lkml.org 
[lkml]   [1999]   [Jul]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    From
    Date
    SubjectRe: low priority soft RT?
    Benny Amorsen <amorsen@sscnet.com> wrote:
    > To test the effect, we would need a benchmark for which the kernel
    > currently scales badly due to lock contention, but which also does a
    > significant amount of user space processing.

    I'm not sure how this would even theoretically be able to improve
    performance, not that I think about it. If there is a task in user space
    ready to do something, then it will be done and there's no point to
    kicking another task out of kernel space (by boosting its priority). If
    there is no task in user space and all the tasks are contending for the
    kernel, then the task holding the kernel lock will be scheduled anyway,
    and there is no point to trying to artificially favor the one task that
    has the lock.

    I'm probably missing something, but I don't know what. Anyway, if it
    would be easy to extend Rik's patch to do that for all processes, it
    wouldn't be too much trouble to benchmark.

    Chris Smith <cd_smith@ou.edu>


    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 13:53    [W:3.500 / U:0.416 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site