svn commit: r328313 - head/sys/netpfil/pf

[Kristof Provost]

On 25 Jan 2018, at 11:34, Ian Lepore wrote:
> On Wed, 2018-01-24 at 16:13 -0800, Gleb Smirnoff wrote:
>> (r328313)
>> K> @@ -1613,6 +1613,7 @@ int
>> K>  pf_unlink_state(struct pf_state *s, u_int flags)
>> K>  {
>> K>  	struct pf_idhash *ih = &V_pf_idhash[PF_IDHASH(s)];
>> K> +	int last;
>> K>  
>> K>  	if ((flags & PF_ENTER_LOCKED) == 0)
>> K>  		PF_HASHROW_LOCK(ih);
>> K> @@ -1653,7 +1654,8 @@ pf_unlink_state(struct pf_state *s, u_int
>> flags)
>> K>  	PF_HASHROW_UNLOCK(ih);
>> K>  
>> K>  	pf_detach_state(s);
>> K> -	refcount_release(&s->refs);
>> K> +	last = refcount_release(&s->refs);
>> K> +	KASSERT(last == 0, ("Incorrect state reference count"));
>> K>  
>> K>  	return (pf_release_state(s));
>> K>  }
>>
>> IMHO, we shouldn't emit extra code to please Coverity. We can mark it
>> as a false positive in the interface. It may make sense to add a
>> comment
>> for a human to explain why return isn't checked here.
>>
>
> Not to mention that when KASSERT compiles to nothing, what you're left
> with is a "defined but not used" warning for 'last'.
>
I’d really like to keep the KASSERT(), because this is the sort of 
thing that could go wrong, and the assertion would be helpful.

I suppose I could wrap last in #ifdef INVARIANTS, but that’s rather 
ugly too.

Asserting that the refcount is at least 1 when entering 
pf_release_state() would express the same, but that’s also 
problematic.
Of course, errors should trigger the KASSERT() in refcount_release(), so 
I think I may have convinced myself that the KASSERT() can in fact be 
removed and replaced with (void)refcount_release() and a comment 
explaining why this refcount_release() can never return 1.

Regards,
Kristof