Auto merge of #2517 - saethlin:zst-field-retagging, r=RalfJung
Skip field retagging on ZSTs, it can take forever I just tried running the `alloc`'s tests with `miri-test-libstd` with field retagging enabled. The test suite eventually hangs on a few tests which pass around ZSTs that have a lot of fields. I don't really know how to test this effectively. The test passes, but if you remove this fast-path it effectively just hangs the interpreter. And since it hangs _inside_ a step, there's no hope for doing some kind of timeout within the test.
This commit is contained in:
commit
284b59c4dc
2 changed files with 14 additions and 0 deletions
|
|
@ -973,6 +973,14 @@ pub trait EvalContextExt<'mir, 'tcx: 'mir>: crate::MiriEvalContextExt<'mir, 'tcx
|
|||
}
|
||||
|
||||
fn visit_value(&mut self, place: &PlaceTy<'tcx, Provenance>) -> InterpResult<'tcx> {
|
||||
// If this place is smaller than a pointer, we know that it can't contain any
|
||||
// pointers we need to retag, so we can stop recursion early.
|
||||
// This optimization is crucial for ZSTs, because they can contain way more fields
|
||||
// than we can ever visit.
|
||||
if !place.layout.is_unsized() && place.layout.size < self.ecx.pointer_size() {
|
||||
return Ok(());
|
||||
}
|
||||
|
||||
if let Some((ref_kind, protector)) = qualify(place.layout.ty, self.kind) {
|
||||
self.retag_place(place, ref_kind, self.retag_cause, protector)?;
|
||||
} else if matches!(place.layout.ty.kind(), ty::RawPtr(..)) {
|
||||
|
|
|
|||
|
|
@ -0,0 +1,6 @@
|
|||
//@compile-flags: -Zmiri-retag-fields
|
||||
// Checks that the test does not run forever (which relies on a fast path).
|
||||
fn main() {
|
||||
let array = [(); usize::MAX];
|
||||
drop(array); // Pass the array to a function, retagging its fields
|
||||
}
|
||||
Loading…
Add table
Add a link
Reference in a new issue