videod fatal error, no devices found, Where have they gone???? - SGI

This is a discussion on videod fatal error, no devices found, Where have they gone???? - SGI ; From what i can find on here, videod errors, no devices found only seem to happen when new hardware is installed. We have not changed the graphics card on our box. What we tried to do, was use a Fiber ...

+ Reply to Thread
Results 1 to 7 of 7

Thread: videod fatal error, no devices found, Where have they gone????

  1. videod fatal error, no devices found, Where have they gone????

    From what i can find on here, videod errors, no devices found only
    seem to happen when new hardware is installed. We have not changed the
    graphics card on our box. What we tried to do, was use a Fiber KVM, to
    extend the monitor.
    So, we had the onyx2 box working, unpluged the monitor, and ran it
    thru the KVM fiber to test. Didnt work, no signal, so plugged monitor
    directly back to onyx2, no problems, worked fine.

    So, we rebooted box to see if that would get the KVM to work, plugged
    the KVM in bewteen box and monitor, waited for reboot, and after
    wards, still no signal, so plugged monitor back into box, and oh dear,
    problems, so signal
    Rebooted box once more, with monitor directly attached. Display was
    working, got the boot up screen, and the screen with the box that has
    all the start up messages, then it went to what should be the login
    screen and all we got was a black screen, with the mouse icon changed
    to a small watch.

    logged in over the network and this is what syslog has to say

    Feb 24 12:46:42 6Anyx3 unix: IR0: ARM: Welcome to ARMLand - 0/0x0d00
    Feb 24 12:46:42 6Anyx3 unix: IR0: ARM: running...(sherwood-root
    0307070344)
    Feb 24 12:46:42 6Anyx3 unix: IR0: ARM:
    ************************************************** ****
    Feb 24 12:46:42 6Anyx3 unix: IR0: ARM: * InfiniteReality/Reality
    Software, IRIX 6.5 release *
    Feb 24 12:46:42 6Anyx3 unix: IR0: ARM:
    ************************************************** ****
    Feb 24 12:46:50 6Anyx3 unix: IR0: GE2: Welcome to GELand!
    Feb 24 12:46:50 6Anyx3 unix: IR0: GE3: Welcome to GELand!
    Feb 24 12:46:50 6Anyx3 unix: IR0: GE0: Welcome to GELand!
    Feb 24 12:46:50 6Anyx3 unix: IR0: GE1: Welcome to GELand!
    Feb 24 12:46:51 3Dnyx3 videod[1487]: Fatal server error:
    Feb 24 12:46:51 3Dnyx3 videod[1487]: no devices found
    Feb 24 12:46:51 3Dnyx3 videod[1487]:
    Feb 24 12:47:55 4Anyx3 unix: |$(0x836)WARNING: IR0: Timeout waiting
    for fifo to drain (level == 0x70c).
    Feb 24 12:47:55 4Anyx3 unix: |$(0x842)WARNING: IR0: Fatal error.
    bdata = 0xa800000000b28000 death_reason = 6
    Feb 24 12:47:55 6Anyx3 unix: Graphics FIFO depth is 1804 words.

    some more output

    $ /usr/gfx/gfxinfo
    Graphics board 0 is "KONAL" graphics.
    Unmanaged 1280x1024
    Display has 2 channels
    4 GEs (of 4), occmask = 0x0f
    4MB external BEF ram, 32bit path
    1 RM10 board (of 1) 1/0/0/0
    Texture Memory: 256MB/-/-/-
    Medium pixel depth
    32K cmap
    GVO option detected
    Xvc info not available for unmanaged boards

    $ hinv
    4 400 MHZ IP27 Processors
    CPU: MIPS R12000 Processor Chip Revision: 3.5
    FPU: MIPS R12010 Floating Point Chip Revision: 3.5
    Main memory size: 1024 Mbytes
    Instruction cache size: 32 Kbytes
    Data cache size: 32 Kbytes
    Secondary unified instruction/data cache size: 8 Mbytes
    Integral SCSI controller 0: Version QL1040B (rev. 2), single ended
    Disk drive: unit 1 on SCSI controller 0
    CDROM: unit 6 on SCSI controller 0
    Integral SCSI controller 1: Version QL1040B (rev. 2), single ended
    IOC3/IOC4 serial port: tty1
    IOC3/IOC4 serial port: tty2
    IOC3/IOC4 serial port: tty3
    IOC3/IOC4 serial port: tty4
    IOC3 parallel port: plp1
    Graphics board: InfiniteReality3
    Integral Fast Ethernet: ef0, version 1, module 1, slot io1, pci 2
    Iris Audio Processor: version RAD revision 7.0, number 1
    Origin BASEIO board, module 1 slot 1: Revision 4
    IOC3/IOC4 external interrupts: 1


    please help!!!!

