Re: [PATCH v4 1/2] dt-bindings: arm: xen: Convert to DT schema

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Critical] The `$nodename` pattern requires a unit address, which breaks Linux's hardcoded Xen node detection.
--

commit e4969fb4c5852849c9ecbfff9b63719a0d10dc05
Author: Tejas Mutalikdesai <tejasmutalikdesai@gmail.com>

dt-bindings: arm: xen: Convert to DT schema

Convert the Xen ARM device tree binding documentation from the legacy
plain-text format to the DT schema format.

> diff --git a/Documentation/devicetree/bindings/arm/xen.yaml b/Documentation/devicetree/bindings/arm/xen.yaml
> new file mode 100644
> index 0000000000000..a22e950566c23
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/arm/xen.yaml
> @@ -0,0 +1,109 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +%YAML 1.2
> +---

[ ... ]

> +description:
> +  Xen ARM virtual platforms shall have a top-level "hypervisor" node with
> +  the properties defined below.
> +
> +properties:
> +  $nodename:
> +    pattern: "^hypervisor@[0-9a-f]+$"

[Severity: Critical]
Does enforcing a unit address here break existing kernel expectations?

The Linux kernel seems to strictly expect the node to be named exactly
"hypervisor" without a unit address.

In arch/arm/xen/enlighten.c:fdt_find_hyper_node(), the code uses:

    if (depth != 1 || strcmp(uname, "hypervisor") != 0)

And in drivers/firmware/efi/fdtparams.c:dt_params[], the path is hardcoded:

    .path = "/hypervisor/uefi",

If a hypervisor or devicetree generator updates its output to comply with this
new schema and adds a unit address (e.g., hypervisor@b0000000), won't the
kernel silently fail to detect Xen and its UEFI runtime services?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260617185518.43500-1-tejasmutalikdesai@gmail.com?part=1