"Alice Ryhl" <aliceryhl@xxxxxxxxxx> writes:
> On Fri, Aug 15, 2025 at 09:30:40AM +0200, Andreas Hindborg wrote:
<cut>
>> +pub fn kstrtobool(string: &CStr) -> Result<bool> {
>> + let mut result: bool = false;
>> +
>> + // SAFETY: `string` is a valid null-terminated C string, and `result` is a valid
>> + // pointer to a bool that we own.
>> + let ret =
>> + unsafe { bindings::kstrtobool(string.as_char_ptr(), core::ptr::from_mut(&mut result)) };
>
> Using ptr::from_mut here seesm excessive IMO. I think that function
> makes sense when it replaces an explicit `as` cast, but now when it can
> be done by a coercion. This is perfectly readable:
>
> let ret = unsafe { bindings::kstrtobool(string.as_char_ptr(), &mut result) };
>
> Or if you insist, you could directly create a raw pointer:
>
> let ret = unsafe { bindings::kstrtobool(string.as_char_ptr(), &raw mut result) };
Oh, I'll do the implicit coercion then.
I was using `&mut result as *mut bool` but clippy told me to do the `from_mut`
instead.
Best regards,
Andreas Hindborg
