1 INFO-VAX	Thu, 20 Jul 2006	Volume 2006 : Issue 400       Contents:* C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE. Re: C or DCL routine to examine EXE$GQ_CPUTYPE Re: DYNDNS client for OpenVMS? Re: DYNDNS client for OpenVMS? Re: DYNDNS client for OpenVMS? Re: DYNDNS client for OpenVMS? gnu gmp on OpenVMS/Alpha ? Re: gnu gmp on OpenVMS/Alpha ?C Re: Greenhorn question: XON/XOFF control chars (when in MicroEMACS) D Re: How to use kitinstal.com procedure to build RZDISK application??# Re: HP to cut down on telecommuting % New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ) Re: New itaniums out at 2.5x perform gain ! PIPE redirection as stream file ? ( Question about the 497.1-day-uptime bug., Re: Question about the 497.1-day-uptime bug., Re: Question about the 497.1-day-uptime bug.( Re: Tomcat user authentication question.( Re: Tomcat user authentication question.E Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing) E Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing) E Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing) E Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing) E Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing)  Re: XML on Expat  F ----------------------------------------------------------------------    Date: 19 Jul 2006 12:45:46 -0700 From: "R Boyd" <bob@hax.com>3 Subject: C or DCL routine to examine EXE$GQ_CPUTYPE C Message-ID: <1153338346.676276.233800@b28g2000cwb.googlegroups.com>   C The gnu config.sub and related routines have various strategies for E determining what model of cpu they are running on.  I'm trying to fix E one for the GMP kit so that it can distinguish which variant of Alpha B it's running on.  As of now the routine blows up when it fails theD attempt to compile the assembler routine that works for Tru64 and/or Linux on Alpha.   G I'd like to find a bit of code that would do the same using either DECC D or DCL on VMS.  There's an assembler routine that is part of the GMP? conf routines that apparently does it, but the assembler is not G available on my system and I don't know if it would actually compile if D I had it.   I do have a C compiler.  Doing it from DCL would be evenF slicker.  I can get the information from ANALYZE/SYSTEM CLUE CONFIG --E I would like to think there might be a lighter weight path to getting  the information.  - Anyone have something they've done like this?    Thanks!    Robert   ------------------------------  % Date: Wed, 19 Jul 2006 15:59:21 -0400 # From: sol gongola <sol@adldata.com> 7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE 0 Message-ID: <1153339238.748470@nntp.acecape.com>  
 R Boyd wrote: E > The gnu config.sub and related routines have various strategies for G > determining what model of cpu they are running on.  I'm trying to fix G > one for the GMP kit so that it can distinguish which variant of Alpha D > it's running on.  As of now the routine blows up when it fails theF > attempt to compile the assembler routine that works for Tru64 and/or > Linux on Alpha.  > I > I'd like to find a bit of code that would do the same using either DECC F > or DCL on VMS.  There's an assembler routine that is part of the GMPA > conf routines that apparently does it, but the assembler is not I > available on my system and I don't know if it would actually compile if F > I had it.   I do have a C compiler.  Doing it from DCL would be evenH > slicker.  I can get the information from ANALYZE/SYSTEM CLUE CONFIG --G > I would like to think there might be a lighter weight path to getting  > the information. > / > Anyone have something they've done like this?  > 	 > Thanks!  >  > Robert > $ write sys$output f$getsyi("hw_name")   ------------------------------    Date: 19 Jul 2006 13:19:13 -0700 From: "R Boyd" <bob@hax.com>7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE C Message-ID: <1153340353.051167.224560@m79g2000cwm.googlegroups.com>   ) Perhaps I should have been more explicit.   F What I need is the EV level of the cpu, not the "hardware name".  Take& a look at the output of ANALYZE/SYSTEM
 > CLUE CONFIG   ! and look at the CPU TYPE field(s)   G I have looked at all of the F$GETSYI bits and there may be something in > the SYSTEM ID, or ARCHFLAG but I am not sure how do decode the necessary subfield(s).   sol gongola wrote:  & > write sys$output f$getsyi("hw_name")   ------------------------------  % Date: Wed, 19 Jul 2006 16:54:24 -0400 # From: sol gongola <sol@adldata.com> 7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE / Message-ID: <1153342541.47845@nntp.acecape.com>   
 R Boyd wrote: + > Perhaps I should have been more explicit.  > H > What I need is the EV level of the cpu, not the "hardware name".  Take( > a look at the output of ANALYZE/SYSTEM >> CLUE CONFIG > # > and look at the CPU TYPE field(s)  > I > I have looked at all of the F$GETSYI bits and there may be something in @ > the SYSTEM ID, or ARCHFLAG but I am not sure how do decode the > necessary subfield(s). >  > sol gongola wrote: > ' >> write sys$output f$getsyi("hw_name")  > 2 The following refers to a C program call whatami.c) It help you get what you are looking for.   ? http://h21007.www2.hp.com/dspp/files/unprotected/tru64/tips.pdf    ------------------------------    Date: 19 Jul 2006 14:38:56 -0700 From: "R Boyd" <bob@hax.com>7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE B Message-ID: <1153345136.014140.290690@i3g2000cwc.googlegroups.com>  G Here's a bit of perl code that does pretty much what is called for.  If E someone has a prettier bit of code to offer, I'll be happy to look at  it.     ! my $cpu_word = qr/[a-zA-Z0-9]+/ ;  my $number = qr/[0-9]+/ ;  my @cpu_types = () ; # 2 # Find the appropriate records in the include file #   foreach (`library/text ? SYS\$SHARE:SYS\$LIB_C.TLB/extract=HWRPBDEF/output=sys\$output`)   {
