Re: [UPDATE] new pid allocation patch

From: Andre Oppermann (andre_at_freebsd.org)
Date: 02/05/04

  • Next message: Kevin Oberman: "Re: HEADS UP: MAJOR changes to the ports system"
    Date: Thu, 05 Feb 2004 16:41:29 +0100
    To: Jun Su <junsu@delphij.net>
    
    

    Jun Su wrote:
    > Hi All,
    >
    > John Baldwin's proc lock commit today made my patch broken, so I have
    > revised the patch, and uploaded it to the same position:
    > http://www.arbornet.org/~junsu/pid.diff
    >
    > The revised patch has been improved based on the feedback. Basically,
    >
    > 1. Added a new sysctl value kern.pidmax to control the current pid range.
    > This value can influence the reuse pid period. The initial value is 4096.

    This sysctl name looks very misleading to me. Unless the highest PID
    I can get here is actually 4096.

    > 2. Change the meaning of PID_MAX back to what it was, and sainty-check
    > whether the pid is 5 digit based on PID_MAX. (I think it is impossible that
    > pid expand PID_MAX for the algorithm nature, though.)
    >
    > 3. The initial pid table size is bumped from 1<<5 to 1<<7.

    -- 
    Andre
    _______________________________________________
    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: Kevin Oberman: "Re: HEADS UP: MAJOR changes to the ports system"

    Relevant Pages

    • Re: Call for testers: New PID allocator patch for -CURRENT
      ... Subject: Call for testers: New PID allocator patch for -CURRENT ... Your patch uses a very clever idea. ... > prevent reuse from happening too quickly in wall time. ... Reusing the proc slot doesn't mean reusing the pid. ...
      (freebsd-current)
    • Re: Call for testers: New PID allocator patch for -CURRENT
      ... Subject: Call for testers: New PID allocator patch for -CURRENT ... Your patch uses a very clever idea. ... > prevent reuse from happening too quickly in wall time. ... Reusing the proc slot doesn't mean reusing the pid. ...
      (freebsd-hackers)
    • Re: [PATCH 2/2] Replace pid_t in autofs with struct pid reference
      ... daemon we are sending it to. ... The context here is the automount daemon only for expire runs. ... here is a replacement patch which sends the pid number ...
      (Linux-Kernel)
    • [ANNOUNCE] JUne LTP release now available
      ... Fix genload in runalltests.sh, it was trying to run it in all caps, but ... Patch from Alastair McKinstry to allow LTP to build on Linux/HPPA ... Made changes to get thread ID vs get PID for NPTL threads for unique ... Changed clone02 to use tid instead of pid to eliminate failures on NPTL ...
      (Linux-Kernel)
    • Re: [PATCH][FAT] FAT dirent scan with hin take #3
      ... at the last weekend I looked into the patch you post and run it. ... same pid and dir. ... * AVG_MSDOS_SLOTS is selected so that it can store 48 chars file name length. ... struct msdos_dir_entry *de) ...
      (Linux-Kernel)