memory-model/Documentation/explanation.txt

206 	P0 stores 1 to buf before storing 1 to flag, since it executes
270 	W: P0 stores 1 to flag   executes before
271 	X: P1 loads 1 from flag  executes before
272 	Y: P1 loads 0 from buf   executes before
273 	Z: P0 stores 1 to buf    executes before
502 executes on a separate CPU before the main program runs.
736 When CPU C executes a store instruction, it tells the memory subsystem
745 When a CPU executes a load instruction R, it first checks to see
768 premature executions; we simply say that the load executes at the
1188 has executed, which is possible if W executes shortly before R.
1196 second event executes.
1337 store to z does (the second cumul-fence), and P0's load executes after the
1356 before F executes.  The formal definition requires that E be linked to
1370 propagate to Y's CPU before X does, hence before Y executes and hence
1371 before the strong fence executes.  Because this fence is strong, we
1372 know that W will propagate to every CPU and to RAM before Z executes.
1375 propagate to every CPU and to RAM before F executes.
1500 executes.
1585 that G starts before C does, but also that any write which executes on
1604 executes before Y, but also (if X is a store) that X propagates to
1605 every CPU before Y executes.  Thus rcu-fence is sort of a
1653 at the time that Z executes.  From this, it can be shown (see the
1958 P1's store to x propagates to P0 before P0's load from x executes.
1975 test against NULL has been made but before the READ_ONCE() executes.
1987 program executes, there will not be any data races.  A "data race"
2022 (together referred to as xb, for "executes before").  However, there
2025 If X is a load and X executes before a store Y, then indeed there is
2029 some time after Y executes and thus after X executes.  But if X is a
2030 store, then even if X executes before Y it is still possible that X
2037 propagate to Y's CPU before Y executes.  (Or vice versa, of course, if
2038 Y executes before X -- then Y must propagate to X's CPU before X
2039 executes if Y is a store.)  This is expressed by the visibility
2070 	barrier executes, to propagate to all CPUs before any events
2099 executes.  Next, Z and Y are on the same CPU and the smp_rmb() fence
2102 executes.
2204 executes (assuming V does execute), ruling out the possibility of a
2209 executes before some marked access E.  We can do this by finding a
2424 	executes before, even if one or both is plain).