On 1/24/24 17:20, Richard Purdie wrote:
> On Wed, 2024-01-24 at 15:29 +0100, Alexis Lothoré via
> lists.openembedded.org wrote:
>> From: Alexis Lothoré <[email protected]>
[...]
>> +
>> +# When any test fails, TESTIMAGE_FAILED_QA ARTIFACTS will be parsed and for
>> +# each entry in it, if artifact pointed by path description exists on
>> target,
>> +# it will be retrieved onto host
>> +TESTIMAGE_FAILED_QA_ARTIFACTS ?= "\
>> + ${localstatedir}/log \
>> + ${sysconfdir}/version \
>> + ${sysconfdir}/os-release"
>> \ No newline at end of file
>
> I'm a little puzzled by this. Doesn't this mean it won't work for any
> image that isn't based upon minimal (which is most of them)?
Arg, I may have made wrong assumptions there. I focused on ptest images (ptest
fast and ptest all), observed that at some point they inherited from minimal,
and assumed it would be the case for any image used for runtime tests.
Checking autobuilder config.json, I see indeed that a lot of runtime tests are
for example done on core-image-sato, which does not depend on
core-image-minimal.So indeed this default value needs to move elsewhere, maybe
remain in testimage (but I will have to find a fix for issues mentioned in the
commit). I'll check how to properly fix this.
Thanks,
Alexis
>
> Cheers,
>
> Richard
>
--
Alexis Lothoré, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#194297):
https://lists.openembedded.org/g/openembedded-core/message/194297
Mute This Topic: https://lists.openembedded.org/mt/103932521/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
