mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Fabian Pflug <f.pflug@pengutronix.de>
To: Ahmad Fatoum <a.fatoum@pengutronix.de>, barebox@lists.infradead.org
Cc: fpg@pengutronix.de, chalianis1@gmail.com
Subject: Re: [PATCH RFT 1/4] efi: payload: ignore ESP state.dtb if device tree is populated
Date: Tue, 01 Sep 2026 17:16:41 +0200	[thread overview]
Message-ID: <0f9f2f7007672dbceb1fc53f2c8705115d8d5677.camel@pengutronix.de> (raw)
In-Reply-To: <20260826121640.2936023-2-a.fatoum@pengutronix.de>

Hey,

On Wed, 2026-08-26 at 14:15 +0200, Ahmad Fatoum wrote:
> In preparation for allowing the state nodes to ship inside the barebox
> binary (as we do for non-EFI-payload already), add provisions to skip
> the external file if barebox already has a populated DT.
> 
> Also for the case that a state had already been set up, skip manual
> creation of the state node.
> 
> Assisted-by: Claude:opus-5
> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> ---
>  efi/payload/init.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/efi/payload/init.c b/efi/payload/init.c
> index 1e5f2eb8cf56..db4bc25682a3 100644
> --- a/efi/payload/init.c
> +++ b/efi/payload/init.c
> @@ -284,6 +284,7 @@ static int efi_late_init(void)
>  {
>  	const char *state_desc = "/boot/EFI/barebox/state.dtb";
>  	struct device_node *state_root = NULL;
> +	struct device_node *root;
>  	size_t size;
>  	void *fdt;
>  	int ret;
> @@ -303,6 +304,17 @@ static int efi_late_init(void)
>  		return 0;
>  	}
>  
> +	/*
> +	 * A device tree built into barebox takes precedence: any state it
> +	 * describes has been instantiated by the state driver already.
> +	 */
> +	root = of_get_root_node();
> +	if (root && !list_empty(&root->children)) {
> +		pr_warn("device tree is populated, ignoring %s\n", state_desc);
> +		free(fdt);
> +		return 0;
> +	}

I would put the check a bit earlier and more sophisticated. Currently I either get an info message, that the state.dtb
could not be found, or a warning, if it could be found, but the compiled in dts takes precedence.

The first is confusing when looking at dmesg.

/Fabian

> +
>  	state_root = of_unflatten_dtb(fdt, size);
>  	free(fdt);
>  
> @@ -320,6 +332,10 @@ static int efi_late_init(void)
>  			return 0;
>  		}
>  
> +		/* probing the registered tree may have instantiated it already */
> +		if (state_by_node(np))
> +			return 0;
> +
>  		state = state_new_from_node(np, false);
>  		if (IS_ERR(state))
>  			return PTR_ERR(state);



  reply	other threads:[~2026-09-01 15:17 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-26 12:15 [PATCH RFT 0/4] efi: payload: allow extension via fragments Ahmad Fatoum
2026-08-26 12:15 ` [PATCH RFT 1/4] efi: payload: ignore ESP state.dtb if device tree is populated Ahmad Fatoum
2026-09-01 15:16   ` Fabian Pflug [this message]
2026-08-26 12:15 ` [PATCH RFT 2/4] kbuild: dtc: introduce empty fallback device tree Ahmad Fatoum
2026-08-26 12:15 ` [PATCH RFT 3/4] efi: payload: export device tree in barebox-dtb EFI variable Ahmad Fatoum
2026-08-26 12:15 ` [PATCH RFT 4/4] Documentation: efi: describe device tree handling Ahmad Fatoum
2026-08-26 22:56 ` [PATCH RFT 0/4] efi: payload: allow extension via fragments chalianis1
2026-08-27  7:31   ` Ahmad Fatoum

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=0f9f2f7007672dbceb1fc53f2c8705115d8d5677.camel@pengutronix.de \
    --to=f.pflug@pengutronix.de \
    --cc=a.fatoum@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=chalianis1@gmail.com \
    --cc=fpg@pengutronix.de \
    /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