Mac Pro (Late 2013) GPU (Driver) Issues

Discussion in 'Mac Pro' started by bax2003, Mar 28, 2015.

  1. mikeboss macrumors 65816

    mikeboss

    Joined:
    Aug 13, 2009
    Location:
    switzerland
    #901
    on my nMP the USB ports all of a sudden started to behave erratically and a few weeks later the whole I/O board died...
     
  2. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #902
    Can you please show us your working evironment with 6 x 30 ACDs :)

    I don't know why Apple used Fresco Logic USB chip instead Intel or NEC.
     
  3. Idolum macrumors member

    Joined:
    Jun 10, 2016
    #903
    How can I deny a request from the legendary thread starter? ;)

    The first picture illustrates how useful the Mac Pro is for certain setups despite its shortcomings. The seconds picture shows how two curved shower curtain rods from Home Depot make an excellent 'monitor mounting solution'. Obviously the weight of the monitors is on the tables. An old backside of a 23 inch Cinema Display functions as a platform for the MacPro to hide all the cables. I love the design of the old Apple monitors just as much as I love the design of the 2013 Mac Pro.

    IMG_5568.JPG IMG_5569.JPG
     
  4. filmak macrumors 6502a

    Joined:
    Jun 21, 2012
    Location:
    between earth and heaven
    #904
    Very nice, congratulations.:)
     
  5. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #905
    I've lost for words, FANTASTIC !
     
  6. Pardo83 macrumors member

    Joined:
    Oct 6, 2015
  7. bradl001 macrumors newbie

    Joined:
    Jul 25, 2002
    #907
    I don't know if it's the 5 monitors that impresses me or the fact that the desk isn't full of clutter like mine.

    Back to the topic at hand... I've been waiting 3 days for Apple support to respond to my latest update. Are they all out looking for Hatchimals?
     
  8. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #908
    They are busy creating FirePro emoji :D
     
  9. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #909
    Two weeks later - still no problems on El Capitan 10.11.6.
     
  10. iigsie macrumors regular

    Joined:
    Nov 17, 2007
    #910
    getting same issues as everyone else here. in July took my computer to Apple Store, they had it for 3 weeks. Initially they wanted to replace the GPU's, but someone blocked it. THey ended up replacing the ribbon cables that connect the GPU's. It's been absolutely fine for 6 months, then started crashing again a few weeks ago. This was on 10.11.6. I rolled back to Yosemite, still had the issue. Been trying Sierra, had 2 GPU crashes in the past week. Not updated to 10.12.2 yet, might as well give it a try though...


    it really can't just be drivers/software/OS, as I've had 6 months of no issues after they replaced the cables on El Capitan.
     
  11. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #911
    I partially agree with you, but one cannot ignore the fact that with the same machine has problems with 10.12, and does not with 10.11.

    I am going to put 10.12.2 on external drive and do some heavy multitasking and report.
     
  12. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #912
    I just had total system freeze in 10.12.2...end of story for me and Sierra.
     
  13. filmak macrumors 6502a

    Joined:
    Jun 21, 2012
    Location:
    between earth and heaven
    #913
    Thanks, good to know.
    So, have you used an external HD?
    This way eliminates the possibility that the SSD which is installed on the GPU has anything to do with the issues...
     
  14. koyoot macrumors 601

    koyoot

    Joined:
    Jun 5, 2012
    #914
    And also confirms, that the freezes are absolutely, completely, utterly non-reproducible...
     
  15. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #915
    External drive brand new (few days old) LaCie USB 3.0 drive with SSD inside.

    I can reproduce them at least once a day with 10.12.
     
  16. koyoot macrumors 601

    koyoot

    Joined:
    Jun 5, 2012
  17. bax2003, Dec 19, 2016
    Last edited: Dec 20, 2016

    bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #917
    Few moments before the freeze I notices graphics gitching on VMWare library GUI, the same 8.5.3 version which works perfectly on 10.11.6 for hours and days.....so...it shure wasn't about AirPort.

    Hm...
    https://discussions.apple.com/thread/7682253?start=15&tstart=0
     
  18. bbindon macrumors newbie

    bbindon

    Joined:
    Dec 21, 2016
    #918

    I am having the same problem - Have been since I got it in may 2014 with Mavericks on it. After years of back and forth with apple, they finally diagnosed the problem as GPU hangs. Replaced both GPUs.

    The problem persisted, so they replaced the whole computer a few weeks ago.

    On Sierra now - just got my first full lockup, pinwheel, hard restart.

    Not sure what to do at this point, as its clearly got to be a design flaw with the D300, or the computer itself.

    For reference, I have some USB 3 SSDs hooked up,, TB sound cards (UAD Apollo's) and also a 34" Ultrawide monitor on Display Port.

    Do owners with the D500 and D700 have this issue?
     
  19. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #919
    o_O (El Capitan 10.11.6)

    Kernel_2016-12-21-190516_Mac-Pro.gpuRestart
    Kernel_2016-12-21-190517_Mac-Pro.gpuRestart

    ******************************************************************************

    Wed Dec 21 19:05:16 2016

    Event: GPU Reset
    Date/Time: Wed Dec 21 19:05:16 2016
    Application:
    Path:
    OS Version: Mac OS X Version 10.11.6 (Build 15G1212)
    Graphics Hardware: AMD FirePro D300
    Signature: e

    Report Data:

    GPURestartReportStart
    ------------------------
    Hung Channels: GFX
    ------------------------
    [00] AccelChannel: GFX
    Pending Command from : UnknownCtx
    PendingCommandTimestamp: 0x00097878, TotalDWords: 0x000001a0, GART Offset=0x0000000406e0f100, stamp_idx=0, estamp=0x00097878
    PendingCommandStart:
    PendingCommandEnd
    ------------------------
    [00] GFX HWChannel : Enabled, NotIdle
    LastReadTimestamp : 0x00097877
    NextSubmitTimestamp : 0x00097888
    [GFX, ts:0x00097878]: semaphore wait[0]: 0xffffff8066af3800
    HWSemaphore 0xffffff8066af3800 Signal Event: [DMA1 channel[4] TS:0x0000f37b lastReadTS:0x0000f376]
    HWSemaphore 0xffffff8066af3800 Wait Event : [GFX channel[0] TS:0x00097878 lastReadTS:0x00097877]
    [GFX, ts:0x00097878]: semaphore signal: 0xffffff8066a85400
    HWSemaphore 0xffffff8066a85400 Signal Event: [GFX channel[0] TS:0x00097878 lastReadTS:0x00097877]
    ------------------------
    [0] PFP_HEADER_DUMP: 0xc0004200 0xc0013900 0xc0053c00 0xc0034300 0x00002010
    [0] ME_HEADER_DUMP : 0xc0013900 0xc0053c00 0xc0034300 0x00002010 0x0000a2a4
    [1] PFP_HEADER_DUMP: 0xd238e852 0xd25d5ae6 0x8448d318 0x018f3682 0xed6a22d7
    [1] ME_HEADER_DUMP : 0x21147c2b 0xedc3461b 0x8f7c6dd3 0x3ca6c172 0xeaf8dc5a
    [2] PFP_HEADER_DUMP: 0x336d760a 0x28d590e1 0x79cfe997 0xf5e8dcb1 0x901ba0e9
    [2] ME_HEADER_DUMP : 0xd3483400 0x0f35199c 0xc7b239d5 0x77396de9 0xb1437795
    ------------------------
    SRBM_CNTL : 0x000003ff | CP_ME_CNTL : 0x00000000
    SRBM_STATUS: 0x200046c0 | SRBM_STATUS2: 0x00000000
    GRBM_STATUS: 0xa0003028 | GRBM_STATUS2: 0x00000000
    ------------------------
    GPU HangFlags 0x00000005: AsicHangState 0x00000005, AsicResetRequirement 0x00000001
    IndividualEngineHang: 1
    NonEngineBlockHang : 0
    FenceNotRetired : 1
    PerEngineReset : 1
    FullAsicReset : 0
    HangEngineBitmap : 0x0000000e
    Hung Engines Detected by Cail: GRAPHICS COMPUTE0 COMPUTE1
    ------------------------
    AMDPitcairnGraphicsAccelerator PCIe Device: [6:0:0] State: ENABLED
    Configuration: deviceBits: 0x50116810, capabilityBits: 0x30840300
    TotalVideoRAMBytes: 0x0000000080000000 (2147483648)
    ------------------------