  2. Re: videod fatal error, no devices found, Where have they gone????

    On 24 Feb 2004 10:17:17 -0800, paulb2791@yahoo.com (Paul) wrote:

    >Rebooted box once more, with monitor directly attached. Display was
    >working, got the boot up screen, and the screen with the box that has
    >all the start up messages, then it went to what should be the login
    >screen and all we got was a black screen, with the mouse icon changed
    >to a small watch.


    Oh oh, I'd try the magic lifesaver: restart, go to the prom and type
    'resetenv' and 'reboot'.

    Most really weird problems have a nice chance of getting fixed that
    way. I sometime get non-booting HD's that have their bootfile env
    param scrambled, that sort of killing-the-prom scary stuff...

    Hope that helps,
    Ruud van Gaal
    Free car sim: http://www.racer.nl/
    Pencil art : http://www.marketgraph.nl/gallery/

  3. Re: videod fatal error, no devices found, Where have they gone????

    _Damn. didnt work

    Still no luck

    Here is what happens, screen wise, after the startup messages, screen
    goes black, mouse icon becomes a red cross, then changes to a white
    watch.
    At that time, syslog is this
    Feb 25 10:06:24 3Dnyx3 videod[1470]: Fatal server error:
    Feb 25 10:06:24 3Dnyx3 videod[1470]: no devices found
    Feb 25 10:06:24 3Dnyx3 videod[1470]:

    Then if i move the mouse, after a few seconds, it freezes in one
    place, and i then get this error in syslog

    Feb 25 10:07:28 4Anyx3 unix: |$(0x836)WARNING: IR0: Timeout waiting
    for fifo to drain (level == 0x70c).
    Feb 25 10:07:28 4Anyx3 unix: |$(0x842)WARNING: IR0: Fatal error.
    bdata = 0xa800000001328000 death_reason = 6
    Feb 25 10:07:28 6Anyx3 unix: Graphics FIFO depth is 1804 words.

    Someelse suggested a factory reset of the card, but gave me
    instructions using /usr/dmedia/bin/DIVO, when i have no /usr/dmedia
    directory.


    I would love to just be able to "reset" this card.

  4. Re: videod fatal error, no devices found, Where have they gone????

    Paul wrote:
    > _Damn. didnt work
    >
    > Still no luck
    >
    > Here is what happens, screen wise, after the startup messages, screen
    > goes black, mouse icon becomes a red cross, then changes to a white
    > watch.
    > At that time, syslog is this
    > Feb 25 10:06:24 3Dnyx3 videod[1470]: Fatal server error:
    > Feb 25 10:06:24 3Dnyx3 videod[1470]: no devices found
    > Feb 25 10:06:24 3Dnyx3 videod[1470]:
    >
    > Then if i move the mouse, after a few seconds, it freezes in one
    > place, and i then get this error in syslog
    >
    > Feb 25 10:07:28 4Anyx3 unix: |$(0x836)WARNING: IR0: Timeout waiting
    > for fifo to drain (level == 0x70c).
    > Feb 25 10:07:28 4Anyx3 unix: |$(0x842)WARNING: IR0: Fatal error.
    > bdata = 0xa800000001328000 death_reason = 6
    > Feb 25 10:07:28 6Anyx3 unix: Graphics FIFO depth is 1804 words.
    >
    > Someelse suggested a factory reset of the card, but gave me
    > instructions using /usr/dmedia/bin/DIVO, when i have no /usr/dmedia
    > directory.
    >
    >
    > I would love to just be able to "reset" this card.


    /usr/gfx/stopgfx
    /usr/diags/IR/bin/irsaudit -c

    I suspect something dead on the GE or further down the pipeline at the
    RM. See what irsaudit says anyway.

