mbox series

[V2,0/7] i2c-imx-lpi2c: add IPG clock

Message ID 20220816125526.2978895-1-peng.fan@oss.nxp.com
Headers show
Series i2c-imx-lpi2c: add IPG clock | expand

Message

Peng Fan (OSS) Aug. 16, 2022, 12:55 p.m. UTC
From: Peng Fan <peng.fan@nxp.com>

V2:
 use clk bulk API in driver to support backward compatibility.
 Include a new patch, patch 1.

The i.MX LPI2C needs PER and IPG clock, not just PER or IPG clock.
The current driver/dts/bindings use one CLK. Although it works with
with upstream kernel, but it not match the hardware design. If IPG
clock is disabled, the LPI2C will not work.

There are changes made to ARM32 i.MX7ULP dts, ARM64 i.MX8 dts, dt-bindings,
and the lpi2c driver.

The driver is updated to use bulk clk API to avoid break backward
compatibility. But it is hard to avoid dtbs_check pass, because the dts
and binding update are in separate patches.

Peng Fan (7):
  ARM: dts: imx7ulp: update the LPI2C clock-names
  dt-bindings: i2c: i2c-imx-lpi2c: add ipg clk
  dt-bindings: i2c: i2c-imx-lpi2c: add dmas property
  dt-bindings: i2c: i2c-imx-lpi2c: add i.MX93
  arm64: dts: imx8-ss-dma: add IPG clock for i2c
  ARM: dts: imx7ulp: Add IPG clock for lpi2c
  i2c: imx-lpi2c: use bulk clk API

 .../bindings/i2c/i2c-imx-lpi2c.yaml           | 20 +++++++++++++---
 arch/arm/boot/dts/imx7ulp.dtsi                | 10 ++++----
 .../arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 20 +++++++++-------
 drivers/i2c/busses/i2c-imx-lpi2c.c            | 24 +++++++++----------
 4 files changed, 47 insertions(+), 27 deletions(-)

Comments

Peng Fan Aug. 22, 2022, 8:46 a.m. UTC | #1
> Subject: [PATCH V2 0/7] i2c-imx-lpi2c: add IPG clock

+ Wolfram Sang I2C maintainer.

Krzysztof,

Do you have time to give a look whether this patchset is ok for you?
Please forgive if this is too early ping. Some i.MX93 dts update pending

Thanks,
Peng.

> 
> From: Peng Fan <peng.fan@nxp.com>
> 
> V2:
>  use clk bulk API in driver to support backward compatibility.
>  Include a new patch, patch 1.
> 
> The i.MX LPI2C needs PER and IPG clock, not just PER or IPG clock.
> The current driver/dts/bindings use one CLK. Although it works with with
> upstream kernel, but it not match the hardware design. If IPG clock is
> disabled, the LPI2C will not work.
> 
> There are changes made to ARM32 i.MX7ULP dts, ARM64 i.MX8 dts, dt-
> bindings, and the lpi2c driver.
> 
> The driver is updated to use bulk clk API to avoid break backward
> compatibility. But it is hard to avoid dtbs_check pass, because the dts and
> binding update are in separate patches.
> 
> Peng Fan (7):
>   ARM: dts: imx7ulp: update the LPI2C clock-names
>   dt-bindings: i2c: i2c-imx-lpi2c: add ipg clk
>   dt-bindings: i2c: i2c-imx-lpi2c: add dmas property
>   dt-bindings: i2c: i2c-imx-lpi2c: add i.MX93
>   arm64: dts: imx8-ss-dma: add IPG clock for i2c
>   ARM: dts: imx7ulp: Add IPG clock for lpi2c
>   i2c: imx-lpi2c: use bulk clk API
> 
>  .../bindings/i2c/i2c-imx-lpi2c.yaml           | 20 +++++++++++++---
>  arch/arm/boot/dts/imx7ulp.dtsi                | 10 ++++----
>  .../arm64/boot/dts/freescale/imx8-ss-dma.dtsi | 20 +++++++++-------
>  drivers/i2c/busses/i2c-imx-lpi2c.c            | 24 +++++++++----------
>  4 files changed, 47 insertions(+), 27 deletions(-)
> 
> --
> 2.37.1
Krzysztof Kozlowski Aug. 23, 2022, 11:20 a.m. UTC | #2
On 22/08/2022 11:46, Peng Fan wrote:
>> Subject: [PATCH V2 0/7] i2c-imx-lpi2c: add IPG clock
> 
> + Wolfram Sang I2C maintainer.
> 
> Krzysztof,
> 
> Do you have time to give a look whether this patchset is ok for you?
> Please forgive if this is too early ping. Some i.MX93 dts update pending
> 

I don't understand the ping. You got everything needed from us, why
still pinging? What that DTS has anything to do with us?

Best regards,
Krzysztof
Wolfram Sang Aug. 24, 2022, 5:47 a.m. UTC | #3
Hi Krzysztof,

> Awesome! But you don't need my ack after such update. The ack or review
> is expected from maintainers and I am not the maintainer of IMX I2C
> driver or IMX platform/DTS.

Despite the fact that the ping was in deed too early, I think Peng Fan
has a point. If you suggest a change (Thanks a ton for that!), a quick
comment on the new version if the change reflects what you had in mind,
is fair, I'd say. Something like "Yeah, looks basically good, but the
maintainers have to check the details." I understand the "all too busy"
part, of course...

Thanks,

   Wolfram