Lines Matching full:operation
113 memory operation can happen on any thread between the load and store.
116 another operation; it is normally used along with Monotonic memory operations.
183 guarantees the operation to be lock-free, so it does not depend on the data
186 operations; if you need such an operation, use explicit locking.
248 to appear as a single operation.
263 operation.
268 operation to after it, and move non-Acquire loads from before an Acquire
269 operation to after it.
277 such a fence after the equivalent Monotonic operation is sufficient to
278 maintain Acquire semantics for a memory operation.
292 operation.
297 operation to before it, and move non-Release stores from after an Release
298 operation to before it.
301 See the section on Acquire; a fence before the relevant operation is usually
318 operation, and vice versa.
325 This operation has Acquire and Release semantics; see the sections on Acquire
368 operation.
371 that they return true for any operation which is volatile or at least
380 Release, and for the address accessed by any Monotonic operation.
385 it doesn't replace an atomic load or store with a non-atomic operation.
390 * ``memcpyopt``: An atomic operation cannot be optimized into part of a
401 atomic operation, but it is fairly tricky. DSE delegates this reasoning to
421 lock-free ``cmpxchg``; such an operation can be used to implement all the other
425 error when given an operation which cannot be implemented. (The LLVM code
440 operation. Loads and stores generate normal instructions. ``cmpxchg`` and
