home *** CD-ROM | disk | FTP | other *** search
- Path: sparky.navsea.navy.mil!matuli_a
- From: matuli_a@marlin.navsea.navy.mil (Alex Matulich 03T1 602-6691)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Messages vs. Semaphores for external clocking
- Date: 8 Apr 1996 13:52:53 GMT
- Organization: Naval Sea Systems Command Code 03T1
- Message-ID: <4kb5nl$c8u@sparky.navsea.navy.mil>
- References: <4ju349$r1e@sparky.navsea.navy.mil> <4jumqg$ss7@serpens.rhein.de>
- NNTP-Posting-Host: marlin.navsea.navy.mil
-
- Michael van Elst <mlelstv@serpens.rhein.de> wrote:
- >matuli_a@marlin.navsea.navy.mil (Alex Matulich 03T1 602-6691) writes:
- >
- >>The time units number will be stored either in a Message or Semaphore.
- >>Then my primary application, running at the same time, will be able to
- >>access this time number any time it needs to.
- >
- >This depends on when the primary application needs the data, how fast
- >it has to get access and wether the data aquisition task is allowed
- >to be blocked.
-
- It needs access within 1/30 to 1/60 of a second from the time it decides
- that it needs access. From email replies I've received, the consensus
- seems to be that using semaphores might involve less overhead than
- messages, and any delays associated with either may come out the same.
-
- >For example, if the data aquisition tasks must not be blocked and
- >the primary task can handle some delay then you would use N messages.
-
- That would work, but I don't need to send a message for every data item
- acquired. The main application just has to poll the data acquisition
- appication (an external clock) whenever it needs to.
-
- Thanks.
-
- --
- /|
- Alex Matulich __. __=#|| ___ _o--
- matuli_a@marlin.navsea.navy.mil ____##_/_____|==###===###____
- \____________________________\
-