IDE errors with 2.6.0-test* and SIS5513

From: Thomas Kjær (lkml_at_lnx.dk)
Date: 10/06/03

  • Next message: Richard B. Johnson: "Re: getting timestamp of last interrupt?"
    Date:	Mon, 6 Oct 2003 21:34:01 +0200
    To: linux-kernel@vger.kernel.org
    
    
    

    Hi!

    I'm having problems with the 2.6 kernels and the SIS5513 IDE chipset.

    The following things produces the error:

    1. ripping a cd with cdparanoia.
    This causes a "Spurious 8259A interrupt: IRQ7." after a few seconds,
    and cdparanoia keep on ripping another couple of minutes after which
    the system freezes completely (ie. no mouse/keyboard in X or console).
    The system doesn't recover after this, and when I try to eject the
    still spinning CD the system just reboots. About a minute before the
    total freeze vmstat stops updating the buffer sizes, and just keeps on
    outputting the same numbers.

    2. making large filesystems with mkfs.
    I don't really have any debug output from this, since I don't have any
    large partitions free to run mkfs on. The symptoms were however the
    same as with cdparanoia.

    I don't see this problem with 2.4 kernels however, haven't tried any
    in the 2.5-series.

    I've tried running with a combination of (no) SIS5513 driver, with
    (no) DMA on the harddrives, and with (no) ACPI. None of the
    combinations solves the problem though.

    I've attached output from dmesg and lspci. Any other things I could
    try to debug this problem?

    -- 
      Thomas Kjær
    
    
    

    -
    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/


    • text/plain attachment: dmesg

    • text/plain attachment: lspci

  • Next message: Richard B. Johnson: "Re: getting timestamp of last interrupt?"

    Relevant Pages

    • Re: 2.6.4-rc2-mm1
      ... > increase while ripping with cdparanoia. ... should see system time go way down though. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: 2.6.4-rc2-mm1
      ... > Was the DMA patch for CDROMREADAUDIO reading? ... > increase while ripping with cdparanoia. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: 2.6.4-rc2-mm1
      ... >> Was the DMA patch for CDROMREADAUDIO reading? ... >> increase while ripping with cdparanoia. ... > If your CPU was fast enough you will not see a speed increase, ...
      (Linux-Kernel)
    • Re: IDE dma support for audio extraction
      ... Ripping like this relies on ATAPI. ... Cdparanoia uses very careful reading and is able to "repair" ... > audio extraction. ... Apparently Linux uses DMA to access the disk. ...
      (comp.os.linux.development.system)
    • Re: What am I doing wrong?
      ... /dev/scd0 - which i believe is the correct device for ripping audio from ... individual tracks) and therefore save worrying about cdparanoia command ...
      (comp.os.linux.setup)