 #  print ;:   if ( m/HWRPB_CPU_TYPE\$K_($cpu_word)\s*($number)\s*/ ) {         @cpu_types[$2] = $1 ;  #       print $2,"=",$1,"\n"      } }  # # # find out the cpu type from GETSYI  # 9 my $cpu_type = `write sys\$output f\$getsyi("cputype")` ; 5 #print "Cpu type $cpu_type = @cpu_types[$cpu_type]" ;  print "@cpu_types[$cpu_type]";   ------------------------------  # Date: Thu, 20 Jul 2006 01:42:53 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE 0 Message-ID: <xcBvg.809$yR4.593@news.cpqcorp.net>  
 R Boyd wrote: E > The gnu config.sub and related routines have various strategies for G > determining what model of cpu they are running on.  I'm trying to fix G > one for the GMP kit so that it can distinguish which variant of Alpha D > it's running on.  As of now the routine blows up when it fails theF > attempt to compile the assembler routine that works for Tru64 and/or > Linux on Alpha.   D The C code available via the following URL works on OpenVMS, and is , known to be portable to at least Tru64 UNIX:  5   http://h71000.www7.hp.com/wizard/swdev/implver.html    ------------------------------    Date: 19 Jul 2006 15:55:37 -0500. From: brooks@cuebid.zko.hp.nospam (Rob Brooks)7 Subject: Re: C or DCL routine to examine EXE$GQ_CPUTYPE , Message-ID: <HTf5qd9VWxvP@cuebid.zko.hp.com>   "R Boyd" <bob@hax.com> writes:  E > The gnu config.sub and related routines have various strategies for G > determining what model of cpu they are running on.  I'm trying to fix G > one for the GMP kit so that it can distinguish which variant of Alpha D > it's running on.  As of now the routine blows up when it fails theF > attempt to compile the assembler routine that works for Tru64 and/or > Linux on Alpha.  > I > I'd like to find a bit of code that would do the same using either DECC F > or DCL on VMS.  There's an assembler routine that is part of the GMP  ( $ write sys$output f$getsyi( "cputype" ) 8    --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------  % Date: Wed, 19 Jul 2006 13:45:27 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ' Subject: Re: DYNDNS client for OpenVMS? , Message-ID: <44BE6FAF.C55F194C@teksavvy.com>  G One thing to consider: VMS is not well suited for dynamic IPs. Its DHCP 1 client doesn't seem to be so easy to get working.   A So, when the VMS DHCP client renewes a lease and is told to use a F different IP address, can you intercept this and immediatly update theE DYNDNS entry, or will you be doing this at regular intervals and hope G there isn't too long a time between your DYNDNS entry not pointing to a  newly assigned IP ?   D It is much easier and much more robust to get a router that does the@ DHCP/PPPoE and automatic DYNDNS interface so that your DYNDNS isF automatically yupdated whenever the pullic IP changes, and this allowsP your VMS host to operate with a fixed IP address which is much easier to manage.   ------------------------------    Date: 19 Jul 2006 22:15:24 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)' Subject: Re: DYNDNS client for OpenVMS? * Message-ID: <44beaefc@news.langstoeger.at>  \ In article <44BE6FAF.C55F194C@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:H >One thing to consider: VMS is not well suited for dynamic IPs. Its DHCP2 >client doesn't seem to be so easy to get working.  N The OpenVMS TCPIP DHCP client is not the real problem (and is easy to config).  H OpenVMS TCPIP services which refuse to start while the DHCP client stillK does try to get a valid IP address and services which don't restart/survive J while the DHCP client renews its address (releases the IP address and getsJ another address instead) are the real problem. The PWIP Driver is just oneK of them (and for me the most important) and makes DHCP clients on VMS still K a nightmare. Better see VMS as a server and give it a fixed address and let ; the firewall/router to the DHCP (and DynDNS) client part...    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Thu, 20 Jul 2006 01:31:00 GMT   From: John Santos <john@egh.com>' Subject: Re: DYNDNS client for OpenVMS? * Message-ID: <o1Bvg.6273$Ss2.4148@trnddc01>   JF Mezei wrote: I > One thing to consider: VMS is not well suited for dynamic IPs. Its DHCP 3 > client doesn't seem to be so easy to get working.  > C > So, when the VMS DHCP client renewes a lease and is told to use a H > different IP address, can you intercept this and immediatly update theG > DYNDNS entry, or will you be doing this at regular intervals and hope I > there isn't too long a time between your DYNDNS entry not pointing to a  > newly assigned IP ?  >   F This should *never* happen.  When you renew a lease, you should alwaysD get the same address back.  The only time you should get a differentB address is if you crash or shut down and stay down longer than theB lease period.  Then when you come back up, and ask for an address,E the DHCP server will give you a new one, but only if your old address D has been assigned to someone else or if the DHCP server also crashedE and forgot about you.  In these cases your client has rebooted, so it 1 knows it should register its address with DYNDNS.   F > It is much easier and much more robust to get a router that does theB > DHCP/PPPoE and automatic DYNDNS interface so that your DYNDNS isH > automatically yupdated whenever the pullic IP changes, and this allowsR > your VMS host to operate with a fixed IP address which is much easier to manage.  D There are other advantages to having a fixed IP on your LAN, such asB having a NAT server in your firewall/router that knows to forward,C e.g. HTTP to your VAX and SMTP services to your other VAX and block ; all of the other TCP ports to keep bad guys out of your PC.      --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Wed, 19 Jul 2006 22:29:07 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ' Subject: Re: DYNDNS client for OpenVMS? , Message-ID: <44BEEA70.5729B885@teksavvy.com>   John Santos wrote:H > This should *never* happen.  When you renew a lease, you should always > get the same address back.   Wrong.  H Consider the ISP planning some router/network changes. They will programB the DHCP server to stop renewing/giving IPs in a certain block andH instead move all customers to a new IP block. Wait for all leases in theA old IP block to expire and they can then make the router changes.    ------------------------------    Date: 19 Jul 2006 11:55:45 -0700 From: "R Boyd" <bob@hax.com># Subject: gnu gmp on OpenVMS/Alpha ? B Message-ID: <1153335345.560107.119350@75g2000cwc.googlegroups.com>   I found this topic:    http://groups.google.com/group/comp.os.vms/browse_frm/thread/9111b97c0b2bd690/4a27948554776286?q=gnu+gmp+openvms+alpha&lnk=gst&rnum=1#4a27948554776286  G I tried downloading the gmp kit and building it under GNV -- however -- C it looks like there's some buggy stuff happening with the configure G step in the config.sub script where it's trying to figure out precisely " which type of Alpha is being used.  G I'm willing to work on getting this thing to work, but I'd love to know F if anyone has already cracked this nut and it just hasn't made its way back into the distribution.   G Ultimately my goal is to build the perl Net::SSH2 module on the system. G   It requires one include file from the gmp kit, gmp.h.  I haven't been E able to figure out yet if there's somewhere in the gmp kit I can just  yank out the gmp.h or not.  + Thanks for any suggestions/pointers/etc....    Robert   ------------------------------    Date: 19 Jul 2006 12:02:34 -0700  From: "Ian Miller" <ijm@uk2.net>' Subject: Re: gnu gmp on OpenVMS/Alpha ? C Message-ID: <1153335754.781591.322960@i42g2000cwa.googlegroups.com>   C the forums on the VAMP site is can be helpful but I think its still  down$ http://www.issinoho.com:8080/phpbb2/   Try the O.O port to VMS people http://www.oooovms.dyndns.org/   ------------------------------  % Date: Wed, 19 Jul 2006 14:31:01 -0600  From: Kevin Handy <kth@srv.net> L Subject: Re: Greenhorn question: XON/XOFF control chars (when in MicroEMACS)3 Message-ID: <1153339599_69561@sp6iad.superfeed.net>    Bob Koehler wrote:  G >    I'm not even sure if there are enough wires on an MMJ or RJ11 jack C >    to handle hardware flow control.  I'm quite sure I couldn't be E >    bothered to spend my customer's money make it work just because   >    emacs wanted it that way.  6 MMJ has (DSR, RTS), but a lot of the physical hardware6 doesn't pass them, like the wall plates that DEC sold.4 There were punch-down points for the wires, but they didn't connect to the jack.   8 Many RS-232 cables were 3-wire only. Trying to use emacs8 on one of these was a mess. You either got the commands,: you got to see all your data, or you ran slow. Most people, I know just used something other than emacs.  Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Wed, 19 Jul 2006 21:42:53 -0500 . From: Bob Blunt <RobertDOTblunt@digitalDOTcom>M Subject: Re: How to use kitinstal.com procedure to build RZDISK application?? : Message-ID: <uY6dncZl8dYPcSPZnZ2dnUVZ_t6dnZ2d@comcast.com>   schneider wrote:H > I just downloaded the "RZDISK030.A" freeware application to an AS4100  <snip> >> $ exit vmi$_failure  A VMSINSTAL.COM, in a nutshell, handles all the required setup and  D manipulation of the savesets included in a software kit.  Leave the H files in their respective savesets on tape, magnetic disk or CD and let G the combination of command procedures do their magic.  There have been  E very few situations in my experience where there was any need to rip  ? open kit savesets, just let 'em do that voodoo they do so well.    bob    ------------------------------  % Date: Thu, 20 Jul 2006 00:09:51 -0400 + From: Steve Matzura <number6@speakeasy.net> , Subject: Re: HP to cut down on telecommuting8 Message-ID: <l30ub2lvlrdmhhkas01oogagn5s06bqukb@4ax.com>  
 Hi, Kerry:E On Sun, 16 Jul 2006 09:06:13 -0400, "Main, Kerry" <Kerry.Main@hp.com>  wrote:  E >There is always going to be some debate which is better, but this is D >only one division in HP doing this that I know of. Others have veryF >openly embraced the virtual work from home strategy. Indeed, facilityG >closures are usually based on this virtual work closure concept as the G >cost of facilities is just way to big of a cost. It assumes a relative H >ratio of 3-5 (or higher) workers per virtual work "pod" and if everyoneG >had to go back into the office, then new facilities would be required.  > B >For DR scenario's, you need to have an established work from homeC >infrastructure as workers may not be able to physically get to the I >backup site. In addition, if everyone is in the office at the same time, I >it further raises the risk of a significant event impacting the majority  >of your workers.  > F >Think of a single person walking into your office facility and HealthH >authorities find out he/she was exposed to SARS. They may not even haveE >it - simply being exposed is enough to close the entire building and F >send *everyone* in that entire building home for 10 days minimum. ForH >those who have not lived through SARS type events, that is the accepted	 >process.  > C >This SARS / pandemic worry is not theory - it actually happened in F >Toronto. The entire facility was closed and everyone sent home for 10H >days. It just so happened that a data center was also in that building.G >The DC had to be vacated immediately (no backups, no gathering tapes - H >just leave now!) and closed to all physical access for 10 days as well.I >Fortunately, the DC was part of a dual site design and the workers could G >continue working from home along with staff at the other facility, but @ >how many companies have this multi-site + remote work from home >capability today? > I >It is also why new DC's being built today should be dedicated structures D >away from major centers and that do not have any shared access with >other tenants or office space.   @ Well said.  Now if only we the workers could get some of the taxE breaks Big Corporate got for building some of those big office towers F filled with us who no longer work there, then it'd be about as perfectC a world as one could get!  There are tax advantages if you own your D own home and dedicate a portion of it entirely to your job function,? but if you live in a small apartment, you can't just wall off a D section of it and say nobody can set foot in it unless the office isA open and functioning as such.  In my apartment, for instance, you F couldn't leave the living room if what used to be my office was walledB off and locked up.  Well, if you turned sideways and made yourselfA really really thin, maybe you could, but the point is that if you @ don't regularly go into an office to perform your work function,B certain space allotment and tax abatement should be granted to youB because you are essentially paying for your own office space.  TheF rules in this country are, I think, far too strict just now to make itF possible to do things strictly by the book.  This will probably change over the next couple decades.    ------------------------------    Date: 19 Jul 2006 12:50:31 -0700 From: bob@instantwhip.com . Subject: New itaniums out at 2.5x perform gainB Message-ID: <1153338631.693974.80440@m73g2000cwd.googlegroups.com>  I http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823    ------------------------------    Date: 19 Jul 2006 13:34:32 -0700- From: "Doug Phillips" <dphill46@netscape.net> 2 Subject: Re: New itaniums out at 2.5x perform gainC Message-ID: <1153341272.166611.189890@i42g2000cwa.googlegroups.com>    bob@instantwhip.com wrote:K > http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823    ##C Itanium was supposed to be a "panacea" for high-end computing, says E Cox. Then "reality set in" that customers were unwilling to reprogram A much or their x86 software to take advantage of Itanium, he says.  ##  B Now we know; it's reality's fault! Darned reality, setting in like that, screwing everything up.    ------------------------------  % Date: Wed, 19 Jul 2006 18:30:38 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> 2 Subject: Re: New itaniums out at 2.5x perform gain: Message-ID: <44beb1c0$0$989$9a6e19ea@news.newshosting.com>  & <bob@instantwhip.com> wrote in message< news:1153338631.693974.80440@m73g2000cwd.googlegroups.com...K > http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823  > 
 <quote-on>G "We're working pretty hard to get it to a profitable product," said Pat I Gelsinger, senior VP of Intel's digital enterprise group, in an interview J following a press conference in San Francisco Tuesday. "If we could unwindF the clock, I would have just built a RAS version of Xeon to attack theI market," he said, using an industry term for "reliable, highly available, F and scalable" chips, and referring to Intel's Xeon server chips, whichG employ the widely used x86 instruction set. Itanium uses a less popular  design called EPIC.  <quote-off>   K When a company makes this kind of statement in public, they are testing the D waters. Now let's see what the investment community says about this.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html9 http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  % Date: Wed, 19 Jul 2006 20:16:54 -0400 ( From: Bill Todd <billtodd@metrocast.net>2 Subject: Re: New itaniums out at 2.5x perform gainG Message-ID: <koGdncJoyNPrViPZnZ2dnUVZ_qudnZ2d@metrocastcablevision.com>    Neil Rieck wrote: ( > <bob@instantwhip.com> wrote in message> > news:1153338631.693974.80440@m73g2000cwd.googlegroups.com...L >> http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823 >> > <quote-on>I > "We're working pretty hard to get it to a profitable product," said Pat K > Gelsinger, senior VP of Intel's digital enterprise group, in an interview L > following a press conference in San Francisco Tuesday. "If we could unwindH > the clock, I would have just built a RAS version of Xeon to attack theK > market," he said, using an industry term for "reliable, highly available, H > and scalable" chips, and referring to Intel's Xeon server chips, whichI > employ the widely used x86 instruction set. Itanium uses a less popular  > design called EPIC. 