    Cheers,
    Simon


  5. Re: videod fatal error, no devices found, Where have they gone????

    I am wondering now if videoD is even necessary? I assumed it was,
    because the other Onyx box we have has it turned on in chkconfig, but
    then, the other box is using the DIVo stuff, which this box isnt.
    ( as an asside, I am not an irix admin (suprise). HP-UX is my game, we
    have maybve 250 HP servers, and 2 irix, hence the confusion)
    videod has no man page, so is it even a problem?

    but also, i still have no display

    working ( slowly) with SGI support who gave me this


    ./ireeprom -w -F -f /usr/gfx/ucode/KONA/tport.bin

    which worked, in that gfxinfo went from

    Graphics board 0 is "KONAL" graphics.
    Unmanaged 1280x1024
    Display has 2 channels
    4 GEs (of 4), occmask = 0x0f
    4MB external BEF ram, 32bit path
    1 RM10 board (of 1) 1/0/0/0
    Texture Memory: 256MB/-/-/-
    Medium pixel depth
    32K cmap
    GVO option detected
    Xvc info not available for unmanaged boards

    to

    Graphics board 0 is "KONAL" graphics.
    Managed (":0.0") 1280x1024
    Display has 2 channels
    4 GEs (of 4), occmask = 0x0f
    4MB external BEF ram, 32bit path
    1 RM10 board (of 1) 1/0/0/0
    Texture Memory: 256MB/-/-/-
    Medium pixel depth
    32K cmap
    GVO option detected
    Channel 0:
    Origin = (0,0)
    Video Output: 1280 pixels, 1024 lines, 60.00Hz
    (1280x1024_60.vfo)
    Channel 1:
    Channel's Source Size: 1280 pixels, 1024 lines; Origin =
    (0,0)
    Video Output: 640 pixels, 486 lines, 29.97Hz
    (640x486_30i.vfo)


    But still no display, and a reboot takes us back to square1, no
    display and unmanaged baord in gfxinfo.

    I am thinking XDm might be the problem. When the screen goes black it
    is not completly black, in top left there is a white square, very
    small, looks like a dot. I read someone on groups.google of someone
    having problem where the kernel had one resolution, but ARM had
    another much tinier 6x7 resolution and that was causing problems.

    I am now, with videod turned off, not seeing any errors in syslog.
    last entries on a reboot are
    Feb 27 11:28:47 6Anyx3 unix: IR0: ARM:
    ************************************************** ****
    Feb 27 11:28:47 6Anyx3 unix: IR0: ARM: * InfiniteReality/Reality
    Software, IRIX 6.5 release *
    Feb 27 11:28:47 6Anyx3 unix: IR0: ARM:
    ************************************************** ****
    Feb 27 11:28:55 6Anyx3 unix: IR0: GE2: Welcome to GELand!
    Feb 27 11:28:55 6Anyx3 unix: IR0: GE3: Welcome to GELand!
    Feb 27 11:28:55 6Anyx3 unix: IR0: GE0: Welcome to GELand!
    Feb 27 11:28:55 6Anyx3 unix: IR0: GE1: Welcome to GELand!


    but still no log in screen, and mouse has chnaged from the white wrist
    watch thing to a square, an inch by an inch, that looks like corrupted
    graphics.
    xdm-errors says this

    xdm error (pid 1427): /usr/bin/X11/xrdb: Connection refused
    /usr/bin/X11/xrdb: Can't open display ':0'


    irsaudit fails with these errors, so maybe it does look like a
    hardware problem



    TRCE Run internal BIST on RM #0 ASICs
    INFO Internal BIST failure for tfp in ram
    INFO utop.uib.urbusinf.rbus_fifo during forcing BIST
    failure
    INFO Internal BIST failure for tfq in ram
    INFO utop.uib.urbusinf.rbus_fifo during forcing BIST
    failure
    DIAG Replace RM board #0; repair by reworking flawed
    ASICs.
    DIAG Replace faulty RM boards; BEFORE replacing any
    ASICS,
    DIAG ensure clocks on the RM are running properly
    (like the
    DIAG TBUS clocks for the PG & TG failures). Faulty
    or not
    DIAG running clocks _will_ cause BIST failures.
    Otherwise,
    DIAG repair by reworking flawed ASICs.
    CODE rm_intbist BF R PNTM-030-1588-001:SN-MFS119(0000.0059.a382)
    CODE rm_intbist BF R PNRM-030-1402-001:SN-MHA829(0000.0059.9d0e)
    RSLT rm_intbist FAIL Number of BIST failures: 2
    INFO Maximum error count (1) reached
    META ITERATION=1 PASSES NON-PASSES
    META reset 1 0
    .....
    META rm_intbist 0 1
    META TOTAL 36 1
    CODE SUM Hardware error


