[Letux-kernel] LetuxOS: Kernel: m-rc 6.18-rc2 success [iMac.local]

[Andreas Kemnade]

On Sat, 1 Nov 2025 20:22:30 +0100
"H. Nikolaus Schaller" <hns at goldelico.com> wrote:

> Hi,
> 
> > Am 01.11.2025 um 19:46 schrieb Andreas Kemnade <andreas at kemnade.info>:
> > 
> > On Tue, 21 Oct 2025 08:14:42 +0200 (CEST)
> > kernel-build-bot at letux.org wrote:
> >   
> >> KVERSION: v6.18-rc2 - http://www.kernel.org
> >> CONFIGS:  arch/arm/configs/letux_defconfig arch/arm/configs/letux_lpae_defconfig arch/arm64/configs/letux_defconfig arch/riscv/configs/letux_defconfig arch/mips/configs/letux_defconfig arch/x86/configs/letux_defconfig
> >> 
> >> Infos
> >> =====
> >> Project Home: http://projects.goldelico.com/p/gta04-kernel/
> >> Issues: http://projects.goldelico.com/p/gta04-kernel/issues/
> >> Original: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/?h=v6.18-rc2
> >> Elixir: https://elixir.bootlin.com/linux/v6.18-rc2/source
> >> 
> >> Downloads
> >> =========
> >> git: http://git.goldelico.com/?p=letux-kernel.git;a=shortlog;h=letux-6.18-rc2
> >> Buildhost: iMac.local
> >> ARM: http://download.goldelico.com/letux-kernel/letux-6.18-rc2
> >> ARM-LPAE: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-lpae
> >> ARM64: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-arm64
> >> X86: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-i386
> >> RISCV64: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-riscv64
> >> MIPS: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-mips
> >> Alpha400: http://download.goldelico.com/letux-kernel/letux-6.18-rc2-l400
> >>   
> > 
> > Why does this contain two drivers for the bq24297? I think one should be
> > enough.
> > One is behind CONFIG_CHARGER_BQ24190 and one is behind CONFIG_CHARGER_BQ2429X.  
> 
> Well, that is because there is one that works and is officially supported by LetuxOS (CONFIG_CHARGER_BQ2429X)
> for the Pyra Handheld:
> 
> ca90eb89df6b76 (H. Nikolaus Schaller     2014-09-08 20:18:00 +0200 1910) MODULE_DESCRIPTION("TI BQ24296/7 charger driver");
> 
> It originated in some rock-chips code because back then there was no upstream driver.
> See commit log. Later, it was extended (e.g. to support some mp2624).
> 
> And then is CONFIG_CHARGER_BQ24190. It originally appeared here:
> 
> d7bf353fd0aa3e (Mark A. Greer            2013-08-23 19:21:03 -0700 2358) MODULE_DESCRIPTION("TI BQ24190 Charger Driver");
> 
> and nobody recognized, if they are similar because it did not support the bq2429x variants.
> 
> Recently, it was extended to support the bq24296/297 as well (without any discussion we were aware of).
> 
> b150a703b56fb6 (Hermes Zhang             2023-12-08 11:47:08 +0800  212) BQ24296,
> 3c7c176b10ccff (Hans de Goede            2024-11-16 21:36:48 +0100  213) BQ24297,
> 
and that commit is extremely simple.
I thought they might be some known shortages of the upstream driver.

> For letux_defconfig there is no conflict:
> 
> arch/arm/configs/letux_defconfig:# CONFIG_CHARGER_BQ24190 is not set
> 
> It has not been evaluated yet to have the same features and if it is really a replacement.
> 
> So should we
> a) submit the letux driver and request that upstream removes the duplicate work?

I guess that the existing one in upstream probably meets more standards,
so we would need much cleanup work and a good argument why it is better.
 
> b1) remove the tested driver from LetuxOS and replace it by an untested one?

probably not until confirmed that it works. 

> b2) remove the tested driver and evaluate/test/extend the bq24190 driver for missing features?

I think having a look is good, maybe updating the dts to work with both
and using thu upstream driver in the pyra-minimal branch. 

> c) never touch a running system and keep both in source and keep the bq24190 disabled in letux_defconfig?
> (upstream will then never see the bq2429x driver and have no conflicts).
> 
this includes more maintenance in the longer term.

> The decision by doing nothing (and not requiring any additional efforts) was c)
> 
> Ideas how to proceed?
> 
I think as the effort is probably low to test it, I think I am
curious enough to do it. If it just works, then we have something less
to upstream.

Regards,
Andreas