Implement `Neg` for signed non-zero integers.

Negating a non-zero integer currently requires unpacking to a
primitive and re-wrapping. Since negation of non-zero signed
integers always produces a non-zero result, it is safe to
implement `Neg` for `NonZeroI{N}`.

The new `impl` is marked as stable because trait implementations
for two stable types can't be marked unstable.

1 files changed, 12 insertions, 0 deletions

diff --git a/tests/ui/numbers-arithmetic/overflowing-neg-nonzero.rs b/tests/ui/numbers-arithmetic/overflowing-neg-nonzero.rs
new file mode 100644
index 00000000000..565b7e86fc4
--- /dev/null
+++ b/tests/ui/numbers-arithmetic/overflowing-neg-nonzero.rs
@@ -0,0 +1,12 @@
+// run-fail
+// error-pattern:thread 'main' panicked at 'attempt to negate with overflow'
+// ignore-emscripten no processes
+// compile-flags: -C debug-assertions
+
+#![allow(arithmetic_overflow)]
+
+use std::num::NonZeroI8;
+
+fn main() {
+    let _x = -NonZeroI8::new(i8::MIN).unwrap();
+}