Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
From:   Eelco Chaudron <echaudro@redhat.com>
To:     Vlad Buslov <vladbu@nvidia.com>,
        Toms Atteka <cpp.code.lv@gmail.com>
Cc:     Roi Dayan <roid@nvidia.com>, Ilya Maximets <i.maximets@ovn.org>,
        Aaron Conole <aconole@redhat.com>,
        Jakub Kicinski <kuba@kernel.org>,
        "David S. Miller" <davem@davemloft.net>,
        Pravin B Shelar <pshelar@ovn.org>, netdev@vger.kernel.org,
        dev@openvswitch.org, linux-kernel@vger.kernel.org,
        Johannes Berg <johannes@sipsolutions.net>,
        Maor Dickman <maord@nvidia.com>
Subject: Re: [PATCH net-next v2] net: openvswitch: fix uAPI incompatibility
 with existing user space
Date:   Mon, 23 May 2022 14:54:09 +0200
Message-ID: <CD6F1EAA-7A32-46D2-9806-98CDB98DC537@redhat.com>
In-Reply-To: <FFBEB52B-FA8C-4989-BDC1-1F3908F024B8@redhat.com>
References: <20220309222033.3018976-1-i.maximets@ovn.org>
 <f7ty21hir5v.fsf@redhat.com>
 <44eeb550-3310-d579-91cc-ec18b59966d2@nvidia.com>
 <1a185332-3693-2750-fef2-f6938bbc8500@ovn.org> <87k0c171ml.fsf@nvidia.com>
 <9cc34fbc-3fd6-b529-7a05-554224510452@ovn.org>
 <4778B505-DBF5-4F57-90AF-87F12C1E0311@redhat.com> <87lev783k8.fsf@nvidia.com>
 <FFBEB52B-FA8C-4989-BDC1-1F3908F024B8@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk


On 17 May 2022, at 13:10, Eelco Chaudron wrote:

> On 12 May 2022, at 12:08, Vlad Buslov wrote:
>
>> On Thu 12 May 2022 at 12:19, Eelco Chaudron <echaudro@redhat.com> wrot=
e:
>>> On 7 Apr 2022, at 12:22, Ilya Maximets wrote:
>>>
>>>> On 4/7/22 10:02, Vlad Buslov wrote:
>>>>> On Mon 14 Mar 2022 at 20:40, Ilya Maximets <i.maximets@ovn.org> wro=
te:
>>>>>> On 3/14/22 19:33, Roi Dayan wrote:
>>>>>>>
>>>>>>>
>>>>>>> On 2022-03-10 8:44 PM, Aaron Conole wrote:
>>>>>>>> Ilya Maximets <i.maximets@ovn.org> writes:
>>>>>>>>
>>>>>>>>> Few years ago OVS user space made a strange choice in the commi=
t [1]
>>>>>>>>> to define types only valid for the user space inside the copy o=
f a
>>>>>>>>> kernel uAPI header.=C2=A0 '#ifndef __KERNEL__' and another attr=
ibute was
>>>>>>>>> added later.
>>>>>>>>>
>>>>>>>>> This leads to the inevitable clash between user space and kerne=
l types
>>>>>>>>> when the kernel uAPI is extended.=C2=A0 The issue was unveiled =
with the
>>>>>>>>> addition of a new type for IPv6 extension header in kernel uAPI=
=2E
>>>>>>>>>
>>>>>>>>> When kernel provides the OVS_KEY_ATTR_IPV6_EXTHDRS attribute to=
 the
>>>>>>>>> older user space application, application tries to parse it as
>>>>>>>>> OVS_KEY_ATTR_PACKET_TYPE and discards the whole netlink message=
 as
>>>>>>>>> malformed.=C2=A0 Since OVS_KEY_ATTR_IPV6_EXTHDRS is supplied al=
ong with
>>>>>>>>> every IPv6 packet that goes to the user space, IPv6 support is =
fully
>>>>>>>>> broken.
>>>>>>>>>
>>>>>>>>> Fixing that by bringing these user space attributes to the kern=
el
>>>>>>>>> uAPI to avoid the clash.=C2=A0 Strictly speaking this is not th=
e problem
>>>>>>>>> of the kernel uAPI, but changing it is the only way to avoid br=
eakage
>>>>>>>>> of the older user space applications at this point.
>>>>>>>>>
>>>>>>>>> These 2 types are explicitly rejected now since they should not=
 be
>>>>>>>>> passed to the kernel.=C2=A0 Additionally, OVS_KEY_ATTR_TUNNEL_I=
NFO moved
>>>>>>>>> out from the '#ifdef __KERNEL__' as there is no good reason to =
hide
>>>>>>>>> it from the userspace.=C2=A0 And it's also explicitly rejected =
now, because
>>>>>>>>> it's for in-kernel use only.
>>>>>>>>>
>>>>>>>>> Comments with warnings were added to avoid the problem coming b=
ack.
>>>>>>>>>
>>>>>>>>> (1 << type) converted to (1ULL << type) to avoid integer overfl=
ow on
>>>>>>>>> OVS_KEY_ATTR_IPV6_EXTHDRS, since it equals 32 now.
>>>>>>>>>
>>>>>>>>> =C2=A0 [1] beb75a40fdc2 ("userspace: Switching of L3 packets in=
 L2 pipeline")
>>>>>>>>>
>>>>>>>>> Fixes: 28a3f0601727 ("net: openvswitch: IPv6: Add IPv6 extensio=
n header support")
>>>>>>>>> Link: https://lore.kernel.org/netdev/3adf00c7-fe65-3ef4-b6d7-6d=
8a0cad8a5f@nvidia.com
>>>>>>>>> Link: https://github.com/openvswitch/ovs/commit/beb75a40fdc295b=
fd6521b0068b4cd12f6de507c
>>>>>>>>> Reported-by: Roi Dayan <roid@nvidia.com>
>>>>>>>>> Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
>>>>>>>>> ---
>>>>>>>>
>>>>>>>> Acked-by: Aaron Conole <aconole@redhat.com>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> I got to check traffic with the fix and I do get some traffic
>>>>>>> but something is broken. I didn't investigate much but the quick
>>>>>>> test shows me rules are not offloaded and dumping ovs rules gives=

