your signature on this form" and "the VAX OpenVMS"   F $!      Upon your qualification for this license and your signature on thisF $!     form, Compaq Computer Corporation ("COMPAQ") will grant you theC $!     right to use the VAX OpenVMS on a single computer ("Licensed  $!     Computer").    C I guess all the Alphas are illegal, then, and I'm illegal because I  didn't sign the emails. :^)   E Paging Doctor Kafka...Doctor Kafka, please pick up the white courtesy  phone...   WWWebb  G =======================================================================   C Sorry. I was not clear. I should have written "you may not, but you F can". The OpenVMS Hobbyist licence expressively (sp?) mention that youH are not allowed to use another VMS version than the one available on the Hobbyist CD.  C >>>Or does this license only work with the CDs in the hobbyist kit?  >> >>Yes, it does.  >>H > No the licenses should work with any VMS distribution you can get your hands  on.   G I was again talking about the VMS licence. For the Layered Products, he E should register his system on the Montagar site to get valid "normal" @ licences PAK to use "normal" software from the Software Library.   D.   ========================  William W. Webb / DSSC/RLM, USPS OpenVMS Support Services& 4924 Green Road Raleigh, NC 27616-2800: 919.874.3043 <FirstInitialDotLastNameAtEmailDotUSPSDotGov>   ------------------------------  % Date: Wed, 15 Jan 2003 18:32:26 +0000 + From: John Laird <john@laird-towers.org.uk> ; Subject: Re: Getting started with an AlphaStation 200 4/233 8 Message-ID: <2t9b2vcmsgf271d2qrflpnr4a5j9m94rka@4ax.com>  I On Wed, 15 Jan 2003 18:06:20 +0000 (UTC), david20@alpha2.mdx.ac.uk (David  Webb) wrote:  [ >In article <3E259A5D.3090209@Free.fr>, SAP Trainee <Didier.Morandi.nospam@Free.fr> writes:  >>David Webb a crit:  >>H >>>>>2. And can I use the hobbyist license with commercial distribution? >>>>D >>>>No, you cannot (but, really, who will care for a dying product?) >>>>O >>> Yes you can. VMS is VMS is VMS. There is no "hobbyist" version of VMS it ise- >>> just a particular CD distribution of VMS.-R >>> There have been many recomendations on this newsgroup from Compaq/HP employeesQ >>> for hobbyists to borrow "commercial distributions" from friends who work withFR >>> VMS if there has been a holdup in getting the hobbyist distribution or if theyQ >>> wanted a layered product which wasn't put on the hobbyist CD because of spacee6 >>> rsstrictions but for which a license was provided. >>F >>Sorry. I was not clear. I should have written "you may not, but you I >>can". The OpenVMS Hobbyist licence expressively (sp?) mention that you EK >>are not allowed to use another VMS version than the one available on the u >>Hobbyist CD. >>K >I've never used a system under the hobbyist license however looking at ther7 >Montagar site I can't see anything remotely like that.o  K Was that not a feature of the first hobbyist licence ? It was restricted toeE a particular version of VMS (possibly 6.2), which was at the time notuJ current.  I recall this was relaxed to the present agreement which is that any version is ok.  : >It certainly isn't in the COMPAQ HOBBY License agreement  >see >e: >http://www.montagar.com/hobbyist/terms_and_conditions.txt >s >orn >r7 >http://www.montagar.com/hobbyist/register_license.htmle >g. >In fact the site explicitly tells you that :- >  >"O >At this time, OpenVMS Vax media kits are sold out. If you are still interestedeJ >in getting OpenVMS for your VAX,you may try borrowing a CD from a friend,7 >co-worker, local Encompass LUG, Ebay, or other places. L >Please note that any OpenVMS distribution is legal to use with the Hobbyist( >Program (as long as it's not stolen!)."  I A licence (as most are aware) is a right to *use*.  As long as no laws oroF contracts are broken in acquiring the software, the hobbyist is legal.J Borrowing an official distribution kit should be okay, and I would think aH backup CD would be regarded as legitimate (even MS let you do this), butJ clearly selling one would not be.  I'm not entirely sure you can sell on aI legally-acquired distribution kit.  The key thing is to be sensible aboutn it.h     	Johno   ------------------------------    Date: 15 Jan 2003 22:13:18 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>; Subject: RE: Getting started with an AlphaStation 200 4/233a6 Message-ID: <20030115221318.19394.qmail@nym.alias.net>  7 On Wed, 15 Jan 2003, VAXVMS <bounce@notmail.com> wrote:r  F >Paging Doctor Kafka...Doctor Kafka, please pick up the white courtesy	 >phone...   1 ROTFLMAO!!!!<insert random "1" instead of "!">!!!:     Doc. -- s6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                             https://vmsbox.cjb.nettK                                                    http://althacker.cjb.netI   ------------------------------  % Date: Wed, 15 Jan 2003 21:47:59 -0600a@ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>; Subject: Re: Getting started with an AlphaStation 200 4/233t6 Message-ID: <3E262B6F.74484B83@earthlink.spamfree.net>   SAP Trainee wrote: >  > David Webb a crit:  > H > >>>2. And can I use the hobbyist license with commercial distribution? > >>D > >>No, you cannot (but, really, who will care for a dying product?) > >>O > > Yes you can. VMS is VMS is VMS. There is no "hobbyist" version of VMS it is - > > just a particular CD distribution of VMS.,R > > There have been many recomendations on this newsgroup from Compaq/HP employeesQ > > for hobbyists to borrow "commercial distributions" from friends who work withaR > > VMS if there has been a holdup in getting the hobbyist distribution or if theyQ > > wanted a layered product which wasn't put on the hobbyist CD because of space.6 > > rsstrictions but for which a license was provided. > E > Sorry. I was not clear. I should have written "you may not, but you H > can". The OpenVMS Hobbyist licence expressively (sp?) mention that youJ > are not allowed to use another VMS version than the one available on the > Hobbyist CD.    That information is not correct.  H At one time, OVMS-Hobby was VAX-only and limited to some version(s), I'm' not sure which. This has since changed.   G Currently, the OVMS-Hobby license is not version-restricted, AFAIK, andE9 even includes those versions which are not LMF-compliant.    ...again, AFAIK.   -- d David J. Dachterat dba DJE Systemsn http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Wed, 15 Jan 2003 21:54:01 -0600i@ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>; Subject: Re: Getting started with an AlphaStation 200 4/233.6 Message-ID: <3E262CD9.BAC02F31@earthlink.spamfree.net>  
 VAXVMS wrote:y >  > You may not, but you can???  > $ > That statement contradicts itself.  F Well, no, not really. What one is allowed to do (what you "may do") is9 usually a subset of what is possible (what you "can do").L  G Some may remember parents and/or teachers responding confusingly to theiF query of, "Can I ... ?" ("Am I capable of ...?") when what is actually5 meant is "May I ... ?" ("Will you allow me to ... ?")b  G Don't really want to get into a pedantic war here, but it's rather likeeG the difference between what CAN be done and what SHOULD be done: more a 1 question of values and/or morals than capability.    --   David J. Dachtera0 dba DJE Systemsd http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/-   ------------------------------  % Date: Wed, 15 Jan 2003 21:44:17 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>; Subject: Re: Getting started with an AlphaStation 200 4/233e6 Message-ID: <3E262A91.FDCEC40F@earthlink.spamfree.net>   Otto wrote:n > [snip]H > 2. And can I use the hobbyist license with commercial distribution? Or? > does this license only work with the CDs in the hobbyist kit?g  3 Again, VMS is VMS. There is no "hobbyist version". -  3 Hoff: This probably needs to get into the FAQ, huh?m  G There is, however, a hobbyist's distribution which has been massaged to C contain as many often-requested layered products as can fit. That's % really the only important difference.w  E The terms of the license are what determine what you can or cannot do  LEGALLY.   --   David J. DachteraB dba DJE Systemsy http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/u   ------------------------------  # Date: Wed, 15 Jan 2003 20:28:19 GMTs/ From: "Robin Schipper" <robinschip@hotmail.com>  Subject: Hardwares. Message-ID: <DvjV9.202566$xm4.9072530@Flipper>  ' i have the following hardware availeblei  I 16 x Vax station 3100 / VS42A-DA / 12MB / RZ23(104 Mb) / TZ30(tape drive)e  
 Microvax 3100e2 DV-31-ETA-A / 2 X RZ24(209 Mb) / TZ30 (tape drive)6 DV-31-ETA-A-A01 / 2 X RZ24(209 Mb) / TZ30 (tape drive)6 DV-31-ETA-A-A01 / 2 X RZ24(209 Mb) / TZ30 (tape drive)  
 DEC Repeatersr DEC Microserver1 Dec Thinwire multiport repeaterk  
 Codex 6003
 Codex 6015	 Codex DSDi AMP 10Base2 Multiport repeater   Textronix XP 400 Textronix XP 200   DEC laserprinter LN15s  LA75 Accesoires (inkt, cleaning) 2 x HP 500cA  * US Robotics 33.6 Faxmodem (with voicemail)   Hayes 2400 modem hayes 9600 modem visa DMD 124 2400 modeme  # 2 x RAD MBE-8 Lanranger Lanextendert# 2 X RAD MBE-1 Lanranger Lanextenderi 2 x RAD MBE  Lanextender   2 X Blackbox SME v.35 HS Modem   2 KPN moduvox 324 Modem- 1 KPN moduvox 397 Modem   # 11 X 3com Etherlink II networkcardsM  
 DLT I tape DLT II tape  DLT III Tape   the hardware is in Amsterdam  6 for information please mail me at rschipper@tribase.nl   ------------------------------    Date: 16 Jan 2003 02:35:46 +0800, From: Paul Repacholi <prep@prep.synonet.com>S Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?n- Message-ID: <87d6myntvh.fsf@prep.synonet.com>E  & rcbryan@hotmail.com (RC Bryan) writes:   ...   - > Is a limit of 64K reasonable for V7.1 on?  l' > Can I go with a much greater limit?  s  D You can be SURE that 64K will work on any disk. That is in the ODS-2B spec, so you are safe. Larger may work with the right software and* drives, but will fail on older HW, and SW.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.p@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 15 Jan 2003 21:32:36 GMT# From: hoff@hp.nospam (Hoff Hoffman) S Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?0* Message-ID: <b04k1k$ja6$1@web1.cup.hp.com>  c In article <fbcf38dc.0301131705.b838357@posting.google.com>, rcbryan@hotmail.com (RC Bryan) writes:n  G   $qio and $io_perform are the usual interfaces for low-level disk I/O.:  G   Most disk devices permit 127 blocks or 256 block transfers, depending2F   on the particular device and the particular device driver.  SCSI andF   Fibre Channel (FC) disk device drivers will typically transfer up toE   256 blocks, depending on the particular port.  ATA (IDE) CD support,G   deliberately restricts its disk I/O transfers to 120 blocks, to avoid !   blocking and deblocking issues."  H   And a plea: Please consider refraining from cross-posting questions to   the OpenVMS Wizard.  Thanks!  N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com0   ------------------------------  % Date: Wed, 15 Jan 2003 23:03:12 +0100t6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>S Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?2) Message-ID: <3E25DAA0.9080207@vajhoej.dk>r   RC Bryan wrote:>E > I never heard of Fast I/O before.  I guess I overlooked it since iteF > does not appear in the paper V6.0 manuals I have on my desk.  If you: > have an example, I would be interested in looking at it.  1 It was introduced with VMS 7.0 and is Alpha only.i   Arne   ------------------------------  + Date: Thu, 16 Jan 2003 06:51:17 +0000 (UTC) 3 From: "Richard Maher" <maher_rj@hotspamnotmail.com>pS Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?o0 Message-ID: <b05kp2$t9a$1@sparta.btinternet.com>    This will wrap terribly but FWIW   $ on warning then exit: $ if .not. f$privilege("cmkrnl,sysprv")  then goto no_priv9 $ if f$getsyi("arch_name") .nes. "Alpha" then goto no_vaxn $! $ create big_buffs.mar ;++a ;o4 ; (c) Copyright Tier3 Software. All rights reserved. ; @ ;     Ownership of this software and all associated intellectual@ ;     property rights remain vested in Tier3 Software Ltd.  This@ ;     software  or any other copies thereof  may not be provided6 ;     or otherwise made available to any other person. ; * ;     Do not remove this copyright notice. ;, ;     Author: Richard Maher- ;- ;---=         .macro define_service,name,narg=0,mode=exec,?endmacroV  3         'mode'_routine_count='mode'_routine_count+1w  0         .call_entry     max_args=narg,         -0                         home_args=true,        -"                         label=name  "         .save_psect    local_block  2         .psect  'mode'_list,long,wrt,noexe,pic,con           .address        name           .restore_psect           .if not_equal narg            cmpb    (ap),#nargm          bgeq    endmacrol           movzwl  #ss$_insfarg,r0          ret  	 endmacro:V  
