INFO-VAX Tue, 05 Aug 2008 Volume 2008 : Issue 424 Contents: Re: HP buys EDS Re: HP buys EDS Re: HP buys EDS Re: I need help Re: I need help Monitor Utility Re: OT: Marketing OpenVMS Re: OT: Marketing OpenVMS Re: Playing with pipes Re: Playing with pipes RE: Playing with pipes Re: Python for VMS RE: Questions i need answered Re: Questions i need answered Re: Show of support for Distributed NetBeans Re: TCPIP SET CONF SMTP/OPT=TOP_HEADERS Re: TCPIP SET CONF SMTP/OPT=TOP_HEADERS Re: VAX C copying sign bit on unsigned int bit shift operations? Re: VAX C copying sign bit on unsigned int bit shift operations? ---------------------------------------------------------------------- Date: Mon, 4 Aug 2008 12:33:23 -0700 (PDT) From: AEF Subject: Re: HP buys EDS Message-ID: <126ff51f-253c-4c14-8767-bf203175dece@r66g2000hsg.googlegroups.com> On Aug 3, 2:11 pm, Arne Vajh=F8j wrote: > Bill Gunshannon wrote: > > In article <489526d0$0$18579$c3e8...@news.astraweb.com>, > > JF Mezei writes: > >> Bill Gunshannon wrote: > >>> Anybody here care to comment on wether this recent acquisition by > >>> HP will bode well for VMS? > >> HP mentality will flow to EDS, not the other way around. > > > As I said in my recent reply to David, EDS employees don't see it > > this way. And, apparently neither did the BOD of EDS who approved > > the merger. > > My guess would be that HP and EDS already have the same mentality. > > There is only one thing that counts: the bottom line. > > Arne \ Well, that is pretty much true for any company. (Not a big surprise -- really.) It reminds me of a joke I heard on the radio that an employee or executive made about his/her company's mission: "To make as much money as we can without going to jail." AEF ------------------------------ Date: Mon, 04 Aug 2008 22:13:13 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: HP buys EDS Message-ID: <4897b734$0$90274$14726298@news.sunsite.dk> John Smith wrote: > "Arne Vajhøj" wrote in message > news:4895d3a6$0$90262$14726298@news.sunsite.dk... >> They must know something since they have become as big as they are. > > A lot of companies that bid on government and large private-sector work > survive because of the self-selection processes that government/business > have. > > You have to be a certain size company in order to bid on many contracts, > irrespective of the quality, or on-time, on-budget nature of your > deliverables. This means that large IT outsourcers like EDS will continue to > get business despite the merits of smaller companies who are shut out of the > bidding process. Sure. But there are other big players than EDS. Arne ------------------------------ Date: Mon, 04 Aug 2008 22:18:24 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: HP buys EDS Message-ID: <4897b86c$0$90270$14726298@news.sunsite.dk> AEF wrote: > On Aug 3, 2:11 pm, Arne Vajhøj wrote: >> Bill Gunshannon wrote: >>> In article <489526d0$0$18579$c3e8...@news.astraweb.com>, >>> JF Mezei writes: >>>> Bill Gunshannon wrote: >>>>> Anybody here care to comment on wether this recent acquisition by >>>>> HP will bode well for VMS? >>>> HP mentality will flow to EDS, not the other way around. >>> As I said in my recent reply to David, EDS employees don't see it >>> this way. And, apparently neither did the BOD of EDS who approved >>> the merger. >> My guess would be that HP and EDS already have the same mentality. >> >> There is only one thing that counts: the bottom line. > Well, that is pretty much true for any company. (Not a big surprise -- > really.) > > It reminds me of a joke I heard on the radio that an employee or > executive made about his/her company's mission: > > "To make as much money as we can without going to jail." Yep. http://dilbert.com/strips/comic/2008-07-16/ http://dilbert.com/strips/comic/2008-07-18/ Arne ------------------------------ Date: Mon, 04 Aug 2008 16:21:18 -0400 From: "Stanley F. Quayle" Subject: Re: I need help Message-ID: <48972C7E.5323.1BCE2AF@infovax.stanq.com> On 3 Aug 2008 at 21:22, Richard