Bug#1109344: /boot/vmlinuz-6.12.35+deb13-amd64: uevent field power_supply/BAT1/uevent:POWER_SUPPLY_CURRENT_NOW now -ve
Hi Ian,
On Wed, Jul 16, 2025 at 03:40:09PM +0100, Ian Jackson wrote:
> Salvatore Bonaccorso writes ("Re: Bug#1109344: /boot/vmlinuz-6.12.35+deb13-amd64: uevent field power_supply/BAT1/uevent:POWER_SUPPLY_CURRENT_NOW now -ve"):
> > Control: tags -1 + moreinfo
> ...
> > I suspect this is a deliberate change maybe e8cd4a8d5b88 ("ACPI:
> > battery: negate current when discharging"), can you try with that
> > commit reverted?
>
> That sounds very likely.
>
> I'm at Debconf and building kernels and rebooting my machine is quite
> disruptive. When I'm back home I'll have a spare laptop I could try
> this out on; that would be over a week from now at the earliest.
It is alright, enjoy Debconf :)
> (Also please would you point me at the right instructions for trying
> such a thing? I can get the kernel source package and patch it but it
> wants to build a lot of flavours so the build is very slow, and
> nowadays I suspect I may have some kind of restricted boot problem?)
It has a quite simple approach for signle patch testing we have
documented here:
https://kernel-team.pages.debian.net/kernel-handbook/ch-common-tasks.html#id-1.6.6.4
that is, with a single patch you want to test on top of all, which
would be a revert patch of the mentioned commit, it should be fairly
more easy with the above and only building what you need. In case you
are still struggling with it let me know and might be able to provide
you (unsigned!) builds for amd64.
> > If that is sucessfull we would appreciate if you can take it to
> > upstream involved people to ask if the side effect you are seeing is
> > considered a regression bug or to be kept deliberately.
> >
> > Can you please do that?
>
> My experience with Linux upstream hasn't always been great but I can
> give it a go. Am I just to post to lmkl, or what ?
Okay understoond. IMHO still better if we do not relay through us back
questions and they can reach you directly. In case the revert commit
helps, then the report can go to:
Peter Marheine <pmarheine@chromium.org>
Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Additionally the stable@vger.kernel.org list and
regressions@lists.linux.dev .
and with help of the get_maintainers.pl script a CC to:
Len Brown <lenb@kernel.org> (reviewer:ACPI)
linux-acpi@vger.kernel.org (open list:ACPI)
linux-kernel@vger.kernel.org (open list)
Does this helps?
Regards,
Salvatore
Reply to: