This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

4AJ.2.5 Release on the OMAP4 GP device

Hi All,

I am trying to get Android  4AJ.2.5 release from omapedia working on the Blaze with the OMAP 4470 ES1.0. Using the 0005-wakeupgen-avoid-crash-in-early-boot patch from http://e2e.ti.com/support/omap/f/849/p/253953/888544.aspx#888544 i am able to get the kernel booted. 

However the backlight turns on but i do not see anything on the display. After bootup I see the following error continuously displayed on the serial console. 

[   12.645294] PVR_K:(Error): BridgedDispatchKM: Initialisation failed.  Driver unusable. [0, ]
[   13.155151] PVR_K:(Error): BridgedDispatchKM: Initialisation failed.  Driver unusable. [0, ]

The complete boot log is attached for reference, looking forward for your inputs.

Thanks,

Vinod



  • Hi Vinod.

    It looks like you have an earlier problem with the framebuffer and I believe that leads to the SGX error: "invalid framebuffer size".  You need to add "vram=32M omapfb.vram=0:32M" to the boot args.  Or whatever memory assignment is appropriate for the framebuffers in your situation.

    Regards,

        Steve

  • Steve,

    Thanks for your input. I tried adding the options to the boot args but it did not make any difference. The console log is attached.

    Thanks,

    Vinod

  • Hi Vinod.

    From drivers/video/omap2/omaplfb/omaplfb_displayclass.c, it appears as though you have either the xres or yres of the display as 0 and they need to be valid values.  I'd track down where that is coming from.  Perhaps this is a result of your omapfb.fb_opt setting?

    Regards,

        Steve

  • Hi Steve,

    I tried adding some debug messages in omaplfb_displayclass.c and it looks like the size is getting passed correctly for the device 0

    From the logs it also looks like the PVR error is seen before FBInitDev is called

    /****************************************/

    [ 8.493316] cfg80211: (2457000 KHz - 2482000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
    [ 8.501983] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
    [ 8.511230] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
    [ 8.520782] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
    [ 8.718841] PVR_K:(Error): BridgedDispatchKM: Driver initialisation not completed yet. [0, ]
    [ 8.812652] PVR_K:(Error): BridgedDispatchKM: Driver initialisation not completed yet. [0, ]
    [ 8.883941] PVR: Installing MISR with cookie bf04787c
    [ 8.890930] PVR: Installing device LISR SGX ISR on IRQ 53 with cookie d5342780
    [ 8.898834] PVR: SysFinalise: Version string: SGX revision = 1.1.2
    [ 8.905731] For Device 2 : x = 0, y = 0
    [ 8.905731] omaplfb: OMAPLFBInitFBDev: Device 2: invalid framebuffer size
    [ 8.918182] For Device 1 : x = 0, y = 0
    [ 8.918182] omaplfb: OMAPLFBInitFBDev: Device 1: invalid framebuffer size
    [ 8.929992] For Device 0 : x = 864, y = 480
    [ 8.934722] omaplfb OMAPLFBRegisterPVRDriver: Could not create ion client
    [ 8.942169] PVR_K:(Error): SysFinalise: Failed to register PVR driver with omaplfb [0, ]
    [ 8.951080] PVR_K:(Error): PVRSRVFinaliseSystem: SysFinalise failed (4) [0, ]
    [ 8.959075] PVR_K:(Error): BridgedDispatchKM: Initialisation failed. Driver unusable. [0, ]
    [ 8.998168] virtio_rpmsg_bus virtio1: creating channel rpmsg-resmgr addr 0x66
    [ 9.065734] EXT4-fs (mmcblk0p8): re-mounted. Opts: (null)
    [ 9.158294] EXT4-fs (mmcblk0p8): re-mounted. Opts: (null)

    /*********************************************************/

    Thanks,

    Vinod

  • Hi Vinod.

    I do see there are PVR messages before the omapfb messages, but they are of the "not completed yet" type.  What seems to cause the fatal PVR failure are the omaplfb calls. 

    Digging a little further into the source, it is the call to OMAPLFBInit() made by OMAPLFBRegisterPVRDriver() that is failing.  You can add some further debug message, but likely that means OMAPLFBInitDev() is failing.  When I had a similar problem, that was because I had no vmem assigned for the FB (the earlier suggestion I had).  But you may be failing in OMAPLFBInitDev() for another reason.  I'd suggest printks to find out.

    Regards,

        Steve

  • Steve,

    Thanks for your inputs. I was able to get the display working. The issue was with the vram buffers not being set for the blaze platform.

     

    Thanks,

    Vinod

  • I have attached the patch just in case anyone else is looking at getting the blaze working with the 4AJ.2.5 release

  • Hi vinod,

    I am kind of facing the same problem as you were and i tried applying your patch to the kernel and building it, but still i cant get the display coming up.

    I can only see the backlight coming up. I am discussing this with steve on this thread here http://e2e.ti.com/support/omap/f/849/p/266210/935351.aspx#935351. Your valuable inputs would be much appreciated.

    Thank You

    Prakash

  • Prakash,

    Have you added "vram=32M omapfb.vram=0:32M" to the kernel bootargs. You will also need this for the display to work.


    Thanks,

    Vinod  

  • Hi vinod,

    Thanks for the quick reply.

    Yes i have tried adding them in the bootargs but still no success.

    Thank You

    Prakash

  • Hi,

    How do we change the kernel bootargs ? Can this be done using the fastboot.sh script ?

    --Kiran