[PATCH v6] ARM: dts: aspeed: Adding Facebook Bletchley BMC

Hi Joel

Once we've fixed the issues with the bindings, we can put the driver
in the openbmc tree. Let's spend our effort fixing that.

It means I can commit Bletchley DTS with fan support to OpenBMC tree once Billy's driver got approved, right?

Howard

-----Original Message-----
From: Joel Stanley <joel@jms.id.au>
Sent: Tuesday, December 7, 2021 2:37 PM
To: Howard Chiu (???) <redacted>
Cc: Patrick Williams <redacted>; Howard Chiu
[off-list ref]; Arnd Bergmann [off-list ref]; Olof
Johansson [off-list ref]; SoC Team [off-list ref]; Rob Herring
[off-list ref]; Andrew Jeffery [off-list ref]; Linux ARM
[off-list ref]; devicetree
[off-list ref]; Linux Kernel Mailing List
[off-list ref]; linux-aspeed [off-list ref]
Subject: Re: [PATCH v6] ARM: dts: aspeed: Adding Facebook Bletchley BMC

On Tue, 7 Dec 2021 at 06:34, Howard Chiu (???)
[off-list ref] wrote:

quoted

Hi Joel

quoted

I'd suggest no, omit those features for now, and you can add them in
once we have a driver. The device tree bindings for the tach/pwm need
some work before they can be approved.

The driver you mentioned is for OpenBMC Linux or upstream?
If it is for OpenBMC, the upstream robot will reject my DTS because the

aspeed-g6.dtsi did not contain pwm/tach node.

quoted

If it is for upstream, it will be a long time that OpenBMC Linux won't have

tach driver for ast2600 series, and my patch to commit with Billy's early driver
for OpenBMC Linux is also meaningless.

Once we've fixed the issues with the bindings, we can put the driver
in the openbmc tree. Let's spend our effort fixing that.

quoted

quoted

When I asked you to fix the patch, I was referring to the usb-c
warnings that I posted about the other week.

I will fix it and others you mentioned in the v7 patch.

Howard

quoted

-----Original Message-----
From: Joel Stanley <joel@jms.id.au>
Sent: Tuesday, December 7, 2021 11:30 AM
To: Howard Chiu (???) <redacted>
Cc: Patrick Williams <redacted>; Howard Chiu
[off-list ref]; Arnd Bergmann [off-list ref]; Olof
Johansson [off-list ref]; SoC Team [off-list ref]; Rob Herring
[off-list ref]; Andrew Jeffery [off-list ref]; Linux ARM
[off-list ref]; devicetree
[off-list ref]; Linux Kernel Mailing List
[off-list ref]; linux-aspeed

[off-list ref]

quoted

quoted

Subject: Re: [PATCH v6] ARM: dts: aspeed: Adding Facebook Bletchley BMC

On Tue, 7 Dec 2021 at 02:53, Howard Chiu (???)
[off-list ref] wrote:

quoted

Hi Joel

Bletchley platform needs to support fan control.
Since Billy's patch for upstream is still under review, I can't add pwm/tach

node into dts or the build would fail.

quoted

As you know, I had committed the pwm/tach driver for ast26xx to

OpenBMC

quoted

quoted

linux, could I commit the dts to OpenBMC linux instead of upstream for

early

quoted

quoted

build?

I'd suggest no, omit those features for now, and you can add them in
once we have a driver. The device tree bindings for the tach/pwm need
some work before they can be approved.

When I asked you to fix the patch, I was referring to the usb-c
warnings that I posted about the other week.

quoted

Howard

quoted

-----Original Message-----
From: Joel Stanley <joel@jms.id.au>
Sent: Tuesday, December 7, 2021 6:52 AM
To: Patrick Williams <redacted>
Cc: Howard Chiu <redacted>; Arnd Bergmann
[off-list ref]; Olof Johansson [off-list ref]; SoC Team
[off-list ref]; Rob Herring [off-list ref]; Andrew Jeffery
[off-list ref]; Linux ARM

[off-list ref];

quoted

quoted

quoted

quoted

devicetree [off-list ref]; Linux Kernel Mailing List
[off-list ref]; linux-aspeed

[off-list ref];

quoted

quoted

Howard Chiu (???) [off-list ref]
Subject: Re: [PATCH v6] ARM: dts: aspeed: Adding Facebook Bletchley

BMC

quoted

quoted

quoted

quoted

On Mon, 6 Dec 2021 at 21:25, Patrick Williams [off-list ref]

wrote:

quoted

quoted

quoted

quoted

quoted

On Wed, Dec 01, 2021 at 11:37:38AM +0800, Howard Chiu wrote:

quoted

Initial introduction of Facebook Bletchley equipped with
Aspeed 2600 BMC SoC.

Signed-off-by: Howard Chiu <redacted>

...

quoted

Change since v2:
- Remove uart5 workaround
- Remove gpio nodes of pca9552/pca9539
- Modify gpio-line-name of led/power/presence pins with openbmc

pattern

quoted

A number of the GPIOs do not have defined openbmc patterns for

them

quoted

quoted

yet.

quoted

quoted

The

quoted

names you have chosen are ok for now, but we will be changing them

as

quoted

quoted

we

quoted

quoted

refine

quoted

development of this machine further.

quoted

+             gpio-line-names =
+

"SLED0_MS_DETECT1","SLED0_VBUS_BMC_EN","SLED0_INA230_ALERT","SLE

quoted

quoted

quoted

quoted

D0_P12V_STBY_ALERT",

quoted

quoted

+

"SLED0_SSD_ALERT","SLED0_MS_DETECT0","SLED0_RST_CCG5","SLED0_FUSB

quoted

quoted

quoted

quoted

302_INT",

quoted

quoted

+

"SLED0_MD_STBY_RESET","SLED0_MD_IOEXP_EN_FAULT","SLED0_MD_DIR","

quoted

quoted

quoted

quoted

SLED0_MD_DECAY",

quoted

quoted

+

"SLED0_MD_MODE1","SLED0_MD_MODE2","SLED0_MD_MODE3","power-ho

quoted

quoted

quoted

quoted

st0";

quoted

Such as these...

quoted

--
2.25.1

Otherwise,

Reviewed-by: Patrick Williams <redacted>

Thanks for reviewing.

Howard, I'm merging patches for v5.17 but I can't apply this until you
fix the warnings I mentioned.

Cheers,

Jeol

quoted

--
Patrick Williams