         .endcp
         .endmc  8         .title  BIG_BUFFS - Demo VLM & Fast I/O routines         .ident  "V2.0"  &         .library "sys$library:lib.mlb"           $vadef         $psldefn         $ssdef         $iodef         $iosadef         $prvdef          $plvdefl         $cbodef'         $syidef          $lnmdefh         $secdef   E         one_megabyte=^x100000           ;Divisible by all bage sizes?h         block_len=512=&         max_vbn=one_megabyte/block_len"         gb_version_from_database=1         enable=1         disable=0t           kernel_routine_count=0         exec_routine_count=0  0         .psect  exec_list,long,wrt,noexe,pic,con exec_table:r  2         .psect  kernel_list,long,wrt,noexe,pic,con
 kernel_table:s  B         .psect  maher$data,pic,con,rel,lcl,noshr,noexe,rd,wrt,quad  ) map_start_va:           .blkq           1 ) map_length:             .blkq           1 4 section_id:             .long           sec$k_matequ) section_version:        .long           0r) region_id:              .blkq           1g) region_start_va:        .blkq           1r) region_length:          .blkq           1 / iosa_region_id:         .quad           va$c_p2 ) iosa_start_va:          .blkq           19) iosa_length:            .blkq           17) iosaobj_start_va:       .blkq           1P) iosaobj_length:         .blkq           1 ) iosa_handle:            .blkq           1r) bufobj_start_va:        .blkq           1:) bufobj_length:          .blkq           10) buffer_handle:          .blkq           1l) read_fandle:            .blkq           1r) write_fandle:           .blkq           1  mgblsc_prv:             .quady( <prv$m_sysprv!prv$m_sysgbl!prv$m_prmgbl>6 io_setup_prv:           .quad           <prv$m_altpri>) del_prv:                .blkq           1 ) old_prv:                .blkq           1 9 syi_itmlst:             .word           4, syi$_page_sizeg5                         .address        cpu_page_size )                         .long           0n=                         .word           4, syi$_galaxy_membera5                         .address        galaxy_member ,                         .long           0, 0) cpu_page_size:          .long           0 )                         .long           0i) galaxy_member:          .long           0h) syi_iosb:               .quad           0uL lnm_attr:               .long           <lnm$m_case_blind!lnm$m_interlocked>6 lnm_itmlst:             .word           1, lnm$_string8                         .address        expedite_io_flag,                         .long           0, 0) flush_vbn:              .quad           1o sec_flags:              .longu/ <sec$m_expreg!sec$m_wrt!sec$m_sysgbl!sec$m_gbl>h) io_setup_flags:         .long           0 ) db_chan:                .word           0o8                         .byte           0, 0, 0, 0, 0, 0  2 gs_name:                .ascii          "MY_BUFFS" gs_name_len=.-gs_name=J cr_fao_ctl:             .ascii          "Shared Page Table Region Created, VA = !@XQ, Size = !@ZQ"b cr_fao_len=.-cr_fao_ctl K ms_fao_ctl:             .ascii          "Section !AS mapped, Section Lengthn = !ZL!/"L                         .ascii          "!_Mapped VA = !@XQ, Mapped Length = !@ZQ"i ms_fao_len=.-ms_fao_ctloF bo_fao_ctl:             .ascii          "Buffer Object Created, Buffer Length = !@ZQ!/"L                         .ascii          "!_Buffer VA = !@XQ, Buffer Handle = !@XQ"h bo_fao_len=.-bo_fao_ctloK iosa_fao_ctl:           .ascii          "IOSA Object Created, IOSA Length =w !@ZQ!/"tH                         .ascii          "!_IOSA VA = !@XQ, IOSA Handle = !@XQ"M iosa_fao_len=.-iosa_fao_ctlc+ fmt_string:             .blkb           2551 fmt_string_len=.-fmt_stringsE gb_error:               .ascii          "%RDB-E-GBVINTAGE, Can't opene* database. Global Section version mismatch" gb_error_len=.-gb_errorr8 def_fnm:                .ascii          "sys$login:.vlm" def_fnm_len=.-def_fnm 9 abnormal_msg:           .ascii          "(Abnormal Exit)"J abnormal_msg_len=.-abnormal_msgi6 lnm_tabnam:             .ascii          "LNM$FILE_DEV" lnm_tabnam_len=.-lnm_tabnam D lnm_lognam:             .ascii          "RDMS$BIND_EXPEDITE_FAST_IO" lnm_lognam_len=.-lnm_lognam ) expedite_io_flag:       .blkb           1s,                         .align          long3 gs_name_desc:           .long           gs_name_leni/                         .address        gs_namea2 cr_fao_ctl_desc:        .long           cr_fao_len2                         .address        cr_fao_ctl2 ms_fao_ctl_desc:        .long           ms_fao_len2                         .address        ms_fao_ctl2 bo_fao_ctl_desc:        .long           bo_fao_len2                         .address        bo_fao_ctl4 iosa_fao_ctl_desc:      .long           iosa_fao_len4                         .address        iosa_fao_ctl6 fmt_string_desc:        .long           fmt_string_len2                         .address        fmt_string) out_string_len:         .long           042                         .address        fmt_string4 somewhat_dated:         .long           gb_error_len0                         .address        gb_error8 abnormal_msg_desc:      .long           abnormal_msg_len4                         .address        abnormal_msg6 lnm_tabnam_desc:        .long           lnm_tabnam_len2                         .address        lnm_tabnam6 lnm_lognam_desc:        .long           lnm_lognam_len2                         .address        lnm_lognam  $                         .align  quad db_fab:o:                         $fab            fnm=<ondisk_db>, -8                                         dna=<def_fnm>, -<                                         dns=<def_fnm_len>, -@                                         fac=<get,put,upd,del>, -A                                         chan_mode=<psl$c_exec>, -'@                                         lnm_mode=<psl$c_exec>, -8                                         fop=<ufo,cif>, -8                                         alq=<max_vbn>, -A                                         shr=<upi,get,put,upd,del>n  F                                         ; Locking ommitted for brevity  @         .psect  maher$code,pic,con,rel,lcl,shr,exe,rd,nowrt,quad  %         define_service  maher$db_openn   10$:    $setup_call64   63(         movaq           region_length,r6         $push_arg64     r6*         movaq           region_start_va,r6         $push_arg64     r6$         movaq           region_id,r6         $push_arg64     r6(         $push_arg64     #va$m_shared_pts1         $push_arg64     #va$c_region_ecreate_eowndI         $push_arg64     #one_megabyte                   ;Page tables will  round to 8MB,         $call64         sys$create_region_64         blbs            r0,20$         ret   1 20$:    $fao_s          ctrstr=cr_fao_ctl_desc, - 0                         outlen=out_string_len, -1                         outbuf=fmt_string_desc, - .                         p1=#region_start_va, -)                         p2=#region_lengthc         blbs            r0,25$         ret  25$:&         pushal          out_string_len+         calls           #1,g^lib$put_output          blbs            r0,30$         ret- 30$:+         $getsyiw_s      itmlst=syi_itmlst,--%                         iosb=syi_iosb0         blbc            r0,32$#         movzwl          syi_iosb,r0          blbs            r0,34$ 32$:         ret4 34$:A         movl            #gb_version_from_database,section_version   %         tstl            galaxy_member          beqlu           36$s.         bisl2           #sec$m_shmgs,sec_flags 36$:(         $setprv_s       enbflg=#enable,-+                         prvadr=mgblsc_prv,-S&                         prvprv=old_prv           $setup_call64   9r%         movaq           map_length,r6R         $push_arg64     r6'         movaq           map_start_va,r6          $push_arg64     r6$         movl            sec_flags,r6         $push_arg64     r6#         $push_arg64     #psl$c_exec:%         $push_arg64     #one_megabytec         $push_arg64     #0$         movaq           region_id,r6         $push_arg64     r6%         movaq           section_id,r6e         $push_arg64     r6'         movaq           gs_name_desc,r6u         $push_arg64     r6%         $call64         sys$mgblsc_64i           movl            r0,r5=2         evax_bic        mgblsc_prv,old_prv,del_prv)         $setprv_s       enbflg=#disable,-i&                         prvadr=del_prv           movl            r5,r0l         blbs            r0,50$)         cmpw            #ss$_nosuchsec,r0r         beqlu           40$w         ret/ 40$:&         pushal          somewhat_dated+         calls           #1,g^lib$put_output-)         movzwl          #ss$_nosuchsec,r01         retr 50$:1         $fao_s          ctrstr=ms_fao_ctl_desc, -e0                         outlen=out_string_len, -1                         outbuf=fmt_string_desc, -o+                         p1=#gs_name_desc, -c+                         p2=#one_megabyte, -a+                         p3=#map_start_va, -i&                         p4=#map_length         blbs            r0,60$         retc 60$:&         pushal          out_string_len+         calls           #1,g^lib$put_outputl         blbs            r0,65$         retC 65$:         $setup_call64   7e(         movaq           buffer_handle,r6         $push_arg64     r6(         movaq           bufobj_length,r6         $push_arg64     r6*         movaq           bufobj_start_va,r6         $push_arg64     r6         $push_arg64     #0#         $push_arg64     #psl$c_exec "         $push_arg64     map_length$         $push_arg64     map_start_va,         $call64         sys$create_bufobj_64           blbs            r0,70$         retc 70$:1         $fao_s          ctrstr=bo_fao_ctl_desc, -p0                         outlen=out_string_len, -1                         outbuf=fmt_string_desc, -o,                         p1=#bufobj_length, -.                         p2=#bufobj_start_va, -)                         p3=#buffer_handle          blbs            r0,80$         ret  80$:&         pushal          out_string_len+         calls           #1,g^lib$put_outputs         blbs            r0,140$          reth 140$:          $setup_call64   6 &         movaq           iosa_length,r6         $push_arg64     r6(         movaq           iosa_start_va,r6         $push_arg64     r6         $push_arg64     #0#         $push_arg64     #psl$c_exec H         $push_arg64     cpu_page_size           ; 2 x 32 bytes + (space))         movaq           iosa_region_id,r6p         $push_arg64     r6%         $call64         sys$expreg_64-         blbs            r0,150$(         retm 150$:e         $setup_call64   7h&         movaq           iosa_handle,r6         $push_arg64     r6)         movaq           iosaobj_length,r6y         $push_arg64     r6+         movaq           iosaobj_start_va,r6@         $push_arg64     r6         $push_arg64     #0#         $push_arg64     #psl$c_exece#         $push_arg64     iosa_length %         $push_arg64     iosa_start_va ,         $call64         sys$create_bufobj_64           blbs            r0,160$d         rett 160$:l3         $fao_s          ctrstr=iosa_fao_ctl_desc, - 0                         outlen=out_string_len, -1                         outbuf=fmt_string_desc, -C-                         p1=#iosaobj_length, - /                         p2=#iosaobj_start_va, - '                         p3=#iosa_handleo         blbs            r0,170$l         retC 170$:r&         pushal          out_string_len+         calls           #1,g^lib$put_outpute         blbs            r0,180$-         ret- 180$: '         $trnlnm_s       attr=lnm_attr,-C0                         tabnam=lnm_tabnam_desc,-0                         lognam=lnm_lognam_desc,-,                         acmode=#psl$c_exec,-)                         itmlst=lnm_itmlstg         blbc            r0,185$t  /         cmpb            expedite_io_flag,#^a"1"3         bnequ           185$6         movl            #fio$m_expedite,io_setup_flags   185$: (         $setprv_s       enbflg=#enable,--                         prvadr=io_setup_prv,-s&                         prvprv=old_prv  #         pushaq          read_fandlen&         pushl           io_setup_flags         pushl           #0#         pushaq          iosa_handle %         pushaq          buffer_handle 5         pushl           #<io$_readvblk!io$m_novcache>t)         calls           #6,g^sys$io_setup          blbc            r0,190$h  $         pushaq          write_fandle&         pushl           io_setup_flags         pushl           #0#         pushaq          iosa_handle %         pushaq          buffer_handlet6         pushl           #<io$_writevblk!io$m_novcache>)         calls           #6,g^sys$io_setup  190$:,         movl            r0,r5t4         evax_bic        io_setup_prv,old_prv,del_prv)         $setprv_s       enbflg=#disable,-i&                         prvadr=del_prv           movl            r5,r0r         blbs            r0,200$          rete 200$: "         $create         fab=db_fab         blbs            r0,210$r         retm 210$:o0         movw            db_fab+fab$l_stv,db_chan         rets  
