[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20210326011310.GA2090957@robh.at.kernel.org>
Date: Thu, 25 Mar 2021 19:13:10 -0600
From: Rob Herring <robh@...nel.org>
To: Alain Volmat <alain.volmat@...s.st.com>
Cc: wsa@...nel.org, mark.rutland@....com,
pierre-yves.mordret@...s.st.com, mcoquelin.stm32@...il.com,
alexandre.torgue@...s.st.com, linux-i2c@...r.kernel.org,
devicetree@...r.kernel.org,
linux-stm32@...md-mailman.stormreply.com,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
fabrice.gasnier@...s.st.com
Subject: Re: [PATCH v2 1/2] dt-bindings: i2c: stm32f7: add st,smbus-alert
binding for SMBus Alert
On Thu, Mar 18, 2021 at 02:44:48PM +0100, Alain Volmat wrote:
> Based on the SMBus specification, SMBus Alert active state is low.
> As often on SoC, the SMBus Alert pin is not only dedicated to this
> feature and can also be used for another purpose (by configuring it
> as alternate function for other functions via pinctrl).
>
> "smbus" dt-binding has been introduced recently [1], however it is also
> used to indicate usage of host-notify feature.
> Relying on 'smbus' binding for SMBus-Alert as well as it was discussed
> previously [2] would lead to requiring the SMBALERT# pin to be configured
> as alternate function for i2c/smbus controller even if only host-notify is
> needed.
> Indeed, not doing so would lead to spurious SMBus Alert interrupts since
> the i2c/smbus controller would see the (not configured) SMBA pin as low
> level.
>
> For that reason, SMBus-Alert needs to have its own binding in order
> to only be enabled whenever SMBALERT# pin is configured as alternate
> function for i2c/smbus controller.
>
> [1] https://marc.info/?l=linux-i2c&m=159531254413805&w=2
> [2] https://marc.info/?l=linux-renesas-soc&m=159361426409817&w=2
Please use lore.kernel.org links.
>
> Signed-off-by: Alain Volmat <alain.volmat@...s.st.com>
>
> ---
> v2: introduce st,smbus-alert property
> ---
> Documentation/devicetree/bindings/i2c/st,stm32-i2c.yaml | 5 +++++
> 1 file changed, 5 insertions(+)
With that,
Reviewed-by: Rob Herring <robh@...nel.org>
Though I defer to Wolfram whether this could/should be common instead.
Rob
Powered by blists - more mailing lists