[qa-build-notification] QA notification for completed autobuilder build (yocto-4.1_M1.rc1)

Richard Purdie

On Wed, 2022-06-01 at 15:17 +0100, Richard Purdie via
lists.yoctoproject.org wrote:
A build flagged for QA (yocto-4.1_M1.rc1) was completed on the
autobuilder and is available at:


Build hash information:

bitbake: 6a346df51b96a6c0e1ee516df36eb0b6c292b063
meta-agl: 3a0b7a965ba370ca1fbe2ca0e2ac3babace5204d
meta-arm: 0c4c33de09aa921cafcea2ad4b7bc0e19f844213
meta-aws: b2f5ef7e724d3a2271ef99d748734578cf8fcb1a
meta-gplv2: d2f8b5cdb285b72a4ed93450f6703ca27aa42e8a
meta-intel: ebb8c1c26e57e78563760431a57b6da388b82be2
meta-mingw: a90614a6498c3345704e9611f2842eb933dc51c1
meta-openembedded: 90ff53b8df1e3259cbc201c658a4f3f4dddf3aa8
meta-virtualization: 8e8f59d007ca8d60ec77565663cf6285b8acbbd4
oecore: 18a0c31b3386aa5a04eb8ee8e804c2415a61eaaf
poky: 95066dde6861ee08fdb505ab3e0422156cc24fae

[there was one build failure on qemumip64 on the autobuilder in
we don't have any fix for that and investigation will take time, I
don't think it should block M1 and we should proceed with the other QA]
I was able to work this one out. It was a zero length bb_unihashes.dat
file being placed in the eSDK, probably from a race on bitbake's cache
files. I have an idea of what is happening and some patches which
should fix it but they need testing. The bug has been there since at
least dunfell so I don't think it should block release, it is very