 > <quote-off>  > M > When a company makes this kind of statement in public, they are testing the 	 > waters.   A Interesting thought.  My own reaction was merely that this blunt  H admission by Gelsinger should finally put paid to the Itanic apologists G who keep bleating that it at least has a safe high-end RAS refuge that   x86 can never assail.    - bill   ------------------------------  % Date: Wed, 19 Jul 2006 20:33:38 -0400 ( From: Bill Todd <billtodd@metrocast.net>2 Subject: Re: New itaniums out at 2.5x perform gainG Message-ID: <W_idncN2X4j-UiPZnZ2dnUVZ_umdnZ2d@metrocastcablevision.com>    bob@instantwhip.com wrote:K > http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823   H Ah, me - even an article that starts by bending over backward to try to G make Itanic look good has a lot of difficulty doing so these days - at  B least if it doesn't play somewhat fast and loose with its wording.  	 Examples:   H "Say this for Intelafter five years of [initially negligible and later F only] modest sales and [dramatically] scaled-down ambitions, it's not B [at least not yet] giving up on Itanium [at least not publicly]."   (bracketed clarifications added)  ? "The chips deliver more than twice the database performance of  H previous-generation Itaniums [at the *chip* level, though only slightly I more *per-core* performance despite years of new development, 5x as much  G L2 cache and 1.33x as much L3 cache, and significant core enhancements  C like multi-threading:  the rest comes from having two cores on the  A chip], and draw 2.5 times less electric power [if you nimbly and  H ever-so-tacitly switch the subject from the *chip* to *per-core* power: 1   the per-chip power is only moderately reduced]"   J "But..." it then continues, and actually isn't all that biased thereafter.   - bill   ------------------------------   Date: 20 Jul 2006 01:02:18 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)2 Subject: Re: New itaniums out at 2.5x perform gain+ Message-ID: <4i830qF2k5j6U1@individual.net>   G In article <koGdncJoyNPrViPZnZ2dnUVZ_qudnZ2d@metrocastcablevision.com>, + 	Bill Todd <billtodd@metrocast.net> writes:  > Neil Rieck wrote: ) >> <bob@instantwhip.com> wrote in message ? >> news:1153338631.693974.80440@m73g2000cwd.googlegroups.com... M >>> http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823  >>> 
 >> <quote-on> J >> "We're working pretty hard to get it to a profitable product," said PatL >> Gelsinger, senior VP of Intel's digital enterprise group, in an interviewM >> following a press conference in San Francisco Tuesday. "If we could unwind I >> the clock, I would have just built a RAS version of Xeon to attack the L >> market," he said, using an industry term for "reliable, highly available,I >> and scalable" chips, and referring to Intel's Xeon server chips, which J >> employ the widely used x86 instruction set. Itanium uses a less popular >> design called EPIC. >> <quote-off> >>  N >> When a company makes this kind of statement in public, they are testing the
 >> waters. > C > Interesting thought.  My own reaction was merely that this blunt  J > admission by Gelsinger should finally put paid to the Itanic apologists I > who keep bleating that it at least has a safe high-end RAS refuge that   > x86 can never assail.  >   G In the corporate world isn't swallowing your ego and admitting that you I made a mistake the first and in most cases the hardest step to correcting 
 that mistake?    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Wed, 19 Jul 2006 20:57:21 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> 2 Subject: Re: New itaniums out at 2.5x perform gain< Message-ID: <44bed422$0$18476$9a6e19ea@news.newshosting.com>  5 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message  4 news:44beb1c0$0$989$9a6e19ea@news.newshosting.com... > ( > <bob@instantwhip.com> wrote in message> > news:1153338631.693974.80440@m73g2000cwd.googlegroups.com...L >> http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823 >> > <quote-on>I > "We're working pretty hard to get it to a profitable product," said Pat K > Gelsinger, senior VP of Intel's digital enterprise group, in an interview L > following a press conference in San Francisco Tuesday. "If we could unwindH > the clock, I would have just built a RAS version of Xeon to attack theK > market," he said, using an industry term for "reliable, highly available, H > and scalable" chips, and referring to Intel's Xeon server chips, whichI > employ the widely used x86 instruction set. Itanium uses a less popular  > design called EPIC. 
 > <quote-off>  > J > When a company makes this kind of statement in public, they are testing  > the F > waters. Now let's see what the investment community says about this. >  Yikes!; http://news.zdnet.com/2100-9595_22-6096207.html?tag=nl.e589 5 Intel has just announced a 57 percent drop in income. * I wonder what those suits will chop first?  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html9 http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  % Date: Wed, 19 Jul 2006 22:11:10 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: New itaniums out at 2.5x perform gain, Message-ID: <44BEE63D.311660E0@teksavvy.com>   Doug Phillips wrote: >  > bob@instantwhip.com wrote:M > > http://www.informationweek.com/news/showArticle.jhtml?articleID=190500823  >  > ##E > Itanium was supposed to be a "panacea" for high-end computing, says G > Cox. Then "reality set in" that customers were unwilling to reprogram C > much or their x86 software to take advantage of Itanium, he says.  > ##    F What I found more telling is the admission that that IA64 thing is not profitable for Intel.   F Me thinks that the second the contract runs out with HP, Intel will be9 more than happy to put that IA64 thing out of its misery.   G So far, I have seen nothing to indicate that such an announcement won't 0 come in 2007. The path was set in February 2004.   ------------------------------  % Date: Wed, 19 Jul 2006 22:22:57 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: New itaniums out at 2.5x perform gain, Message-ID: <44BEE900.A45B5062@teksavvy.com>   Neil Rieck wrote: = > http://news.zdnet.com/2100-9595_22-6096207.html?tag=nl.e589 7 > Intel has just announced a 57 percent drop in income. , > I wonder what those suits will chop first?  E IA64 is safe. It is a fair bet that HP is subsidizing IA64 to keep it C alive for now. And if an exit strategy has already been decided and F slowly being implemented, by next year, the 8086 will probably be muchF faster than that IA64 contraption and it will be much easier for HP toD announce it is moving to simplify its platforms and standardizing on industry standard 8086s.    