         .pagem         .sbttl  Read Buff   (         define_service maher$buff_read,2           clrq            r8F         movl            4(ap),r8                       ; Check VBN > 0         blequ           98$rH         cmpl            #max_vbn,r8                    ;       and < ALQ         blssu           98$e         clrq            r9          movl            8(ap),r9F         ifnowrt         #block_len,(r9),99$            ; Can buffer be written          brb             100$  G 98$:    movzwl          #ss$_badparam,r0               ; Indicate dodgy- argument         ret   H 99$:    movzwl          #ss$_accvio,r0                 ; Indicate access	 violationR         ret.   100$:   decl            r8L         evax_mulq       #block_len,r8,r6               ; Get offset to block-         evax_addq       r6,bufobj_start_va,r7e,         movc3           #block_len,(r7),(r9)  &         movzwl          #ss$_normal,r0         ret   
         .pagei         .sbttl  Write Buff  )         define_service maher$buff_write,2            clrq            r8F         movl            4(ap),r8                       ; Check VBN > 0         blequ           98$tH         cmpl            #max_vbn,r8                    ;       and < ALQ         blssu           98$          clrq            r9          movl            8(ap),r9K         ifnord          #block_len,(r9),99$            ; Can buffer be read          brb             100$  G 98$:    movzwl          #ss$_badparam,r0               ; Indicate dodgyo argument         reta  H 99$:    movzwl          #ss$_accvio,r0                 ; Indicate access	 violationh         reti   100$:   decl            r8L         evax_mulq       #block_len,r8,r6               ; Get offset to block-         evax_addq       r6,bufobj_start_va,r7 ,         movc3           #block_len,(r9),(r7)  &         movzwl          #ss$_normal,r0         rete  
         .pagey         .sbttl  Read Databasee  &         define_service maher$db_read,2           clrq            r8F         movl            4(ap),r8                       ; Check VBN > 0         blequ           98$oH         cmpl            #max_vbn,r8                    ;       and < ALQ         blssu           98$_         clrq            r9          movl            8(ap),r9F         ifnowrt         #block_len,(r9),99$            ; Can buffer be written          brb             100$  G 98$:    movzwl          #ss$_badparam,r0               ; Indicate dodgyn argument         rets  H 99$:    movzwl          #ss$_accvio,r0                 ; Indicate access	 violation          ret   ! 100$:   evax_subq       r8,#1,r10-L         evax_mulq       #block_len,r10,r6              ; Get offset to block-         evax_addq       r6,bufobj_start_va,r7a           $setup_call64   6n         $push_arg64     r8"         $push_arg64     #block_len         $push_arg64     r7+         evax_ldq        r5,iosaobj_start_va-,         evax_addq       r5,#iosa$k_length,r5         $push_arg64     r5         $push_arg64     db_chanc#         $push_arg64     read_fandleq'         $call64         sys$io_performwc           blbc            r0,999$0,         movl            iosa$l_status(r5),r0         blbc            r0,999$   ,         movc3           #block_len,(r7),(r9)&         movzwl          #ss$_normal,r0 999$:n         reta  
         .page          .sbttl  Write Database  '         define_service maher$db_write,2            clrq            r8F         movl            4(ap),r8                       ; Check VBN > 0         blequ           98$ H         cmpl            #max_vbn,r8                    ;       and < ALQ         blssu           98$y         clrq            r9          movl            8(ap),r9K         ifnord          #block_len,(r9),99$            ; Can buffer be reade         brb             100$  G 98$:    movzwl          #ss$_badparam,r0               ; Indicate dodgye argument         reta  H 99$:    movzwl          #ss$_accvio,r0                 ; Indicate access	 violatione         reta  ! 100$:   evax_subq       r8,#1,r10nL         evax_mulq       #block_len,r10,r6              ; Get offset to block-         evax_addq       r6,bufobj_start_va,r7c,         movc3           #block_len,(r9),(r7)           $setup_call64   6o         $push_arg64     r8"         $push_arg64     #block_len         $push_arg64     r7+         evax_ldq        r5,iosaobj_start_vat         $push_arg64     r5         $push_arg64     db_chanH$         $push_arg64     write_fandle'         $call64         sys$io_performw            blbc            r0,999$s,         movl            iosa$l_status(r5),r0         blbc            r0,999$  999$:          rets  
         .pagen!         .sbttl  Flush All Buffersy  %         define_service maher$db_flush            $setup_call64   6e!         $push_arg64     flush_vbnu%         $push_arg64     bufobj_lengthS'         $push_arg64     bufobj_start_va +         evax_ldq        r5,iosaobj_start_var         $push_arg64     r5         $push_arg64     db_chanf$         $push_arg64     write_fandle'         $call64         sys$io_performwf           blbc            r0,99$,         movl            iosa$l_status(r5),r0         blbc            r0,99$ 99$:         retn  
         .pages         .sbttl  Close Database  %         define_service maher$db_closeb           $setup_call64   1l#         $push_arg64     read_fandlew&         $call64         sys$io_cleanup         blbc            r0,99$           $setup_call64   1 $         $push_arg64     write_fandle&         $call64         sys$io_cleanup         blbc            r0,99$           $delete_bufobj_s -3                         buffer_handle=buffer_handler         blbc            r0,99$           $delete_bufobj_s -1                         buffer_handle=iosa_handle6         blbc            r0,99$  $         $dassgn_s       chan=db_chan         blbc            r0,99$           $setup_call64   4 (         movaq           region_length,r6         $push_arg64     r6*         movaq           region_start_va,r6         $push_arg64     r6#         $push_arg64     #psl$c_exect$         movaq           region_id,r6         $push_arg64     r6,         $call64         sys$delete_region_64           blbc            r0,99$         clrw            db_chan- 99$:         ret   
 exec_rundown:H  I         .jsb_entry                              ; Entry point for rundown-9                                                 ; handlerB         tstw            db_chan K ;        beqlu           99$                    ; Removed to prevent *BOOM*.K                                                 ; Must clean-up EVERY time!s$         $cancel_s       chan=db_chan           $setup_call64   1 #         $push_arg64     read_fandlel&         $call64         sys$io_cleanup           $setup_call64   1 $         $push_arg64     write_fandle&         $call64         sys$io_cleanup           $delete_bufobj_s -3                         buffer_handle=buffer_handled           $delete_bufobj_s -1                         buffer_handle=iosa_handlen  $         $dassgn_s       chan=db_chan           $setup_call64   4 (         movaq           region_length,r6         $push_arg64     r6*         movaq           region_start_va,r6         $push_arg64     r6#         $push_arg64     #psl$c_exect$         movaq           region_id,r6         $push_arg64     r6,         $call64         sys$delete_region_64  D ;        pushal          abnormal_msg_desc      ; For the moment allB ;        calls           #1,g^lib$put_output    ; roads lead here. 99$:         rsbu  
         .PAGEd)         .SBTTL  Privileged Library Vector    ;+K ; Any psect with the VEC attribute will be automatically moved to the startu
 of the image.  ;-6         .psect  dickie$services,page,vec,pic,nowrt,exe  L         .long    plv$c_typ_cmod          ; Set type of vector to change mode
 dispatcher3         .long    0                       ; ReservedcD         .long    kernel_routine_count    ; # of Kernel mode routinesG         .long    exec_routine_count      ; # of Executive mode routinesI>         .address kernel_table            ; Kernel routine list<         .address exec_table              ; Exec routine listA         .long    0                       ; Kernel rundown handler A         .address exec_rundown            ; Exec   rundown handlerc9         .long    0                       ; RMS Dispatcher ?         .long    0                       ; Kernel routine flags =         .long    0                       ; Exec routine flagse           .end $!& $ macro/list/enable=quad big_buffs.mar $! $ link  /share=big_buffs -         /sysexe -e         /map -         /cross -         /full -t         /notrace -         /section_binding -         big_buffs.obj, -         sys$input:/options   gsmatch=lequal,2,0  * symbol_vector = (maher$db_open=procedure,--                  maher$buff_write=procedure,-U,                  maher$buff_read=procedure,-+                  maher$db_write=procedure,- *                  maher$db_read=procedure,-+                  maher$db_flush=procedure,-s*                  maher$db_close=procedure)   protect=yesr collect=safe,_maher$data   $!+ $copy/log big_buffs.exe sys$common:[syslib]  $! $install:==$install/commanda8 $if f$file_attributes("sys$share:big_buffs.exe","KNOWN") $thene/ $       install replace sys$share:big_buffs.exeo $else F $       install add sys$share:big_buffs.exe /open/header/share/protect $!C $! If you have your GH_RSRVPGCNT SYSGEN parameter geared up for it,-2 $! you can install maher$share as /RESIDENT as in: $!B $!      install add sys$share:big_buffs.exe /open/protect/resident $endif $! $purge sys$share:big_buffs.exe3 $set file/protection=(w:re) sys$share:big_buffs.exey $! $create test_buffs.cob identification division., program-id.    test_buffs with ident "V2.0". *o data division. working-storage section.K 01  rms$_normal         pic 9(9)        comp    value external rms$_normal. L 01  rms$_created        pic 9(9)        comp    value external rms$_created.J 01  ss$_normal          pic 9(9)        comp    value external ss$_normal.- 01  sys_status          pic 9(9)        comp.h *e- 01  user_opt            pic 9(4)        comp.  01  my_record.-     03  rec_vers        pic 9(9)        comp.a-     03  rec_len         pic 9(4)        comp.c#     03  rec_data        pic x(506).t" 01  in_data             pic x(60).- 01  db_key              pic 9(9)        comp.i *  procedure division.e kick_off section.  00. 2     display "Initializing. . ." bold erase screen.+     call "maher$db_open" giving sys_status.r4     if sys_status not = rms$_normal and rms$_created2         call "lib$stop" using by value sys_status.       perform display_menu. '     perform test_it until user_opt = 0.,  ,     call "maher$db_close" giving sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by valuec sys_status.t  )     display "Good-Bye. . ." erase screen.M
     stop run.b *  test_it section. 00.      if user_opt <= 5         go to   write_buff,r                 write_db,                  read_buff,                 read_db,6                 flush_db        depending on user_opt.       go to fini.u *h write_buff.y     perform get_data.      if rec_vers = zerost<         display "Inserting New Record" erase to end of line.     add 1 to rec_vers.       call "maher$buff_write"t&         using   by value        db_key)                 by reference    my_recorda         giving  sys_status.eA     if sys_status not = ss$_normal call "lib$stop" using by valuen sys_status.-       go to fini.t *r	 write_db.c     perform get_data.i     if rec_vers = zerosi<         display "Inserting New Record" erase to end of line.     add 1 to rec_vers.       call "maher$db_write"/&         using   by value        db_key)                 by reference    my_record-         giving  sys_status. A     if sys_status not = ss$_normal call "lib$stop" using by value  sys_status.m       go to fini.. *>
