Re: [patch, 2.6.10-rc2] sched: fix ->nr_uninterruptible handling bugs

From: Ingo Molnar (mingo_at_elte.hu)
Date: 11/18/04

  • Next message: Christian Meder: "Re: [patch] Real-Time Preemption, -RT-2.6.10-rc2-mm1-V0.7.28-1" 
    Date:	Thu, 18 Nov 2004 17:21:51 +0100
To: Linus Torvalds <torvalds@osdl.org>

    * Linus Torvalds <torvalds@osdl.org> wrote:

    > And it's not just P4's either. It shows up on at least P-M's too, even
    > though that's a PPro-based core like a PIII. It's at least 22 cycles
    > according to my tests (certainly not 10), but it seems to have a
    > bigger impact than that, likely because it also ends up serializing
    > the pipeline around it (ie it makes the instructions _around_ it
    > slower too, something you don't end up seeing if you just do rdtsc
    > which _also_ serializes).
    >
    > Try lmbench some time. Just pick a UP machine, compile a UP kernel on
    > it, run lmbench, then compile a SMP kernel, and run it again. There's
    > literally a 10-20% performance hit on a lot of things. Just from a
    > _couple_ of locks on the paths.
    >
    > Sure, some of it is actual different paths, like the VM teardown,
    > which on SMP is just fundamentally more expensive because we have to
    > be more careful. So mmap latency (delayed page frees) and file delete
    > (delayed RCU delete of dentry) ends up having differences of 30-40%.
    > But the 10-20% differences are things where the _only_ difference
    > really is just the totally uncontended locking.
    >
    > That's on a P-M. On a P4 it is _worse_, I think.

    oh well. The Athlon64 really shines doing atomic ops though, and this
    fooled me into thinking that the LOCK prefix has been properly taken
    care of forever. What the hell is Intel doing ... atomic ops are so
    fundamental to spinlocks. Thank God we can do the movb $1,%0 trick for
    spin-unlock.

            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: Christian Meder: "Re: [patch] Real-Time Preemption, -RT-2.6.10-rc2-mm1-V0.7.28-1"

    Relevant Pages

    • Re: Kernel 2.4.18-3smp is a SMP kernel or non-smp kernel
      ... > If it is a smp kernel, How can I write a Makefile to compile a Device ... of other ways to help yourself prior to asking questions here. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Question on atomic_inc/dec
      ... >> use directly the kernel stuff unless you are coding kernel stuff. ... > Is there any reason NOT to use the atomic ops in user-space? ... send the line "unsubscribe linux-kernel" in ... Please read the FAQ at http://www.tux.org/lkml/ ...
      (Linux-Kernel)
    • Re: Possible dcache BUG
      ... During the compile, the above line output this warning: ... Copyright 2004 by Maurice Eugene Heskett, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Linux GPL and binary module exception clause?
      ... of them for every exported symbol in turn, ... hacked -E:) we don't need to compile for this, ... `...some suburbanite DSL customer who thinks kernel patches are some ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Compiling cardbus devices monolithic doesnt work?
      ... I should compile a monolithic 2.4 kernel and see what it does...) ... > IIRC hermes provides the low level interface to the device, ... > provides a bridge between the PCMCIA subsystem and orinoco. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)