B. Gilbert wrote: > > Whenever I try to run the edit program, it says must supported crt, > > also how do I get some compilers on my system? > > You get compilers on your system by using VMSINSTAL. You must purchase > or borrow the media. You must also purchase a license for each compiler > you wish to use unless you are eligible for hobbyist licenses. You can get the editor to detect your terminal type with the command SET TERM/INQUIRE. Media is pretty easy to borrow. If you're doing something commercial, you should join HP's DSPP program -- it provides free annually-renewable licenses. --Stan Quayle Quayle Consulting Inc. ---------- Stanley F. Quayle, P.E. N8SQ Toll free: 1-888-I-LUV-VAX 8572 North Spring Ct., Pickerington, OH 43147 USA stan-at-stanq-dot-com http://www.stanq.com/charon-vax.html "OpenVMS, when downtime is not an option" ------------------------------ Date: Mon, 04 Aug 2008 16:33:57 -0400 From: "Richard B. Gilbert" Subject: Re: I need help Message-ID: Stanley F. Quayle wrote: > On 3 Aug 2008 at 21:22, Richard B. Gilbert wrote: >>> Whenever I try to run the edit program, it says must supported crt, >>> also how do I get some compilers on my system? >> You get compilers on your system by using VMSINSTAL. You must purchase >> or borrow the media. You must also purchase a license for each compiler >> you wish to use unless you are eligible for hobbyist licenses. > > You can get the editor to detect your terminal type with the command SET TERM/INQUIRE. > > Media is pretty easy to borrow. If you're doing something commercial, you should join > HP's DSPP program -- it provides free annually-renewable licenses. > > --Stan Quayle > Quayle Consulting Inc. > > ---------- > Stanley F. Quayle, P.E. N8SQ Toll free: 1-888-I-LUV-VAX > 8572 North Spring Ct., Pickerington, OH 43147 USA > stan-at-stanq-dot-com http://www.stanq.com/charon-vax.html > "OpenVMS, when downtime is not an option" > > Stan, PLEASE watch your attributions and be careful how you trim things. The way the above message has been mangled, I could be mistaken for the original poster! I DID NOT WRITE "Whenever I try to run the edit program. . . . " ------------------------------ Date: Mon, 4 Aug 2008 20:12:02 -0700 (PDT) From: "James J. O'Shea" Subject: Monitor Utility Message-ID: <525817.47530.qm@web83908.mail.sp1.yahoo.com> I have an ES45 running VMS 8.3. When I type $MONITOR SYSTEM/ALL one of the rows is "Idle Time". And one of it's colum's is "MIN". But the value of that column never changes from "0.00" Can anyone tell me why? Thanks, Jim O'Shea Chicago ------------------------------ Date: Mon, 04 Aug 2008 14:25:15 -0400 From: JF Mezei Subject: Re: OT: Marketing OpenVMS Message-ID: <489749fc$0$18596$c3e8da3@news.astraweb.com> John Smith wrote: > I marked this topic as OT because I know that many people (HP employees and > management mostly) still here consider marketing OpenVMS as being > sacrilegious and not germaine to C.O.V., or to HP for that matter. When you have a dying family member, is it best to remain at his/her side all the time and tell him/her to fight on and that he/she will come out of it, or should we just leave them die in peace and just notice the signs that the death is approaching ? At one point, when it becomes apparent that HP has no intentions to prop up VMS, is there a point in fighthing to try to get HP to change its policy ? All that there is left in my opinion is to point to the signs that the death is approaching. (for instance, not publishing the critical fix to the DNS server). Even if HP has no malice against VMS, the fact is that VMS has suffered from the Palmer malaise and nobody has been willing to fix it. And HP might not realise VMS needs simple medication to fix it and bring it back to health. (medication is called MarketingRx :-). And since HP isn't doing anythng to fix VMS, VMS is going down. HP may argue that it is treating VMS the same as Tandem/HP-UX. But neither of those were made sick by the Palmer policies (continued under Curly at Compaq