INFO-VAX Sun, 27 Apr 2008 Volume 2008 : Issue 234 Contents: Any one out there have VAX APL Manuals Re: Intel Itanium RAS Comparison with X86 Re: Intel Itanium RAS Comparison with X86 Re: Intel Itanium RAS Comparison with X86 Re: Longtime VMS system manager/programmer available Re: Longtime VMS system manager/programmer available Re: Longtime VMS system manager/programmer available Re: Longtime VMS system manager/programmer available Re: Php invented by a former DECcie ? Re: Watching P2 memory usage on an Alpha ---------------------------------------------------------------------- Date: Sat, 26 Apr 2008 13:08:45 -0700 (PDT) From: urbancamo Subject: Any one out there have VAX APL Manuals Message-ID: <583355d7-a3c2-49b5-b1c4-1db86f2166c2@e39g2000hsf.googlegroups.com> Sorry for repeating myself, but does anyone out there have a set of the VAX APL Manuals? Thanks for your time, Mark. ------------------------------ Date: Sat, 26 Apr 2008 19:02:43 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: <4813b490$0$90273$14726298@news.sunsite.dk> Bill Gunshannon wrote: > In article , > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: >> In article <6718lhF2m4eapU2@mid.individual.net>, billg999@cs.uofs.edu (Bill Gunshannon) writes: >>> Somebody tell me again how DOD is one of VMS's biggest customers! >> I'm, sorry, we were unable to ascertain need to know. > > This red herring is getting as tired as Kerry's "5-20 patches" mantra. It seems obvious to me that DoD has VMS systems. They must have hundreds of thousands of servers running. Acquired and configured during a number of decades. There has to be lot of those systems acquired between the early 80's and mid 90's. And a good portion of those system must have been VMS systems. And I don't believe that they have replaced all of them. Systems tend to keep running and running forever. > How many times do I have to tell you, I am on the inside. It is not my impression that the entire DoD is a transparent organization with good overview of everything. > I know there are still VMS systems in existance. But they > appear to be the very best examples of legacy systems. They run one > or two fixed programs. They see no new development and very limited > maintenance (probably due to a fear that anything done may result in > breaking the running application.) I think that description fit a huge portion of all remaining VMS systems. DoD or non-DoD. Arne ------------------------------ Date: Sat, 26 Apr 2008 21:17:00 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: <4813d408$0$90263$14726298@news.sunsite.dk> Bill Gunshannon wrote: > In article , > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: >> In article <673tb7F2lqjjfU2@mid.individual.net>, billg999@cs.uofs.edu (Bill Gunshannon) writes: >>> In article , >>> koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: >>>> In article <6718lhF2m4eapU2@mid.individual.net>, billg999@cs.uofs.edu (Bill Gunshannon) writes: >>>>> Somebody tell me again how DOD is one of VMS's biggest customers! >>>> I'm, sorry, we were unable to ascertain need to know. >>> This red herring is getting as tired as Kerry's "5-20 patches" mantra. >>> How many times do I have to tell you, I am on the inside. >> Tired of the facts becaudse you can't make them go away? Everyone >> one the inside knows about need to know. You job as a subject >> matter expert is limitted to those subjects they want to be able to >> ask you about. > > Bob, this is just getting funnier all the time. "Need to know" > applies to classified information. VMS is not classified. The fact that VMS is not a secret OS (even though it can be argued that DEC/CPQ/HP marketing have tried to make it so) does not imply that the existence of all system running VMS are not secret. Arne ------------------------------ Date: Sat, 26 Apr 2008 23:25:32 -0400 From: JF Mezei Subject: Re: Intel Itanium RAS Comparison with X86 Message-ID: <4813f286$0$12291$c3e8da3@news.astraweb.com> Arne Vajhøj wrote: > The fact that VMS is not a secret OS (even though it can be argued > that DEC/CPQ/HP marketing have tried to make it so) does not imply > that the existence of all system running VMS are not secret. Those secret VMS DOD systems that "don't exist" may still be bringing HP some revenu, but they don't help attract new ISVs nor do they help shore up VMS' image because they "don't exist". Those secret system might still benefit normal VMS systems though. Say the "non existant" DOD systems still pay maintenant to Tom Linden for PL1, it means that normal customers benefit from PL1 continuing to be available on Alpha/VAX VMS. IF DOD demanded PL1 on those IA64 contraptions, HP would go through hoops and loops with Intel to give Mr Linden access to the GEM backend. So lack of PL1 on IA64 is an indication that DOD isn't using PL1 on IA64. What would benefit VMS greatly is if DOD were to admit that they are still developing NEW applications on VMS. If all they got left are mature VMS systems that are just being maintained with no new VMS development of new apps/systems, then it is just a matter of time before they are eventually replaced. And systems that are just on maintanance don't generate much to the VMS ecosystems in terms of buying new applications and supporting/attracting new ISVs. ------------------------------ Date: Sat, 26 Apr 2008 22:31:06 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Longtime VMS system manager/programmer available Message-ID: <4813e566$0$90267$14726298@news.sunsite.