On 3/10/2026 6:14 PM, Bruce Ashfield wrote:
**
*CAUTION: This email comes from a non Wind River email account!*
Do not click links or open attachments unless you recognize the sender
and know the content is safe.
On Tue, Mar 10, 2026 at 6:58 AM Harish Sadineni
<[email protected]> wrote:
On 3/10/2026 12:50 AM, Richard Purdie wrote:
> CAUTION: This email comes from a non Wind River email account!
> Do not click links or open attachments unless you recognize the
sender and know the content is safe.
>
> On Mon, 2026-03-09 at 14:01 -0400, Bruce Ashfield wrote:
>>
>>
>> On Mon, Mar 9, 2026 at 12:54 PM Harish Sadineni
>> <[email protected]> wrote:
>>>
>>>
>>>
>>>
>>> On 3/9/2026 9:59 PM, Bruce Ashfield wrote:
>>>
>>>
>>>> CAUTION: This email comes from a non Wind River email account!
>>>> Do not click links or open attachments unless you recognize the
>>>> sender and know the content is safe.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Mon, Mar 9, 2026 at 12:10 PM Harish Sadineni
>>>> <[email protected]> wrote:
>>>>
>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 3/9/2026 4:56 PM, Richard Purdie wrote:
>>>>>
>>>>>
>>>>>> CAUTION: This email comes from a non Wind River email
>>>>>> account!
>>>>>> Do not click links or open attachments unless you recognize
>>>>>> the sender and know the content is safe.
>>>>>>
>>>>>> On Wed, 2026-03-04 at 09:43 -0800, Sadineni, Harish via
>>>>>> lists.openembedded.org
<https://urldefense.com/v3/__http://lists.openembedded.org__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8am36JgQ$>
wrote:
>>>>>>
>>>>>>> From: Harish Sadineni <[email protected]>
>>>>>>>
>>>>>>> Conditionally add 'clang-native', 'rust-native' and
>>>>>>> 'bindgen-cli-native' to 'DEPENDS'
>>>>>>> when Kernel Rust Support is enabled.
>>>>>>>
>>>>>>> These tools are required for building Rust-enabled kernels
>>>>>>> and for
>>>>>>> generating Rust FFI bindings via bindgen during the kernel
>>>>>>> build.
>>>>>>>
>>>>>>> This ensures the additional dependencies are only pulled in
>>>>>>> when
>>>>>>> Rust support is explicitly enabled, avoiding unnecessary
>>>>>>> native
>>>>>>> dependencies for non-Rust kernel builds.
>>>>>>>
>>>>>>> Signed-off-by: Harish Sadineni
>>>>>>> <[email protected]>
>>>>>>> ---
>>>>>>> meta/recipes-kernel/linux/linux-yocto.inc
<https://urldefense.com/v3/__http://linux-yocto.inc__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8KLPsVX0$>
| 3 +++
>>>>>>> 1 file changed, 3 insertions(+)
>>>>>>>
>>>>>>> diff --git a/meta/recipes-kernel/linux/linux-yocto.inc
<https://urldefense.com/v3/__http://linux-yocto.inc__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8KLPsVX0$>
>>>>>>> b/meta/recipes-kernel/linux/linux-yocto.inc
<https://urldefense.com/v3/__http://linux-yocto.inc__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8KLPsVX0$>
>>>>>>> index 4d0a726bb6..b7961bbcdf 100644
>>>>>>> --- a/meta/recipes-kernel/linux/linux-yocto.inc
<https://urldefense.com/v3/__http://linux-yocto.inc__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8KLPsVX0$>
>>>>>>> +++ b/meta/recipes-kernel/linux/linux-yocto.inc
<https://urldefense.com/v3/__http://linux-yocto.inc__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8KLPsVX0$>
>>>>>>> @@ -76,11 +76,14 @@ do_install:append(){
>>>>>>> KERNEL_FEATURES:append:qemuall = " features/kernel-
>>>>>>> sample/kernel-sample.scc"
>>>>>>>
>>>>>>> KERNEL_DEBUG ?= ""
>>>>>>> +KERNEL_RUST_SUPPORT ?= ""
>>>>>>> # These used to be version specific, but are now common
>>>>>>> dependencies. New
>>>>>>> # tools / dependencies will continue to be added in
>>>>>>> version specific recipes.