D Right now, the cost of prematurely cancelling IA64 would be probablyF higher for Intel than the costs of continuing it. But it is a fair betE that Intel will not be hiring extra engineers to speed up development G work of IA64 and the opposite may happen which will make each remaining D iteration of IA64 even later and thus less relevant to the industry.  F The big question is at what level within HP are they aware of the true7 strategy and working to plan on the transition to 8086.    ------------------------------  % Date: Wed, 19 Jul 2006 23:25:30 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: New itaniums out at 2.5x perform gain, Message-ID: <44BEF7A2.70EB9F9A@teksavvy.com>  G In the news today as well: Intel will deliver quadcore 8086s early (end E of 2006 instead of mid 2007). Why ? because they MUST compete against E AMD and regain marklet share against AMD. That is where the resources  are being put.   ------------------------------    Date: 19 Jul 2006 16:51:41 -0700% From: "Pierre" <pierre.bru@gmail.com> * Subject: PIPE redirection as stream file ?A Message-ID: <1153353100.976519.4720@m79g2000cwm.googlegroups.com>    hi,   > PIPE redirection are "Record format: VFC, 2 byte header" files  A I got a program the output some text followed by <CR><LF> and the ? result is the the redirected output contains 2 records: the 1st & contains the data and the 2nd is empty  ( is it possible to change this behavior ?   TIA, Pierre.    ------------------------------    Date: 19 Jul 2006 10:58:16 -0700$ From: "AEF" <spamsink2001@yahoo.com>1 Subject: Question about the 497.1-day-uptime bug. B Message-ID: <1153331896.770693.58710@m73g2000cwd.googlegroups.com>   Hello!  F In VAX/VMS V6.2, there is the 497.1-day-uptime bug. Does anyone happenC to know if V6.1 also has this bug? I looked on the Web but couldn't 8 find anything. A description of the bug follows. Thanks!  ' Problems addressed in VAXSYSA02_062 kit     B o  If a system has been up for 497.1 days without  rebooting,  theB    system  cell  EXE$GL_ABSTIM_TICS (number of 10 millisecond ticsB    since  boot)  will  overflow.   This  problem  can  cause  someB    processes   to   remain  indefinitely  in  the  RWMPB  or  COMOB    scheduling state.  Furthermore, candidate processes for Virtual3    Balance Set Slot (VBSS) selection can be missed.   B    Code using the EXE$GL_ABSTIM_TICS cell to time  other  activityB    could  get stuck if a reference counter is large (FFFFxxxx) andB    the  EXE$GL_ABSTIM_TICS  value  is  low  due  to  the  overflowB    (0000xxxx).   Most  such checks compare to see if the number ofB    tics is larger than the reference value, i.e.   has  the  event    come due?   ------------------------------  % Date: Wed, 19 Jul 2006 14:57:48 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 5 Subject: Re: Question about the 497.1-day-uptime bug. , Message-ID: <44BE80A0.6AF76574@teksavvy.com>  
 AEF wrote: >  > Hello! > H > In VAX/VMS V6.2, there is the 497.1-day-uptime bug. Does anyone happenE > to know if V6.1 also has this bug? I looked on the Web but couldn't : > find anything. A description of the bug follows. Thanks! > ) > Problems addressed in VAXSYSA02_062 kit  > D > o  If a system has been up for 497.1 days without  rebooting,  theD >    system  cell  EXE$GL_ABSTIM_TICS (number of 10 millisecond tics$ >    since  boot)  will  overflow.      E Not sure about that particular bug, but there is a most excellent FAQ / item on the care and feeding of VAX TOY clocks.   F Basically, the toy clock accumulates time since jan 1 of any year. The; number of bits in that register only allows up to 497 days.   G Whenever you issue a SET TIME command to change the time, the TOY clock F is updated (so if you set time in January 1 at 00:00:01, the toy clockA gets back to a value near "1", and the SYS.EXE file is updated to . contain that fact as well as the current year.  F So when you reboot, the system checks the time in SYS.EXE and the time@ in the TOY clock. If the time in TOY is smaller than the time in" SYS.EXE, it is considered invalid.  G So if you do not issue a SET TIME command within 3/4 months of january, F the TOY clock overflows and goes back to 0, at which point it is lower1 than the value in SYS.EXE and considered invalid.    ------------------------------    Date: 19 Jul 2006 12:02:49 -0700$ From: "AEF" <spamsink2001@yahoo.com>5 Subject: Re: Question about the 497.1-day-uptime bug. B Message-ID: <1153335769.610034.155580@75g2000cwc.googlegroups.com>   JF Mezei wrote:  > AEF wrote: > > 
 > > Hello! > > J > > In VAX/VMS V6.2, there is the 497.1-day-uptime bug. Does anyone happenG > > to know if V6.1 also has this bug? I looked on the Web but couldn't < > > find anything. A description of the bug follows. Thanks! > > + > > Problems addressed in VAXSYSA02_062 kit  > > F > > o  If a system has been up for 497.1 days without  rebooting,  theF > >    system  cell  EXE$GL_ABSTIM_TICS (number of 10 millisecond tics$ > >    since  boot)  will  overflow. >  > G > Not sure about that particular bug, but there is a most excellent FAQ 1 > item on the care and feeding of VAX TOY clocks.   ' [more about the VAX TOY clocks omitted]   F I know that one. I run SET TIME between Jan 1 and Apr 1 on each VAX to/ take care of that. This is something different.    ------------------------------  % Date: Wed, 19 Jul 2006 21:58:57 -0400 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>1 Subject: Re: Tomcat user authentication question. . Message-ID: <CrBvg.70275$fG3.27290@dukeread09>   Dave McNeil wrote:H > We are running an OpenVMS 7.3-2 on an Alpha, with CSWS 2.0  and Tomcat > 5.5.9.  @ > I am trying to implement user authentication with Tomcat (JavaA > servlets) and I would like to use a similar method where Tomcat 5 > validates their web logon against their VMS userid.  > D > I'm hoping that HP made some OpenVMS specific tags for web.xml for1 > BASIC authentication, but I haven't found them.   C Write a java class that extends org.apache.catalina.realm.RealmBase > and use JNI to call some C code that do the validation and add that to server.xml.    Arne   ------------------------------  % Date: Wed, 19 Jul 2006 17:04:46 -0400 ) From: "Rick Barry" <richard.barry@hp.com> 1 Subject: Re: Tomcat user authentication question. , Message-ID: <44be9e6f$1@usenet01.boi.hp.com>  K CSWS_JAVA (Tomcat) does not yet contain a built-in authentication mechanism I for OpenVMS. You could architect such a mechanism using JAAS and Tomcat's G JAAS Realm plug-in, but you'll need to do a bit of coding using the JNI C interface. We may add this to a future version of CSWS_JAVA. In the $ meantime, you need to roll your own.  
