Re: Process stats wrong under ULE

From: Ian Freislich (ianf_at_za.uu.net)
Date: 07/15/03

  • Next message: Tinderbox: "[-CURRENT tinderbox] failure on i386/i386" 
    To: Kris Kennaway <kris@obsecurity.org>
Date: Tue, 15 Jul 2003 09:08:24 +0200

    Kris Kennaway wrote:
    > I'm seeing the following kind of behaviour under ULE on a UP machine
    > (kernel updated earlier this evening). Notice that the total CPU%
    > adds up to way more than 100%; indeed one single process is allegedly
    > using more than 100% CPU, and (not clear from the top(1) output) the
    > processes that are sleeping do not have their CPU% updated until the
    > next time they run.

    Jeff is aware of this and has said it is on his list of things to
    fix. Not sure where on his list it is though.

    Ian
    _______________________________________________
    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: Tinderbox: "[-CURRENT tinderbox] failure on i386/i386"

    Relevant Pages

    • Re: Is this code MT-Safe?
      ... > (context switches are still possible when a critical section is entered) ... > a single process. ... > kernel call will be made and the thread can continue. ... > contention (another thread has already entered the same critical ...
      (comp.lang.cpp)
    • Re: ULE vs. 4BSD in RELENG_7
      ... single process that's multi-threaded). ... as it shows an N-process workload on an ... N-processor system is the same between ULE and 4BSD, ...
      (freebsd-performance)
    • Re: binaries not portable from Debian to RedHat
      ... >> multiple KSEs are actually a single process. ... >> sufficient details of the kernel and threading libraries (and anything ... > the information given by the kernel. ...
      (comp.os.linux.development.apps)
    • Re: [SLE] maximum nproc value
      ... Compiling the kernel _requires_ just a single process, ... the kernel to stress a machine, one way is to build it with 'make -j' ...
      (SuSE)
    • Re: ULE vs. 4BSD in RELENG_7
      ... single process that's multi-threaded). ... as it shows an N-process workload on an ... N-processor system is the same between ULE and 4BSD, ...
      (freebsd-performance)