    ******************************************************************************

    Wed Dec 21 19:05:17 2016

    Event: GPU Reset
    Date/Time: Wed Dec 21 19:05:17 2016
    Application:
    Path:
    OS Version: Mac OS X Version 10.11.6 (Build 15G1212)
    Graphics Hardware: AMD FirePro D300
    Signature: 0

    Report Data:

    GPURestartReportStart
    ------------------------
    Hung Channels: DMA1
    ------------------------
    [04] AccelChannel: DMA1
    Pending Command from : UnknownCtx
    PendingCommandTimestamp: 0x0000f377, TotalDWords: 0x00000009, GART Offset=0x00000000801d0000, stamp_idx=4, estamp=0x0000f377
    PendingCommandStart:
    PendingCommandEnd
    ------------------------
    [04] DMA1 HWChannel : Enabled, Idle
    LastReadTimestamp : 0x0000f376
    NextSubmitTimestamp : 0x0000f37c
    [DMA1, ts:0x0000f377]: No semaphore wait
    [DMA1, ts:0x0000f377]: semaphore signal: 0xffffff8066af2e00
    HWSemaphore 0xffffff8066af2e00 Signal Event: [DMA1 channel[4] TS:0x0000f377 lastReadTS:0x0000f376]
    ------------------------
    [0] PFP_HEADER_DUMP: 0xc0013900 0xc0044700 0xc0001700 0xc0034300 0xc0004600
    [0] ME_HEADER_DUMP : 0xc0013900 0xc0044700 0xc0001700 0xc0034300 0xc0004600
    [1] PFP_HEADER_DUMP: 0xd238e852 0xd25d5ae6 0x8448d318 0x018f3682 0xed6a22d7
    [1] ME_HEADER_DUMP : 0x21147c2b 0xedc3461b 0x8f7c6dd3 0x3ca6c172 0xeaf8dc5a
    [2] PFP_HEADER_DUMP: 0x336d760a 0x28d590e1 0x79cfe997 0xf5e8dcb1 0x901ba0e9
    [2] ME_HEADER_DUMP : 0xd3483400 0x0f35199c 0xc7b239d5 0x77396de9 0xb1437795
    ------------------------
    SRBM_CNTL : 0x000103ff | CP_ME_CNTL : 0x00000000
    SRBM_STATUS: 0x200000c0 | SRBM_STATUS2: 0x00000000
    GRBM_STATUS: 0x00003028 | GRBM_STATUS2: 0x00000000
    ------------------------
    GPU HangFlags 0x00000004: AsicHangState 0x00000004, AsicResetRequirement 0x00000002
    IndividualEngineHang: 0
    NonEngineBlockHang : 0
    FenceNotRetired : 1
    PerEngineReset : 0
    FullAsicReset : 1
    HangEngineBitmap : 0x00000000
    ------------------------
    AMDPitcairnGraphicsAccelerator PCIe Device: [6:0:0] State: ENABLED
    Configuration: deviceBits: 0x50116810, capabilityBits: 0x30840300
    TotalVideoRAMBytes: 0x0000000080000000 (2147483648)
    ------------------------

    ******************************************************************************
     
  20. filmak macrumors 6502a

    Joined:
    Jun 21, 2012
    Location:
    between earth and heaven
    #920
    Do you think that sierra install may have broken something in 10.11 installation?
     
  21. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #921
    No, that was clean install on separate disk.
     
  22. filmak macrumors 6502a

    Joined:
    Jun 21, 2012
    Location:
    between earth and heaven
    #922
    It was, I know, but pram etc are the same, even an unseen firmware update can happen...
     
  23. bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #923
    Sierra was already on Trashcan before I starter testing it with El Capitan.
    It seems that it was only a matter of time when it it going to explode on 10.11.6.

    Of course, I am not giving up :D
     
  24. bax2003, Dec 25, 2016
    Last edited: Dec 30, 2016

    bax2003 thread starter macrumors 6502a

    bax2003

    Joined:
    Dec 25, 2011
    Location:
    Belgrade, Serbia
    #925
    Examples show that WindiwServer is crashing at least once a day on Sierra....
     

Share This Page