Messages in this thread | | | Date | Tue, 26 Jun 2018 09:49:57 -0400 (EDT) | From | Alan Stern <> | Subject | Re: [PATCH] doc: Update wake_up() & co. memory-barrier guarantees |
| |
On Tue, 26 Jun 2018, Andrea Parri wrote:
> > > -A write memory barrier is implied by wake_up() and co. if and only if they > > > -wake something up. The barrier occurs before the task state is cleared, and so > > > -sits between the STORE to indicate the event and the STORE to set TASK_RUNNING: > > > +A general memory barrier is executed by wake_up() if it wakes something up. > > > +If it doesn't wake anything up then a memory barrier may or may not be > > > +executed; you must not rely on it. The barrier occurs before the task state > > > +is accessed, in part., it sits between the STORE to indicate the event and > > > +the STORE to set TASK_RUNNING: > > > > Minor suggestion: Instead of "in part.", how about "that is"? > > > > (I generally find "in part." to be at least a little confusing, > > probably because "part" is itself a word and "in part" is a > > reasonably common phrase in English.) > > Mmh, the fact is that that "before the task state is accessed" does want > to include the LOAD from ->state to check for the task state (recall the > pattern in [1])...; how about if I expand "in part." to "in particular"?
That would be acceptable.
Alan
| |