Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uwm.edu!psuvax1!rutgers!mcnc!ecsvax.uncecs.edu!uccjcm From: uccjcm@uncecs.edu (John McLendon) Newsgroups: comp.sys.hp Subject: HP-UX 7.0 experience Message-ID: <1990Mar1.005350.5060@uncecs.edu> Date: 1 Mar 90 00:53:50 GMT Organization: UNC Educational Computing Service Lines: 61 In article <681@mmlai.UUCP> burzio@mmlai.UUCP (Tony Burzio) writes: >In article , wayne@dsndata.uucp (Wayne Schlitt) writes: >> In article <680@mmlai.UUCP> burzio@mmlai.UUCP (Tony Burzio) writes: >> > but I truly don't believe that V7 actually exists, just the manuals >> >> and then wanted to know if they should install it. we had to tell >> them no 'cause we have no idea what, if any, problems our software >> will have on the new release. *sigh*. you would think that software >> developers could get the releases first... > >From past experiences, I expected that V7 would not break any 3rd party >software, but this seems to be incorrect this time. My sales rep and >local SE are comming by today to discuss the update (sounds >ominous). I DO know that LISP doesn't work under V7 on the 300s (of >course, from the number of bug reports that our local LISP hacker has >put in recently, I expect the problem is that LISP didn't work so well >to begin with :-) Well.... We installed it a week ago, on a third party SCSI disk (first time I've tried this!!!). Everything pretty much seems to work ok. The ANSI-C compiler is NOT included yet. The response center said to wait a month or so. We actually got uucp up and running in two days or so... Not too bad for a coupla virgins. We also managed to completely fake out the lp spooler and hang the laserjet off a cnode; no way to do this with the trailblazer. I'll be glad to publish our hack if there's enough interest... We're actually using the spooler directly! Motif is just like the 1.0 OSF distribution as far as I can tell except that the mwm cursors (hourglasss, etc..) are twice as big. Also, the mwm.h and MwmUtil.h includes are missing. The HP widget set doesn't want to handle cascades properly with mwm, but the hpwm included works just fine. I was able to bring up our old X11R2 application using the HP widget set by using a -lXtR2 compile option to cc and a -I/usr/include/XtR2 to get the old Core.h and stuff. The X windows performance is definitely better. We do get whines from the compiler about SIGCLD being redefined when including , but to no effect. Mkmf complains vociferously about not finding things of the form ../h/xxxx.h on those same includes. We also get complaints from ld of the form "Possible floating point incompatibility, recompile with +O1 option." Anyone know what this is about? We're running plain-jane 360/370's using the cc -g option. There's a scsi tape driver included in /usr/lib/drivers. Does anyone know if it works? All-in-all, I'm pleased with the better X performance. Things are running pretty smoothly. No crash and burns yet. I've got five cnodes on a single server (I know, i know). Of course, we don't use third party stuff, we write third party stuff. Obviously, I'm going for it. I expect the whines about includes to go away once I learn the incantation from the as yet unreceived PE manual set. Hope this helps... John -- Signed: John McLendon uunet\ - or - uunet!wgate.UUCP!jcm (919) 846-7931 (home) >mcnc!ecsvax.uncecs.edu!uccjcm (919) 941-5730 (play) gatech/