Hi Dmitry, Sorry for the breakage, and thanks for catching it and helping with the partial revert to fix the regression.
This was an oversight on my end. I had verified that the build compiled fine for an RPI4 target with core-image-minimal (qemuarm with GCC and CLANG) , but since that was only a compile-time check and I didn’t flash it on real hardware running a mainline kernel, the missing ConfigFS scripts slipped through. I want to avoid missing this kind of issue on future upgrades. - Could you share the testing or sanity-checking workflow you usually follow for userspace/kernel integrations like this? - Are there specific QEMU configs, runtime checks, or review habits you rely on to make sure upstream compatibility doesn’t break? I’d like to adapt my own workflow so I can catch this kind of runtime regression earlier. Thanks again for the guidance. Ashish
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#127485): https://lists.openembedded.org/g/openembedded-devel/message/127485 Mute This Topic: https://lists.openembedded.org/mt/119463946/21656 Group Owner: [email protected] Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