and Carly at HP, and now Hurd). ------------------------------ Date: Mon, 04 Aug 2008 15:39:13 -0400 From: "Richard B. Gilbert" Subject: Re: OT: Marketing OpenVMS Message-ID: JF Mezei wrote: > John Smith wrote: >> I marked this topic as OT because I know that many people (HP employees and >> management mostly) still here consider marketing OpenVMS as being >> sacrilegious and not germaine to C.O.V., or to HP for that matter. > > When you have a dying family member, is it best to remain at his/her > side all the time and tell him/her to fight on and that he/she will come > out of it, or should we just leave them die in peace and just notice the > signs that the death is approaching ? > > At one point, when it becomes apparent that HP has no intentions to prop > up VMS, is there a point in fighthing to try to get HP to change its > policy ? All that there is left in my opinion is to point to the signs > that the death is approaching. (for instance, not publishing the > critical fix to the DNS server). > > > Even if HP has no malice against VMS, the fact is that VMS has suffered > from the Palmer malaise and nobody has been willing to fix it. And HP > might not realise VMS needs simple medication to fix it and bring it > back to health. (medication is called MarketingRx :-). And since HP > isn't doing anythng to fix VMS, VMS is going down. > > HP may argue that it is treating VMS the same as Tandem/HP-UX. But > neither of those were made sick by the Palmer policies (continued under > Curly at Compaq and Carly at HP, and now Hurd). > The problem goes back long before Palmer. DEC didn't want the workstation market, they wanted to go head to head with IBM in the main frame market. IBM is still there, DEC is not! The Decline and Fall of Digital Equipment Corporation is a long and complex story of one bad move after another! If DEC had been willing and able to compete with Sun, Silicon Graphics and the other companies in the workstation market, there might still be a Digital Equipment Corporation! ------------------------------ Date: Mon, 04 Aug 2008 14:16:29 -0400 From: JF Mezei Subject: Re: Playing with pipes Message-ID: <4897482c$0$5416$c3e8da3@news.astraweb.com> EJunkermann@googlemail.com wrote: > If I do this: > > pipe myprog sys$output | rsh unixbox "cat - > xxxx" > > I get the two lots of output mixed in file xxxx on the unixbox. myprog would have to write its messages to another file. You could fopen "TT:" for instance, and fprintf to it instead of doing printf. Then, you could redefine TT to point to some logfile instead of a terminal, and then the above command would work because only the binary data would be sent to sys$output. Also, you might want to use SYS$PIPE which is the official "pipe". (sys$output gets redirected to sys$pipe). ------------------------------ Date: Mon, 4 Aug 2008 20:21:46 +0100 From: Eric Junkermann Subject: Re: Playing with pipes Message-ID: On 2008-08-04, JF Mezei wrote: > EJunkermann@googlemail.com wrote: >> If I do this: >> >> pipe myprog sys$output | rsh unixbox "cat - > xxxx" >> >> I get the two lots of output mixed in file xxxx on the unixbox. > > > myprog would have to write its messages to another file. You could fopen > "TT:" for instance, and fprintf to it instead of doing printf. Then, you > could redefine TT to point to some logfile instead of a terminal, and > then the above command would work because only the binary data would be > sent to sys$output. Darn, shouln't have called it "myprog", because it isn't mine, and I can't do that. > > Also, you might want to use SYS$PIPE which is the official "pipe". > (sys$output gets redirected to sys$pipe). Tried that (I think - too many possible combinations), but I need to send what should be sys$output somewhere else, and what should be a named file to sys$output, and do both of these at once. :( Eric ------------------------------ Date: Mon, 4 Aug 2008 16:53:42 -0400 From: "Peter Weaver" Subject: RE: Playing with pipes Message-ID: <46b401c8f674$2e6d3f90$2802a8c0@CHARONLAP> > > I