linux-kernel - Re: [PATCH net-next v2 03/10] net: ppp: Remove unnecessary (void*) conversions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZK+4udwfawcJq5qC@corigine.com>
Date:   Thu, 13 Jul 2023 09:41:29 +0100
From:   Simon Horman <simon.horman@...igine.com>
To:     yunchuan <yunchuan@...china.com>
Cc:     mostrows@...thlink.net, davem@...emloft.net, edumazet@...gle.com,
        kuba@...nel.org, pabeni@...hat.com, xeb@...l.ru,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        kernel-janitors@...r.kernel.org
Subject: Re: [PATCH net-next v2 03/10] net: ppp: Remove unnecessary (void*)
 conversions

On Wed, Jul 12, 2023 at 09:44:40AM +0800, yunchuan wrote:
> On 2023/7/12 01:50, Simon Horman wrote:
> > > -	struct sock *sk = (struct sock *)chan->private;
> > > +	struct sock *sk = chan->private;
> > >   	struct pppox_sock *po = pppox_sk(sk);
> > >   	struct net_device *dev = po->pppoe_dev;
> > Hi,
> > 
> > Please don't break reverse xmas tree ordering - longest line to shortest -
> > of local variable declarations in Networking code.
> 
> Hi,
> 
> This can't be reversed because it depends on the first declaration.
> Should I change it like this?
> 
> -	struct sock *sk = (struct sock *)chan->private;
> - 	struct pppox_sock *po = pppox_sk(sk);
> +	struct pppox_sock *po = pppox_sk(chan->private);
>  	struct net_device *dev = po->pppoe_dev;
> +	struct sock *sk = chan->private;
> 
> But this seems to be bad. As the advice of Andrew[1] and Dan[2]:
> 
> "
> 
> When dealing with existing broken reverse Christmas tree, please don't
> make it worse with a change. But actually fixing it should be in a
> different patch.
> 
> We want patches to be obviously correct. By removing the cast and
> moving variables around, it is less obvious it is correct, than having
> two patches.
> 
> "

Thanks, I agree this is a good approach.