>>>>>>> error like this
>>>>>>>
>>>>>>> recirc_id(0),in_port(enp8s0f0_1),ct_state(-trk),eth(),eth_type(0x=
86dd),ipv6(frag=3Dno)(bad
>>>>>>> key length 2, expected -1)(00 00/(bad mask length 2, expected -1)=
(00 00),
>>>>>>> packets:2453, bytes:211594, used:0.004s, flags:S., actions:ct,rec=
irc(0x2)
>>>>>>
>>>>>> Such a dump is expected, because kernel parses fields that current=

>>>>>> userspace doesn't understand, and at the same time OVS by design i=
s
>>>>>> using kernel provided key/mask while installing datapath rules, II=
RC.
>>>>>> It should be possible to make these dumps a bit more friendly thou=
gh.
>>>>>>
>>>>>> For the offloading not working, see my comment in the v2 patch ema=
il
>>>>>> I sent (top email of this thread).  In short, it's a problem in us=
er
>>>>>> space and it can not be fixed from the kernel side, unless we reve=
rt
>>>>>> IPv6 extension header support and never add any new types, which i=
s
>>>>>> unreasonable.  I didn't test any actual offloading, but I had a
>>>>>> successful run of 'make check-offloads' with my quick'n'dirty fix =
from
>>>>>> the top email.
>>>>>
>>>>> Hi Ilya,
>>>>>
>>>>> I can confirm that with latest OvS master IPv6 rules offload still =
fails
>>>>> without your pastebin code applied.
>>>>>
>>>>>>
>>>>>> Since we're here:
>>>>>>
>>>>>> Toms, do you plan to submit user space patches for this feature?
>>>>>
>>>>> I see there is a patch from you that is supposed to fix compatibili=
ty
>>>>> issues caused by this change in OvS d96d14b14733 ("openvswitch.h: A=
lign
>>>>> uAPI definition with the kernel."), but it doesn't fix offload for =
me
>>>>> without pastebin patch.
>>>>
>>>> Yes.  OVS commit d96d14b14733 is intended to only fix the uAPI.
>>>> Issue with offload is an OVS bug that should be fixed separately.
>>>> The fix will also need to be backported to OVS stable branches.
>>>>
>>>>> Do you plan to merge that code into OvS or you
>>>>> require some help from our side?
>>>>
>>>> I could do that, but I don't really have enough time.  So, if you
>>>> can work on that fix, it would be great.  Note that comments inside
>>>> the OVS's lib/odp-util.c:parse_key_and_mask_to_match() was blindly
>>>> copied from the userspace datapath and are incorrect for the general=

>>>> case, so has to be fixed alongside the logic of that function.
>>>
>>> Tom or Vlad, are you working on this? Asking, as the release of a ker=
nel with
>>> Tom=E2=80=99s =E2=80=9Cnet: openvswitch: IPv6: Add IPv6 extension hea=
der support=E2=80=9D patch will
>>> break OVS.
>>>
>>> //Eelco
>>
>> Hi Eelco,
>>
>> My simple fix for OvS was rejected and I don't have time to rework it =
at
>> the moment.
>
> That=E2=80=99s a pity, Tom do you maybe have time as your patch left OV=
S in this error state?

Looks like everybody is busy, and as the patched kernel is now available,=
 let me try to fix this on the OVS side.

//Eelco