Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 24 Jan 2018 17:34:58 -0700
From:      Ian Lepore <ian@freebsd.org>
To:        Gleb Smirnoff <glebius@FreeBSD.org>, Kristof Provost <kp@FreeBSD.org>
Cc:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   Re: svn commit: r328313 - head/sys/netpfil/pf
Message-ID:  <1516840498.42536.213.camel@freebsd.org>
In-Reply-To: <20180125001310.GJ8113@FreeBSD.org>
References:  <201801240429.w0O4THIl059440@repo.freebsd.org> <20180125001310.GJ8113@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 2018-01-24 at 16:13 -0800, Gleb Smirnoff wrote:
>   Hi Kristof,
> 
> On Wed, Jan 24, 2018 at 04:29:17AM +0000, Kristof Provost wrote:
> K> Author: kp
> K> Date: Wed Jan 24 04:29:16 2018
> K> New Revision: 328313
> K> URL: https://svnweb.freebsd.org/changeset/base/328313
> K> 
> K> Log:
> K>   pf: States have at least two references
> K>   
> K>   pf_unlink_state() releases a reference to the state without
> checking if
> K>   this is the last reference. It can't be, because
> pf_state_insert()
> K>   initialises it to two. KASSERT() that this is always the case.
> K>   
> K>   CID:	1347140
> K> 
> K> Modified:
> K>   head/sys/netpfil/pf/pf.c
> K> 
> K> Modified: head/sys/netpfil/pf/pf.c
> K>
> =====================================================================
> =========
> K> --- head/sys/netpfil/pf/pf.c	Wed Jan 24 03:09:56 2018	
> (r328312)
> K> +++ head/sys/netpfil/pf/pf.c	Wed Jan 24 04:29:16 2018	
> (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'.

-- Ian



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1516840498.42536.213.camel>