From: Alexandr Nedvedicky <sashan@fastmail.net>
Subject: Re: pf: u_int32_t conn counter underflow in pf_src_tree_remove_state()
To: Alexander Bluhm <alexander.bluhm@gmx.net>
Cc: tech@openbsd.org
Date: Tue, 18 Aug 2026 18:53:21 +0200

On Tue, Aug 18, 2026 at 05:57:23PM +0200, Alexander Bluhm wrote:
> On Tue, Aug 18, 2026 at 05:01:58PM +0200, Alexandr Nedvedicky wrote:
</snip>
> > > 
> > > The overflow check is reading the sn->conn value independent from the
> > > atomic increment.  Something like this is better:
> > > 
> > > 	uint32_t sn_conn;
> > > 
> > > 	sn_conn = atomic_inc_int_nv(&sn->conn);
> >     do you mind if I do '(u_int32_t)atomic_inc_int_nv(..);'
> 
> The compiler does the right thing.  The cast is unnecessary.

    OK I will commit the diff without cast.

> 
> > updated diff is below.
> 
> OK bluhm@

thanks and
regards
sashan