memory-model/Documentation/explanation.txt

246 this outcome to occur, and in fact it does sometimes occur on x86 and
374 accessed normally without READ_ONCE() or WRITE_ONCE().  In fact, they
415 Another issue arises from the fact that in C, arguments to many
423 between them.  (In fact, if the function calls are inlined then the
500 loaded.  But in fact, the value stored does not really depend on
808 (In fact, the picture needs to be a little more complicated than this.
910 maintaining cache coherence and the fact that a CPU can't operate on a
1029 to each CPU before the store X does.  Then the fact that X and Y are
1207 architecture.  In fact, as of version 4.15, the kernel automatically
1255 The equivalent fact need not hold if W ->rfi R (i.e., W and R are on
1385 followed by two cumul-fences and an rfe link, utilizing the fact that
1423 In summary, the fact that the hb relation links memory access events
1467 coherence order, contradicting the fact that E ->coe W.  If E was a
1470 contradicting the fact that E ->fre W.
1508 In summary, the fact that the pb relation links events in the order
1523 be nested, although we won't make use of this fact.
1606 	E ->rcu-gp F means that E and F are in fact the same event,
1618 grace period which ends before Z begins.  (In fact it covers more than
1645 imply that E must execute before F; in fact, each synchronize_rcu()
1691 CPUs.  (Perhaps this fact should lead us to say that rcu-fence isn't
1708 Justifying the axiom isn't easy, but it is in fact a valid
1881 	srcu_read_unlock().  In fact, it is possible for two SRCU
1959 shared variable temporarily.  In fact, this is the only way for
2020 The LKMM includes locking.  In fact, there is special code for locking
2348 The second complicating factor mentioned above arises from the fact
2460 In fact, we go one step further: Since R is a read, we say that R is