 read_buff.     perform get_data.      if rec_vers = zeros.5         display "No such record" erase to end of line>         display " " erase line.        go to fini.  *  read_db.H     display "Enter DBKEY 1 to 2048: " erase to end of line no advancing.5     accept db_key with conversion protected reversed.i       call "maher$db_read"&         using   by value        db_key)                 by reference    my_recordu         giving  sys_status.cA     if sys_status not = ss$_normal call "lib$stop" using by valuel sys_status.e       if rec_vers = zerosy5         display "No such record" erase to end of linel         display " " erase line     elseL         display "Current version is ", rec_vers with conversion erase to end of line L         display "Data is [", rec_data (1:rec_len), "]" erase to end of line.       go to fini.p *M	 flush_db.:,     call "maher$db_flush" giving sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by valuer sys_status. ?     display "All buffers are now on disk" erase to end of line.t     display " " erase line.t *  fini.3     perform display_menu.D *4 display_menu section.e 00. $     display " " at line 10 column 1.A     display "*** SYSTEM GLOBAL SECTION & FAST I/O DEMO ***" bold.t     display " ".     display "    (0) Exit".i(     display "    (1) Write Buffer Only".,     display "    (2) Write Buffer and Disk".(     display "    (3) Read  Buffer Only".,     display "    (4) Read  Buffer and Disk".+     display "    (5) Flush Global Section".e     display " ".3     display "    Enter Option: " bold no advancing.        accept      user_opt         with    conversion                 protectedd                 reversed7                 at end          move zeros to user_opt.  *n get_data section.  00.MH     display "Enter DBKEY 1 to 2048: " erase to end of line no advancing.5     accept db_key with conversion protected reversed.R     call "maher$buff_read"&         using   by value        db_key)                 by reference    my_record          giving  sys_status. A     if sys_status not = ss$_normal call "lib$stop" using by value) sys_status.d       if rec_vers not = zeros-L         display "Current version is ", rec_vers with conversion erase to end of line L         display "Data is [", rec_data (1:rec_len), "]" erase to end of line.       if user_opt = 1 or 2C         display "Enter New Data " no advancing erase to end of linei)         accept in_data reversed protectedi         call "str$trim"09             using       by descriptor   rec_data, in_data)/                         by reference    rec_len #             giving      sys_status.  *0 end program test_buffs.  $! $cobol/lis test_buffs.cobe5 $link/exe=test_buffs.exe test_buffs.obj,sys$input/opt    sys$library:big_buffs.exe/shareb   $!* $set file/protection=(w:re) test_buffs.exe $exito $!	 $no_priv: K $       write sys$output "Insufficient privilege. You need (CMKRNL,SYSPRV)"  $       exit 44a $no_vax:8 $       write sys$output "This code only works on alpha" $       exit 44-  / RC Bryan <rcbryan@hotmail.com> wrote in message:7 news:fbcf38dc.0301141521.56c85e33@posting.google.com...u@ > "Richard Maher" <maher_rj@hotspamnotmail.com> wrote in message, news:<b01kn2$cav$1@sparta.btinternet.com>... > > Hi,n > >mL > > Does Fast I/O (and whatever device restrictions go with it) interest you at2 > > all? "buflen" says it can be the full 64-bits. > >tH > > Although the documentation says Alpha only, somewhere I saw that, on VAX,K > > the single $io call was translated automagically into multiple $qios. I  > > don't know if it's true. > >-. > > I'll post an example if you're interested. > >e > > Regards Richard Maher. > E > I never heard of Fast I/O before.  I guess I overlooked it since it F > does not appear in the paper V6.0 manuals I have on my desk.  If you: > have an example, I would be interested in looking at it. >rE > I am not sure I am going to replace the $QIOs I have in the program-C > currently.  The application is in the field in two or three dozen.F > locations and if I screw something up, they are going to hang me out	 > to dry._ >T
 > Regards, > /RC Bryans   ------------------------------  + Date: Thu, 16 Jan 2003 06:55:25 +0000 (UTC)@3 From: "Richard Maher" <maher_rj@hotspamnotmail.com>rS Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK? 1 Message-ID: <b05l0s$7lt$1@knossos.btinternet.com>t  & This is the other part of the example:  @ (It's VLM *and* fast i/o but the two do go rather well together)   $!A $! Before running RMU_INIT_GB you should have done the following:e $!
 $! $mc sysmanu+ $! SYSMAN> reserved_memory add "MY_BUFFS" -e8 $!              /allocate/page_tables/zero/size=1/sysgbl $!C $! Then autogen and reboot a couple of times. This will effectivelyeA $! remove 1MB of memory from your system and create the in-memoryaB $! database MY_BUFFS. Do $SHOW MEMORY after reboot to see results. $! $ on warning then exit: $ if .not. f$privilege("cmkrnl,sysprv")  then goto no_privC $ set rights/enable vms$mem_resident_user/attr=(subsystem,resource)- $!9 $ if f$getsyi("arch_name") .nes. "Alpha" then goto no_vaxn $! $ create rmu_init_gb.mar ;++- ;-4 ; (c) Copyright Tier3 Software. All rights reserved. ;0@ ;     Ownership of this software and all associated intellectual@ ;     property rights remain vested in Tier3 Software Ltd.  This@ ;     software  or any other copies thereof  may not be provided6 ;     or otherwise made available to any other person. ;o* ;     Do not remove this copyright notice. ;e ;     Author: Richard Maher  ;d ;-- I         .title  rmu_init_gb - RMU/INITIALIZE GLOBAL_BUFFER root_file_specs         .ident  "V1.0"  &         .library "sys$library:lib.mlb"           $vadef         $psldef1         $ssdef         $syidef          $secdefh           one_megabyte=^x100000p         no_reg=44m  C         .psect  _maher$data,pic,con,rel,lcl,noshr,noexe,rd,wrt,quads  ) arg_list_1:             .long           3 4                         .long           sec$m_sysgbl9                         .address        section_name_descn7                         .address        wild_section_ido  ) arg_list_2:             .long           8o9                         .address        section_name_desc 2                         .address        section_id.                         .long           ^xfff04                         .long           one_megabyte2                         .long           psl$c_execJ sec_flags:              .long           <sec$m_dzro!sec$m_gbl!sec$m_perm -K                                         !sec$m_sysgbl!sec$m_mres!sec$m_wrt>o4                         .address        reserved_len)                         .long           0o  8 section_name_desc:      .long           section_name_len4                         .address        section_name4 section_id:             .long           sec$k_matequ) section_version:        .long           0s7 wild_section_id:        .long           sec$k_matall, 0 = syi_itmlst:             .word           4, syi$_galaxy_member 5                         .address        galaxy_member ,                         .long           0, 0) syi_iosb:               .quad           0c) galaxy_member:          .long           0F  ) reserved_len:           .blkq           1   3 fao_ctl_desc:           .long           fao_ctl_len /                         .address        fao_ctl 6 fmt_string_desc:        .long           fmt_string_len2                         .address        fmt_string) out_string_len:         .long           012                         .address        fmt_string5 nosec_desc:             .long           nosec_msg_lenl1                         .address        nosec_msgb  D fao_ctl:                .ascii          "Global Section !AS Created, Registry Size = !@ZQ"  fao_ctl_len=.-fao_ctloF nosec_msg:              .ascii          "No existing Global Section to delete. . ." nosec_msg_len=.-nosec_msg 2 section_name:           .ascii          "MY_BUFFS" section_name_len=.-section_name + fmt_string:             .blkb           255  fmt_string_len=.-fmt_stringd  A         .psect  _maher$code,pic,con,rel,lcl,shr,exe,rd,nowrt,quadl  $         .entry          init_gb,^m<>  +         $getsyiw_s      itmlst=syi_itmlst,-b%                         iosb=syi_iosb          blbc            r0,fini_#         movzwl          syi_iosb,r0          blbc            r0,fini   %         tstl            galaxy_member          beqlu           old_sec .         bisl2           #sec$m_shmgs,sec_flags ;+< ; Only one version of the Global Section is valid at a time.8 ; Obviously the database must be closed on this node and7 ; whatever exclusive locks required have been obtained.  ;- old_sec:+         $cmexec_s       routin=sys$dgblsc,-t)                         arglst=arg_list_1l  "         blbs            r0,new_sec)         cmpw            #ss$_nosuchsec,r0v         bnequ           fini  "         pushal          nosec_desc+         calls           #1,g^lib$put_outputb         blbc            r0,fini    new_sec:J         movzbl          #1,section_version              ;Version read from root  1         $cmexec_s       routin=sys$create_gdzro,- )                         arglst=arg_list_2_         blbc            r0,fini   .         $fao_s          ctrstr=fao_ctl_desc, -0                         outlen=out_string_len, -1                         outbuf=fmt_string_desc, - 0                         p1=#section_name_desc, -(                         p2=#reserved_len         blbc            r0,fini   &         pushal          out_string_len+         calls           #1,g^lib$put_outputr         blbc            r0,fini   '         movq            reserved_len,r1i         bnequ           fini"         movzwl          #no_reg,r0 fini:          $exit_s         code=r0:           .end            init_gb!   $! $ macro/list rmu_init_gb.mar $! $ link  /exe=rmu_init_gb.exe -         /notrace -         /section_binding -         rmu_init_gb.obj  $!. $ copy/log rmu_init_gb.exe sys$common:[sysexe] $!C $! This way you won't have to give the identifier to everyone. Only ( $! enable temporarily while running RMU. $!+ $ set security sys$system:rmu_init_gb.exe -r  G /acl=(subsystem,identifier=vms$mem_resident_user,attributes=resource) -S         /protection=(w:re) $! $install:==$install/commanda; $if f$file_attributes("sys$system:rmu_init_gb.exe","KNOWN")s $thenn2 $       install replace sys$system:rmu_init_gb.exe $elsef0 $       install add sys$system:rmu_init_gb.exe -E                 /open/header/share/priv=(sysgbl,prmgbl,cmexec,sysprv)= $!C $! If you have your GH_RSRVPGCNT SYSGEN parameter geared up for it, + $! you can install rmu_init_gb as /RESIDENT/ $! $endif $!! $purge sys$system:rmu_init_gb.exe  $! $exitI $!	 $no_priv:@K $       write sys$output "Insufficient privilege. You need (CMKRNL,SYSPRV)"5 $       exit 44n $no_vax:8 $       write sys$output "This code only works on alpha" $       exit 44e   ------------------------------  % Date: Wed, 15 Jan 2003 16:44:34 -0500_* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: HP Sets the Stage for Alpha's Last Hurrah2 Message-ID: <o3qdnUR6r5rOS7ijXTWcqg@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:UjZoKfVaE$kr@eisner.encompasserve.org...n@ > In article <A86cnauXEu3By7mjXTWcqg@metrocast.net>, "Bill Todd"  <billtodd@metrocast.net> writes: > >e< > > "Rob Young" <young_r@encompasserve.org> wrote in message1 > > news:2jhv5O8PYazj@eisner.encompasserve.org...  > >  > > ..._ > > # > >> Look, at least it is shipping.  > >n/ > > Always one to look on the bright side, Rob.f > >n$ > >   Go back and read some opinions > >> a year or two ago.l > >:K > > You should go back and do so yourself, since you clearly don't remembers them > > correctly. > >  > 