>>>>>>> DEPENDS += '${@bb.utils.contains_any("ARCH", [ "x86",
>>>>>>> "arm64", "powerpc" ], "elfutils-native", "", d)}'
>>>>>>> DEPENDS += "openssl-native util-linux-native"
>>>>>>> DEPENDS += "gmp-native libmpc-native"
>>>>>>> +RUST_KERNEL_DEPENDS ?=
>>>>>>> "${@bb.utils.contains('KERNEL_RUST_SUPPORT', 'True',
>>>>>>> 'clang-native rust-native bindgen-cli-native', '', d)}"
>>>>>>> +DEPENDS += "${RUST_KERNEL_DEPENDS}"
>>>>>>>
>>>>>>> # Some options depend on CONFIG_PAHOLE_VERSION, so need to
>>>>>>> make pahole-native available before do_kernel_configme
>>>>>>> do_kernel_configme[depends] +=
>>>>>>> '${@bb.utils.contains("KERNEL_DEBUG", "True", "pahole-
>>>>>>> native:do_populate_sysroot", "", d)}'
>>>>>>>
>>>>>> I thought we were going to try and use KERNEL_FEATURES for
>>>>>> this?
>>>>>>
>>>>> Hi Richard,
>>>>>
>>>>> I have tried this but i got circular dependency error with
>>>>> KERNEL_FEATURES. So, I opted using variable
>>>>> KERNEL_RUST_SUPPORT.
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> I'm curious about that. Can you share the details ? (diff and the
>>>> bitbake error)
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>> As discussed previously, I attempted to use the following in
>>> meta/recipes-kernel/linux/linux-yocto_6.18.bb
<https://urldefense.com/v3/__http://linux-yocto_6.18.bb__;!!AjveYdw8EvQ!ekkiIdFR1oUUFgkUmXVgRrlc4N10osQLVSoHUP7tqKnyuU2FDh5UdcpyUkPIW6qIyPNmuAqUaeJv3wtCoY4XeJd8cBMffSY$>:
>>>
>>> KERNEL_FEATURES:append = "
${@bb.utils.contains('KERNEL_FEATURES',
>>> 'rust-kernel', ' features/kernel-rust/kernel-rust.scc', '', d)}"
>>>
>>> This resulted in a BitBake variable dependency chain
failure. The
>>> issue occurs because the expression is trying to read
>>> KERNEL_FEATURES
>>> while simultaneously modifying KERNEL_FEATURES, which creates a
>>> circular dependency.
>>>
>>
>> Yes, that's clearly not going to work, but that wouldn't have been
>> the suggestion.
>>
>> Wasn't Richard asking about setting the RUST_KERNEL_DEPENDS
based on
>> the KERNEL_FEATURES containing rust.scc ?
We understood like Richard suggestion is to use KERNEL_FEATURES
everywhere instead of KERNEL_RUST_SUPPORT variable.
We can use the KERNEL_RUST_SUPPORT variable to append rust.scc to
KERNEL_FEATURES and then we could make other checks depends on
KERNEL_FEATURES.
But as the other recipes (make-mod-scripts, kernel-devsrc,
module-rust.bbclass) do not know about KERNEL_FEATURES, so again
we've
to depend on KERNEL_RUST_SUPPORT Variable.
I'm curious as to what you think is different about KERNEL_RUST_SUPPORT
and KERNEL_FEATURES as variables ?
There is no difference i guess.
They are both defined in the kernel recipe namespace, so any
limitations about
the visibility of one, is true about the other.
We will set KERNEL_RUST_SUPPORT = "true" in local.conf, so it is
globally visible.
In contrast, KERNEL_FEATURES is conditionally set based on
KERNEL_RUST_SUPPORT inside the kernel recipe,
meaning its value is only determined at the recipe level and not
accessible elsewhere.
Thanks,
Harish
or am I missing something ?
Bruce
So, for consistency we used everywhere KERNEL_RUST_SUPPORT.
Is this implementation ok or we should switch to use -
KERNEL_FEATURES in kernel recipes and
KERNEL_RUST_SUPPORT in other recipes.
Thanks,
Harish
> Yes, along with maybe with a tweak to the KERNEL_FEATURES code which
> could expand " rust " or maybe " kernel-rust "in KERNEL_FEATURES
into
> "features/kernel-rust/kernel-rust.scc" too just so there was a nice
> short cut for it?
>
> Cheers,
>
> Richard
>
>
>
--
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#232783):
https://lists.openembedded.org/g/openembedded-core/message/232783
Mute This Topic: https://lists.openembedded.org/mt/118136909/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
