Re: 2.6.17-rc5-mm2

On Fri, 2 Jun 2006 19:38:36 +0100 (IST)
Mel Gorman <mel@xxxxxxxxx> wrote:

This reliably goes kablam on an x86_64 machine with a tg3 network card
which was also happening for 2.6.17-rc5-mm1. A patch bisect found that
reversing git-net.patch and git-net-git-klibc-fixup.patch on top of
the -mm3 got rid of the problem. Don't ask me why.

The console log I have of the most common oops is below and the .config
used is attached. The oops happens reliably but at varying times and not
always the same oops either. Usually sshing into the machine and compiling
the kernel is enough. On at least one occasion, sshing to the machine
triggered it.

Yeah, sorry. I _knew_ LLC was buggy but I forgot to mention it in the
release notes and wasted heaps of lots of people's time. Feel free to
bill me :(

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc5/2.6.17-rc5-mm2/hot-fixes/git-net-llc-fix.patch

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

Relevant Pages

  • Re: 2.6.17-rc5-mm2
    ... Mel Gorman wrote: ... The console log I have of the most common oops is below and the .config ... The oops happens reliably but at varying times and not ... Usually sshing into the machine and compiling ...
    (Linux-Kernel)
  • Re: Microsoft SUS..
    ... I'll spare you the "get rid of them" speech. ... Oops, just did it;?) ... > Does the shavlik tool work for Win9X machines? ...
    (microsoft.public.backoffice.smallbiz2000)
  • Re: [PATCH] Remove final traces of long-deprecated "ramdisk" kernel parm.
    ... might as well finally get rid of it. ... m68k part is ... Oops, I meant ... Please read the FAQ at http://www.tux.org/lkml/ ...
    (Linux-Kernel)
  • Re: Lazy "for line in f" ?
    ... Duncan Booth wrote: ... Oops, I forgot the other obvious variant on this, which has the benefit of ... getting rid of the test I said was 'required' while still leaving the data ...
    (comp.lang.python)
  • Re: [MSH] Current provider?
    ... >> Oops, that's not the provider, one step more needed: ... > Since we're supposed to get rid of string parsing with MSH, ...
    (microsoft.public.windows.server.scripting)