Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Duncan Roe <duncan.roe2@gmail.com>
From:   Duncan Roe <duncan_roe@optusnet.com.au>
Date:   Wed, 8 Sep 2021 12:22:44 +1000
To:     Jan Engelhardt <jengelh@inai.de>
Cc:     Florian Westphal <fw@strlen.de>,
        Cole Dishington <Cole.Dishington@alliedtelesis.co.nz>,
        pablo@netfilter.org, kadlec@netfilter.org, davem@davemloft.net,
        kuba@kernel.org, shuah@kernel.org, linux-kernel@vger.kernel.org,
        netfilter-devel@vger.kernel.org, coreteam@netfilter.org,
        netdev@vger.kernel.org,
        Anthony Lineham <anthony.lineham@alliedtelesis.co.nz>,
        Scott Parlane <scott.parlane@alliedtelesis.co.nz>,
        Blair Steven <blair.steven@alliedtelesis.co.nz>
Subject: Re: [PATCH net v2] net: netfilter: Fix port selection of FTP for
 NF_NAT_RANGE_PROTO_SPECIFIED
Message-ID: <YTgedODOPAQboQlm@slk1.local.net>
Mail-Followup-To: Jan Engelhardt <jengelh@inai.de>,
        Florian Westphal <fw@strlen.de>,
        Cole Dishington <Cole.Dishington@alliedtelesis.co.nz>,
        pablo@netfilter.org, kadlec@netfilter.org, davem@davemloft.net,
        kuba@kernel.org, shuah@kernel.org, linux-kernel@vger.kernel.org,
        netfilter-devel@vger.kernel.org, coreteam@netfilter.org,
        netdev@vger.kernel.org,
        Anthony Lineham <anthony.lineham@alliedtelesis.co.nz>,
        Scott Parlane <scott.parlane@alliedtelesis.co.nz>,
        Blair Steven <blair.steven@alliedtelesis.co.nz>
References: <20210907021415.962-1-Cole.Dishington@alliedtelesis.co.nz>
 <20210907135458.GF23554@breakpoint.cc>
 <r46nn4-n993-rs28-84sr-o1qop429rr9@vanv.qr>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <r46nn4-n993-rs28-84sr-o1qop429rr9@vanv.qr>
Precedence: bulk

On Tue, Sep 07, 2021 at 05:11:42PM +0200, Jan Engelhardt wrote:
>
> On Tuesday 2021-09-07 15:54, Florian Westphal wrote:
> >> -	/* Try to get same port: if not, try to change it. */
> >> -	for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) {
> >> -		int ret;
> >> +	if (htons(nat->range_info.min_proto.all) == 0 ||
> >> +	    htons(nat->range_info.max_proto.all) == 0) {
> >
> >Either use if (nat->range_info.min_proto.all || ...
> >
> >or use ntohs().  I will leave it up to you if you prefer
> >ntohs(nat->range_info.min_proto.all) == 0 or
> >nat->range_info.min_proto.all == ntohs(0).
>
> If one has the option, one should always prefer to put htons/htonl on
> the side with the constant literal;
> Propagation of constants and compile-time evaluation is the target.
>
> That works for some other functions as well (e.g.
> strlen("fixedstring")).

When comparing against constant zero, why use htons/htonl at all?

Cheers ... Duncan.