Re: qla2xxx: frequent total lockups (2.6.8, 2.6.9-rc{1-mm5,2})

From: Andrew Vasquez (andrew.vasquez_at_qlogic.com)
Date: 09/16/04

  • Next message: Christoph Lameter: "Re: device driver for the SGI system clock, mmtimer"
    Date:	Thu, 16 Sep 2004 09:30:29 -0700
    To: "Oliver M. Bolzer" <oliver@fakeroot.net>
    
    

    On Thu, 16 Sep 2004, Oliver M. Bolzer wrote:

    > I'm currently setting up a new Dual Opteron box (Tyan Transport
    > GX28) equipped with a QLogic QLA2340 fibre channel HBA.
    >
    > As soon as there is I/O load on the HBA, I start seeing
    >

    Could you provide some details on the type of I/O load?

    > I've tested and reproduced the error on the following kernels, all
    > compiled for x86_64.
    > 2.6.8.1
    > 2.6.9-rc1-mm5 (with dma_fixups patch posted by Andrew Vasquez on 13.9)
    > 2.6.9-rc2
    >

    For another datapoint, did you have any problems with earlier driver
    versions (pre b21)? 2.6.8.1 had b14k...

    > Without any I/O on the HBA (nothing mounted), I have yet to capture a
    > crash, but the driver still ocasionally reports
    > qla2300 0000:01:03.0: cmd_timeout: LOST command state = 0x6
    >
    > Any help would be greatly appreciated. If there are any tests I could
    > run, just let me know.
    >

    Hmm, could you enable some additional debug settings in the driver:

      in qla_settings.h:

        modify the following line:

            #define DEBUG_QLA2100 0 /* For Debug of qla2x00 */

        to read as:

            #define DEBUG_QLA2100 1 /* For Debug of qla2x00 */

      and in qla_dbg.h

        modify the following line:

            /* #define QL_DEBUG_LEVEL_2 */ /* Output error msgs to COM1 */

        to read as:

            #define QL_DEBUG_LEVEL_2 /* Output error msgs to COM1 */

    Rerun your test, then forward over the log.

    Regards,
    Andrew Vasquez

    -
    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: Christoph Lameter: "Re: device driver for the SGI system clock, mmtimer"

    Relevant Pages

    • Re: Incorrect driver getting loaded for Qlogic FC-HBA
      ... On 7/26/05, Greg KH wrote: ... When ever I hot-plug my HBA into the system, the driver "qla2300" gets ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: My thoughts on the "new development model"
      ... |>>> 2.6 tree is great for gentoo users who like gcc consuming all CPU ... |>> that after a month or so of fixes etc it will be a very stable kernel ... driver for 2.6.7 for an ADSL card; a development driver for 2.6.5 for a ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: sata related hang with linux-2.6
      ... IMHO there's something not quite right with the Silicon Image libata ... Perhaps the driver is enabling the hardware to generate interrupts ... before setting up the interrupt routine for it? ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: SL811 problem on mach-pxa
      ... It was tested with _both_ workarounds for IRQ issues; ... unlike the predecessors to this driver). ... I've had reports that one of the ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • [RFC] removal of legacy cdrom drivers (Re: [PATCH] mcdx.c insanity removal)
      ... bar and baz and fairly long expressions. ... if we want to keep the FPOS in the tree. ... Driver is obviously ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)