about summary refs log tree commit diff
path: root/include
diff options
context:
space:
mode:
authorTorvald Riegel <triegel@redhat.com>2014-09-14 20:04:54 +0200
committerTorvald Riegel <triegel@redhat.com>2014-11-20 11:58:42 +0100
commitff8714269c9312d9164456279a56b6f6c47e2771 (patch)
treeda8862c802a71cbb1269ba79cc7766edb5a03d8e /include
parent1ea339b69725cb2f30b5a84cb7ca96111c9a637b (diff)
downloadglibc-ff8714269c9312d9164456279a56b6f6c47e2771.tar.gz
glibc-ff8714269c9312d9164456279a56b6f6c47e2771.tar.xz
glibc-ff8714269c9312d9164456279a56b6f6c47e2771.zip
Add atomic operations similar to those provided by C11.
Diffstat (limited to 'include')
-rw-r--r--include/atomic.h212
1 files changed, 212 insertions, 0 deletions
diff --git a/include/atomic.h b/include/atomic.h
index 3e82b6a777..c4df8f10a1 100644
--- a/include/atomic.h
+++ b/include/atomic.h
@@ -542,6 +542,218 @@
   ({ __typeof (x) __x; __asm ("" : "=r" (__x) : "0" (x)); __x; })
 #endif
 
+/* This is equal to 1 iff the architecture supports 64b atomic operations.  */
+#ifndef __HAVE_64B_ATOMICS
+#error Unable to determine if 64-bit atomics are present.
+#endif
+
+/* The following functions are a subset of the atomic operations provided by
+   C11.  Usually, a function named atomic_OP_MO(args) is equivalent to C11's
+   atomic_OP_explicit(args, memory_order_MO); exceptions noted below.  */
+
+/* Each arch can request to use compiler built-ins for C11 atomics.  If it
+   does, all atomics will be based on these.  */
+#if USE_ATOMIC_COMPILER_BUILTINS
+
+/* We require 32b atomic operations; some archs also support 64b atomic
+   operations.  */
+void __atomic_link_error (void);
+# if __HAVE_64B_ATOMICS == 1
+#  define __atomic_check_size(mem) \
+   if ((sizeof (*mem) != 4) && (sizeof (*mem) != 8))			      \
+     __atomic_link_error ();
+# else
+#  define __atomic_check_size(mem) \
+   if (sizeof (*mem) != 4)						      \
+     __atomic_link_error ();
+# endif
+
+# define atomic_thread_fence_acquire() \
+  __atomic_thread_fence (__ATOMIC_ACQUIRE)
+# define atomic_thread_fence_release() \
+  __atomic_thread_fence (__ATOMIC_RELEASE)
+# define atomic_thread_fence_seq_cst() \
+  __atomic_thread_fence (__ATOMIC_SEQ_CST)
+
+# define atomic_load_relaxed(mem) \
+  ({ __atomic_check_size((mem)); __atomic_load_n ((mem), __ATOMIC_RELAXED); })
+# define atomic_load_acquire(mem) \
+  ({ __atomic_check_size((mem)); __atomic_load_n ((mem), __ATOMIC_ACQUIRE); })
+
+# define atomic_store_relaxed(mem, val) \
+  do {									      \
+    __atomic_check_size((mem));						      \
+    __atomic_store_n ((mem), (val), __ATOMIC_RELAXED);			      \
+  } while (0)
+# define atomic_store_release(mem, val) \
+  do {									      \
+    __atomic_check_size((mem));						      \
+    __atomic_store_n ((mem), (val), __ATOMIC_RELEASE);			      \
+  } while (0)
+
+/* On failure, this CAS has memory_order_relaxed semantics.  */
+# define atomic_compare_exchange_weak_relaxed(mem, expected, desired) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_compare_exchange_n ((mem), (expected), (desired), 1,		      \
+    __ATOMIC_RELAXED, __ATOMIC_RELAXED); })
+# define atomic_compare_exchange_weak_acquire(mem, expected, desired) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_compare_exchange_n ((mem), (expected), (desired), 1,		      \
+    __ATOMIC_ACQUIRE, __ATOMIC_RELAXED); })
+# define atomic_compare_exchange_weak_release(mem, expected, desired) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_compare_exchange_n ((mem), (expected), (desired), 1,		      \
+    __ATOMIC_RELEASE, __ATOMIC_RELAXED); })
+
+# define atomic_exchange_acquire(mem, desired) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_exchange_n ((mem), (desired), __ATOMIC_ACQUIRE); })
+# define atomic_exchange_release(mem, desired) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_exchange_n ((mem), (desired), __ATOMIC_RELEASE); })
+
+# define atomic_fetch_add_relaxed(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_add ((mem), (operand), __ATOMIC_RELAXED); })
+# define atomic_fetch_add_acquire(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_add ((mem), (operand), __ATOMIC_ACQUIRE); })
+# define atomic_fetch_add_release(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_add ((mem), (operand), __ATOMIC_RELEASE); })
+# define atomic_fetch_add_acq_rel(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_add ((mem), (operand), __ATOMIC_ACQ_REL); })
+
+# define atomic_fetch_and_acquire(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_and ((mem), (operand), __ATOMIC_ACQUIRE); })
+
+# define atomic_fetch_or_relaxed(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_or ((mem), (operand), __ATOMIC_RELAXED); })
+# define atomic_fetch_or_acquire(mem, operand) \
