Re: [PATCH v3] rust: Add support for feeding entropy to randomness pool

[Matthew Maurer]

On Sun, Dec 28, 2025 at 1:13 AM Brendan Shephard <bshephar@xxxxxxxxxxxx> wrote:
>
> On Fri, Dec 26, 2025 at 07:15:10PM +0000, Matthew Maurer wrote:
> > +/// Adds the given buffer to the entropy pool, but does not credit any entropy.
> > +///
> > +/// This is intended for use mixing in data that is likely to differ between devices or boots, but
> Yeah, I agree with Alexandre here. I think this would probably be better
> written as: "This function mixes in data that is likely to differ between
> devices or boots". Or, "add_device_randomness mixes in data that is
> likely to differ between devices or boots." If you would prefer to avoid
> saying "this function".

I believe the grammar to be correct here - if it helps, the "use" in
this case is the "yoos" variant of use, not the "yooz" variant.

That said, if it's confusing, it's confusing. I'll reword it as suggested.

> > +/// may otherwise be predictable. Examples include MAC addresses or RTC values. This slightly
> > +/// improves randomness in entropy-constrained environments (especially common for embedded
> > +/// devices).
> > +pub fn add_device_randomness(buf: &[u8]) {
> > +    // SAFETY: We just need the pointer to be valid for the length, which a slice provides.
> > +    unsafe { bindings::add_device_randomness(buf.as_ptr().cast::<c_void>(), buf.len()) };
> > +}
> >
> > ---
> > base-commit: 008d3547aae5bc86fac3eda317489169c3fda112
> > change-id: 20251029-add-entropy-f57e12ebe110
> >
> > Best regards,
> > --
> > Matthew Maurer <mmaurer@xxxxxxxxxx>
> >
> >