From 05bea95f17e4aea03d1781d5409cb844b272e2a2 Mon Sep 17 00:00:00 2001
From: Allard Hendriksen <ahendriksen@nvidia.com>
Date: Thu, 27 Jul 2023 17:09:51 +0200
Subject: [PATCH] Do not move arrival_token in barrier::try_wait

Take a mutable reference instead.

Moving the token into try_wait makes no sense. The caller has to be able
to retry the try wait with the same token.

For more details and C++ spec, see:

https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/p2643r1.html
---
 .../std/detail/libcxx/include/__cuda/barrier.h   | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/include/cuda/std/detail/libcxx/include/__cuda/barrier.h b/include/cuda/std/detail/libcxx/include/__cuda/barrier.h
index 90a13f658f..fc7a3ea9ea 100644
--- a/include/cuda/std/detail/libcxx/include/__cuda/barrier.h
+++ b/include/cuda/std/detail/libcxx/include/__cuda/barrier.h
@@ -274,7 +274,7 @@ friend class _CUDA_VSTD::__barrier_poll_tester_parity;
             NV_PROVIDES_SM_90, (
                 int32_t __ready = 0;
                 if (!__isClusterShared(&__barrier)) {
-                    return _CUDA_VSTD::__call_try_wait(__barrier, _CUDA_VSTD::move(__token));
+                    return _CUDA_VSTD::__call_try_wait(__barrier, __token);
                 }
                 else if (!__isShared(&__barrier)) {
                     __trap();
@@ -291,11 +291,11 @@ friend class _CUDA_VSTD::__barrier_poll_tester_parity;
                 return __ready;
             ), NV_PROVIDES_SM_80, (
                 if (!__isShared(&__barrier)) {
-                    return _CUDA_VSTD::__call_try_wait(__barrier, _CUDA_VSTD::move(__token));
+                    return _CUDA_VSTD::__call_try_wait(__barrier, __token);
                 }
                 return __test_wait_sm_80(__token);
             ), NV_ANY_TARGET, (
-                    return _CUDA_VSTD::__call_try_wait(__barrier, _CUDA_VSTD::move(__token));
+                    return _CUDA_VSTD::__call_try_wait(__barrier, __token);
             )
         )
     }
@@ -304,7 +304,7 @@ friend class _CUDA_VSTD::__barrier_poll_tester_parity;
     _LIBCUDACXX_INLINE_VISIBILITY
     bool __try_wait(arrival_token __token, _CUDA_VSTD::chrono::nanoseconds __nanosec) const {
         if (__nanosec.count() < 1) {
-            return __try_wait(_CUDA_VSTD::move(__token));
+            return __try_wait(__token);
         }
 
         NV_DISPATCH_TARGET(
@@ -525,16 +525,16 @@ friend class _CUDA_VSTD::__barrier_poll_tester_parity;
 
     template<class _Rep, class _Period>
     _LIBCUDACXX_NODISCARD_ATTRIBUTE _LIBCUDACXX_INLINE_VISIBILITY
-    bool try_wait_for(arrival_token && __token, const _CUDA_VSTD::chrono::duration<_Rep, _Period>& __dur) {
+    bool try_wait_for(arrival_token & __token, const _CUDA_VSTD::chrono::duration<_Rep, _Period>& __dur) {
         auto __nanosec = _CUDA_VSTD::chrono::duration_cast<_CUDA_VSTD::chrono::nanoseconds>(__dur);
 
-        return __try_wait(_CUDA_VSTD::move(__token), __nanosec);
+        return __try_wait(__token, __nanosec);
     }
 
     template<class _Clock, class _Duration>
     _LIBCUDACXX_NODISCARD_ATTRIBUTE _LIBCUDACXX_INLINE_VISIBILITY
-    bool try_wait_until(arrival_token && __token, const _CUDA_VSTD::chrono::time_point<_Clock, _Duration>& __time) {
-        return try_wait_for(_CUDA_VSTD::move(__token), (__time - _Clock::now()));
+    bool try_wait_until(arrival_token & __token, const _CUDA_VSTD::chrono::time_point<_Clock, _Duration>& __time) {
+        return try_wait_for(__token, (__time - _Clock::now()));
     }
 
     template<class _Rep, class _Period>