lint on core::ffi::c_void as an &/&mut referent - #159986
Conversation
|
cc @rust-lang/miri |
rustbot
added
S-waiting-on-review
|
r? @fee1-dead rustbot has assigned @fee1-dead. Use Why was this reviewer chosen?The reviewer was selected based on:
|
rustbot
added
A-FFI
This comment has been minimized.
This comment has been minimized.
Also, fix false negatives in `c_void_returns`.
33e8e23 to
2949956
Compare
| unsafe extern "C" fn libffi_closure_callback<'tcx>( | ||
| _cif: &libffi::low::ffi_cif, | ||
| _result: &mut c_void, | ||
| _result: &mut (), |
There was a problem hiding this comment.
Good catch!
There was a problem hiding this comment.
Shouldn't this be a raw pointer? Presumably it's mutating more than 0 bytes
There was a problem hiding this comment.
The type is forced on us by libffi.
Also we're not doing anything with this reference.
There was a problem hiding this comment.
I'm not seeing how libffi itself forces &mut c_void? If I look at the repo, it only uses *mut c_void.
There was a problem hiding this comment.
It forces &mut <something>.
7 participants
core::ffi::c_voidhas size 1 and a highly non-trivial validity invariant. Creating an&or&mutreference to it is a recipe for UB. It's also widespread, e.g. here in Miri, here across GitHub. Let's lint on it!Also fixes false negatives in
c_void_returns. Like that lint, this should probably become deny-by-default eventually.See also #159935
@rustbot label T-lang needs-fcp A-FFI