Re: [PATCH v3 1/7] rust: firmware: add request_into_buf()

[Timur Tabi <ttabi@nvidia.com>]

On Fri, 2026-07-03 at 10:51 +0800, Alvin Sun wrote:


> > +pub fn request_into_buf(name: &CStr, dev: &Device, buf: &mut [u8]) -> Result {
> > +    // `as_mut_ptr()` on an empty slice returns a non-NULL pointer to
> > +    // memory which the loader does not own. Passing that pointer with `size == 0`
> > +    // makes the loader believe that it is buffer it allocated itself, so when
> > +    // `release_firmware()` is called, it will vfree the pointer and trigger a
> > +    // bug. Reject empty slices to avoid this situation.
> > +    if buf.is_empty() {
> > +        return Err(crate::error::code::EINVAL);
> 
> `EINVAL` is already in prelude, you can use it directly.

I would have to add it to the list of imports.

> > +    // SAFETY: `pfw` is a valid pointer to a NULL initialized `bindings::firmware` pointer.
> > +    // `name` and `dev` are valid as by their type invariants. `buf` is a valid writable
> > +    // buffer of `buf.len()` bytes.
> > +    let ret = unsafe {
> > +        bindings::request_firmware_into_buf(
> > +            pfw,
> > +            name.as_char_ptr(),
> > +            dev.as_raw(),
> > +            buf.as_mut_ptr().cast(),
> > +            buf.len(),
> > +        )
> > +    };
> > +    if ret != 0 {
> > +        return Err(Error::from_errno(ret));
> > +    }
> 
> `to_result` can be used here to simplify.

Will add in v4.