Re: atapicam breakage (was: ata breakage [STILL])

From: Bob Willcox (bob_at_immure.com)
Date: 07/13/05

  • Next message: Kenneth Culver: "Re: AMD64 + Nvidia Display Card"
    Date: Wed, 13 Jul 2005 08:57:34 -0500
    To: Jonathan Noack <noackjr@alumni.rice.edu>
    
    

    On Thu, May 12, 2005 at 02:03:44AM -0500, Jonathan Noack wrote:
    > On 05/12/05 01:36, S?ren Schmidt wrote:
    > >Jonathan Noack wrote:
    > >>I am still getting breakage from ata with sources from around midnight
    > >>CDT (after the latest updates). These result in odd messages being
    > >>printed to the screen and eventual panics before the machine mounts
    > >>root. See the attached horrific.txt for the boot messages and some
    > >>brief ddb action. Note the garbage right before the panic. The
    > >>garbage was printed out before a 30 second pause prior to the panic.
    > >>
    > >>I backed out src/sys/dev/ata to '2005-05-03 07:55:00 UTC' and
    > >>everything works perfectly. See the attached normal.txt for these
    > >>boot messages.
    > >>
    > >>The panic seems to happen during probing of the CD drive. I use
    > >>atapicam *without* atapicd. The kernel config is also attached.
    > >
    > >OK, loose atapicam from the kernel and let me know how that goes..
    >
    > Yup, everything works fine without atapicam but with atapicd. Looks
    > like atapicam is the source of the breakage.

    I'm seeing this same problem (well, the symptoms are the same) on a
    6-stable amd64 system (cvsup'd today). If I include atapicam in the
    kernel the system panics on boot (after the "ATA PseudoRAID loaded"
    message and before the "Trying to mount root ..." message). If I remove
    atapicam from the kernel everything is fine unless I kldload atapicam.
    A short while after loading atapicam (takes a few minutes to panic) it
    then panics with the same panic message:

    panic: g_read_data(): invalid length 3179371452

    Bob

    >
    > --
    > Jonathan Noack | noackjr@alumni.rice.edu | OpenPGP: 0x991D8195
    >

    -- 
    Bob Willcox                Reality is nothing but a collective hunch.
    bob@immure.com                     -- Lily Tomlin
    Austin, TX
    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
    

  • Next message: Kenneth Culver: "Re: AMD64 + Nvidia Display Card"

    Relevant Pages

    • atapicam breakage (was: ata breakage [STILL])
      ... >> printed to the screen and eventual panics before the machine mounts ... Note the garbage right before the panic. ... everything works fine without atapicam but with atapicd. ... like atapicam is the source of the breakage. ...
      (freebsd-current)
    • Re: umtx/libthr SMP fixes.
      ... > panic the system whether I use atapicam or not. ... a matter of the memory layout as to what gets stepped on. ... If the panics appear with the addition of the variable, ... probably is a memory stepping thing and a bug in some other piece of code ...
      (freebsd-current)
    • Re: umtx/libthr SMP fixes.
      ... >> there is no panic at all, with or without atapicam. ... Without DDB, it panics. ... Savecore refuses to grab the dump: ... > probably is a memory stepping thing and a bug in some other piece of code ...
      (freebsd-current)
    • Re: umtx/libthr SMP fixes.
      ... >>If you have had issues with libthr on SMP or umtx panics, ... >>This patch fixes several race conditions and other issues with umtx. ... panic the system whether I use atapicam or not. ...
      (freebsd-current)
    • Re: recent 5.1-CURRENT kernel panics on acd0 probe/attach, IBM T30
      ... email I saw the message about atapicam from Kenneth to Martin. ... Removing atapicam from my kernel configuration fixed the problem ... The cry has been that when war is declared, ...
      (freebsd-current)