[v2,0/2] iio: adc: meson: add MPLL clock workaround for GXLX

Message

Martin Blumenstingl March 30, 2025, 10:19 a.m. UTC

Hello,

Amlogic GXLX SoCs seem to be mostly the same silicon as GXL. The only
known differences are:
- one less Mali-450 GPU core
- no VP9 codec
- and an odd one: the three MPLL clocks need a bit toggled in the SAR
  ADC register space

This series attempt to fix audio output (which relies on the MPLL
clocks) on the GXLX boards. Unfortunately all we have is a downstream
commit [0] without any further explanation (or anyone who wants to
provide details on this). Since it's not clear if this is a gate, a
reset or some other hardware fix: the driver side includes a warning
for users to update their .dtb along with kernel images in case we
ever figure out what these bits do and how to model them properly.


Changes since v1 at [1]:
- added Krzysztof's Acked-by to the dt-bindings patch (thank you)
- added Neil's Reviewed-by (thank you!)
- fixed meson_sar_adc_gxlx_param to be independent of future
  to-be-upstreamed patches (fixes a build error)


[0] https://github.com/khadas/linux/commit/d1d98f2ed8c83eb42af8880ed8e206aa402dd70a#diff-c5aaf54323ef93777c5083de37f933058ea8d0af79a1941e0b5a0667dc0f89b3
[1] https://lore.kernel.org/linux-amlogic/[e-mail]/


Martin Blumenstingl (2):
  dt-bindings: iio: adc: amlogic,meson-saradc: Add GXLX SoC compatible
  iio: adc: meson: add support for the GXLX SoC

 .../iio/adc/amlogic,meson-saradc.yaml         |  1 +
 drivers/iio/adc/meson_saradc.c                | 34 +++++++++++++++++++
 2 files changed, 35 insertions(+)

Comments

Jonathan Cameron March 30, 2025, 2:08 p.m. UTC | #1

On Sun, 30 Mar 2025 12:19:20 +0200
Martin Blumenstingl <[e-mail]> wrote:

> Hello,
> 
> Amlogic GXLX SoCs seem to be mostly the same silicon as GXL. The only
> known differences are:
> - one less Mali-450 GPU core
> - no VP9 codec
> - and an odd one: the three MPLL clocks need a bit toggled in the SAR
>   ADC register space
> 
> This series attempt to fix audio output (which relies on the MPLL
> clocks) on the GXLX boards. Unfortunately all we have is a downstream
> commit [0] without any further explanation (or anyone who wants to
> provide details on this). Since it's not clear if this is a gate, a
> reset or some other hardware fix: the driver side includes a warning
> for users to update their .dtb along with kernel images in case we
> ever figure out what these bits do and how to model them properly.
Applied to the togreg branch of iio.git. Initially pushed out as testing
to get some early testing before I rebase on rc1 next weekend.

Thanks,

Jonathan

> 
> 
> Changes since v1 at [1]:
> - added Krzysztof's Acked-by to the dt-bindings patch (thank you)
> - added Neil's Reviewed-by (thank you!)
> - fixed meson_sar_adc_gxlx_param to be independent of future
>   to-be-upstreamed patches (fixes a build error)
> 
> 
> [0] https://github.com/khadas/linux/commit/d1d98f2ed8c83eb42af8880ed8e206aa402dd70a#diff-c5aaf54323ef93777c5083de37f933058ea8d0af79a1941e0b5a0667dc0f89b3
> [1] https://lore.kernel.org/linux-amlogic/[e-mail]/
> 
> 
> Martin Blumenstingl (2):
>   dt-bindings: iio: adc: amlogic,meson-saradc: Add GXLX SoC compatible
>   iio: adc: meson: add support for the GXLX SoC
> 
>  .../iio/adc/amlogic,meson-saradc.yaml         |  1 +
>  drivers/iio/adc/meson_saradc.c                | 34 +++++++++++++++++++
>  2 files changed, 35 insertions(+)
>