__acpi_node_...ty_reference(9)

📦 version 7.3.1, in nixpkgs unstable

💖 __acpi_node_...ty_reference(9) 💖

okay bestie, welcome to the Kernel API Manual — yes that's the manual name, we're being fancy today ✨ this one's a mouthful of a name (__acpi_node_...ty_reference(9), the truncated little icon that it is), so grab a strawberry drink 🍓 and let's get into it.

🌸 NAME 🌸

__acpi_node_get_property_reference — literally returns a handle to the referenced object. that's the whole vibe. 🫶

✨ SYNOPSIS ✨

int __acpi_node_get_property_reference (const struct fwnode_handle
*fwnode , const char *propname , size_t index , size_t num_args , struct
fwnode_reference_args *args );

like, look at all those parameters. main character energy fr fr 👑

🎀 ARGUMENTS 🎀

  • fwnode — the firmware node we're pulling the property from, iykyk 🌈
  • propname — the name of the property. yes, this is the one we're hunting for 💅
  • index — which reference in the lineup you want back. pick a number, any number 🦋
  • num_args — the max number of arguments allowed after each reference. cap it, periodt ✋
  • args — where the returned reference (plus any optional arguments) gets stored. it may be NULL, no pressure 😌

🧚 DESCRIPTION 🧚

so here's the tea ☕ — this function finds the property with that name, checks that it's a package containing at least one object reference, and if it is (slay), it stores the ACPI device object pointer to the target object in args->adev. 🎯 and if the reference comes with arguments? those get stashed in the args->args[] array. obsessed. 💖

if there's more than one reference chillin' in the property value package, index is what picks which one you get back. lowkey the whole point of it. 🔥

and here's the cute part — you can literally leave holes in the property value set, like this:

Package () { "cs-gpios", Package () { ^GPIO, 19, 0, 0, ^GPIO, 20, 0, 0, 0, ^GPIO, 21, 0, 0, } }

call this function with index 2 or index 3 and you get -ENOENT bestie 😭 it's not a bug, it's a gap. and if the property doesn't contain any more values, -ENOENT is returned too. the NULL entry must be a single integer and preferably contain the value 0. noted, no cap. 📝

⭐ RETURN ⭐

  • 0 on success 🥳
  • a negative error code on failure 💀

🦋 SEE ALSO 🦋

kernel file ./drivers/acpi/property.c, plus the whole friend group:

ok real talk, that args parameter being allowed to be NULL is genuinely iconic — do with that what you will 🫶

and that's the page!! signed, sealed, delivered from ./drivers/acpi, dated October 2026, still rocking the __acpi_node_...ty_reference(9) title like the queen it is. mwah 💋💜

📦 where this page comes from