diff --git a/src/libstd/panicking.rs b/src/libstd/panicking.rs index 97d62d958ca4f..9542e7209b4cf 100644 --- a/src/libstd/panicking.rs +++ b/src/libstd/panicking.rs @@ -229,10 +229,10 @@ pub mod panic_count { thread_local! { static LOCAL_PANIC_COUNT: Cell<usize> = Cell::new(0) } // Sum of panic counts from all threads. The purpose of this is to have - // a fast path in `is_zero` (which is used by `panicking`). Access to - // this variable can be always be done with relaxed ordering because - // it is always guaranteed that, if `GLOBAL_PANIC_COUNT` is zero, - // `LOCAL_PANIC_COUNT` will be zero. + // a fast path in `is_zero` (which is used by `panicking`). In any particular + // thread, if that thread currently views `GLOBAL_PANIC_COUNT` as being zero, + // then `LOCAL_PANIC_COUNT` in that thread is zero. This invariant holds before + // and after increase and decrease, but not necessarily during their execution. static GLOBAL_PANIC_COUNT: AtomicUsize = AtomicUsize::new(0); pub fn increase() -> usize { @@ -263,6 +263,12 @@ pub mod panic_count { // Fast path: if `GLOBAL_PANIC_COUNT` is zero, all threads // (including the current one) will have `LOCAL_PANIC_COUNT` // equal to zero, so TLS access can be avoided. + // + // In terms of performance, a relaxed atomic load is similar to a normal + // aligned memory read (e.g., a mov instruction in x86), but with some + // compiler optimization restrictions. On the other hand, a TLS access + // might require calling a non-inlinable function (such as `__tls_get_addr` + // when using the GD TLS model). true } else { is_zero_slow_path()