have a program which writes binary data to a file specified on the > command line, and miscellaneous information to the screen. > > If I do this: > > pipe myprog sys$output | rsh unixbox "cat - > xxxx" > > I get the two lots of output mixed in file xxxx on the unixbox. > > What I would like is to send the screen output somewhere else > (anywhere else - I don't want it) - there is no space to put the > binary data on the VMS machine. > >... Have you tried pipe myprog sys$error: 2> | rsh unixbox "cat - > xxxx" SYS$OUTPUT should still go to your terminal, SYS$ERROR should go to the PIPE, but you would have to test it with your actual program to make sure it does. Peter Weaver www.weaverconsulting.ca www.openvmsvirtualization.com www.vaxvirtualization.com www.alphavirtualization.com Winner of the 2007 OpenVMS.org Readers' Choice Award for System Management/Performance ------------------------------ Date: Mon, 04 Aug 2008 18:25:42 -0700 From: Marty Kuhrt Subject: Re: Python for VMS Message-ID: Jean-François Piéronne wrote: > As requested by someone Python can, now, be used with Apache (CSWS). > One of my friend has successfully ported mod_python and mod_wsgi on > OpenVMS IA64. Some simple scripts and a more large application like > django run without any problem. > If someone want to try it, let me know. > > JFP I'd be interested in an Alpha version. If you can get me the sources and build procedure(s) used to make the IA64 version, I'd be happy to do the Alpha version. I'd really like to try the Python stuff you've done, but, alas, I only have an Alpha running Apache at the moment for a test system. Worst case, I could install WASD, but I'd rather do Apache because I don't have to explain that web server on a resume. Thanks again for all the great work you have been doing with porting and modifying open source stuff to VMS! Regards, Marty (remove the obvious to email me) ------------------------------ Date: Mon, 4 Aug 2008 17:59:54 +0000 From: "Main, Kerry" Subject: RE: Questions i need answered Message-ID: > -----Original Message----- > From: Bob Koehler [mailto:koehler@eisner.nospam.encompasserve.org] > Sent: August 4, 2008 1:23 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Questions i need answered > > In article , > gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann) writes: > > > > Again from a privileged account do > > mc authorize > > and then use the command "help". > > Please note that there is file called SYSUAF.DAT where all the users > are > > stored. You should define a logical SYSUAF to point to this file. > > If that's all he does, his users won't be happy. He really needs to > see the system manager's manual. In addition, if you want a Windows based point-click OpenVMS user /disk / queue mgmt solution, reference: (free) http://www.openvms.compaq.com/openvms/products/argus/ (see sample screen shots on web pages) Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-254-8911 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Mon, 04 Aug 2008 14:32:49 -0400 From: JF Mezei Subject: Re: Questions i need answered Message-ID: <48974bc2$0$18534$c3e8da3@news.astraweb.com> hamilton_n@encompasserve.org wrote: > 1.How do i call decwindows, I have it installed but I dont know how to > call it from the command line Normally, DECwindows starts as part of the boot sequence on a workstation. MC SYSGEN SHOW $ mc sysgen show window_system Parameter Name Current Default Min. Max. Unit Dynamic -------------- ------- ------- ------- ------- ---- ------- WINDOW_SYSTEM 1 0 0 2 Pure-numbe D $ mc sysgen SYSGEN> HELP SYS WINDOW_SYSTEM Sys_Parameters WINDOW_SYSTEM WINDOW_SYSTEM specifies the windowing system to be used on a workstation. Specify one of the following values: Value Description 1 Load the DECwindows Motif for OpenVMS workstation environment. 