Re: [RFC v2 3/4] Documentation/locking/atomic: Add a litmus test for atomic_set()

[Boqun Feng <boqun.feng@gmail.com>]

On Wed, Feb 19, 2020 at 10:07:09AM -0500, Alan Stern wrote:
> On Wed, 19 Feb 2020, Boqun Feng wrote:
> 
> > We already use a litmus test in atomic_t.txt to describe the behavior of
> > an atomic_set() with the an atomic RMW, so add it into atomic-tests
> > directory to make it easily accessible for anyone who cares about the
> > semantics of our atomic APIs.
> > 
> > Additionally, change the sentences describing the test in atomic_t.txt
> > with better wording.
> 
> One very minor point about the new working in atomic_t.txt:
> 
> > diff --git a/Documentation/atomic_t.txt b/Documentation/atomic_t.txt
> > index ceb85ada378e..d30cb3d87375 100644
> > --- a/Documentation/atomic_t.txt
> > +++ b/Documentation/atomic_t.txt
> > @@ -85,10 +85,10 @@ smp_store_release() respectively. Therefore, if you find yourself only using
> >  the Non-RMW operations of atomic_t, you do not in fact need atomic_t at all
> >  and are doing it wrong.
> >  
> > -A subtle detail of atomic_set{}() is that it should be observable to the RMW
> > -ops. That is:
> > +A note for the implementation of atomic_set{}() is that it cannot break the
> > +atomicity of the RMW ops. That is:
> 
> This would be slightly better if you changed it to: "it must not break".
> 

Got it. Indeed it's the better wording, thanks!

Regards,
Boqun

> The comments in the litmus test and README file are okay as they stand.
> 
> Alan
>