netdev - Re: mlx5 XDP redirect leaking memory on kernel 6.3

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] [day] [month] [year] [list]

Date: Fri, 28 Jul 2023 15:14:06 +0200
From: Jesper Dangaard Brouer <jbrouer@...hat.com>
To: Dragos Tatulea <dtatulea@...dia.com>,
 "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>
Cc: brouer@...hat.com, "atzin@...hat.com" <atzin@...hat.com>,
 "linyunsheng@...wei.com" <linyunsheng@...wei.com>,
 "saeed@...nel.org" <saeed@...nel.org>,
 "ttoukan.linux@...il.com" <ttoukan.linux@...il.com>,
 "maxtram95@...il.com" <maxtram95@...il.com>,
 "jbrouer@...hat.com" <jbrouer@...hat.com>,
 "kheib@...hat.com" <kheib@...hat.com>, "jbenc@...hat.com"
 <jbenc@...hat.com>, "alexander.duyck@...il.com" <alexander.duyck@...il.com>,
 "bpf@...r.kernel.org" <bpf@...r.kernel.org>,
 "ilias.apalodimas@...aro.org" <ilias.apalodimas@...aro.org>,
 Saeed Mahameed <saeedm@...dia.com>,
 "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
 "fmaurer@...hat.com" <fmaurer@...hat.com>,
 "mkabat@...hat.com" <mkabat@...hat.com>, Tariq Toukan <tariqt@...dia.com>,
 "lorenzo@...nel.org" <lorenzo@...nel.org>
Subject: Re: mlx5 XDP redirect leaking memory on kernel 6.3


Correction: mlx5 XDP redirect leaking memory on kernel 6.4
  - Explained below

On 17/07/2023 17.15, Dragos Tatulea wrote:
> On Mon, 2023-07-17 at 16:42 +0200, gregkh@...uxfoundation.org wrote:
>> On Mon, Jul 17, 2023 at 02:37:44PM +0000, Dragos Tatulea wrote:
>>> On Thu, 2023-07-13 at 17:31 +0200, Greg KH wrote:
>>>> On Thu, Jul 13, 2023 at 04:58:04PM +0200, Jesper Dangaard Brouer wrote:
>>>>>
>>>>>
>>>>> On 13/07/2023 12.11, Dragos Tatulea wrote:
>>>>>> Gi Jesper,
>>>>>> On Thu, 2023-07-13 at 11:20 +0200, Jesper Dangaard Brouer wrote:
>>>>>>> Hi Dragos,
>>>>>>>
>>>>>>> Below you promised to work on a fix for XDP redirect memory leak...
>>>>>>> What is the status?
>>>>>>>
>>>>>> The fix got merged into net a week ago:
>>>>>> https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/drivers/net/ethernet/mellanox/mlx5/core?id=7abd955a58fb0fcd4e756fa2065c03ae488fcfa7
>>>>>>
>>>>>> Just forgot to follow up on this thread. Sorry about that...
>>>>>>
>>>>>
>>>>> Good to see it being fixed in net.git commit:
>>>>>   7abd955a58fb ("net/mlx5e: RX, Fix page_pool page fragment tracking for XDP")
>>>>>
>>>>> This need to be backported into stable tree 6.3, but I can see 6.3.13 is
>>>>> marked EOL (End-of-Life).
>>>>> Can we still get this fix applied? (Cc. GregKH)
>>>>
>>>> <formletter>
>>>>
>>>> This is not the correct way to submit patches for inclusion in the
>>>> stable kernel tree.  Please read:
>>>>     
>>>> https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
>>>> for how to do this properly.
>>>>
>>>> </formletter>
[...]
>>> to the stable submission rules or is it too late?
>>
[...]
>> If you mean the "6.3.y" tree, yes, there is nothing to do here as that
>> tree is end-of-life and you should have moved to the 6.4.y kernel tree
>> at this point in time.
>>
>> What is preventing you from moving?
>>
> I am fine with the state of things. But Jesper was asking. I suppose
> the answer to his question is "it's too late".
I was looking for the answer "it is too late for 6.3.y".

This exercise is just to make Google searches and distro people (like
myself and cc) aware that mlx5 XDP *redirect* is (intermediately) broken
and we should remember to backport 7abd955a58fb ("net/mlx5e: RX, Fix
page_pool page fragment tracking for XDP") to fix this.

Looking at git details, I notice that I (and subject) were wrong. The
buggy feature first landed in v6.4 and not v6.3.
Thus, true subject "mlx5 XDP redirect leaking memory on kernel 6.4".
Sorry for the stable confusion.  Greg will "automatically" pickup
Dragos's fix commit for 6.4.y as it have correct fixes tag.  Guess, I
damaged the Google search parameter, but I trying to correct it with
this update.

--Jesper