 > Sure do.  D Well, since you weren't able to produce a single piece of supporting; evidence to that effect below, I'd say you're full of shit.    > 6 > >>  Several thought it wouldn't ship.  What a crock. > > H > > Wrong, Rob:  several people thought *that there was a non-negligible chance > > that EV7 wouldn't ship*n > : > And I suppose no one said it won't ship?  Of course they > did.  L Where?  You're so fond of saying "Google is your friend", and if anyone said/ it (here) it'll be right there for the finding.   H But remember, you need to find *several* people who said they thought it. wouldn't ship, since that's what you asserted.   >nH > However, in repeated attempts to get you to say so *definitively*, ...B > we hear terms of possibly, probably , maybe ... I didn't get theF > sweet lift quote I was after, because you weren't about to say that.  J Of course I wasn't about to say that, because that wasn't my opinion.  TheK opinion I *did* have is exactly the one you reproduce below, is stated with H complete clarity, and, of course, does not reflect your assertion above.  3 > However, Terry stated definitively it would ship.   L Certitude is like faith:  often of comfort to the holder, but not worth warm spit to anyone else.   > D > Like Gartner, the best you would do is assign a probability factor > (FUD factor) to it:m >v >,L http://groups.google.com/groups?selm=9nb577%24ofg%241%40pyrite.mv.net&output =gplain  > ) > From: "Bill Todd" <billtodd@foo.mv.com>r& > Newsgroups: comp.os.vms,comp.sys.dec# > Subject: Re: EV7 will never ship? & > Date: Fri, 7 Sep 2001 14:56:33 -0400 >  > > A > > So.. how about this for a statement that you can sign off on:  > >p" > > "I highly doubt EV7 will ship" > > 3 > > Is that your opinion?  We are entitled to them.  >oJ > And you shall have them, in great copiousness and detail, merely for the	 > asking:h >dL > My opinion is that there is sufficient probability (SWAG 30%, but it couldJ > be half or double that) that EV7 will never ship (due to HP's commitment toI > a weak Itanic product that EV7 would make look even weaker coupled with K > demand-erosion in the Alpha customer base over the next year) that people J > need to take this into account in any decisions they might be making forI > anything other than very-short-term use of Alpha and the systems on it.  >  > ---i >l4 > "Sufficient probability that EV7 will never ship." >:, > Sufficient for what?  FUD purposes?  Okay. > > > Like Gartner, and (mentioned elsewhere in that thread) - Our= > British Champion - you were cagey by tip-toeing and statingmE > generalities.  Sounds good and loud, but in this case mostly wrong.   L Rob, you really are an idiot.  If you read the statement above, it indicatesI that my guess was that EV7 *would* ship (with a mean probability of 70% -eK over 2 chances in 3) - and, by George, that's just what happened.  However, J I felt that there was sufficient possibility that it would *not* ship thatL people should factor that possibility into their planning, and the purportedJ insider observation to that effect in the recent Inquirer article suggestsE that I wasn't being unduly unduly alarmist in that suggestion either.a  ? You will also note that my suggestion that HP would find EV7 anoG embarrassment to Itanic seems to be panning out as well (did the silent6' launch actually take place yesterday?).   : > But not totally!  Remember, we were talking probability! > E > Any predictions for Power5?  I say it will ship.  Any probabilitiese? > you would care to assign to it or would you be definitive and > > go out on a limb and say it will ship?  Remember, just after+ > an opinion.  We are entitled to them. ;-)   G Wrong:  I'm entitled to *give* them, if I choose (as I did w.r.t. EV7).rK Since I have considerably less interest in (and knowledge of) the situation J with POWER5 than I had with Alpha, and since I know of no reason to assumeK it might be in anything like the jeopardy that Alpha futures were and still , are, I expect, as you do, that it will ship.   - bill   ------------------------------  % Date: Wed, 15 Jan 2003 21:07:35 -0500  From: "ICUSC" <sales@hpaq.net>" Subject: Incoming !  Ds20e Systems/ Message-ID: <v2c4phc1ajmja8@news.supernews.com>   - We are the only Dealer with DS20e's incoming.o  F Call or email sales@hpaq.net  with your configuration requirements !!!  ( Pricing for DS20e starting at only $4995  A The best prices and of course the highest quality... guaranteed !    Island Computers US Corp.  2700 Gregory St, Ste 180 Savannah GA 31404, USA Tel: (00) 1 912 447 6622 Fax: (00) 1 912 201 0402 sales@hpaq.net www.islandco.com   ------------------------------  # Date: Wed, 15 Jan 2003 18:46:37 GMTr" From: Alfred Falk <falk@arc.ab.ca>= Subject: Re: incomprehensible mountVerifyTimeout in a cluster69 Message-ID: <Xns930477CCD5292falkarcabca@205.233.108.180>   / Pierre.Bru@spotimage.fr (Bru, Pierre) wrote in  5 news:a39f53b1.0301150656.43ec4649@posting.google.com:    > hello, > 0 > I have 3 alpha in a cluster (let say A, B & C)5 > A mount disks of B and C. sometimes, these disks go  > mountverifytimeout.c > H > as MVTIMEOUT=3600 and RECNXINTERVAL=120 on the 3 alpha, I do not thinkG > the problem is a networks pb as none of the alphas exist the cluster._ > E > what if not a network pb can make the disks go mountVerifyTimeout ?   E You might try increasing MSCP_CREDITS in SYSGEN.  Unfortunately this  + will require re-boots to do the experiment.0   --  @ ----------------------------------------------------------------A   A L B E R T A         Alfred Falk               falk@arc.ab.ca $@ R E S E A R C H         Information Systems Dept   (780)450-5185+   C O U N C I L         250 Karl Clark Roada1                         Edmonton, Alberta, Canada  http://www.arc.ab.ca/   T6N 1E4   http://www.arc.ab.ca/staff/falk/   ------------------------------    Date: 15 Jan 2003 14:28:03 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)p= Subject: Re: incomprehensible mountVerifyTimeout in a cluster 3 Message-ID: <WxxlxIFh3fqj@eisner.encompasserve.org>   k In article <a39f53b1.0301150656.43ec4649@posting.google.com>, Pierre.Bru@spotimage.fr (Bru, Pierre) writes:  > hello, > 0 > I have 3 alpha in a cluster (let say A, B & C)5 > A mount disks of B and C. sometimes, these disks go  > mountverifytimeout.6 > H > as MVTIMEOUT=3600 and RECNXINTERVAL=120 on the 3 alpha, I do not thinkG > the problem is a networks pb as none of the alphas exist the cluster.y > E > what if not a network pb can make the disks go mountVerifyTimeout ?l  C    Was one of the systems down for a while?  I've seen this on some D    cluster members when looking at the disks served by a member thatA    was shut down for a while.  I can't recall if there's a simple B    method to get the mount verify to retry after it has timed out.   ------------------------------  % Date: Wed, 15 Jan 2003 16:27:15 -0500  From: norm.raphael@metso.com= Subject: Re: incomprehensible mountVerifyTimeout in a cluster ? Message-ID: <OF0942D6A8.5CDB4919-ON85256CAF.00759766@metso.com>   I MVTIMEOUT=3600 is 1 hour.  After that timeout occurs, then the disks must  be! DISMOUNT/ABORT and MOUNTed again.eK I do not understand "a networks pb as none of the alphas exist the cluster.t "t? after "I have 3 alpha in a cluster", so I cannot comment on the  configuration.  J From:  koehler@eisner.nospam.encompasserve.org (Bob Koehler) on 01/15/2003        03:28 PMp  G Please respond to koehler@eisner.nospam.encompasserve.org (Bob Koehler)    To:    Info-VAX@Mvb.Saic.Com cc:   @ Subject:    Re: incomprehensible mountVerifyTimeout in a cluster    = In article <a39f53b1.0301150656.43ec4649@posting.google.com>, - Pierre.Bru@spotimage.fr (Bru, Pierre) writes:  > hello, > 0 > I have 3 alpha in a cluster (let say A, B & C)5 > A mount disks of B and C. sometimes, these disks gog > mountverifytimeout.  >cH > as MVTIMEOUT=3600 and RECNXINTERVAL=120 on the 3 alpha, I do not thinkG > the problem is a networks pb as none of the alphas exist the cluster.  > E > what if not a network pb can make the disks go mountVerifyTimeout ?   C    Was one of the systems down for a while?  I've seen this on somesD    cluster members when looking at the disks served by a member thatA    was shut down for a while.  I can't recall if there's a simple#B    method to get the mount verify to retry after it has timed out.   ------------------------------  # Date: Thu, 16 Jan 2003 03:40:14 GMT # From: "John Smith" <a@nonymous.com> : Subject: It's 'open source', so it must be good.....right?J Message-ID: <yQpV9.348623$F2h1.89233@news01.bloor.is.net.cable.rogers.com>  F http://news.yahoo.com/news?tmpl=story2&ncid=1209&e=9&u=/nm/20030115/tc- _nm/tech_microsoft_government_dc&sid=95573713   2 Microsoft Opens Guarded Source Code to Governments   Tue Jan 14,10:58 PM ET By Reed Stevenson   E SEATTLE (Reuters) - In an unprecedented move, Microsoft Corp. said on B Tuesday that it would open its source code -- the software giant's= closely guarded blueprints for programs -- to governments andC& international organizations worldwide.  A Under the initiative, called the Government Security Program, the B world's largest software maker said governments and their agenciesC would be able to examine its source code to enhance the security of D their software, used for tasks such as tracking personal data, taxes and ensuring national security.e  E The move is aimed at strengthening Microsoft's position in government D markets where it is already the largest or second-largest government supplier of software.   = "We have a business interest in having people feel completelymF comfortable with our software, whether it is mission critical or not,"A Craig Mundie, Microsoft's Chief Technology Officer, told Reuters.   D The NATO defense alliance, as well as Russia, have already agreed toC participate in the program, and discussions with more than 60 other9E governments and agencies are being held, Mundie said, adding that any(B cooperation with Microsoft would be disclosed at the discretion of each government agency.   B Microsoft is facing pressure from free software, which is becoming@ ever more tempting to governments worldwide as an alternative to? costly proprietary software from Microsoft and other companies.   F Since free software, such as Linux, is "open source" and by definition< available for close scrutiny, its advocates argue that it is inherently more secure.o  @ Last June, Microsoft rival International Business Machines Corp.? signed a deal with the German government to provide Linux-based  systems to some agencies.d  > Open source software is also appealing to some governments andC companies because it is free, and can be copied and modified unlike + Microsoft's Windows and its other programs.   B "They're trying to block the entry of competing platforms by theseB governments," said Rob Enderle of Giga Information Group Inc.. "IfF they can get governments to feel comfortable with their software, thenF they can effectively hold off the influx of those competing platforms.. It can pretty much lock you in as a standard."     ONGOING CAMPAIGNA Up to now, many governments developed their own software to store C information and run their agencies, but the mounting complexity and 3 cost has pushed them toward off-the-shelf programs.b  ? To woo governments, Microsoft said it will make its source codetD available to government agencies over the Internet, provided they doE not disclose that code. They will also be required to sign contracts,j$ but will not have to pay for access.  E Opening its software blueprints is not a new idea from Microsoft, butb* the scope of the government disclosure is.  A In 2000, Microsoft began its Shared Source Initiative to open its B source code to other companies as well as research and educationalF institutions. The program, which was later expanded to include a widerD range of technology companies, allows Microsoft's corporate partners= to refer to Windows code, but not change and redistribute it.e  C "We also plan to provide source access, technical documentation and C resources that we generally have never made available before," said  Mundie.   D Microsoft said it will also make its software engineers available atF its Redmond, Washington headquarters for consultations with government representatives.  @ But questions will always remain over the new initiative becauseF governments have no way of being sure that the source code they see isF the blueprint used to run software, said Dan Kusnetzky, Vice PresidentD of systems software at Framingham, Massachusetts-based research firm International Data Corp.  A That's because source code is compiled, or turned into electronicbF language that only machines can understand. Inner workings of compiledB programs are extremely difficult to decipher without access to the underlying code, he said.6  C Microsoft, which has gotten mixed reviews for its efforts to ensure F security, has also been trying hard to convince existing and potential? customers that its software is more trustworthy. Early in 2002, : Microsoft's co-founder and Chairman Bill Gates announced a> company-wide initiative to improve the security of Microsoft's	 products.    ------------------------------    Date: 16 Jan 2003 00:27:43 +0800, From: Paul Repacholi <prep@prep.synonet.com>0 Subject: Re: Looks like AMD has deep pockets now- Message-ID: <873cnupedc.fsf@prep.synonet.com>   2 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  E > Question: to produce something this small, would they need to build ? > a new FAB plant, or could they retrofit existing fab plant top > produce such chips ?  