RE: [PATCH v1 2/8] ARM: dts: aspeed: yosemite4: Add i2c-mux for four NICs

From: Delphine_CC_Chiu/WYHQ/Wiwynn
Date: Tue Oct 01 2024 - 03:53:01 EST




> -----Original Message-----
> From: Andrew Jeffery <andrew@xxxxxxxxxxxxxxxxxxxx>
> Sent: Tuesday, October 1, 2024 8:37 AM
> To: Delphine_CC_Chiu/WYHQ/Wiwynn <Delphine_CC_Chiu@xxxxxxxxxx>;
> patrick@xxxxxxxxx; Rob Herring <robh@xxxxxxxxxx>; Krzysztof Kozlowski
> <krzk+dt@xxxxxxxxxx>; Conor Dooley <conor+dt@xxxxxxxxxx>; Joel Stanley
> <joel@xxxxxxxxx>
> Cc: Ricky CX Wu <ricky.cx.wu.wiwynn@xxxxxxxxx>;
> devicetree@xxxxxxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx;
> linux-aspeed@xxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v1 2/8] ARM: dts: aspeed: yosemite4: Add i2c-mux for four
> NICs
>
> [External Sender]
>
> [External Sender]
>
> On Mon, 2024-09-30 at 15:04 +0800, Delphine CC Chiu wrote:
> > From: Ricky CX Wu <ricky.cx.wu.wiwynn@xxxxxxxxx>
> >
> > Add i2c-mux on Spider board for four NICs and add the temperature
> > sensor and EEPROM for the NICs.
> >
> > Signed-off-by: Ricky CX Wu <ricky.cx.wu.wiwynn@xxxxxxxxx>
> > Signed-off-by: Delphine CC Chiu <Delphine_CC_Chiu@xxxxxxxxxx>
> > ---
> > .../aspeed/aspeed-bmc-facebook-yosemite4.dts | 75
> > ++++++++++++++++++-
> > 1 file changed, 72 insertions(+), 3 deletions(-)
> >
> > diff --git
> > a/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-yosemite4.dts
> > b/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-yosemite4.dts
> > index dbc992a625b7..b813140b3c5b 100644
> > --- a/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-yosemite4.dts
> > +++ b/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-yosemite4.dts
> > @@ -26,6 +26,10 @@ aliases {
> > i2c21 = &imux21;
> > i2c22 = &imux22;
> > i2c23 = &imux23;
> > + i2c24 = &imux24;
> > + i2c25 = &imux25;
> > + i2c26 = &imux26;
> > + i2c27 = &imux27;
> > i2c34 = &imux34;
> > i2c35 = &imux35;
> > };
> > @@ -1196,8 +1200,9 @@ adc@35 {
> > };
> >
> > &i2c15 {
> > + #address-cells = <1>;
> > + #size-cells = <0>;
> > status = "okay";
> > - mctp-controller;
>
> Why are you deleting this if you're not also deleting the MCTP endpoint node?
>
> Unless there's some reason this is related to the NICs, this should be its own
> patch with its own justification.
>
> Andrew
Hi Andrew,

We have discussed with Jeremy from Code Construct about removing the
mctp-controller on bus 15 before.

He thought it would be better to add the mctp-controller on each i2c mux port
so the MCTP driver would ensure that the mux was configured properly before
communicating with the device.

I'll split a patch to remove the mctp-controller on bus 15 in next serial of patches.

Regards,
Ricky