Slight docs update

qadapt-macro/src/lib.rs

@@ -152,9 +152,10 @@ fn escape_return(ts: TokenStream) -> TokenStream {
 /// Set up the QADAPT allocator to trigger a panic if any allocations happen during
 /// calls to this function.
 ///
-/// QADAPT will only track allocations in the thread that calls this function;
+/// QADAPT will only track allocations in the current function call;
 /// if (for example) this function receives the results of an allocation in a
-/// separate thread, QADAPT will not trigger a panic.
+/// separate thread, or defers allocations via closure/Future, those results
+/// will not trigger an error.
 #[proc_macro_attribute]
 pub fn allocate_panic(_attr: TokenStream, item: TokenStream) -> TokenStream {
     let mut protected_fn: Vec<TokenTree> = Vec::new();

src/lib.rs

@@ -1,9 +1,9 @@
-//! The Quick And Dirty Allocation Profiling Tool
+//! # The Quick And Dirty Allocation Profiling Tool
 //!
 //! This allocator is a helper for writing high-performance code that is allocation/drop free;
 //! for functions annotated with `#[allocate_panic]`, QADAPT will detect when allocations/drops
 //! happen during their execution (and execution of any functions they call) and throw a
-//! thread panic if this occurs. QADAPT-related code is stripped out during release builds,
+//! thread panic if this occurs. QADAPT-related code is *stripped out during release builds*,
 //! so no worries about random allocations crashing in production.
 //!
 //! Currently this crate is Nightly-only, but will work once `const fn` is in Stable.