Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Message-ID: <5AB132C5.5010806@ORACLE.COM>
Date:   Tue, 20 Mar 2018 18:11:49 +0200
From:   Liran Alon <LIRAN.ALON@ORACLE.COM>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0
MIME-Version: 1.0
To:     David Miller <davem@davemloft.net>
CC:     netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        idan.brown@ORACLE.COM, yuval.shaia@ORACLE.COM
Subject: Re: [PATCH] net: dev_forward_skb(): Scrub packet's per-netns info
 only when crossing netns
References: <1520953642-8145-1-git-send-email-liran.alon@oracle.com>    <20180320.104759.796804827689233281.davem@davemloft.net>        <5AB12A0E.2060704@ORACLE.COM> <20180320.120036.1999626754164343704.davem@davemloft.net>
In-Reply-To: <20180320.120036.1999626754164343704.davem@davemloft.net>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 20/03/18 18:00, David Miller wrote:
> From: Liran Alon <LIRAN.ALON@ORACLE.COM>
> Date: Tue, 20 Mar 2018 17:34:38 +0200
>
>> I personally don't understand why we should maintain
>> backwards-comparability to this behaviour.
>
> The reason is because not breaking things is a cornerstone of Linux
> kernel development.
>
>> This feature is not documented to user-mode and I don't see why it
>> is legit for the user to rely on it.
>
> Whether it is documented or not is irrelevant.  A lot of our
> interfaces and behaviors are not documented or poorly documented
> at best.
>
>> In addition, even if we do want to maintain backwards-comparability to
>> this behaviour, I think it is enough to have an opt-in flag in
>> /proc/sys/net/core/ that when set to 1 will activate the fix in
>> dev_forward_skb() provided by this patch. That would also be a very
>> simple change to the patch provided here.
>
> Making it opt-in makes it more palatable, that's for sure.
>

1. Do we want to make a flag for every bug that is user-space visible? I 
think there is place for consideration on a per-case basis. I still 
don't see how a user can utilize this behaviour. He is basically loosing 
information (skb->mark) without this patch.

2. Having said that, I don't mind changing patch to maintain backwards 
compatibility here. However, there was also a discussion here on where 
the flag should sit. I think that a global /proc/sys/net/core/ flag 
should be enough. Do you agree it's sufficient for now?

Thanks,
-Liran