Lines Matching full:before

15  *    barrier before sending the IPI
22  * order to enforce the guarantee that any writes occurring on CPU0 before
41  * so it's possible to have "r1 = x" reordered before "y = 1" at any
48  * before the IPI-induced memory barrier on CPU1.
50  * B) Userspace thread execution before IPI vs membarrier's memory
58  * order to enforce the guarantee that any writes occurring on CPU1 before
78  * before (b) (although not before (a)), so we get "r1 = 0".  This violates
179 	 * ensure that memory on remote CPUs that occur before the IPI  in ipi_sync_core()
196 	 * to the current task before the current task resumes.  We could  in ipi_rseq()
217 	 * before registration.  in ipi_sync_rq_state()
225 	 * Issue a memory barrier before clearing membarrier_state to  in membarrier_exec_mmap()
309 	 * waiting for the last IPI. Matches memory barriers before  in membarrier_global_expedited()
406 		 * task in the same mm just before, during, or after  in membarrier_private_expedited()
430 	 * waiting for the last IPI. Matches memory barriers before  in membarrier_private_expedited()
451 		 * access following registration is reordered before  in sync_runqueues_membarrier_state()