    TRCE RUNNING MinireadR test...
    INFO ODMA FAILED on Primary Channel.
    INFO ERROR! PIX: 000 Exp:11111111 Recv:00000000
    ......
    INFO ERROR! PIX: 019 Exp:aaaaaaaa Recv:00000000
    INFO Total Non-CRC errors = 940. Set MAX_ERRORS to see
    more.
    INFO ERROR: MinireadR test failed, 941 non-crc errs
    INFO ERROR: MinireadR test failed, 310 errs
    INFO Only displaying 5 errors. Set MAX_CRC_ERRS=x to
    see more
    DIAG **** ERROR is most likely on the IMP or TF, NOT
    the PG
    DIAG or the TG ****
    CODE miniread BF R PNTM-030-1588-001:SN-MFS119(0000.0059.a382)
    CODE miniread BF R
    PNRM-030-1402-001:SN-MHA829(0000.0059.9d0e)@IMP0,IMP1,IMP2,IMP3,IMP4,IM P5,IMP6,IMP7,IMP8,IMP9,IMP10,IMP11,IMP12,IMP13,IMP 14,IMP15,IMP16,IMP17,IMP18,IMP19
    RSLT miniread FAIL

  6. Re: videod fatal error, no devices found, Where have they gone????

    Paul wrote:

    > I am wondering now if videoD is even necessary?


    Nope. If it isn't started, video library apps will start it up
    when it's needed - but starting it up at boot time will hide
    some latency.

    It doesn't pertain to graphics, BTW, but to real
    PAL/NTSC/CCIR601/SD/HD video devices (DIVO, DMx, etc.).

    --
    Alexis Cousein Senior Systems Engineer
    alexis@sgi.com SGI/Silicon Graphics Brussels

    Nobody Expects the Belgian Inquisition!