F Mix of the two. First problem is that the tools for exposing the masksC for 65nm are not out there yet. Second is the steppers and the likelF will posibly be not up to the task at 65nm, so they will be downgradedC as well. The positive detail, is that the designs of the plants haseB changed so they don't have to junk the (almost) entire building toF upgrade. The sections of the FAB are often built into a self-containedE sealed unit that can be replaced, and is not trashed when its neibour  is ripped out!   -- c< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.f@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 15 Jan 2003 19:08:37 -0500l  From: John Santos <JOHN@egh.com>$ Subject: Re: Microvax Floating point4 Message-ID: <1030115190631.631C-100000@Ives.egh.com>  / On Wed, 15 Jan 2003, Brian 'Jarai' Chase wrote:n  L > In article <3E2307D8.6030204@vajhoej.dk>, Arne Vajh=F8j  <arne@vajhoej.dk= > wrote: > > Brian 'Jarai' Chase wrote:L > > > In article <3E214480.2020206@vajhoej.dk>, Arne Vajh=F8j <arne@vajhoej= .dk> wrote: " > > > > Brian 'Jarai' Chase wrote: >=20J > > > > > The most sensible option, IMO, for supporting IEEE FP on the VAXL > > > > > would be to create a software library against which apps requirin= go$ > > > > > IEEE FP could be compiled. > > > >  > > > > ???? > > > > B > > > > And write a compiler that generate calls to that library ? > > > L > > > The apps in question would have to be linked against the library, usi= ng0 > > > the IEEE compatible functions it provides. > >=20? > > The problem is that there are no library calls - it is justp > > instructions._ > >=20C > > To replace those instructions with library calls will require ad > > new compiler.  >=20K > Ah.  I'm not suggesting that the instructions be replaced.  For one, it's_L > impossible to replace the microcoded instructions on any but the oldest o= f G > the VAXen.  And though it's possible that one /could/ trap say... the H > MACRO coded H_float instructions and reserved opcodes, bending them toK > one's will, that would require modifying the operating system in additionyF > to the compilers.  Also, I'm not familiar enough with VMS to know ifH > that's even possible for end users.  It would be possible on something> > like NetBSD/vax since the the full source code is available. >=20F > And even given the above, you'd still have to implement some sort ofI > library function to translate persistent VAX FP formated data into IEEEiI > format data.  For example, lets say you've got FP values stored in some J > file.  If they're VAX FP formatted values, you'll have to translate themF > to IEEE before your new instructions can work on them.  Much, as I'dI > imagine, people have had to translate IEEE formated FP values to VAX FPo( > format to process them on their VAXen.  C This part already exists.  See CVT$CONVERT_FLOAT and CVT$FTOF under_ "$ HELP RTL CVT$"; =20 F > My suggestion would be to not even attempt to replace or augment theL > native VAX opcodes.  Leave the machine language completely alone; instead= , L > build a nice math library that operates on IEEE FP formatted data, and us= e L > its functions in your applications where IEEE FP manipulation is needed. = =20EL > It'll be slow, but at least it won't break everything else that depends o= n!	 > VAX FP.s >=20	 > -brian.    --=20s John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Wed, 15 Jan 2003 18:47:37 -0600 . From: Lyndon Bartels <lbartels@pressenter.com>+ Subject: Re: New Marvel machines? New news? - Message-ID: <3E25ACC9.84CF1AF@pressenter.com>    JF Mezei wrote:b > H > I can see a scenario where you might get all the information about theK > wildfire/marvel from a sales humanoid who would then tell you that if yousM > order on the web, you get an extra X5 off and faster delivery because ordere( > goes tdirect to manufacturing etc etc.    E I can remember a seen in the Stallone movie "Demolition Man." Where asB Police Officer is answering the phone: "Welcome to the San-AngelosC Police Hot-line. If you would like an automated response, press '1'f now."o   -- dG My opinions are mine and mine alone. They seldom align with those of mye	 employer.     H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Wed, 15 Jan 2003 18:55:26 -0600c. From: Lyndon Bartels <lbartels@pressenter.com>+ Subject: Re: New Marvel machines? New news?r. Message-ID: <3E25AE9E.138C7D45@pressenter.com>   Larry Kilgallen wrote: >  > In article <rdeininger-0301030906200001@user-2ive194.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:p@ > > In article <3E14B1BB.D664332@pressenter.com>, Lyndon Bartels$ > > <lbartels@pressenter.com> wrote: > > F > >>The last I heard, HP was supposed to announce the new EV7 (Marvel) > >>machines this month. > >  > > The month isn't over yet.  > 1 > Let's see, how can we argue with _that_ ?   :-)     3 Which month... That wasn't specified either.... :-)     3 But the computer world article say January 21st....     , But I don't think it said which year.... :-)         -- cG My opinions are mine and mine alone. They seldom align with those of myl	 employer.o    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------    Date: 16 Jan 2003 02:16:06 +0800, From: Paul Repacholi <prep@prep.synonet.com>5 Subject: Re: Newbie to OpenVMS (contrasting to Linux)l- Message-ID: <87lm1mnus9.fsf@prep.synonet.com>   "  VAXman- @SendSpamHere.ORG writes:  F > http://www.openvms.compaq.com/doc/73final/9996/9996pro_contents.html   $ > actually produces content whereas:   G > http://www.openvms.compaq.com/doc/731final/9996/9996pro_contents.html     > does not.    E > Of course, I haven't look but perhaps DCL was ripped out of V7.3-1?f1 > One more step to being Weendoze like -- no CLI.i  E Two. A big part of the reason you don't get anything out of the 7.3-1.B doc CDs is the use of a mix of upper and lower case file names and2 URLs. With NTs file rules, this does not matter...   -- i< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.t@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 15 Jan 2003 12:49:06 -0600s' From: MrCoffee <mrcoffee73@hotmail.com>"9 Subject: Re: No mention of Digital in official HP historyd* Message-ID: <3E25AD22.2020607@hotmail.com>  E It really surprises me how they can flaunt a clone maker like Compaq  I over an innovative company like Digital.  But don't worry. The HP/Compaq  K megamerge will someday flop.  No digital, but no worry about Compaq either.e   Alan Greig wrote:tH > On Wed, 15 Jan 2003 01:36:19 GMT, "John Smith" <a@nonymous.com> wrote: >  > 
 >>It figures. A >>http://www.hp.com/hpinfo/abouthp/histnfacts/timeline/index.htmlf >>G >>Arguably the most innovative computer company, Digital's achievements F >>in computing are studiously ignored in HP's corporate history, while. >>clonemaker Compaq is trumpeted to the world. >  > G > I just looked. Absolutely and utterly disgusting. The HP timeline has H > been back edited to insert Compaq puffs but none of DEC's achievements4 > are listed. No PDPs, no VAX, no Alpha, no nothing. >  >  > -- > Alan   ------------------------------  % Date: Wed, 15 Jan 2003 16:57:35 -0400g0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>9 Subject: Re: No mention of Digital in official HP historyr/ Message-ID: <3E25CB33.4DE047EB@vl.videotron.ca>D   MrCoffee wrote:s > F > It really surprises me how they can flaunt a clone maker like CompaqJ > over an innovative company like Digital.  But don't worry. The HP/CompaqM > megamerge will someday flop.  No digital, but no worry about Compaq either.)  & Does this document talk about Tandem ?  L Compaq's only achievement is having made IBM PC clones respectable. Osbornes1 had created the concept of luggables, not Compaq.     H Right now, it is sad to say, but it looks like HP has done a fair job inN buying Compaq. Seems that Gateway is the one hurting. So it might be HP versusK Gateway, until Sony start to really put on the power and crush them both ina the desktop market.   I The one thing taht puzzles me is why HP still brands some of the recentlycL announced  products with "Compaq". I can understand keepong the compaq brandI for legacy products such as proliant, deskpro and Ipaq, but not for newlyr designed products.  N As far as I am concerned, Comapq should be burried very deep and forgotten. ItK has done more harm to the computer industry in the past 5 years than it didr good in its first 15 years.e   ------------------------------  % Date: Wed, 15 Jan 2003 13:55:30 -0800l$ From: Shane Smith <ssmith@icius.com>9 Subject: RE: No mention of Digital in official HP historya0 Message-ID: <01C2BC9D.CD7242A0@sulfer.icius.com>  H I heard they decided to keep the Compaq brand name for products destinedG for the home market, and use the HP brand for business. Apparently home G users trust the Compaq name more than HP, and it's the other way aroundf in the business arena.   Shaneo   -----Original Message-----7 From: JF Mezei [mailto:jfmezei.spamnot@vl.videotron.ca]/* Sent: Wednesday, January 15, 2003 12:58 PM To: Info-VAX@Mvb.Saic.Comn9 Subject: Re: No mention of Digital in official HP history      MrCoffee wrote:R > F > It really surprises me how they can flaunt a clone maker like CompaqJ > over an innovative company like Digital.  But don't worry. The HP/CompaqM > megamerge will someday flop.  No digital, but no worry about Compaq either.M  & Does this document talk about Tandem ?  C Compaq's only achievement is having made IBM PC clones respectable.m Osbornes1 had created the concept of luggables, not Compaq.     H Right now, it is sad to say, but it looks like HP has done a fair job inG buying Compaq. Seems that Gateway is the one hurting. So it might be HPn versusH Gateway, until Sony start to really put on the power and crush them both in the desktop market.s  @ The one thing taht puzzles me is why HP still brands some of the recentlyF announced  products with "Compaq". I can understand keepong the compaq brand C for legacy products such as proliant, deskpro and Ipaq, but not fory newly  designed products.  @ As far as I am concerned, Comapq should be burried very deep and
 forgotten. IteG has done more harm to the computer industry in the past 5 years than iti did  good in its first 15 years.    ------------------------------   Date: 15 Jan 2003 23:43:07 GMT# From: hoff@hp.nospam (Hoff Hoffman) / Subject: Re: OpenVMS for AlphaStation 200 4/233 * Message-ID: <b04rmb$nnt$4@web1.cup.hp.com>  f In article <meea2v053qeiee36oi2f9sej9md8c2bnbv@4ax.com>, John Laird <john@laird-towers.org.uk> writes:D :On Wed, 15 Jan 2003 00:55:49 GMT, VAXman-  @SendSpamHere.ORG wrote: ..E :>On an AS200, Shane?  I vaguely remember trying to install V6.1 and t :>there were issues. :tB :Maybe because it needed 6.1-1H1 or -1H2 ?  Release history is at:F :http://www.openvms.compaq.com/openvms/os/openvms-release-history.html :dI :And yes, that is the URL provided by HP's own search utility, as well as! :Google.    I   This pointer is also in the OpenVMS Frequently Asked Questions (FAQ).  m  D   The FAQ also has a pointer to a chart specifically referencing theF   required minimum OpenVMS version for most of (all of?) the supported*   (and announced) VAX and Alpha platforms:  ;     http://www.openvms.compaq.com/openvms/supportchart.htmlr    N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Wed, 15 Jan 2003 20:28:35 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>/ Subject: Re: OpenVMS for AlphaStation 200 4/233 . Message-ID: <3E25FCB2.19243EF@vl.videotron.ca>   Hoff Hoffman wrote: G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com     M Wow, Hoff got a "vanity" email address at HP... Was that hard to negotiate to  get permission ?  M Are you allowed to get your emails delivered to a VMS workstation, or are you  forced to use Windows ?t   ------------------------------  % Date: Wed, 15 Jan 2003 20:40:39 +0100 1 From: SAP Trainee <Didier.Morandi.nospam@Free.fr> Y Subject: OpenVMS Hobbyist licence (was: Getting started with an AlphaStation 200 4/233) 2 & Message-ID: <3E25B937.8000401@Free.fr>  F 1. As far as I know, an AlphaStation 200 4/233 is not a VAX (but this F may have changed too since a few months) so your text is not relevant.  C 2. When I received my Hobbyist CD, it was written somewhere what I  H posted in my previous message, i.e. no commercial OpenVMS distrib. Only  the Montagar one.c  I 3. things may have changed about license management, but I was not aware   of it.   D.   