home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: vmsnet.vms-posix
- Path: sparky!uunet!think.com!yale.edu!ira.uka.de!news.belwue.de!news.uni-ulm.de!news
- From: ORAKEL@rzmain.rz.uni-ulm.de (Framstag)
- Subject: Re: strange posix timing bevavior
- In-Reply-To: moeller@gwdgv1.gwdg.de's message of Sat, 7 Nov 1992 03:00:40 GMT
- Message-ID: <1992Nov8.201431.7023@wega.rz.uni-ulm.de>
- Sender: news@wega.rz.uni-ulm.de (News Net)
- Organization: University of Ulm, Germany
- References: <1992Oct31.003920.17824@wega.rz.uni-ulm.de> <2TN8ZVH@gwdu03.gwdg.de>
- Date: Sun, 8 Nov 1992 20:14:31 GMT
- X-Copyright1: Fuer Veroeffentlichungen AUSSERHALB des Newssystemes gelten MEINE Geschaeftsbedingungen.
- X-Copyright2: Nichtbeachtung wird in der Regel zivil- und strafrechtliche Folgen nach sich ziehen.
- X-News-Reader: VMS NEWS 1.24
- Lines: 21
-
- In <2TN8ZVH@gwdu03.gwdg.de> moeller@gwdgv1.gwdg.de writes:
-
- > If you DEFINE/USER, VAXC I/O will indeed occur; if you DEFINE in supervisor
- > mode, I/O to standard output (SYS$OUTPUT) will be re-directed to the
- > process-permanent file, which *is* buffered the normal way (typically 8 or
- > 16 disk blocks). From the difference in the number of direct I/Os executed,
- > the difference in elapsed time can be easily understood.
-
-
- As a general statement, can one say:
-
- "always use DEFINE SYS$OUTPUT in supervisor-mode than in user-mode (with
- additional DEASSIGN) for C-programs for better performance"
-
- Is this correct?
-
-
- \ Ulli 'Framstag' Horlacher \ psi%0262450502601::rzmain::orakel \
- \ Student consultant VAX/VMS and networks \ orakel@dulruu51.bitnet \
- \ Rechenzentrum Universitaet Ulm Germany \ orakel@rz.uni-ulm.de \
- \ "With mainframes you just can do nonsense, with PCs not even that." \
-