2 Load the UIS workstation environment. WINDOW_SYSTEM is a DYNAMIC parameter. Note that you should also incorporate this into the SYS$SYSTEM:MODPARAMS.DAT file. When you boot, the graphics console will initially be text base during the boot sequence, and you will then switch to decwidnows with a GUI login screen. If your VMS box is a server and you wish to have a Decwidnows application run on another machine, you still want WINDOW_SYSTEM set to 1 (it will load various libraries), and then, once you get to a text session: $SET DISPLAY/CREATE/TRANSPORT=TCPIP/NODE=10.11.12.13 $RUN/PROCCESS=CLOCK SYS$SYSTEM:DECW$CLOCK The MC DECW$CLOCK will pop a clock on your remote workstation. Remember that the remote system must be set to accept X windows from your system. (which si why the DECW$CLOCK is useful for testing) ------------------------------ Date: Mon, 04 Aug 2008 18:40:11 -0700 From: Marty Kuhrt Subject: Re: Show of support for Distributed NetBeans Message-ID: Craig A. Berry wrote: > > Marty Kuhrt wrote: >> IanMiller wrote: >> >>> On Jun 17, 12:08 pm, issinoho wrote: >>> >>>> Chaps, >>>> >>>> Just thought I'd show a bit of support for the Distributed NetBeans >>>> product. Not sure how many of you are using it or have tried it but it >>>> really is a terrific addition to the development options on OpenVMS. >>>> >>>> I recently finished porting a rather large commercial control system >>>> from VAX C to I64 and did it the old-fashioned character-cell way >>>> which although perfectly acceptable felt a bit archaic in this day and >>>> age. >>>> >>>> I installed and ran NetBeans and it handled the entire project >>>> flawlessly; in fact, I was editing and compiling the code base from >>>> within a modern Windows IDE back onto the VMS server. From a standing >>>> start this took all of about 3 hours and involved C, MMS and FMS. >>>> >>>> So, a slap on the back to the team involved and thanks for this. > > I would second (or umpteenth) that. And it's likely to help make the > case for a shiny new Itanium box or two in an environment I'm associated > with that has been limping along with v7.1 on an AlphaServer 2100 for > over a decade. For whatever reason, people find DEC BASIC deep, dark, > and difficult, but I think it's not having a familiar editor with syntax > highlighting that is 90% of the (perception) problem. NetBeans erases > that part of the problem. > >>> I think it's an interesting thing but it was in beta for so long that >>> NetBeans has moved on to V6. I hope they get the next version out the >>> door quicker. > > I'm not overly concerned about its being based on 5.5. Sun Studio is > based on the same thing. Most Microsoft-oriented folks are currently > using Visual Studio 2005 even though 2008 is now out. That said, it > *was* a very long field test and a bit more agility with future releases > would be better. > >> I'd also like to see more documentation on how to debug the IDE Server >> and associated stuff on the VMS side. They have some examples in the >> online docs on how to set stuff up, but only if nothing "bad" happens. >> I've been trying to get it to work on my development system without >> much luck. > > What documentation there is looks pretty good, but more is definitely > needed. A lot of the stuff that is only available in a Windows help file > format needs to be moved into the Quick Start Guide, which implies that > you can only link the source files you've compiled by writing a DCL > procedure to do so, which isn't the case at all. > > And as Marty points out, on the server side you're pretty much working > from first principles after the installation. Is there a log file where > the IDE server indicates any problems it has? Does it have any knobs > and buttons like most VMS-based server processes, where you can set a > logical name or two and get more verbose traces of what's going on? > Nothing I've seen in the docs has any hints along these lines. I > suppose you can peer into the Java bytecodes or trace the RPC traffic, > but that sounds like work. > I've been able to debug some of the stuff on my end. I altered some of the startup procedures to turn on the highest level of debugging. Since I've run