+  ({ __atomic_check_size((mem));					      \
+  __atomic_fetch_or ((mem), (operand), __ATOMIC_ACQUIRE); })
+
+#else /* !USE_ATOMIC_COMPILER_BUILTINS  */
+
+/* By default, we assume that read, write, and full barriers are equivalent
+   to acquire, release, and seq_cst barriers.  Archs for which this does not
+   hold have to provide custom definitions of the fences.  */
+# ifndef atomic_thread_fence_acquire
+#  define atomic_thread_fence_acquire() atomic_read_barrier ()
+# endif
+# ifndef atomic_thread_fence_release
+#  define atomic_thread_fence_release() atomic_write_barrier ()
+# endif
+# ifndef atomic_thread_fence_seq_cst
+#  define atomic_thread_fence_seq_cst() atomic_full_barrier ()
+# endif
+
+# ifndef atomic_load_relaxed
+#  define atomic_load_relaxed(mem) \
+   ({ __typeof (*(mem)) __atg100_val;					      \
+   __asm ("" : "=r" (__atg100_val) : "0" (*(mem)));			      \
+   __atg100_val; })
+# endif
+# ifndef atomic_load_acquire
+#  define atomic_load_acquire(mem) \
+   ({ __typeof (*(mem)) __atg101_val = atomic_load_relaxed (mem);	      \
+   atomic_thread_fence_acquire ();					      \
+   __atg101_val; })
+# endif
+
+# ifndef atomic_store_relaxed
+/* XXX Use inline asm here?  */
+#  define atomic_store_relaxed(mem, val) do { *(mem) = (val); } while (0)
+# endif
+# ifndef atomic_store_release
+#  define atomic_store_release(mem, val) \
+   do {									      \
+     atomic_thread_fence_release ();					      \
+     atomic_store_relaxed ((mem), (val));				      \
+   } while (0)
+# endif
+
+/* On failure, this CAS has memory_order_relaxed semantics.  */
+/* XXX This potentially has one branch more than necessary, but archs
+   currently do not define a CAS that returns both the previous value and
+   the success flag.  */
+# ifndef atomic_compare_exchange_weak_acquire
+#  define atomic_compare_exchange_weak_acquire(mem, expected, desired) \
+   ({ typeof (*(expected)) __atg102_expected = *(expected);		      \
+   *(expected) =							      \
+     atomic_compare_and_exchange_val_acq ((mem), (desired), *(expected));     \
+   *(expected) == __atg102_expected; })
+# endif
+# ifndef atomic_compare_exchange_weak_relaxed
+/* XXX Fall back to CAS with acquire MO because archs do not define a weaker
+   CAS.  */
+#  define atomic_compare_exchange_weak_relaxed(mem, expected, desired) \
+   atomic_compare_exchange_weak_acquire ((mem), (expected), (desired))
+# endif
+# ifndef atomic_compare_exchange_weak_release
+#  define atomic_compare_exchange_weak_release(mem, expected, desired) \
+   ({ typeof (*(expected)) __atg103_expected = *(expected);		      \
+   *(expected) =							      \
+     atomic_compare_and_exchange_val_rel ((mem), (desired), *(expected));     \
+   *(expected) == __atg103_expected; })
+# endif
+
+# ifndef atomic_exchange_acquire
+#  define atomic_exchange_acquire(mem, val) \
+   atomic_exchange_acq ((mem), (val))
+# endif
+# ifndef atomic_exchange_release
+#  define atomic_exchange_release(mem, val) \
+   atomic_exchange_rel ((mem), (val))
+# endif
+
+# ifndef atomic_fetch_add_acquire
+#  define atomic_fetch_add_acquire(mem, operand) \
+   atomic_exchange_and_add_acq ((mem), (operand))
+# endif
+# ifndef atomic_fetch_add_relaxed
+/* XXX Fall back to acquire MO because the MO semantics of
+   atomic_exchange_and_add are not documented; the generic version falls back
+   to atomic_exchange_and_add_acq if atomic_exchange_and_add is not defined,
+   and vice versa.  */
+#  define atomic_fetch_add_relaxed(mem, operand) \
+   atomic_fetch_add_acquire ((mem), (operand))
+# endif
+# ifndef atomic_fetch_add_release
+#  define atomic_fetch_add_release(mem, operand) \
+   atomic_exchange_and_add_rel ((mem), (operand))
+# endif
+# ifndef atomic_fetch_add_acq_rel
+#  define atomic_fetch_add_acq_rel(mem, operand) \
+   ({ atomic_thread_fence_release ();					      \
+   atomic_exchange_and_add_acq ((mem), (operand)); })
+# endif
+
+/* XXX The default for atomic_and_val has acquire semantics, but this is not
+   documented.  */
+# ifndef atomic_fetch_and_acquire
+#  define atomic_fetch_and_acquire(mem, operand) \
+   atomic_and_val ((mem), (operand))
+# endif
+
+/* XXX The default for atomic_or_val has acquire semantics, but this is not
+   documented.  */
+# ifndef atomic_fetch_or_acquire
+#  define atomic_fetch_or_acquire(mem, operand) \
+   atomic_or_val ((mem), (operand))
+# endif
+/* XXX Fall back to acquire MO because archs do not define a weaker
+   atomic_or_val.  */
+# ifndef atomic_fetch_or_relaxed
+#  define atomic_fetch_or_relaxed(mem, operand) \
+   atomic_fetch_or_acquire ((mem), (operand))
+# endif
+
+#endif /* !USE_ATOMIC_COMPILER_BUILTINS  */
+
 
 #ifndef atomic_delay
 # define atomic_delay() do { /* nothing */ } while (0)