mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: gianluca <gianlucarenzi@eurek.it>,
	Barebox List <barebox@lists.infradead.org>
Subject: Re: Barebox 2020.01.0 and Device Tree from Linux Kernel 5.4
Date: Tue, 28 Jan 2020 13:45:13 +0100	[thread overview]
Message-ID: <4f81ce73-ead6-fad4-1336-2f89819227ce@pengutronix.de> (raw)
In-Reply-To: <8f731e2b-9c11-d183-a660-e0a7ea621074@eurek.it>

Hello,

(please keep the barebox mailing list in CC)

On 1/27/20 3:02 PM, gianluca wrote:
> On 01/24/2020 05:13 PM, Ahmad Fatoum wrote:
>> Hello,
>>
>> On 1/24/20 4:32 PM, gianluca wrote:
>>> I was wondering if the device tree from (latest) Linux Kernel can be used when building Barebox 2020.01.0 for iMX6 compatible custom board.
>>> For sure the include path and other stuff are quite different (kernel and barebox), so I am pretty sure it will fail to build with some sort of "foreign" device-tree.
>>
>> barebox already regularly imports Linux's device tree directories into its dts/ directory.
>> v2020.01.0 contains the v5.4-rc7 state.
>>
> 
> Good to hear...
> 
>> Another issue, will be the drivers to have full access from Barebox-Point-Of-View. If the device-tree properties and compatible strings are quite different, Barebox will fail to activating/probing/using the internal driver.
>>
>> barebox driver compatibles should be aligned with the kernel's.
>> If a barebox driver lacks handling for a property, the driver can be extended.
>>
>>> So I am asking:
>>> There is a "official" way to manage those differencies?
>>
>> I recently added a short section about this in the Documentation
>> (https://barebox.org/doc/latest/devicetree/index.html):
>>
>>   "For supporting architectures, barebox device trees are located in arch/$ARCH/dts.
>>    Usually the barebox board.dts imports the upstream device tree under dts/src/$ARCH
>>    with #include "$ARCH/board.dts" and then extends it with barebox-specifics like
>>    Barebox state, environment or boot-time device configuration."
>>
>> Take a look at arch/arm/dts/imx6q-marsboard.dts to see how that looks in practice.
>> The kernel device tree is reused as is and extended slightly for barebox use.
> 
> So, basically the device-tree stuff in Barebox are different from the Linux kernel one only for the #include path?

The barebox device trees for ARM are in arch/arm/dts. Best practice
is that these device trees include the upstream device tree with just
the barebox changes custom.

> The biggest issue I've found here, is the compatibility of the device-tree nodes for my board referring the standard imx6 board. For example:
> 
> &uart3 {
>     status = "okay";
> };
> 
> is good as you are using the standard pinout for uart3.
> 
> But, if I am using a different pinout layout for the same device, this is not working at all.
> 
> Maybe I have to find the device-node parent, then looking for its configuration, and if it fits my needs, ok I will use it.
> 
> Otherwise I have to redefine a new configuration node:
> 
> uart3 {
>     pinctrl = ....
>     interrupts = ...
> };
> 
> But this does not work. uart3 is already defined into the parent node of imx6q.dtsi.
> 
> In that case I need to redefine a new name:
> 
> my_uart3 {
>    pinctrl = ...
>    interrupts = ...
> };
> 
> &my_uart3 {
>    status = "okay";
> };
> 
> I hope you understand what I am trying to tell you.

You don't need to do this, you can just write

&uart3 {
	pinctrl = ...;
	status = "okay";
};

These properties will override the properties from earlier in the preprocessed file.

Cheers
Ahmad

> 
> Best regards,
> Gianluca


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | https://www.pengutronix.de/ |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

      parent reply	other threads:[~2020-01-28 12:45 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-24 15:32 gianluca
2020-01-24 16:13 ` Ahmad Fatoum
     [not found]   ` <8f731e2b-9c11-d183-a660-e0a7ea621074@eurek.it>
2020-01-28 12:45     ` Ahmad Fatoum [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4f81ce73-ead6-fad4-1336-2f89819227ce@pengutronix.de \
    --to=a.fatoum@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=gianlucarenzi@eurek.it \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox