Re: [bitbake][wrynose][2.18][PATCH 0/3] Pull request (cover letter only)

On Wed May 13, 2026 at 11:21 AM CEST, Richard Purdie wrote:
> On Tue, 2026-05-12 at 11:07 +0200, Yoann Congal wrote:
>> Those are the patches from the last patch review:
>> https://lore.kernel.org/bitbake-devel/[email protected]/T/#t
>> 
>> Passed a-full on autobuilder with 3 AB-INT issues:
>> https://autobuilder.yoctoproject.org/valkyrie/#/builders/29/builds/3795
>> * qemuarm-tc: 15698 – AB-INT: runqemu.QemuTest.test_qemu_can_shutdown hangs with "Data received serial thread"
>>   Retried in https://autobuilder.yoctoproject.org/valkyrie/#/builders/42/builds/3689
>> * qemuriscv64-ptest: https://autobuilder.yoctoproject.org/valkyrie/#/builders/56/builds/1562
>>   This is a ptest failure in procps: 'pgrep_match_against_full_process_name'
>>   This particular ptest later passed in https://autobuilder.yoctoproject.org/valkyrie/#/builders/56/builds/1564
>>   (I have yet to properly report this one)
>> * qemuarmv5: 16217 – AB-INT: boot hang in qemu (amba chip errors?)
>>   Retried in https://autobuilder.yoctoproject.org/valkyrie/#/builders/80/builds/3559
>> 
>> The following changes since commit a2dd9be788274d9c7280be5b1be5eb5c3990cf54:
>> 
>>   fetch2/crate: use CDN endpoint for version checking if possible (2026-04-30 10:08:54 +0100)
>> 
>> are available in the Git repository at:
>> 
>>   https://git.openembedded.org/bitbake-contrib stable/2.18-next
>>   https://git.openembedded.org/bitbake-contrib/log/?h=stable/2.18-next
>> 
>> for you to fetch changes up to 87dcc2935137558ac236676ab7bcbc3cc4883c2c:
>> 
>>   lib: Replace codegen with ast.unparse() (2026-05-05 17:36:57 +0200)
>> 
>> ----------------------------------------------------------------
>> 
>> Alexander Kanavin (1):
>>   fetch/wget: in upstream version checks, match versioned directories
>>     exactly
>> 
>> Paul Barker (1):
>>   doc: Fix description of git shallow mirror tarball creation
>> 
>> Richard Purdie (1):
>>   lib: Replace codegen with ast.unparse()
>
> This breaks the hashequiv docker file so we may want to hold off on
> this until we have the other fixes in master?
>
> I'm a little torn on backporting it. Yes, it is a nice cleanup but it
> does introduce risk, as that other issue shows. It is all relatively
> minor but...

Let's drop it. I was a little too enthusiastic with my backport
requests. I'll go back to being more conservative.

Regards,

>
> Cheers,
>
> Richard


-- 
Yoann Congal
Smile ECS