Re: SCHED_ULE and nice still ignored

From: Harald Schmalzbauer (h_at_schmalzbauer.de)
Date: 01/31/04

  • Next message: Ruslan Ermilov: "Re: make on 4.9 for 5.2 broken"
    To: Tom <tom@sdf.com>
    Date: Sat, 31 Jan 2004 23:07:34 +0100
    
    
    

    On Saturday 31 January 2004 22:53, Tom wrote:
    > On Sat, 31 Jan 2004, Harald Schmalzbauer wrote:
    >
    > ...
    >
    > > If I start a process with nice 15 (like seti) it shouldn't slow down my
    > > machine by exponetial factors.
    > > It should take cycles which are almost unused and not block regular
    > > processes (like make)
    >
    > ...
    >
    > Isn't "idprio" the best way of scheduling a process to run when the
    > system is otherwise idle? Can you try with idprio?

    I didn't know that.

    Thank you for that hint, I'll have a look at it.
    But I think SCHED_ULE shouldn't behave that different to 4BSD ragrding nice.

    Thanks,

    -Harry

    >
    > From what I know of "nice", it should reduces the processes scheduling
    > priority. There is no definition in the nice manpage on the exact
    > definition of reduced priority means.
    >
    > Tom
    > _______________________________________________
    > freebsd-current@freebsd.org mailing list
    > http://lists.freebsd.org/mailman/listinfo/freebsd-current
    > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"

    
    



  • Next message: Ruslan Ermilov: "Re: make on 4.9 for 5.2 broken"

    Relevant Pages

    • Re: SCHED_ULE and nice still ignored
      ... On Sat, 31 Jan 2004, Tom wrote: ... There is no definition in the nice manpage on the exact ... "Reduced priority" is subjective. ...
      (freebsd-current)
    • Re: SCHED_ULE and nice still ignored
      ... On Sat, 31 Jan 2004, Harald Schmalzbauer wrote: ... > It should take cycles which are almost unused and not block regular processes ... definition of reduced priority means. ...
      (freebsd-current)
    • Re: More LTC timings...
      ... > Tom St Denis wrote: ... >> My AES key schedule without the decrypt key takes about 240 cycles. ...
      (sci.crypt)
    • Re: Sandbox update
      ... > Tom St Denis wrote: ... E.g. per byte processed/byte produced. ... cycles per byte processed per byte output. ... [All on my Athlon XP with ISO C portable source code from you know where...] ...
      (sci.crypt)
    • Re: More LTC timings...
      ... Tom St Denis wrote: ... > My AES key schedule without the decrypt key takes about 240 cycles. ... encrypt/decrypt keys. ...
      (sci.crypt)