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> And I would say that C is a language where the difference between >> good and bad programmers really show ! > > At the time C was relatively new outside of unix, it was said that > assembler programmers were old enough to remember the absolute need to > write efficient code in a very small memory footprint. My point was about robustness not efficiency. > Today, C programmers are the ones who are old enough to know to make > efficient programs without bloated structures and giganourmous memory > and CPU requirements. It the younger generation who grew up on C++, > perl, php, all the visual microsoft crap who now are so far detached > from the machines that they write bloated inefficient code and they have > no experience in needing to write efficient code since they just buy > more memory and faster CPU when their app needs it. Usually that is more cost efficient. Arne ------------------------------ Date: Sat, 26 Apr 2008 22:32:10 -0400 From: =?ISO-8859-15?Q?Arne_Vajh=F8j?= Subject: Re: Longtime VMS system manager/programmer available Message-ID: <4813e5a6$0$90267$14726298@news.sunsite.dk> Tom Linden wrote: > On Sat, 12 Apr 2008 20:59:19 -0700, Arne Vajhøj wrote: >> I am not a driver person. Far from. But as I understand it then there >> are quite a few things you can not do in a driver and there are >> documentation for C about what you can use of C without breaking >> any of the rules. Does similar documentation exist in the PL/I >> documentation or in the VMS documentation or in a book for PL/I ? >> >> Even if the docs does not exist then you may still be able to >> write a driver in PL/I because you know what it does, but that >> does not cut it as supported. > > If there is a SDL description of the interface, then it should be easy, > otherwise translating a C prototype to PL/I is trivial. I don't think writing a driver is so much a matter of using or providing an API. Arne ------------------------------ Date: Sat, 26 Apr 2008 22:34:11 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Longtime VMS system manager/programmer available Message-ID: <4813e61f$0$90267$14726298@news.sunsite.dk> Main, Kerry wrote: > From: AEF [mailto:spamsink2001@yahoo.com] >> On Apr 14, 11:14 pm, JF Mezei wrote: >>> Today, C programmers are the ones who are old enough to know to make >>> efficient programs without bloated structures and giganourmous memory >>> and CPU requirements. It the younger generation who grew up on C++, >>> perl, php, all the visual microsoft crap who now are so far detached >>> from the machines that they write bloated inefficient code and they have >>> no experience in needing to write efficient code since they just buy >>> more memory and faster CPU when their app needs it. >> Say what? Since when has this made things run acceptably fast? > > This is especially true these days with the advancement of multi-core > cpu's. Applications not written with the concept of threads, SMP, > job scheduling etc will have problems with performance no matter how > many additional CPU's you throw at the solution. But it is so much easier to write the multithreaded apps in some of the newer languages than in C ... Arne ------------------------------ Date: Sat, 26 Apr 2008 20:24:02 -0700 From: "Tom Linden" Subject: Re: Longtime VMS system manager/programmer available Message-ID: On Sat, 26 Apr 2008 19:32:10 -0700, Arne Vajhøj wrote: > If there is a SDL description of the interface, then it should be easy, > otherwise translating a C prototype to PL/I is trivial. > I don't think writing a driver is so much a matter of using or > providing an API. > I am not sure what you mean. -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Sat, 26 Apr 2008 22:19:39 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Php invented by a former DECcie ? Message-ID: <4813e2b8$0$90267$14726298@news.sunsite.dk> Richard B. Gilbert wrote: > Arne Vajhøj wrote: >> A language like Ada prevents various array index out of bounds >> and wild pointers problems resulting in garbage data being >> read, data being overwritten or crashes. > > ISTR that ADA is also strongly typed. Ada is very strongly typed. > It is not easy to store an int > into a float. It can be done but you have to first explain to the > compiler yes, I really, really, mean that. Correct. > If a function takes two > arguments, you had better supply exactly two. And so on. Most languages support that. Arne ------------------------------ Date: Sat, 26 Apr 2008 21:59:56 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Watching P2 memory usage on an Alpha Message-ID: <4813de1a$0$90267$14726298@news.sunsite.dk> Rob Brooks wrote: > VMS is Virus Free writes: >> $ write sys$output f$getjpi(the_programs_pid,"p2_first_free_va_64") >> >> This works fine until it gets to FFFFFFFF then it rolls over to >> 00000000. I know the program that it's monitoring is working because >> its page file quota continues to slowly decrement even after this >> overflow. >> >> Any thoughts or ideas? > > The problem is that DCL does not do 64-bit math, and that value is being > returned as an integer. > > I would not expect to see DCL enhanced to support 64-bit integers. I would call the example above a bug. Silently truncating 64 bit values to 32 bit is not good. Arne ------------------------------ End of INFO-VAX 2008.234 ************************