David Webb a crit: \ > In article <3E259A5D.3090209@Free.fr>, SAP Trainee <Didier.Morandi.nospam@Free.fr> writes: >  >>David Webb a crit:n >> >>H >>>>>2. And can I use the hobbyist license with commercial distribution? >>>>D >>>>No, you cannot (but, really, who will care for a dying product?) >>>> >>> N >>>Yes you can. VMS is VMS is VMS. There is no "hobbyist" version of VMS it is, >>>just a particular CD distribution of VMS.Q >>>There have been many recomendations on this newsgroup from Compaq/HP employeeseP >>>for hobbyists to borrow "commercial distributions" from friends who work withQ >>>VMS if there has been a holdup in getting the hobbyist distribution or if theyeP >>>wanted a layered product which wasn't put on the hobbyist CD because of space5 >>>rsstrictions but for which a license was provided.a >>F >>Sorry. I was not clear. I should have written "you may not, but you I >>can". The OpenVMS Hobbyist licence expressively (sp?) mention that you  K >>are not allowed to use another VMS version than the one available on the   >>Hobbyist CD. >> > L > I've never used a system under the hobbyist license however looking at the8 > Montagar site I can't see anything remotely like that.; > It certainly isn't in the COMPAQ HOBBY License agreement   > see  > ; > http://www.montagar.com/hobbyist/terms_and_conditions.txt  >  > or > 8 > http://www.montagar.com/hobbyist/register_license.html > / > In fact the site explicitly tells you that :-  >  > " P > At this time, OpenVMS Vax media kits are sold out. If you are still interestedK > in getting OpenVMS for your VAX,you may try borrowing a CD from a friend,m8 > co-worker, local Encompass LUG, Ebay, or other places.M > Please note that any OpenVMS distribution is legal to use with the Hobbyistz) > Program (as long as it's not stolen!)."  >  > Seen > - > http://www.montagar.com/hobbyist/mount.html    ------------------------------  % Date: Thu, 16 Jan 2003 00:00:45 -0600 : From: Michael Rice <MichaelARice@knology.n-o--s-p-a-m.net>& Subject: Re: OpenVMS VAX Hobbyist Kits/ Message-ID: <v2cikhaj151s30@corp.supernews.com>s   David L. Cathey wrote:  ( > "David J. Dachtera"  wrote in message 4 > news:<3E1B8D12.ADE256DA@earthlink.spamfree.net>... >lG > >That may be confusing - the available OpenVMS-Alpha distribution for$J > >hobbyists is V7.2 plus selected layered products to fill up a CD image.F > >However, as Robert noted: if you can get the V7.3 distribution fromF > >work, a friend, etc., you certainly can install and run it with the: > >hobbyist license. Likewise, the layered products (SPL). >  >tD > 	The OpenVMS Alpha Hobbyist Kit is in the works, and will be basedH > on OpenVMS V7.3-1 (which should already have the discussed ECO's).  ItJ > looks like it's going to be a 2 CD set, since OpenVMS Engineering really. > packed the base O/S disk with stuff already. >  > - David L. Catheye    I Is there an estimated date for the Alpha kits?  Any chance for a preview f/ as to which software packages will be included?    Thanks.  Michael    ------------------------------    Date: 15 Jan 2003 14:40:30 -08001 From: usenet_vms@lehrerfamily.com (Joshua Lehrer)  Subject: out-of-band ast= Message-ID: <477e0934.0301151440.4ae1a055@posting.google.com>   A We have figured out how to install an out-of-band ast in order to!F install our own custom CTRL-T handler.  This is done using a qio call,B passing in a pointer to the function as P1, and a mask, signifying control-t, as P2.   D When our ast is called, it accepts a void*, which we have determinedA is the character sequence that was hit (control-t, in this case.)s  B Is there any way to get user data passed through to the ast?  MostF ast's allow you to control what is passed in as the void*, making them much more useful.a   Thanks for your help  
 joshua lehrern factset research systems   ------------------------------  % Date: Thu, 16 Jan 2003 00:10:22 +0000t+ From: John Laird <john@laird-towers.org.uk>  Subject: Re: out-of-band ast8 Message-ID: <mqtb2v00tbq18lchpce2ihiq5nfu4cjv7a@4ax.com>  J On 15 Jan 2003 14:40:30 -0800, usenet_vms@lehrerfamily.com (Joshua Lehrer) wrote:  B >We have figured out how to install an out-of-band ast in order toG >install our own custom CTRL-T handler.  This is done using a qio call, C >passing in a pointer to the function as P1, and a mask, signifying$ >control-t, as P2. >wE >When our ast is called, it accepts a void*, which we have determinedaB >is the character sequence that was hit (control-t, in this case.) >cC >Is there any way to get user data passed through to the ast?  MostsG >ast's allow you to control what is passed in as the void*, making themc >much more useful.  L The documentation says the out-of-band character *is* the AST parameter.  ItL would only make sense to specify your own if you could only trap one controlI character per AST routine.  It doesn't work that way - the AST routine is H called for any character that you specify in the mask, when it is typed.  K Long ago, I found that out-of-band ASTs were a much better way of providingdJ a Ctrl/C "interrupt" facility within programs than the official Ctrl/C astI route - on a loaded 11/780 it was a fair bet that a user could hit Ctrl/CtK twice before the AST had managed to re-enable itself, and a Ctrl/Y results.e5 Out-of-band ASTs are repeating, which is much neater.n     	Johne   ------------------------------    Date: 15 Jan 2003 21:43:30 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r Subject: Re: out-of-band ast3 Message-ID: <sDYG6RkZC36r@eisner.encompasserve.org>t  f In article <mqtb2v00tbq18lchpce2ihiq5nfu4cjv7a@4ax.com>, John Laird <john@laird-towers.org.uk> writes:  M > Long ago, I found that out-of-band ASTs were a much better way of providinglL > a Ctrl/C "interrupt" facility within programs than the official Ctrl/C astK > route - on a loaded 11/780 it was a fair bet that a user could hit Ctrl/CeM > twice before the AST had managed to re-enable itself, and a Ctrl/Y results..7 > Out-of-band ASTs are repeating, which is much neater.i  8 As I recall, out-of-band ASTs were not there for VMS V1.F When they were introduced they were described as the preferred method.   ------------------------------  % Date: Wed, 15 Jan 2003 19:18:16 -0500 , From: "Kenneth Block" <krblock@computer.org>! Subject: Re: Product kit questione. Message-ID: <eTmV9.69$Ai4.52@news.cpqcorp.net>  I You cannot compile C++ code with DECC. DECC is a C compiler. You need thet DECCXX compiler.  , "Atcq4eva" <olu-o@mail.com> wrote in message7 news:3b1ab430.0301141203.471b5eb7@posting.google.com...t > Hello, >FG > I working on some legacy code, compiling c++ on vax vms 5.5 with DECC4C > 6.0. I'm getting compiler errors complaining about missing headerlG > files iostream.hxx and exc_handling.h. I'm guessing I need to install:G > a C++ Library kit and possibly DCE Applications developer's kit. Does H > anyone know of a site that will allow me to see what the kit names and > versions are needed? >l	 > Thanks,  >  > Olua   ------------------------------   Date: 15 Jan 2003 22:55:04 GMT# From: hoff@hp.nospam (Hoff Hoffman)l  Subject: Re: Reading file in DCL* Message-ID: <b04os8$nnt$1@web1.cup.hp.com>  e In article <slrnb26a5d.7gp.WADE@DAX.ASUB.ARKNET.EDU>, WADE@DAX.ASUB.ARKNET.EDU (Wade Fincher) writes:eD :Does anyone have any ideas on how I might go about reading the last? :30 bytes of a binary file and assigning the data to a logical?e  E   DCL doesn't really provide for access to binary files and arbitrarynD   binary data -- embedded non-printable characters are comparativelyB   difficult for DCL programs to process.  (Various folks certainlyC   have tried this with DCL and with varying degrees of success, butu    the results can be very ugly.)  F   A small and quite simple program would be the best and most reliableD   approach for the specific task, and would be my prefered solution.  E   What does TYPE/TAIL show for your effort?  (TYPE/TAIL may have some A   difficulty, as it too assumes a printable file and might becomehD   confused by binary data.)  Alternatively, use DUMP/BLOCKS=(...) to'   look at the ending block of the file.   C   And can you provide some background on the particular task?  Also F   please provide some details on exactly what you mean by your "binaryE   file" reference, as this can potentially have various connotations.h  N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.coml   ------------------------------  % Date: Wed, 15 Jan 2003 10:57:50 -08005$ From: Shane Smith <ssmith@icius.com>F Subject: RE: searching for seemingly obscure disk activity light cable0 Message-ID: <01C2BC84.F546D480@sulfer.icius.com>  F You might try going to your local tech shop (I'd use Fry's) and have a= look for alternate cables with the right pin pitch. From youriC description, I'd have a look at internal speaker cables for PCs, ortF maybe some of the audio cables you string between sound cards and CDs.D If the pin assignment's wrong, cut cables and patch them back in the right pattern.   Shane "Heath Robinson" Smith   -----Original Message-----> From: David Spencer [mailto:spencer@spaamfree.pageweavers.com]' Sent: Tuesday, January 14, 2003 9:12 PMu To: Info-VAX@Mvb.Saic.ComsB Subject: searching for seemingly obscure disk activity light cable    F Sorry if this is slightly off topic; but this is for a VMS machine. :)  E I need a disk activity light cable for a Digital Personal Workstation$ 500au. The part numberE I've discovered is 17-04472-01 and is made by Amphenol. It's a simpleU cable - it has a fourSC pin connector on both ends and connects the motherboard to the SCSI0 controller.4  H Mine was wrecked and I'd like to get another. So far I've only found one source and they0C want $40 for it. Considering it's a very simple cable (one 14" wirec between two plasticrH connectors) I consider that excessive. Does anybody know where I can get spares on things like this??a   Thanks very much.r     -- Dave Spencer, PageWeavers   ------------------------------    Date: 15 Jan 2003 16:04:19 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)d8 Subject: Re: ShadowSets, MiniCopy, and SHUTDOWN - Take 2< Message-ID: <cf15391e.0301151604.21852da@posting.google.com>  a Paul Repacholi <prep@prep.synonet.com> wrote in message news:<87smw1q0hn.fsf@prep.synonet.com>... C > I thought you needed a HSC/D/J to get the benefits of mini-merge?i  E This question happened to be about Mini-Copy, rather than Mini-Merge.q   ------------------------------  % Date: Wed, 15 Jan 2003 23:43:24 -0500-; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>l% Subject: Re: Suppressing SMTP headers$$ Message-ID: <3e26386a$1@news.si.com>  < >What I would like to have happen is to have the subject andI >sender/recipients show up in VAXMail, and that is it.  Does anybody knowa how.. >I can get rid of the SMTP header information?  K Sure.  Disable UCX's SMTP, download and install MX.  You can control any oroJ all of the headers that way.  I simply move them to the bottom.  That way,D they're not in anyone's way, but I still have them in case I need to diagnose a routing problem.n -- cI Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot comf5 Smiths Aerospace  Addresses modified to prevent SPAM.a@ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991n8        This opinion doesn't represent that of my company   ------------------------------    Date: 15 Jan 2003 12:17:51 -0600 From: briggs@encompasserve.orgY Subject: Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofW3 Message-ID: <gwrOxVgFnO7T@eisner.encompasserve.org>i  d In article <fbcf38dc.0301150955.189340c4@posting.google.com>, rcbryan@hotmail.com (RC Bryan) writes:U > danco@ns2.pebble.org wrote in message news:<slrnb29lbv.l8k.danco@ns2.pebble.org>...yP >> In article <fbcf38dc.0301141045.25fd47be@posting.google.com>, RC Bryan wrote: > H >> QIO is good.  RMS BLOCK I/O is okay.  The C Run-Time Library file I/OJ >> functions in recent OpenVMS versions are also good (they're also easier@ >> for some programmers to code to and certainly more portable). > G > The C run time library is my first choice for just about everything.  G > I can type in C RTL I/O without thinking about it.  When I want to dorD > anything with QIOs, I have to look in the manuals and even then it- > takes several cycles before I get it right.p > E > We are using QIOs because we had an application that shares data oneG > disk between multiple processes and when we write the data to disk it C > has to be on disk, I played with all kinds of combinations of RTLuF > options, fflush, fsync, etc. and the QIOs were the only way we couldH > GUARANTEE 100% that the data was correct.  I imagine QIOs gave us lessG > overhead as well, another plus.  QIOs are a P.I.T.A. but they get the  > job done.d > H > I got started with this because the manual says "Do not use QIO..."  IE > thought all I/O in the system came down to QIOs in the end.  I havesG > since heard about Fast I/Os.  When I do an upgrade of our software, IoF > will have to check that out. Perhaps the manual means that Fast I/Os" > (sys$io) should be used instead.  @ It seems clear to me that the recommendation is to prevent naive? users from jumping through all sorts of hoops with $QIO and therE ACP interface and parsing file attributes and accessing, for intance,aH a VMS text file and then forgetting to look for the record length fields and pad bytes.  > In almost all situations, RMS record level access or RMS blockA access will let you accomplish the needed tasks more easily, more2& efficiently and more safely then $QIO.  A If you know what you're doing, $QIO can be the right tool to use.2A But if you know what you're doing, you're presumably able to readsC right past a "Do not use QIO" recommendation without worrying aboutT it.R  A If you don't know what you're doing, $QIO is almost certainly thes wrong tool use.    	John Briggs   ------------------------------    Date: 15 Jan 2003 12:25:05 -0600+ From: young_r@encompasserve.org (Rob Young)CY Subject: Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofu3 Message-ID: <$3tSKl5vGw+Q@eisner.encompasserve.org>i  d In article <fbcf38dc.0301150955.189340c4@posting.google.com>, rcbryan@hotmail.com (RC Bryan) writes:   > H > I got started with this because the manual says "Do not use QIO..."  IE > thought all I/O in the system came down to QIOs in the end.  I havesG > since heard about Fast I/Os.  When I do an upgrade of our software, InF > will have to check that out. Perhaps the manual means that Fast I/Os" > (sys$io) should be used instead. >    $ ) $ search sys$examples:io_perform.c sys$  i $c status = SYS$OPEN(&fab);8 status = SYS$CREATE_BUFOBJ(inadr, retadr, 0, 0, handle);C status = SYS$IO_SETUP(IO$_READVBLK, handle, handle, &ast, 0, &fan); C status = SYS$IO_PERFORM(fan, chan, iostat, addr, diskblocksize, i);eD status = SYS$IO_PERFORMW(fan, chan, iostat, addr, diskblocksize, i); status = SYS$IO_CLEANUP(fan);t# status = SYS$DELETE_BUFOBJ(handle);  SYS$DASSGN(chan);  $   
