Re: 2.6.8-rc2 crashes

From: Randy.Dunlap (rddunlap_at_osdl.org)
Date: 07/27/04

  • Next message: Mikael Pettersson: "[PATCH][2.6.8-rc1-mm1] decode local APIC errors"
    Date:	Tue, 27 Jul 2004 10:32:56 -0700
    To: Gene Heskett <gene.heskett@verizon.net>
    
    

    On Tue, 27 Jul 2004 13:43:43 -0400 Gene Heskett wrote:

    | On Tuesday 27 July 2004 12:51, Randy.Dunlap wrote:
    | >On Tue, 27 Jul 2004 13:02:41 -0400 Gene Heskett wrote:
    | >| On Tuesday 27 July 2004 10:29,
    | >| viro@parcelfarce.linux.theplanet.co.uk
    | >|
    | >| wrote:
    | >| >On Tue, Jul 27, 2004 at 12:33:04PM -0400, Gene Heskett wrote:
    | >| >> Greetings everybody;
    | >| >>
    | >| >> I have now had 4 crashes while running 2.6.8-rc2, the last one
    | >| >> requiring a full powerdown before the intel-8x0 could
    | >| >> re-establish control over the sound.
    | >| >>
    | >| >> All have had an initial Opps located in prune_dcache, and were
    | >| >> logged as follows:
    | >| >> Jul 27 07:58:58 coyote kernel: Unable to handle kernel NULL
    | >| >> pointer dereference at virtual address 00000000
    | >| >
    | >| >... which means that dentry_unused list got corrupted, which
    | >| > doesn't really help. Could you try to narrow it down to
    | >| > 2.6.8-rc1-bk<day>?
    | >|
    | >| I don't have bitkeeper installed. I get my patches from
    | >| kernel.org, either from the main Linus tree, or Andrews akpm tree.
    | >| Is there someplace where I can dl a patchkit that leads up to rc1
    | >| by the "daily" snapshot? Or should I just get the 'broken out'
    | >| patch after I verify that it also occurs with rc1 of course. That
    | >| I haven't done, mainly because rc2 has a lot of usb bugfixes and
    | >| my mouse hasn't died all by itself on rc2 like it was for at least
    | >| 2 dozen patches before rc2. IIRC it did a time or 2 on rc1.
    | >
    | >In case you need daily bk snapshots, they are here:
    | > http://www.kernel.org/pub/linux/kernel/v2.6/snapshots/
    |
    | Thanks Randy, got all 6 of them.
    |
    | >| It takes a < 1 day for this to occur usually, but seems to occur
    | >| after a long session of doing something else that fails. (or maybe
    | >| causes the fail, of a make install in konstruct/meta/kde) Thats
    | >| running now, someplace in kdegraphics-3.2.92 from the looks of it.
    |
    | I take it that I should apply these to a 2.6.7 tarballs tree in this
    | order:
    | 1. 2.6.8-rc1
    >>>> 2.6.8-rc2 <<<<<

    | 2. each of these 'rc2-bk' patches by the day and then run each for a
    | couple days, or should I start in the middle, say the 3rd one and
    | work forward or backwards from there depending on the results?

    I'd suggest beginning with -bk3 and doing a binary search.

    | Your (and Viro's) call. I'd imagine you would want to run this to
    | earth as quick as we can.
    |
    | Are these patches cumulative? I presume they are as they grow by the
    | day.

    Sorry, I should have mentioned that. Yes, they are cumulative.

    --
    ~Randy
    -
    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: Mikael Pettersson: "[PATCH][2.6.8-rc1-mm1] decode local APIC errors"

    Relevant Pages

    • Re: 2.6.8-rc2 crashes
      ... I get my patches from ... >| kernel.org, either from the main Linus tree, or Andrews akpm tree. ... Additions to this message made by Gene Heskett are Copyright 2004, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: merge status
      ... are our patches in it" ... closer this tree is to what mainline will be next release, ... > practicality. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: new dev model (was Re: Default cache_hot_time value back to 10ms)
      ... >> Major developers such as David and Al don't have trees that see wide ... >> allow for stabilization of the main tree prior to relesae. ... That's 500 patches which are either non-bugfixes or minor ... 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: RFD: Kernel release numbering
      ... > some patches that belong in the 2.6.X.Y tree go straight to you/Andrew, ... Think of how the -mm tree has evolved - with me and Andrew learning how ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Fw: Re: 2.6.9-rc2-mm4
      ... On Thursday 30 September 2004 17:32, Borislav Petkov wrote: ... if 2.6.9-rc3 fails then it is not my tree... ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)