VMS development tech support departments a number times in my life I always assume I'm doing something wrong. Thus, I've installed various patches and upgraded the OS on the VMS side. Since I just did that today, I have not had the chance to try it yet. One problem I ran across when doing a FTP sync build is that the FTP log on the Mac side never contains anything. I've been cranking up the logging on the VMS side to see if there is something wrong with the FTP interaction, but haven't run into any obvious problems. I do get some error messages in the extended IDE main server and IDE process server logs I have yet to work on, but I figure I'll do everything on the VMS side before I run up the white flag. I have three things going right now that are non-standard. 1. I'm running the client on a Mac. 2. I'm running Multinet. 3. I had not put all the patches into V8.3 on the test Alpha. I've finished number three. I hope don't need to change the other two. We'll see. I plan on trying another "Hello, World!" tomorrow. >> Using an FTP file system project I cannot seem to get the remote >> machine to sync. The IDE server is running and it seems to be >> responding to the diagnostics and compile requests. I can manually >> ftp from the client to the server without problem, but the ide client >> doesn't seem to be able to do it. If the files aren't sync'd then the >> compile just says "FNF". > > I've so far only done a simple "hello world" program but did not have > any of these problems. Check your Windows firewall settings or try > temporarily disabling it outright. See whether you can locate some kind > of bare-bones RPC client or test program to check out whether RPC > generally is a problem or just the IDE server. It's not at all clear > the IDE server would report access or quota troubles in a way that you > could see, so traditional auditing practices may be in order. ------------------------------ Date: Mon, 04 Aug 2008 14:09:06 -0400 From: JF Mezei Subject: Re: TCPIP SET CONF SMTP/OPT=TOP_HEADERS Message-ID: <48974672$0$5424$c3e8da3@news.astraweb.com> Phillip Helbig---remove CLOTHES to reply wrote: > OK, maybe I was confused and the change applies only to NEW messages. That is correct. I am not sure if this is for the receiver or the symbiont. But it applies only to messages that go through one or the other. Once deposited in your personal MAIL.MAI file, those messages are just text files and MAIL is unaware that they are RFC822 messages inside. Remember that you need to have top headers for POP and IMAP to function properly. With headers at bottom or no headers, POP and IMAP will synthetise an RFC822 header from the MAIL.MAI information (to, from, subject and date), and the remote mail client will treat the actual RFC822 header as part of the text message. ------------------------------ Date: Mon, 4 Aug 2008 19:12:48 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: TCPIP SET CONF SMTP/OPT=TOP_HEADERS Message-ID: In article <48974672$0$5424$c3e8da3@news.astraweb.com>, JF Mezei writes: > Phillip Helbig---remove CLOTHES to reply wrote: > > > OK, maybe I was confused and the change applies only to NEW messages. > > That is correct. I am not sure if this is for the receiver or the > symbiont. But it applies only to messages that go through one or the > other. Once deposited in your personal MAIL.MAI file, those messages are > just text files and MAIL is unaware that they are RFC822 messages inside. OK, makes sense. All seems to work OK now. > Remember that you need to have top headers for POP and IMAP to function > properly. I just use normal VMS mail (not even the DECwindows interface to it). I like the headers at the bottom since it makes it easier to page through messages in a folder. ------------------------------ Date: Mon, 4 Aug 2008 19:54:55 -0400 From: "Ed Vogel" Subject: Re: VAX C copying sign bit on unsigned int bit shift operations? Message-ID: "Fatz" wrote in message news:c34bd5ff-aac5-4f86-904e-e99abe068395@k37g2000hsf.googlegroups.com... > HP