Auto merge of #90442 - ChrisDenton:win-tls-dtor, r=alexcrichton

Windows thread-local keyless drop

`#[thread_local]` allows us to maintain a per-thread list of destructors. This also avoids the need to synchronize global data (which is particularly tricky within the TLS callback function).

r? `@alexcrichton`

library/std/src/sys/windows/thread_local_dtor.rs

@@ -1,4 +1,28 @@
+//! Implements thread-local destructors that are not associated with any
+//! particular data.
+
 #![unstable(feature = "thread_local_internals", issue = "none")]
 #![cfg(target_thread_local)]
 
-pub use crate::sys_common::thread_local_dtor::register_dtor_fallback as register_dtor;
+// Using a per-thread list avoids the problems in synchronizing global state.
+#[thread_local]
+static mut DESTRUCTORS: Vec<(*mut u8, unsafe extern "C" fn(*mut u8))> = Vec::new();
+
+pub unsafe fn register_dtor(t: *mut u8, dtor: unsafe extern "C" fn(*mut u8)) {
+    DESTRUCTORS.push((t, dtor));
+}
+
+/// Runs destructors. This should not be called until thread exit.
+pub unsafe fn run_keyless_dtors() {
+    // Drop all the destructors.
+    //
+    // Note: While this is potentially an infinite loop, it *should* be
+    // the case that this loop always terminates because we provide the
+    // guarantee that a TLS key cannot be set after it is flagged for
+    // destruction.
+    while let Some((ptr, dtor)) = DESTRUCTORS.pop() {
+        (dtor)(ptr);
+    }
+    // We're done so free the memory.
+    DESTRUCTORS = Vec::new();
+}

library/std/src/sys/windows/thread_local_key.rs

@@ -196,6 +196,8 @@ pub static p_thread_callback: unsafe extern "system" fn(c::LPVOID, c::DWORD, c::
 unsafe extern "system" fn on_tls_callback(h: c::LPVOID, dwReason: c::DWORD, pv: c::LPVOID) {
     if dwReason == c::DLL_THREAD_DETACH || dwReason == c::DLL_PROCESS_DETACH {
         run_dtors();
+        #[cfg(target_thread_local)]
+        super::thread_local_dtor::run_keyless_dtors();
     }
 
     // See comments above for what this is doing. Note that we don't need this