TDA4VM: TDA4VM-SK : Board Bringup issue with SDK11 edgeai image for TDA4VM evaluation module

Hi Vipul,

4.927986] No soundcards found.
[ 4.933649] mmc1: SDHCI controller on 4fb0000.mmc [4fb0000.mmc] using ADMA 64-bit
[ 4.941233] Waiting for root device PARTUUID=076c4a2a-02...
[ 4.980947] tps6594 0-004c: Error IRQ trap overvoltage for LDO1
[ 5.010409] sdhci-am654 4fb0000.mmc: Failed to find itapdly, fail tuning
[ 5.017099] mmc1: tuning execution failed: -1
[ 5.021447] mmc1: ddr50 tuning failed
[ 5.025103] mmc1: new ultra high speed DDR50 SDXC card at address d555
[ 5.032020] mmcblk1: mmc1:d555 SK64G 59.6 GiB

MMCSD is not coming up. Can you try using a different SD card?

- Keerthy

Hi Keerthy,

Thanks for your response!

I can try that but in the same 64GB SD card I flashed the SDK11 edgeai image downloaded from TI official page and it's booting up perfectly without any issue. How one image come up & other could be an issue here.

One thing i noticed that in my final built image uEnv.txt file was empty by default. Anything w.r.t that could be an issue here?

Any other suggestion on this? 

Thanks & Regards,

Vipul Kumar 

Vipul,

Thanks for the additional details. I believe you are using the create-sd-card.sh script to transfer the boot binaries and the file system contents to SD card. 

From the details given above I assume that the prebuilt boot binaries are working fine but if you are compiling and then using your binaries to boot then you are observing the failure?

I recommend using 

sudo make linux_install 

The above command will directly copy the images that you have built on to the SD card.

You can copy the default uEnv.txt file as well to the SD card boot partition.

Best Regards,

Keerthy 

Hi Keerthy,

Sorry for the delay!

I didn't use create-sd-card.sh, actually i couldn't locate this in my build & even i couldn't find linux_install as well. I used bmaptool to flash the entire image in SD card in single shot. Please find below the bmaptool logs:

automotive@automotive-OptiPlex-5080:~/Automotive/ti/TDA4VM_SDK11_image/TDA4VM_SDK11_edgeai_image/build/deploy-ti/images/j721e-evm$ sudo bmaptool copy tisdk-edgeai-image-j721e-evm.rootfs.wic.xz /dev/sdb
[sudo] password for automotive:
bmaptool: info: discovered bmap file 'tisdk-edgeai-image-j721e-evm.rootfs.wic.bmap'
bmaptool: info: block map format version 2.0
bmaptool: info: 3461982 blocks of size 4096 (13.2 GiB), mapped 1957452 blocks (7.5 GiB or 56.5%)
bmaptool: info: copying image 'tisdk-edgeai-image-j721e-evm.rootfs.wic.xz' to block device '/dev/sdb' using bmap file 'tisdk-edgeai-image-j721e-evm.rootfs.wic.bmap'
bmaptool: info: 100% copied
bmaptool: info: synchronizing '/dev/sdb'
bmaptool: info: copying time: 6m 18.9s, copying speed 20.2 MiB/sec
automotive@automotive-OptiPlex-5080:~/Automotive/ti/TDA4VM_SDK11_image/TDA4VM_SDK11_edgeai_image/build/deploy-ti/images/j721e-evm$

One more point i am not sure if it make sense here, i was running out of space so i included "INHERIT += "rm_work"" in my local.conf to remove work directory which helped me to build the edgeai image in less disk space.

Looking for more inputs from your end.

Thanks & Regards,

Vipul Kumar

Hello,
Thank you for your inquiry. **Keerthy** is currently on leave and will be available on **2026-05-02**.
Your query will be addressed upon their return. We appreciate your patience and understanding.
Best regards,
TI E2E Support Team
---
*This is an automated notification.*

Hi Vipul,

Vipul Kumar Suneja said:

I have build tisdk-edgeai-image with SDK version 11 for the TDA4VM evaluation module. The image build successfully & i flashed the same in SD card with bmap tool also. But the SD card is not booting up, kernel bootup get started but it hangs there only.  I am attaching the bootup logs for reference. 

In the same SD card if i flash the edgeai image downloaded from the TI official page, it's booting up without any issue.

I would highly suggest using the edge ai image from the TI official page itself, as there could be some packaging which could be missing when building tisdk-edgeai-image with SDK11 which is why TI has released a separate edge-ai image on the TI official page.

Best Regards

Gokul Praveen 

Hi Gokul Praveen,

Thanks for your response!

My requirement is to build for development & optimize the image later before it goes to production. So anyhow i have to build the image, TI official image may not serve the purpose for me. Any other suggestion can you give to try it out on the source side.

Thanks & Regards,

Vipul Kumar

Hi Vipul,

Then please start with TI official image. You could optimize it later. 

Regards,

Brijesh