 	Benefits:  M http://www.openvms.compaq.com/doc/731FINAL/6136/6136pro_038.html#index_x_1321t   10.1.1 Fast I/O Benefits  M The performance benefits of Fast I/O result from streamlining high-volume I/O K requests. The Fast I/O system service interfaces are optimized to avoid thevM overhead of general-purpose services. For example, I/O request packets (IRPs)uO are now permanently allocated and used repeatedly for I/O rather than allocateda# and deallocated anew for each I/O. r  L The greatest benefits stem from having user data buffers and user I/O statusM structures permanently locked down and mapped using system space. This allowse Fast I/O to do the following:   < For direct I/O, avoid per-I/O buffer lockdown or unlocking.   H For buffered I/O, avoid allocation and deallocation of a separate system7 buffer, because the user buffer is always addressable. n  N Complete Fast I/O operations at IPL 8, thereby avoiding the interrupt chainingO usually required by the more general-purpose $QIO system service. For each I/O,a= this eliminates the IPL 4 IOPOST interrupt and a kernel AST. e  M In total, Fast I/O services eliminate four spinlock acquisitions per I/O (twoiO for the MMG spinlock and two for the SCHED spinlock). The reduction in CPU costoL per I/O is 20% for uniprocessor systems and 10% for multiprocessor systems.    				Robi   ------------------------------    Date: 15 Jan 2003 17:43:50 -0800$ From: rcbryan@hotmail.com (RC Bryan)Y Subject: Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofe= Message-ID: <fbcf38dc.0301151743.4f72e32b@posting.google.com>i  Y briggs@encompasserve.org wrote in message news:<gwrOxVgFnO7T@eisner.encompasserve.org>...r   >aC > If you don't know what you're doing, $QIO is almost certainly thew > wrong tool use.n  E You said that right.  I think I can count on my fingers the number of F times I have needed to use $QIO to disk.  Now that I think about it, I can count them on one finger.p  	 /RC Bryan    ------------------------------  % Date: Wed, 15 Jan 2003 16:50:35 -0500r< From: "John E. Malmberg" <Malmberg@dskwld.zko.dec.compaq.hp>Y Subject: Re: The manual says "Do not use QIO for reading from a file" has anyone heard ofr. Message-ID: <NIkV9.64$fc4.59@news.cpqcorp.net>   RC Bryan wrote:iG > In the "OpenVMS Programming Concepts Manual" (which can be referenced  > at:gC > http://www.openvms.compaq.com/doc/731FINAL/5841/5841pro_061.html)tG > in section 23.9.1: it says "Note  Do not use the SYS$QIO and SYS$QIOWa? > system services for input from a file or nonterminal device."  > E > I admit I did not read every last bit of the surrounding text but IeD > have to ask, "What does the system use for reading?"  Am I missingD > something here?  What are we supposed to use to read non-RMS data?  F Thank you for your interest in OpenVMS and in using the documentation.  A That appears to be a documentation error, as the use SYS$QIO and sI SYS$QIOW for devices directly supported by OpenVMS are documented in the iH I/O User's Reference Manual.  The SYS$QIO and SYS$QIOW can also be used I with I/O devices supplied by layered products as per their documentation.-  A Care must be taken that you make sure that a SYS$QIO or SYS$QIOW mG operation is valid for the device you are communicating with.  And you  G should use system services such as SYS$GETDVI or SYS$GETDVIW to verify z4 that what type of device you are communicating with.  A The use of logical names may redirect terminals to disk files or  D devices, and with proper care by the programmer, it should all work.  D The use of logical names is encouraged instead of hardcoding device  names into programs.  J I have put in a request for the Programming Concepts manual to be updated.   -Johnp! malmberg@dskwld.zko.dec.compaq.hpb Personal Opinion OnlyT   ------------------------------  % Date: Wed, 15 Jan 2003 20:35:29 +0100 % From: "Jakob Erber" <erberj@yahoo.de>b Subject: Re: THREADCP problemn, Message-ID: <3e25b8d0$1@news.swissonline.ch>  D "Robert TRAWISKI" <robert.trawinski@softax.com.pl> wrote in message( news:b03f5c$va4$1@bozon.softax.com.pl... > Hi,  > L > I develop program that should have kernel thread disabled (OpenVMS v.7.2-1- > and OpenVMS v.7.3-1). I use THREADCP tools:  > A > $ THREADCP/DISABLE=(MULITPE_KERNEL_THREADS,UPCALLS) PROGRAM.EXEl >dK > But the program link few shared libraries. Shuold I do the same for those  > libraries ? For example: >wA > $ THREADCP/DISABLE=(MULITPE_KERNEL_THREADS,UPCALLS) LIB1SHR.EXE  >3 > Thanks in advancem >o > Robert >H >g    
 Hi Robert,  K I think, it should be sufficient, if you perform the command for the (main)o executable onlyJ   regardsi   Jakob'   ------------------------------    Date: 16 Jan 2003 00:53:07 +0800, From: Paul Repacholi <prep@prep.synonet.com>+ Subject: Re: Traceback and installed images4- Message-ID: <87u1ganymk.fsf@prep.synonet.com>h  2 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  ! > briggs@encompasserve.org wrote:p  A > > handler.  Most of them will just trigger a stack dump.  But arC > > signal of SS$_DEBUG will cause the traceback handler to map theE? > > debugger.  And handing a user access to a debugger within a A > > privileged image is tantamount to handing him the keys to the  > > system.p  iC > Ok thanks to the explanation. In this case, wouldn't it have been B > better for the VMS engineers to disable SS$DEBUG handling if the > image is installed ?  ; That will also disable debugging of any image that links torA it. Remember that the image activator will not allow any outgoingeB calls from a priv image, and that must include the traceback/debug calls.     E > I realise that you can't change the exit handler of old images, butaE > they could change the debugger code to prevent it from running whenn? > invoked by an exit handler of an image installed with privs ?d  F And what of another debugger?  This is like the error of thinking that@ changing a directory protection stops someone from accessing the files.   -- a< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.a@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 15 Jan 2003 23:33:41 GMT# From: hoff@hp.nospam (Hoff Hoffman)t. Subject: Re: VMS on Itanic boot report sighted* Message-ID: <b04r4l$nnt$3@web1.cup.hp.com>  b "Tom Linden" <tom@kednos.com> wrote in message news:CIEJLCMNHNNDLLOOGNJIMEJLGGAA.tom@kednos.com...C : The transition to Itanium might have been easier had it been fromC : VAX rather than Alpha.    D   The first platform port is often the most difficult -- OpenVMS VAXD   was tied very tightly into VAX hardware, while the port of OpenVMSI   to Alpha caused the removal of many of the latent hardware assumptions.   E   That said, the Intel Itanium architecture is probably arguably moreeH   similar to that of VAX than it is to Alpha -- but the Itanium platformH   I/O, peripherals, buses, and ancillary constructs are far more similar&   to those of Alpha than those of VAX.    B :                         It looks to me that it would have been aA : lot easier porting VCG rather than GEM, but then, I am probably< : wrong.  C   We did not port either one of the two referenced code generators.D  E   We do have a GEM calling interface wrapped around the Intel Itanium <   code generator, as this made porting the compilers easier.  B   Code generators tend to be highly complex and are generally also.   (for obvious reasons) architecture-specific.  C   As for VCG, I remember the early problems with VCG.  I solidified E   what I know (or once knew :-) of the VAX instruction set when I wasaH   debugging the VCG-generated binary code.  I still distinctly remember E   chasing a gratuitous address dereference within the generated code. D   (I have encountered far less pain with GEM-generated binary code.)  C   The engineering teams are presently actively working with the newt,   compilers and with the new code generator.    N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com0   ------------------------------    Date: 15 Jan 2003 17:29:14 -08001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)E" Subject: Re: What does RWCLU mean?= Message-ID: <cf15391e.0301151729.3ac1f718@posting.google.com>   Q Saminda C F Lam <ccminda@ust.hk> wrote in message news:<3E1E9FE6.60809@ust.hk>...tD > What does RWCLU mean?  The manual says "Resource Wait for Cluster J > Transition".  I found processes in RWCLU state on an Alpha cluster.  Is A > it indicative of some resource problem?  If so, what resources?   > Given that you observed this (and likely couldn't have seen itD dynamically if a state transition were actually in progress), it was> most probably due to activity-based remastering of lock trees.  @ In VMS 7.3 or higher, $MONITOR RLOCK can be used to monitor lock, remastering activity.  In VMS 7.2 or higher,    SDA> SHOW LOCK/SUMMARYaE can provide statistics on lock remastering.  In earlier versions, useo#    SDA> EXAMINE PMS$GL_RM_RBLD_SENTl#    SDA> EXAMINE PMS$GL_RM_RBLD_RCVDX! and watch the counters over time.   C In clusters with multiple nodes of about the same CPU power running ? the same application(s), sometimes VMS remasters lock trees tooc8 frequently.  If this is a problem, you can control it byA redistributing workload (bias the workload toward one node or thepE other), introduce an imbalance of LOCKDIRWT values, or use a non-zero 
 value of PE1.r  Y See http://www.geocities.com/keithparris/decus_presentations/s2002_dist_lock_mgr_perf.pptr for more info.   ------------------------------  % Date: Wed, 15 Jan 2003 11:30:54 -0700 $ From: Lee Y T Mah <lytmah@cha.ab.ca>2 Subject: Re: where are diskeeper logicals defined?) Message-ID: <3E25A8DE.D400B31B@cha.ab.ca>   L The logicals are defined in [ESICM]CONTROL_MASTER_STARTUP.COM located on theK device on which Diskeeper is installed and licensed.  This is for Diskeeperh 7.2-1A.o   frank brown wrote:  A > Running Diskeeper 7.1 on VAX/VMS 5.5-2.  I can't find where theeN > ESI_PROD_CONTROL_MASTER logical is defined; it's not in SYLOGICALS.COM.  Can9 > anyone point me to where this logical is being defined?g   -- LeeA  ; Lee Y T Mah                        Capital Health Authority    ------------------------------    Date: 15 Jan 2003 18:45:27 -0800) From: munroe@csworks.com (Richard Munroe)e Subject: ZMENU - Menus for DCL< Message-ID: <8d09fa7.0301151845.1d867729@posting.google.com>  E is now available from Sourceforge.  This version compiles cleanly fortC both the VAX and Alpha architectures (The original abused the frameaC pointers on Alpha in a way that the Macro-32 compiler couldn't cope  with).  You can get it from:  *     http://sourceforge.net/projects/zmenu/  @ My apologies to anybody doing active development, I did look for you...  @ This is also used by Branches and [coming soon] VirtualBranches.   Dick Munroe   D p.s. I'm looking for work.  Anybody interested can take a look at my= resume it's online at http://www.acornsw.com/resume/dick.html    ------------------------------   End of INFO-VAX 2003.031 ************************