Arch Linux

zse

Member

Registered: 2024-05-28

Posts: 46

New kernel warning: rust_binder: Loaded Rust Binder

A recent kernel update started to cause a new kernel warning on boot, just saying "rust_binder: Loaded Rust Binder." I'm assuming this is about this.

Anyone knows why it warrants a kernel warning? Should Arch really load this per default, given that I'd imagine very few users have a use for it?

Lone_Wolf

Administrator

From: Netherlands, Europe

Registered: 2005-10-04

Posts: 14,795

Re: New kernel warning: rust_binder: Loaded Rust Binder

No warning here, just a line in journalctl. Please post your full journal.

Arch Linux devs aim to stay as close to upstream as possible. If kernel devs enabled the rust_binder by default arch doesn't change that without a very good reason.

---

Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.

clean chroot building not flexible enough ?
Try clean chroot manager by graysky

zse

Member

Registered: 2024-05-28

Posts: 46

Re: New kernel warning: rust_binder: Loaded Rust Binder

Hm well, I haven't looked much into it, so I don't know how upstream this decision to enable per default is. It could just as well be a packaging decision/mistake...

Anyway, for the syslog are the warnings only enough? otherwise it gets quite long

$ journalctl -b -k -o short-monotonic -p warning --no-pager
[    0.759595] archlinux kernel: ACPI: \_SB_.PCI0.GP11: Overriding _PRW sleep state (S4) by S0 from power resources
[    0.759646] archlinux kernel: ACPI: \_SB_.PCI0.GP12: Overriding _PRW sleep state (S4) by S0 from power resources
[    0.759888] archlinux kernel: ACPI: \_SB_.PCI0.GP19.NHI0: Overriding _PRW sleep state (S4) by S0 from power resources
[    0.759919] archlinux kernel: ACPI: \_SB_.PCI0.GP19.NHI1: Overriding _PRW sleep state (S4) by S0 from power resources
[    0.782038] archlinux kernel: i2c_designware AMDI0010:00: [Firmware Bug]: DSDT uses known not-working I2C bus speed 400000, forcing it to 100000
[    0.816464] archlinux kernel: i8042: PNP: PS/2 appears to have AUX port disabled, if this is incorrect please boot with i8042.nopnp
[    0.817555] archlinux kernel: rust_binder: Loaded Rust Binder.
...

EDIT: grepping the full log (so not only warnings) for "rust" also does not show anything more related:

$ journalctl -b -k -o short-monotonic | grep rust
[    0.799291] archlinux kernel: Initialise system trusted keyrings
[    0.817555] archlinux kernel: rust_binder: Loaded Rust Binder.
[    2.856611] archlinux kernel: amdgpu 0000:73:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)

Last edited by zse (2025-12-18 11:47:27)

loqs

Member

Registered: 2014-03-06

Posts: 18,796

Re: New kernel warning: rust_binder: Loaded Rust Binder

https://github.com/torvalds/linux/blame … in.rs#L289 no mention of why loading the module warrants a warning in the source.

@ze does your system use binder for anything such as an Android emulator?

zse

Member

Registered: 2024-05-28

Posts: 46

Re: New kernel warning: rust_binder: Loaded Rust Binder

https://github.com/torvalds/linux/blame … in.rs#L289 no mention of why loading the module warrants a warning in the source.

@ze does your system use binder for anything such as an Android emulator?

Nope, I have no use for it. I'm seeing CONFIG_ANDROID_BINDER_IPC_RUST=y in the /proc/config.gz, so I believe the issue is that this gets compiled directly into the kernel, but it should rather be a loadable module.

Last edited by zse (2025-12-19 15:35:11)

xerxes_

Member

Registered: 2018-04-29

Posts: 1,018

Re: New kernel warning: rust_binder: Loaded Rust Binder

@zse: check if you have loaded binder module: 'lsmod | grep -i binder'.

In kernel 6.12.61-1-lts there is also binder module, but not written in rust and don't appear in logs, it's built in kernel:

$ zgrep -i binder /proc/config.gz 
CONFIG_ANDROID_BINDER_IPC=y
CONFIG_ANDROID_BINDERFS=y
CONFIG_ANDROID_BINDER_DEVICES=""
# CONFIG_ANDROID_BINDER_IPC_SELFTEST is not set
$ modinfo binder
name:           binder
filename:       (builtin)
license:        GPL v2
file:           drivers/android/binder
parm:           debug_mask:uint
parm:           devices:charp

So, in kernel 6.18 and up binder rewritten in rust replaces binder written in C? When binder appear in kernel and what for is used in Linux distros?

zse

Member

Registered: 2024-05-28

Posts: 46

Re: New kernel warning: rust_binder: Loaded Rust Binder

So, in kernel 6.18 and up binder rewritten in rust replaces binder written in C? When binder appear in kernel and what for is used in Linux distros?

Yes, it apparently has been rewritten. As mentioned, binder is an IPC mechanism that is used on Android, so it presumably can be used by Android emulators to run Android apps.

I just checked, the rust_binder driver cannot be compiled as a module for some reason, so it has to be a kernel built-in, which explains why it is always loaded... kind of annoying, oh well.

xerxes_

Member

Registered: 2018-04-29

Posts: 1,018

Re: New kernel warning: rust_binder: Loaded Rust Binder

Maybe this (initcall_blacklist=) ? :
https://unix.stackexchange.com/question … t-a-module
But I didn't test it yet.
Might be something like that:

initcall_blacklist=rust_binder_init
initcall_blacklist=binder_init

as kernel boot parameter.

Last edited by xerxes_ (2025-12-20 22:53:34)

zse

Member

Registered: 2024-05-28

Posts: 46

Re: New kernel warning: rust_binder: Loaded Rust Binder

Thanks for trying to help, but using (and maintaining) a fairly obscure hack to prevent the loading is absolutely not worth it for such a minor annoyance in my book (and will also not prevent the code from sitting in RAM).

xerxes_

Member

Registered: 2018-04-29

Posts: 1,018

Re: New kernel warning: rust_binder: Loaded Rust Binder

Would you mark thread as [SOLVED] by editing first post?

Last edited by xerxes_ (2025-12-23 21:18:57)

zse

Member

Registered: 2024-05-28

Posts: 46

Re: New kernel warning: rust_binder: Loaded Rust Binder

Well, it is technically not resolved, it is still unclear why the module is printing a warning. Though I guess only upstream kernel devs could answer resp. fix...