 Rick Barry OpenVMS System Software Group  Hewlett Packard Company   4 "Dave McNeil" <dave.mcneil@bell.ca> wrote in message= news:1153238326.433552.109210@m79g2000cwm.googlegroups.com...  > Hello all, > H > We are running an OpenVMS 7.3-2 on an Alpha, with CSWS 2.0  and TomcatI > 5.5.9.  Some of our static html pages require user authentication which I > Apache handles by validating their web logon with their VMS userid.  An  > example from httpd.conf: > & > <Directory "/apache$documents/dave"> > AuthType Basic > AuthAuthoritative On* > AuthName "ICSIS Bell-ATS Authentication" > AuthOpenVMSUser On > AuthOpenVMSAuthoritative On  > require valid-user > </Directory> > @ > I am trying to implement user authentication with Tomcat (JavaA > servlets) and I would like to use a similar method where Tomcat 5 > validates their web logon against their VMS userid.  > D > I'm hoping that HP made some OpenVMS specific tags for web.xml for1 > BASIC authentication, but I haven't found them.  > * > Does anyone know how to accomplish this? >  > Dave >    ------------------------------  % Date: Wed, 19 Jul 2006 14:10:23 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> N Subject: Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing), Message-ID: <44BE7585.F609E138@teksavvy.com>   DaveH wrote:H > was under development. OpenVMS will be a Guest O/S on the HP-UX kernelI > (Hypervisor), just like Linux and HP-UX now. We are striving to deliver / > OpenVMS VM in 2H07 as per the latest Roadmap.   ! Thank you for your clarification.   H However, this still leaves the "cell technology" issue unanswered.  WillG cell based computers allow only one instance of an OS, with the ability F to add/remove CPUs for that one instance ? Or will they allow multiple% instances of an OS (or multiple OSs ?    ------------------------------    Date: 19 Jul 2006 13:52:35 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) N Subject: Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing)3 Message-ID: <a0RbKR0VlNsO@eisner.encompasserve.org>   e In article <1153324080.290828.53940@m79g2000cwm.googlegroups.com>, "Ian Miller" <ijm@uk2.net> writes: H > I think the idea is that you can slice your multi cpu box into systemsH > which can use less than one real cpu/core. I think this does look moreH > useful the unfourtunes cursed with lots of windows systems rather than > the typical VMS installation.  >  > F > I think Dave Holt's posting was the first time I have seen it statedI > that this is not intended for mission critical systems. {and if you are F > not using VMS for mission critical systems then what are you using ? > :-) }   F Software Development, where an occasional crash would not be horrible.  > I have also encountered at least one government site whose VMS> machines are currently judged "Low Impact" under FIPS 199.  Of@ course that might get reevaluated under review, but in the short< range it reduces the effort required by them under FIPS 200.   ------------------------------    Date: 19 Jul 2006 13:53:51 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) N Subject: Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing)3 Message-ID: <u7VcXMKhDAG8@eisner.encompasserve.org>   x In article <AIJ35ZdWBbYe@eisner.encompasserve.org>, clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes:d > In article <1153306981.487187.118800@i3g2000cwc.googlegroups.com>, "DaveH" <d.holt@hp.com> writes:H >> Good day, I'm the OpenVMS Business Manager for Virtualization and I'd* >> just like to clarify one or two things. >>  J >> Firstly, we are not porting Galaxy to Integrity. This decision was madeJ >> some time ago as we knew the corporate Virtual Machines (VM) technologyI >> was under development. OpenVMS will be a Guest O/S on the HP-UX kernel J >> (Hypervisor), just like Linux and HP-UX now. We are striving to deliver0 >> OpenVMS VM in 2H07 as per the latest Roadmap. >>  I >> Secondly, we are not recommending deploying buscrit apps on VMs of any I >> description. We know OpenVMS customers will be concerned about running / >> their businesses on top of any type of Unix.  > M > So the obvious question to ask is what is the expected IA64 growth path for * > current business critical Galaxy users ? > K > For the record, I am not personally in that situation, but a satisfactory I > answer seems important in order to ensure a long term viable future for  > all VMS users.  E Presumably HP knows how many customers are actually in this situation C of needing to share hardware between mission critical VMS and other A operating systems.  Perhaps that knowledge affected the decision.    ------------------------------    Date: 19 Jul 2006 15:49:11 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) N Subject: Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing)3 Message-ID: <qaNR1gfLz$I7@eisner.encompasserve.org>   b In article <1153306981.487187.118800@i3g2000cwc.googlegroups.com>, "DaveH" <d.holt@hp.com> writes:G > Good day, I'm the OpenVMS Business Manager for Virtualization and I'd ) > just like to clarify one or two things.  > I > Firstly, we are not porting Galaxy to Integrity. This decision was made I > some time ago as we knew the corporate Virtual Machines (VM) technology H > was under development. OpenVMS will be a Guest O/S on the HP-UX kernelI > (Hypervisor), just like Linux and HP-UX now. We are striving to deliver / > OpenVMS VM in 2H07 as per the latest Roadmap.   H    This reminds me of the decision to drop Tru64 UNIX and put TruClusterG    features into HP-UX.  Which was then dropped in favor of some third  I    party product.  