Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.BERKELEY.EDU Path: utzoo!decvax!ucbvax!JASPER.Palladian.COM!dp From: dp@JASPER.Palladian.COM (Jeffrey Del Papa) Newsgroups: mod.computers.vax Subject: vms charging Message-ID: <860925004129.8.DP@BANFF.Palladian.COM> Date: Thu, 25-Sep-86 00:41:00 EDT Article-I.D.: BANFF.860925004129.8.DP Posted: Thu Sep 25 00:41:00 1986 Date-Received: Wed, 8-Oct-86 03:58:51 EDT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: Jeffrey Del Papa Organization: The ARPA Internet Lines: 13 Approved: info-vax@sri-kl.arpa the vax hardware (780's and 750's at least) have a register that contains the time remaining to the next tick (in units of 100ns) this is (or at least could be) used to accurately calculate for charging. weather vms actualy does this is conjecture, but the harware is ready. as to charging cpu for page faults, qio's etc I remember a magic number (may be real magic on the show/special sysgen list) that was the amount of time to charge quanta (tho I never looked to see if it was acct as well) when ever the user got hit by one of these. (useful for tweaking sched params)