netdev - Re: [systemd-devel] [PATCH] Use dev_port for the ID of a network device.

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 for Android: free password hash cracker in your pocket

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

Message-ID: <53B2B511.8050107@gmail.com>
Date:	Tue, 01 Jul 2014 16:18:09 +0300
From:	Amir Vadai <amirv.mellanox@...il.com>
To:	Kay Sievers <kay@...y.org>, Tom Gundersen <teg@...m.no>
CC:	Thadeu Lima de Souza Cascardo <cascardo@...ux.vnet.ibm.com>,
	netdev <netdev@...r.kernel.org>,
	Or Gerlitz <ogerlitz@...lanox.com>,
	Yevgeny Petrilin <yevgenyp@...lanox.com>
Subject: Re: [systemd-devel] [PATCH] Use dev_port for the ID of a network
 device.

On 7/1/2014 3:59 PM, Kay Sievers wrote:
> On Tue, Jul 1, 2014 at 2:45 PM, Tom Gundersen <teg@...m.no> wrote:
>> On Tue, Jul 1, 2014 at 2:32 PM, Thadeu Lima de Souza Cascardo
>> <cascardo@...ux.vnet.ibm.com> wrote:
>>> On Tue, Jul 01, 2014 at 02:33:19AM +0200, Kay Sievers wrote:
>>>> On Mon, Jun 30, 2014 at 10:36 PM, Thadeu Lima de Souza Cascardo
>>>> <cascardo@...ux.vnet.ibm.com> wrote:
>>>>> For network devices on the same PCI function, dev_id should not be used,
>>>>> since its purpose is for IPv6 support on interfaces with the same MAC
>>>>> address.
>>>>>
>>>>> The new dev_port sysfs attribute should be used when it is found. When
>>>>> it is not, using dev_id might work.
>>>>
>>>> I don't see a problem switching this over, but why would we keep using
>>>> dev_id if it is not the right thing to use?
>>>>
>>>> Kay
>>>>
>>>
>>> Because dev_port has only been introduced into Linux 3.15, and some
>>> drivers used dev_id before Linux 3.10. It's not an ideal situation, so I
>>> think it's important to keep some compatibility for the time being. Or
>>> we could simply drop dev_id, and force everyone to do the right thing
>>> and backport dev_port support and fix their drivers.
>>
>> If this fixes a real issue, I guess it would be nice to push the
>> dev_port stuff to -stable. If it is accepted it would solve the
>> backwards compatibility problem, and we could just drop dev_id from
>> udev.
> 
> Let's drop it now and mention it in NEWS. That way, distros which
> really care can add the compat hack to the code.
> 
> Kay
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Acked-by: Amir Vadai <amirv@...lanox.com>

mlx4_en is one of those drivers that wrongly used dev_id and changed to
dev_port.
I guess that having a bug in driver that used dev_id the right way is
worse than bad interface name for drivers that wrongly used it.
Never the less - we should pull dev_port to stable, to minimize the damage.

Amir

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html