Now the end customer does not have what they wanted:   +    a continuation of UNIX with TruClusters.   H    What does the customer using Galaxy on Alpha want?  A continuation of    VMS with Galaxy.   C    HP business managers seem to have a habit of following a line of D    replacing world class capabilities inherited from DEC with "good F    enough" capabilities developed at HP and then dropping the customer    all together.  I    This does not bode well for the future of VMS, which is a bad business G    decision.  Customers like us will not stay with HP if we are treated G    that way.  We can get "good enough" from vendors who haven't dropped     the software we use.   D    What business case has been made that chasing away customers on a7    profitable software product looks like a good thing?    ------------------------------  % Date: Wed, 19 Jul 2006 18:18:46 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> N Subject: Re: VMS and HPVM (was: Parsec webinar (2006-07-12) OpenVMS Licensing)< Message-ID: <44beaef9$0$18476$9a6e19ea@news.newshosting.com>  I "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message  - news:qaNR1gfLz$I7@eisner.encompasserve.org... M > In article <1153306981.487187.118800@i3g2000cwc.googlegroups.com>, "DaveH"   > <d.holt@hp.com> writes: H >> Good day, I'm the OpenVMS Business Manager for Virtualization and I'd* >> just like to clarify one or two things. >>J >> Firstly, we are not porting Galaxy to Integrity. This decision was madeJ >> some time ago as we knew the corporate Virtual Machines (VM) technologyI >> was under development. OpenVMS will be a Guest O/S on the HP-UX kernel J >> (Hypervisor), just like Linux and HP-UX now. We are striving to deliver0 >> OpenVMS VM in 2H07 as per the latest Roadmap. > I >   This reminds me of the decision to drop Tru64 UNIX and put TruCluster G >   features into HP-UX.  Which was then dropped in favor of some third H >   party product.  Now the end customer does not have what they wanted:, >   a continuation of UNIX with TruClusters. > I >   What does the customer using Galaxy on Alpha want?  A continuation of  >   VMS with Galaxy. > D >   HP business managers seem to have a habit of following a line ofD >   replacing world class capabilities inherited from DEC with "goodG >   enough" capabilities developed at HP and then dropping the customer  >   all together.  > J >   This does not bode well for the future of VMS, which is a bad businessH >   decision.  Customers like us will not stay with HP if we are treatedH >   that way.  We can get "good enough" from vendors who haven't dropped >   the software we use. > E >   What business case has been made that chasing away customers on a 8 >   profitable software product looks like a good thing? >    I agree with you.   L Doing the HP-VM thing is OK for people who want that product but the recent K revelation that "Galaxy is not being ported to Itanium" makes me wonder if  L HP is committed to supporting OpenVMS. After reading this (and also hearing L that Intel management is now more focused on loss of market share to AMD) I E now wonder whether the whole Itanium thing was ever worth the effort.     
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html: http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  # Date: Wed, 19 Jul 2006 19:20:00 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com> Subject: Re: XML on Expat 0 Message-ID: <ABvvg.791$fL4.475@news.cpqcorp.net>   Barry wrote:E > (I figured it's better to piggyback on this one than to start a new  > topic) > H > I'm aware of the XML parser for VMS from Expat.  I recently downloadedD > the files and used the supplied descrip.mms file to build the OLB.H > However, I haven't been able to figure out what else has to be done inF > order to build an executable that will in fact parse an XML file.  IF > have 15 years experience with VMS (and MMS) so that's not a problem.H > If anyone can tell me what's necessary to get a runnable program builtE > from the Expat sources I'll gladly create an MMS file to do it all.   F    If you don't specifically need the expat XML parser, a port of the H libxml2 environebt is available at the Freeware V8.0 staging area, FWIW.  ( <http://www.hp.com/go/openvms/freeware/> <http://www.xmlsoft.org/>   1 The Freeware port includes build procedures, etc.   E I have finished a libxml2 2.6.24 port not to long ago, with yet more  E pieces, and that'll be the newest version that ships on the Freeware   V8.0 distro.   --  H And one question I ask myself when dealing with my own or the makefiles I of others (and whether based on DECset mms, mmk, or otherwise) centrally  H involves the speed of a brute-force build on newer hardware as compared C with the effort involved in maintaining and debugging and using an  H incremental build.  Incremental builds are obviously beneficial and can I be really speedy, but brute-force builds on the hardware I regularly use  G can be very fast, too, and are obviously rather less involved than the  G makefiles, and the brute-force build can obviously entirely avoid some  I of the "fun" of the more subtle bugs that I and others have occasionally  G managed to introduce into our makefiles.  There are always trade-offs,  
 of course.   ------------------------------   End of INFO-VAX 2006.400 ************************              sϧas6Xջ2oo;!@wMRd?Ś_);|ɠg6VߗUY1FnNp)'?h,X.P(ض#[Aqw(6*y	 | e@NCriZ`3VWӤ[AjB4(ZDnev0Nulbfu&	KWA(
0\Rvo$J+&'_?{I(8y&c庼MIZbP[h&b
v+@ie#ٳeα'TDgeT>f흒i	)V0k/ a[)sDBzRCo+BW}Vp#ZGRZt3ߺJSޣԅeCl\ 
Xʜ
&D90^0=EUbâu|݄
4ƒ/QJ.