  7. Re: videod fatal error, no devices found, Where have they gone????

    Paul wrote:
    > I am wondering now if videoD is even necessary? I assumed it was,
    > because the other Onyx box we have has it turned on in chkconfig, but
    > then, the other box is using the DIVo stuff, which this box isnt.
    > ( as an asside, I am not an irix admin (suprise). HP-UX is my game, we
    > have maybve 250 HP servers, and 2 irix, hence the confusion)
    > videod has no man page, so is it even a problem?
    >
    > but also, i still have no display
    >
    > working ( slowly) with SGI support who gave me this
    >
    >
    > ./ireeprom -w -F -f /usr/gfx/ucode/KONA/tport.bin
    >
    > which worked, in that gfxinfo went from
    >
    > Graphics board 0 is "KONAL" graphics.
    > Unmanaged 1280x1024
    > Display has 2 channels
    > 4 GEs (of 4), occmask = 0x0f
    > 4MB external BEF ram, 32bit path
    > 1 RM10 board (of 1) 1/0/0/0
    > Texture Memory: 256MB/-/-/-
    > Medium pixel depth
    > 32K cmap
    > GVO option detected
    > Xvc info not available for unmanaged boards
    >
    > to
    >
    > Graphics board 0 is "KONAL" graphics.
    > Managed (":0.0") 1280x1024
    > Display has 2 channels
    > 4 GEs (of 4), occmask = 0x0f
    > 4MB external BEF ram, 32bit path
    > 1 RM10 board (of 1) 1/0/0/0
    > Texture Memory: 256MB/-/-/-
    > Medium pixel depth
    > 32K cmap
    > GVO option detected
    > Channel 0:
    > Origin = (0,0)
    > Video Output: 1280 pixels, 1024 lines, 60.00Hz
    > (1280x1024_60.vfo)
    > Channel 1:
    > Channel's Source Size: 1280 pixels, 1024 lines; Origin =
    > (0,0)
    > Video Output: 640 pixels, 486 lines, 29.97Hz
    > (640x486_30i.vfo)
    >
    >
    > But still no display, and a reboot takes us back to square1, no
    > display and unmanaged baord in gfxinfo.
    >
    > I am thinking XDm might be the problem. When the screen goes black it
    > is not completly black, in top left there is a white square, very
    > small, looks like a dot. I read someone on groups.google of someone
    > having problem where the kernel had one resolution, but ARM had
    > another much tinier 6x7 resolution and that was causing problems.
    >
    > I am now, with videod turned off, not seeing any errors in syslog.
    > last entries on a reboot are
    > Feb 27 11:28:47 6Anyx3 unix: IR0: ARM:
    > ************************************************** ****
    > Feb 27 11:28:47 6Anyx3 unix: IR0: ARM: * InfiniteReality/Reality
    > Software, IRIX 6.5 release *
    > Feb 27 11:28:47 6Anyx3 unix: IR0: ARM:
    > ************************************************** ****
    > Feb 27 11:28:55 6Anyx3 unix: IR0: GE2: Welcome to GELand!
    > Feb 27 11:28:55 6Anyx3 unix: IR0: GE3: Welcome to GELand!
    > Feb 27 11:28:55 6Anyx3 unix: IR0: GE0: Welcome to GELand!
    > Feb 27 11:28:55 6Anyx3 unix: IR0: GE1: Welcome to GELand!
    >
    >
    > but still no log in screen, and mouse has chnaged from the white wrist
    > watch thing to a square, an inch by an inch, that looks like corrupted
    > graphics.
    > xdm-errors says this
    >
    > xdm error (pid 1427): /usr/bin/X11/xrdb: Connection refused
    > /usr/bin/X11/xrdb: Can't open display ':0'
    >
    >
    > irsaudit fails with these errors, so maybe it does look like a
    > hardware problem
    >
    >
    >
    > TRCE Run internal BIST on RM #0 ASICs
    > INFO Internal BIST failure for tfp in ram
    > INFO utop.uib.urbusinf.rbus_fifo during forcing BIST
    > failure
    > INFO Internal BIST failure for tfq in ram
    > INFO utop.uib.urbusinf.rbus_fifo during forcing BIST
    > failure
    > DIAG Replace RM board #0; repair by reworking flawed
    > ASICs.
    > DIAG Replace faulty RM boards; BEFORE replacing any
    > ASICS,
    > DIAG ensure clocks on the RM are running properly
    > (like the
    > DIAG TBUS clocks for the PG & TG failures). Faulty
    > or not
    > DIAG running clocks _will_ cause BIST failures.
    > Otherwise,
    > DIAG repair by reworking flawed ASICs.
    > CODE rm_intbist BF R PNTM-030-1588-001:SN-MFS119(0000.0059.a382)
    > CODE rm_intbist BF R PNRM-030-1402-001:SN-MHA829(0000.0059.9d0e)
    > RSLT rm_intbist FAIL Number of BIST failures: 2
    > INFO Maximum error count (1) reached
    > META ITERATION=1 PASSES NON-PASSES
    > META reset 1 0
    > ....
    > META rm_intbist 0 1
    > META TOTAL 36 1
    > CODE SUM Hardware error
    >
    >
    > TRCE RUNNING MinireadR test...
    > INFO ODMA FAILED on Primary Channel.
    > INFO ERROR! PIX: 000 Exp:11111111 Recv:00000000
    > .....
    > INFO ERROR! PIX: 019 Exp:aaaaaaaa Recv:00000000
    > INFO Total Non-CRC errors = 940. Set MAX_ERRORS to see
    > more.
    > INFO ERROR: MinireadR test failed, 941 non-crc errs
    > INFO ERROR: MinireadR test failed, 310 errs
    > INFO Only displaying 5 errors. Set MAX_CRC_ERRS=x to
    > see more
    > DIAG **** ERROR is most likely on the IMP or TF, NOT
    > the PG
    > DIAG or the TG ****
    > CODE miniread BF R PNTM-030-1588-001:SN-MFS119(0000.0059.a382)
    > CODE miniread BF R
    > PNRM-030-1402-001:SN-MHA829(0000.0059.9d0e)@IMP0,IMP1,IMP2,IMP3,IMP4,IM P5,IMP6,IMP7,IMP8,IMP9,IMP10,IMP11,IMP12,IMP13,IMP 14,IMP15,IMP16,IMP17,IMP18,IMP19
    > RSLT miniread FAIL


    Doesn't look too good to me - its unfortunate that you only have one RM
    board if you had had two then you could have done some board swapping to
    see if you could isolate it. Then again, if you have an SGI support
    relationship I think you might show them the irsaudit output and see if
    you can get another RM10 sent out.

    Cheers,
    Simon


+ Reply to Thread