Re: RFC: PERCPU_LIST to fix PR kern/52515

[Taylor R Campbell <campbell+netbsd-tech-kern%mumble.net@localhost>]

> Date: Mon, 11 Dec 2017 11:55:27 +0100
> From: Edgar Fuß <ef%math.uni-bonn.de@localhost>
> 
> >  struct psref {
> > -	LIST_ENTRY(psref)		psref_entry;
> > +	SLIST_ENTRY(psref)		psref_entry;
> > +	/* To keep ABI with LIST_ENTRY(psref) version. */
> > +	void				*psref_unused0;
> Isn't it somewhat fishy to manally pad this, knowing how big a [S]LIST_ENTRY 
> is? Wouldn't it be cleaner to use a union?

Also fine by me, but a little more work to implement.  Feel free to
add some ctasserts to raise your confidence in this.  I don't expect
LIST_ENTRY or SLIST_ENTRY to change, but I don't mind assertions to
add additional confidence.