Re: [PATCH 01/13] seqnum_ops: Introduce Sequence Number Ops

[Shuah Khan]

On 11/12/20 9:45 AM, Greg KH wrote:
> On Thu, Nov 12, 2020 at 09:17:27AM -0700, Shuah Khan wrote:
> > On 11/12/20 5:36 AM, Matthew Wilcox wrote:
> > > On Wed, Nov 11, 2020 at 12:23:03PM -0700, Shuah Khan wrote:
> > > > > Agreed: this is a clear wrapping sequence counter. It's only abuse would
> > > > > be using it in a place where wrapping actually is _not_ safe. (bikeshed:
> > > > > can we call it wrap_u32 and wrap_u64?)
> > > >
> > > > Still like seqnum_ops.
> > > >
> > > > There is seqcount_t in seqlock.h which is a totally different feature.
> > >
> > > Yes, and that's why this new thing, whatever it is called should not
> > > have the word "sequence" in it.  People will get it confused.
> >
> > Any suggestions for name. I am bad with coming up with names. How does
> > Statcnt API and struct statcnt along the lines of your name suggestions
> > in your previous email?
>
> What does "stat" mean here?

Stat doesn't really reflect what we are trying to do here and sequence
does. I am just looking to address confusion if any and make a call.

> And I don't understand the hesitation about "sequence" in a name, as
> that's exactly what this is.  seqlock is different, yes.
>  > How about "seqnum_t"?  That's what we call the sequence number that we
> export to uevents, a "SEQNUM".

Good point.
This is what we have currently in patch v1 and let's just go with it.

thanks,
-- Shuah