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.

PROCESSOR-SDK-AM437X: 【am437x-evm-】 Why is there a difference in the URL of the Linux kernel downloaded by yocto and installer?


I have two questions.  I would be happy if somebody could give me an answer.

1. In am437x-evm-, why is there a difference in the URL of the Linux kernel downloaded by yocto and installer?(see appendix)
       URL of linux kernel with installer) git://
       URL of linux kernel with yocto) git://
2. Which is intentional or not about above? If it's intentional, why?

●Ensure URL with installer
1. curl -LO
2. chmod +x ti-processor-sdk-linux-am437x-evm-
3. ./ti-processor-sdk-linux-am437x-evm-
4. cd ti-processor-sdk-linux-am437x-evm-
5. git remote -v
origin git:// (fetch)
origin git:// (push)

●Ensure URL with yocto
1. git clone
2. cd oe-layersetup
3. ./ -f configs/processor-sdk/configs/processor-sdk/processor-sdk-
4. cat sources/meta-ti/recipes-kernel/linux/ | tail -n 20

SRCREV = "e19ba3d996f22ad8cc7187b30c18347aba0d594d"
PV = "3.14.43+git${SRCPV}"

# Append to the MACHINE_KERNEL_PR so that a new SRCREV will cause a rebuild
MACHINE_KERNEL_PR_append = "j"

KERNEL_CONFIG_DIR = "${S}/ti_config_fragments"

KERNEL_CONFIG_FRAGMENTS_append_ti33x = " ${KERNEL_CONFIG_DIR}/am33xx_only.cfg"
KERNEL_CONFIG_FRAGMENTS_append_ti43x = " ${KERNEL_CONFIG_DIR}/am43xx_only.cfg"


KERNEL_GIT_URI = "git://"
file://defconfig \

  • I'm so sorry...
    I should use sources/meta-processor-sdk/recipes-kernel/linux/

    I withdraw Above questions. I have additional one question.
    What's difference between and

  • What's difference between and

    First, let me provide some background that is probably not obvious, for better context. The way our SDK release process works is that there is an initial SDK product being build within TI known as the "Core SDK" (using the linux-ti-staging* recipes for the Kernel, for example). The Core SDKs are mostly focused on solid and stable TI base platform enablement. Then, this "Core SDK" is being consumed by different groups within TI, one of them being the "Processor" group. Those groups ("Processor" group here) will then add/extend the "Core SDK" with additional fixes (often "late fixes") or features specific to their products (AM437x in this case) as they productize the final SDK for customers ("Processor SDK" in this case).

    So long story short, the processor SDK tree(s) may contain additional features or late-fixes over their corresponding base trees. Usually it's just a couple of items, that you can easily review by looking at the git log. In many cases if those are late fixes those will get folded into the base trees (and pushed upstream) for future SDK releases. In case of feature additions you can review and decide if that's something you need based on the git logs of the differences.

    Also note that there's a MUCH NEVER version of SDK available for AM437x, which a much newer kernel and rest of the software stacks. You should consider using/migrating to it (at least for any new projects) as this will like result in better product security and performance.

    Regards, Andreas

  • consumed

    Thank you for answer.
    I understand relation between linux-ti-staging and linux-processor-sdk.