[libc-commits] [libc] [libc][cpp] add `atomic_signal_fence` (PR #82138)

[via libc-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-libc

Author: Schrodinger ZHU Yifan (SchrodingerZhu)

<details>
<summary>Changes</summary>

Add `atomic_signal_fence`. This will be useful in https://gustedt.gitlabpages.inria.fr/c23-library/#memset_explicit.

---
Full diff: https://github.com/llvm/llvm-project/pull/82138.diff


1 Files Affected:

- (modified) libc/src/__support/CPP/atomic.h (+15) 


``````````diff
diff --git a/libc/src/__support/CPP/atomic.h b/libc/src/__support/CPP/atomic.h
index 1c4478dfeab6de..61e2742661f21a 100644
--- a/libc/src/__support/CPP/atomic.h
+++ b/libc/src/__support/CPP/atomic.h
@@ -161,6 +161,21 @@ LIBC_INLINE void atomic_thread_fence(MemoryOrder mem_ord) {
 #endif
 }
 
+// Establishes memory synchronization ordering of non-atomic and relaxed atomic
+// accesses, as instructed by order, between a thread and a signal handler
+// executed on the same thread. This is equivalent to atomic_thread_fence,
+// except no instructions for memory ordering are issued. Only reordering of
+// the instructions by the compiler is suppressed as order instructs.
+LIBC_INLINE void atomic_signal_fence(MemoryOrder mem_ord) {
+#if __has_builtin(__atomic_signal_fence)
+  __atomic_signal_fence(int(mem_ord));
+#else
+  // if the builtin is not ready, use asm as a full compiler barrier.
+  (void)mem_ord;
+  asm volatile("" ::: "memory");
+#endif
+}
+
 } // namespace cpp
 } // namespace LIBC_NAMESPACE
 

``````````

</details>


https://github.com/llvm/llvm-project/pull/82138