Re: Fedora and threading limit

From: Tom 'Needs A Hat' Mitchell (mitch48_at_sbcglobal.net)
Date: 04/30/04

  • Next message: Andreas Z.: "Re: (no subject)" 
    Date: Thu, 29 Apr 2004 22:11:21 -0700
To: For users of Fedora Core releases <fedora-list@redhat.com>

    On Wed, Apr 28, 2004 at 10:16:24AM -0700, Guolin Cheng wrote:
    ....
    > , I still can not quite understand, Since I can only use 2GB stack
    > space. (256*8MB). Not 3GB.
    >
    > The source of the thread-testing C file is from website
    > http://www.linuxquestions.org/questions/showthread.php?s=&threadid=173680.
    >
    >
    > Any other easy ways to change system-wide threading limit? At least from
    > Fedora Core 1's 256 to Redhat 8.0's 1533? Thanks A lot.

    This seems to be stack size related (memory map).

    On this FC1 box if I start with the default ulimits I get about 300
    threads. After ulimit -s 5120 I top out at 611 threads as reported by
    the test 'c' code.

    Since the kernel thread model has changed in nptl kernels I suspect
    that changes in structures to manage threads and processes have
    changed and the net memory footprint of a process will have changed.
    The 2.6.5 kernel seems to permit about 30% more threads than the
    current 2.4.22 kernel.

    Since you seem to be seeing a 256 limit and I get ~300 to start I
    wonder if there is also a physical memory computed limit component as
    well. How much DRAM?

    You may get some more threads by tinkering with link options for text,
    heap and stack but in the end this could prove to be something hard to
    fix. To me, anything with more than 100 threads seems to be in need
    of a rewrite. Like recursion threads are a marvelous model but tend
    to be stack limited. See: recursion. 

    -- 
	T o m  M i t c h e l l 
	/dev/null the ultimate in secure storage.
-- 
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list
  • Next message: Andreas Z.: "Re: (no subject)"

    Relevant Pages

    • Re: Questions about Minix
      ... from the MMU to protect processes from messing up eachother's memory ... Paging is, but segmentation is not. ... in user mode and at kernel level if you need dynamic ... A Stack Fault to be handled in Ring 0 causes a stack switch (unless the ...
      (comp.os.minix)
    • 64-bit c++ application crashing on solaris
      ... We think this could probably be due to sufficient memory not available ... Specifies the default value of the stack size to be used when a kernel ... Illegal stack size, Using N ... Gbytes for 64-bit systems) and if smaller than the minimum or larger ...
      (comp.unix.programmer)
    • Re: [Bug #11342] Linux 2.6.27-rc3: kernel BUG at mm/vmalloc.c - bisected
      ... Savings in few Mb of Kernel stack are not necessarily going to allow ... getting rid of a single memory chip of 64M or so. ...
      (Linux-Kernel)
    • Re: Questions about Minix
      ... Of course you still need to allocate chunks of contiguous memory, ... in user mode and at kernel level if you need dynamic ... I read "data segment". ... a stack from growing into the heap (only a possibility of detecting it ...
      (comp.os.minix)
    • Re: Windows CE Threading questions
      ... The next step is reclaiming RAM where possible. ... For instance, memory ... scavenging can occur to reclaim stack pages that are now out of scope. ... will start failing in the kernel prior to being completely out of ...
      (microsoft.public.windowsce.platbuilder)
    Loading