lkml.org 
[lkml]   [2000]   [Aug]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    SubjectRe: SCO: "thread creation is about a thousand times faster than on
    From
    Date
    Linus Torvalds <torvalds@transmeta.com> writes:
    > Well, CLONE_PARENT is already there. And has been for a long time.

    And it works fine. I've modified LinuxThreads in glibc-2.1.3 to use
    it (patch at <http://www.doc.ic.ac.uk/~dpw/linux/dc/>). The remaining
    loose end I have to address is to do with the setting the scheduling
    parameters for the newly created thread specified in the call the
    pthread_create(). (The scheduling parameters need to be set by the
    time pthread_create() returns, and also by the time the new thread's
    start function gets called. This can be solved by calling
    sched_setscheduler() both in the parent thread and the child thread,
    but then both threads need to wait until the other one has finished
    making the call, and the two sched_setscheduler() syscalls tend to
    contend with each other on SMP, so the performance isn't wonderful.)

    > The CLONE_WAITPID patches I don't remember, so please refresh my memory
    > about what they look like, and why they were needed in the first place..

    POSIX says that any thread can do a wait() et al. for a child process.
    But Linux currently thinks that a child processes is a child of a
    particular thread, and so only that thread can wait() for it.


    David Wragg

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

    \
     
     \ /
      Last update: 2005-03-22 12:38    [W:2.669 / U:0.228 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site