Re: [PATCH] Athlon/Opteron Prefetch Fix for 2.6.0test5 + numbers

From: Nick Piggin (piggin_at_cyberone.com.au)
Date: 09/17/03

  • Next message: Rusty Russell: "Re: [PATCH] Split futex global spinlock futex_lock" 
    Date:	Wed, 17 Sep 2003 15:19:12 +1000
To: Andrew Morton <akpm@osdl.org>

    Andrew Morton wrote:

    >Nick Piggin <piggin@cyberone.com.au> wrote:
    >
    >>What is intriguing to me is the "Its only a 2% slowdown of the page
    >> fault for every cpu other than K[78] for this single workaround. There
    >> is no point to conditional compilation" attitude some people have.
    >> Of course, its only 2% on a pagefault, not anywhere near 2% of kernel
    >> performance as a whole, so maybe that is justified.
    >>
    >
    >Absolutely. But it's a bit of a pain finding a config option which says
    >"this CPU might need the fixup".
    >

    Right. It obviously can't be done using the current system.

    >
    >> Just repeating though, that is a seperate issue and I think Andi's patch
    >> is needed.
    >>
    >
    >It is unquestionably needed - the kernel _has_ to perform the fixup for this
    >CPU erratum.
    >
    >
    >But I would like to see some evidence that prefetch ever provides any
    >performance gain in-kernel. I spent some time fiddling a while back and
    >was unable to demonstrate any difference.
    >
    >
    >

    OK. I just liked this patch because apparently it fixes userspace as
    well. Disabling prefetch for the kernel doesn't.

    -
    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: Rusty Russell: "Re: [PATCH] Split futex global spinlock futex_lock"

    Relevant Pages

    • Re: [2.6 patch] dont allow users to set CONFIG_BROKEN=y
      ... the menus to look for something matching your hardware is nice, even if that something is marked BROKEN at least you've then found a place to start working on. ... or do you intend to have a bunch of patches that remove BROKEN for a config option so that people can test them during the -rc and then add it back for them all before a real release? ... nobody will be able to test that patch unless they are willing ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: log-buf-len dynamic
      ... I really didn't want to say your patch didn't make any good, ... further on the current code to make the config option still usable. ... If you can explain me _why_ you don't want to pass the kernel parameter ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH] make uselib configurable (was Re: uselib() & 2.6.X?)
      ... > Here's an initial stab at such a patch. ... It adds a new config option, ... 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/ ...
      (Linux-Kernel)
    • Re: CONFIG_4KSTACKS in mm2?
      ... here is the patch to reenable the menu config option ... ... over the universities first nuclear research site? ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: keyboard - was: Re: Linux 2.6.0-test4
      ... >> I was able to get the key unstuck by switching back and forth between ... I rebuild my kernel including your patch; ... I'll get back to you once I verify that the problem doesn't occur ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)