[lkml]   [1999]   [Jan]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: [uPATCH] SMP scheduling fix (?)

    On Thu, 14 Jan 1999, Rik van Riel wrote:

    > [...] In fact,
    > you often see that 2 non-niced tasks have to share a
    > CPU while 2 niced tasks get the other CPU...

    (Do you have some (simple or complex) testcase where we fail?)

    > This is probably because the PROC_CHANGE_PENALTY is
    > _larger_ than the priority of the niced task to begin
    > with, so the niced task gets it's priority more than
    > doubled and the normal task can't get through...

    ? it doesnt matter wether it's doubled or not, being on the 'previous' CPU
    means it gets an (absolute) goodness boost. A normal task can get through
    if it was running on this CPU before too, and will be sent to / scheduled
    by another CPU if not.

    > This patch may fix the problem by giving a large bonus
    > to 'normal' processes and a smaller one to niced tasks.

    why? SMP + caching issues are the same no matter what priority the task
    has. Priority is mainly a way to control CPU-bound processes.
    (interactive processes will have maximum priority anyway)

    we already kindof penalize lower priority processes with the mechanizm you
    want to have removed:

    > It also removes the standard 'weight += p->priority'
    > because that one was a workaround for the bug in tty_ioctl.c
    > and we shouldn't need it any more now that that's fixed.

    no. the point as i understand it is to have high statical priority tasks
    getting scheduled before lowprio tasks, even if the higher priority task
    has used up most of it's timeslice. This concept is independent of SMP
    issues. We still schedule exactly along priorities though for CPU-bound

    -- mingo

    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to
    Please read the FAQ at

     \ /
      Last update: 2005-03-22 13:49    [W:0.022 / U:13.044 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site