Re: [RFC] Patch for isolated scheduler domains

From: Ingo Molnar (mingo_at_elte.hu)
Date: 07/22/04

  • Next message: Cam: "linux-2.4.20_mvl31 OOPS ip_route_input called within ip_rt_init" 
    Date:	Thu, 22 Jul 2004 19:54:59 +0200
To: Dimitri Sivanich <sivanich@sgi.com>

    * Dimitri Sivanich <sivanich@sgi.com> wrote:

    > I'm interested in implementing something I'll call isolated sched
    > domains for single cpus (to minimize the latencies involved when doing
    > things like load balancing on certain select cpus) on IA64.
    >
    > Below I've included an initial patch to illustrate what I'd like to
    > do. I know there's been mention of 'platform specific work' in the
    > area of sched domains. This patch only addresses IA64, but could be
    > made generic as well. The code is derived directly from the current
    > default arch_init_sched_domains code.

    it looks good to me - and i'd suggest to put it into sched.c. Every
    architecture benefits from the ability to define isolated CPUs.

    One minor nit wrt. this line:

    + cpu_sd->flags &= ~(SD_BALANCE_NEWIDLE | SD_BALANCE_EXEC |
    + SD_BALANCE_CLONE); /* Probably redundant */

    i'd suggest to set it to 0. You dont want WAKE_AFFINE nor WAKE_BALANCE
    to move your tasks out of the isolated domain.

    > - Assuming boot time configuration is appropriate ('isolcpus=' in my example),
    > is allowing boot time configuration of only completely isolated cpus
    > focusing too narrowly on this one concept, or should a boot time
    > configuration allow for a broader array of configurations, or would other
    > types of sched domain configurations be addressed separately?

    i'd prefer to go with this simple solution and wait for actual usage
    patterns to materialize. If it becomes popular we can define a syscall
    to configure the domain hierarchy (maybe even the parameters) runtime.

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

  • Next message: Cam: "linux-2.4.20_mvl31 OOPS ip_route_input called within ip_rt_init"

    Relevant Pages

    • Re: [PATCH 1/2] enable remote rcu callback processing
      ... any other rcu callback delays that may be lurking, ... configuration of processors for remote rcu ... * Remote callback processing allows specified cpus to have ...
      (Linux-Kernel)
    • Re: It is almost certain now, INTEL will have 64bit x86 !!
      ... > or the differences between this and the configuration of a 8400. ... They are similar in that if you want the maximum CPU count ... maximum I/O, you could not have the maximum CPUs on the 15K. ...
      (comp.os.vms)
    • Re: try, throw, and catch Gets a Linker Error
      ... other CPUs: GetProcAddress, ... > It looks like exceptions are not supported for my configuration. ... >> Michael Salamone [eMVP] ...
      (microsoft.public.windowsce.embedded)
    • Re: [RFC] Patch for isolated scheduler domains
      ... >>things like load balancing on certain select cpus) on IA64. ... >> is allowing boot time configuration of only completely isolated cpus ... >> configuration allow for a broader array of configurations, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [RFC] cpuset: add interface to isolated cpus
      ... need to be able to change which CPUs are ... isolated from the scheduler balancing code on the fly. ... in the root cpuset directory, in a file called 'isolated_cpus', ... and that tasks are not scheduled on such isolated CPUs ...
      (Linux-Kernel)
    Loading