C V7.3-009 on OpenVMS Alpha V7.3-2 > OpenVMS V7.3-2 > > Hello, > > Here's a test prog I used to discover that compiling /STANDARD=VAXC > gives different treatment of unsigned ints from compiling with ANSI > C. Do you agree? I can work around the problem by separating out the > embedded operations in the initialisers of decode_bitfield and do them > as they appear in main. But thought I should show willing and post > my findings in case other people are between standards and haven't > come across this. > I can not reproduce the problem using HP C V7.3-018 on OpenVMS IA64 V8.3: Here's a cut/paste from my screen (hope it posts O.K): $ cc t/stand=vaxc $ link t $ mc []t 9673453 2 5 decode_bitfield: mask = 1111 1111 1111 1111 1111 1111 1111 1111 decode_bitfield: mask >> 25 = 0000 0000 0000 0000 0000 0000 0111 1111 decode_bitfield: mask >> 2 = 0000 0000 0000 0000 0000 0000 0001 1111 decode_bitfield: mask << 2 = 0000 0000 0000 0000 0000 0000 0111 1100 decode_bitfield: field = 0000 0000 0000 0000 0000 0000 0110 1100 decode_bitfield: object_id = 0000 0000 1001 0011 1001 1010 1110 1101 decode_bitfield: value = 27 decode_bitfield: object_id = 9673453, shift = 2, length = 5 decode_bitfield: debug - mask = 124, field = 108, value = 27 decode_bitfield: returned = 27 $ cc/ver HP C V7.3-018 on OpenVMS IA64 V8.3 Would you have CC defined to be something? Can you post a smaller reproducer? This is rather strange. Ed Vogel HP/Compaq/DEC C/C++ for OpenVMS Engineering. ------------------------------ Date: 05 Aug 2008 01:17:18 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: VAX C copying sign bit on unsigned int bit shift operations? Message-ID: <4897aa1e$0$7323$607ed4bc@cv.net> In article , "Ed Vogel" writes: > >"Fatz" wrote in message >news:c34bd5ff-aac5-4f86-904e-e99abe068395@k37g2000hsf.googlegroups.com... >> HP C V7.3-009 on OpenVMS Alpha V7.3-2 >> OpenVMS V7.3-2 >> >> Hello, >> >> Here's a test prog I used to discover that compiling /STANDARD=VAXC >> gives different treatment of unsigned ints from compiling with ANSI >> C. Do you agree? I can work around the problem by separating out the >> embedded operations in the initialisers of decode_bitfield and do them >> as they appear in main. But thought I should show willing and post >> my findings in case other people are between standards and haven't >> come across this. >> > >I can not reproduce the problem using HP C V7.3-018 >on OpenVMS IA64 V8.3: >Here's a cut/paste from my screen (hope it posts O.K): > >$ cc t/stand=vaxc >$ link t >$ mc []t 9673453 2 5 >decode_bitfield: mask = 1111 1111 1111 1111 1111 1111 1111 1111 >decode_bitfield: mask >> 25 = 0000 0000 0000 0000 0000 0000 0111 1111 >decode_bitfield: mask >> 2 = 0000 0000 0000 0000 0000 0000 0001 1111 >decode_bitfield: mask << 2 = 0000 0000 0000 0000 0000 0000 0111 1100 >decode_bitfield: field = 0000 0000 0000 0000 0000 0000 0110 1100 >decode_bitfield: object_id = 0000 0000 1001 0011 1001 1010 1110 1101 >decode_bitfield: value = 27 >decode_bitfield: object_id = 9673453, shift = 2, length = 5 >decode_bitfield: debug - mask = 124, field = 108, value = 27 >decode_bitfield: returned = 27 >$ cc/ver >HP C V7.3-018 on OpenVMS IA64 V8.3 > >Would you have CC defined to be something? >Can you post a smaller reproducer? > >This is rather strange. > >Ed Vogel >HP/Compaq/DEC C/C++ for OpenVMS Engineering. Wouldn't a copy of his listing file be most helpful? You've have all of the qualifiers and macro setting and, if he compiled to show the machine code, you could compare with yours. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM ... pejorative statements of opinion are entitled to constitutional protection no matter how extreme, vituperous, or vigorously expressed they may be. (NJSC) Copr. 2008 Brian Schenkenberger. Publication of _this_ usenet article outside of usenet _must_ include its contents in its entirety including this copyright notice, disclaimer and quotations. ------------------------------ End of INFO-VAX 2008.424 ************************