On 11/20/23 11:42, Vladislav Odintsov wrote:
> Hi Ilya,
>
>> On 15 Nov 2023, at 21:51, Ilya Maximets wrote:
>>
>> On 11/15/23 14:13, Vladislav Odintsov wrote:
>>> Hi Ilya,
>>>
On 13 Nov 2023, at 22:25, Ilya Maximets wrote:
On 11/13/23 13:13, Eelco Chaudron wrote:
>
>
>
Hi Ilya,
> On 15 Nov 2023, at 21:51, Ilya Maximets wrote:
>
> On 11/15/23 14:13, Vladislav Odintsov wrote:
>> Hi Ilya,
>>
>>> On 13 Nov 2023, at 22:25, Ilya Maximets wrote:
>>>
>>> On 11/13/23 13:13, Eelco Chaudron wrote:
On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>>
Hi Marcelo,
PSB.
> On 16 Nov 2023, at 15:02, Marcelo Ricardo Leitner wrote:
>
> Hi Vladislav,
>
> On Wed, Nov 15, 2023 at 04:13:13PM +0300, Vladislav Odintsov wrote:
> ...
>> Final flow:
>> arp,reg11=0x2,reg12=0x6,reg14=0x4,reg15=0x2,tun_id=0x6,tun_src=10.1.0.103,tun_dst=10.1.0.109,tun_ipv6_s
Hi Vladislav,
On Wed, Nov 15, 2023 at 04:13:13PM +0300, Vladislav Odintsov wrote:
...
> Final flow:
> arp,reg11=0x2,reg12=0x6,reg14=0x4,reg15=0x2,tun_id=0x6,tun_src=10.1.0.103,tun_dst=10.1.0.109,tun_ipv6_src=::,tun_ipv6_dst=::,tun_gbp_id=0,tun_gbp_flags=0,tun_tos=0,tun_ttl=64,tun_erspan_ver=0,gtp
On 11/15/23 14:13, Vladislav Odintsov wrote:
> Hi Ilya,
>
>> On 13 Nov 2023, at 22:25, Ilya Maximets wrote:
>>
>> On 11/13/23 13:13, Eelco Chaudron wrote:
>>>
>>>
>>> On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>>>
> On 13 Nov 2023, at 14:17, Eelco Chaudron wrote:
>
>
>
>
Hi Ilya,
> On 13 Nov 2023, at 22:25, Ilya Maximets wrote:
>
> On 11/13/23 13:13, Eelco Chaudron wrote:
>>
>>
>> On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>>
On 13 Nov 2023, at 14:17, Eelco Chaudron wrote:
On 8 Nov 2023, at 14:39, Vladislav Odintsov wrote
On Mon, Oct 30, 2023 at 03:00:29PM +0100, Ilya Maximets wrote:
> There is no TCA_TUNNEL_KEY_ENC_SRC_PORT in the kernel, so the offload
> should not be attempted if OVS_TUNNEL_KEY_ATTR_TP_SRC is requested
> by OVS. Current code just ignores the attribute in the tunnel(set())
> action leading to a f
(re-send as for some reason gmail rejected the previous attempt
and it wasn't delivered to Vladislav)
On 11/13/23 20:25, Ilya Maximets wrote:
> On 11/13/23 13:13, Eelco Chaudron wrote:
>>
>>
>> On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>>
On 13 Nov 2023, at 14:17, Eelco Chaudron wr
On 11/13/23 13:13, Eelco Chaudron wrote:
>
>
> On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>
>>> On 13 Nov 2023, at 14:17, Eelco Chaudron wrote:
>>>
>>>
>>>
>>> On 8 Nov 2023, at 14:39, Vladislav Odintsov wrote:
>>>
Hi Ilya, Eelco,
I’ve tried this patch against 3.1 and la
On 13 Nov 2023, at 12:43, Vladislav Odintsov wrote:
>> On 13 Nov 2023, at 14:17, Eelco Chaudron wrote:
>>
>>
>>
>> On 8 Nov 2023, at 14:39, Vladislav Odintsov wrote:
>>
>>> Hi Ilya, Eelco,
>>>
>>> I’ve tried this patch against 3.1 and latest master branch. There are no
>>> warnings anymore,
>>
> On 13 Nov 2023, at 14:17, Eelco Chaudron wrote:
>
>
>
> On 8 Nov 2023, at 14:39, Vladislav Odintsov wrote:
>
>> Hi Ilya, Eelco,
>>
>> I’ve tried this patch against 3.1 and latest master branch. There are no
>> warnings anymore,
>> but it seems that in my installation it has broken offloa
On 8 Nov 2023, at 14:39, Vladislav Odintsov wrote:
> Hi Ilya, Eelco,
>
> I’ve tried this patch against 3.1 and latest master branch. There are no
> warnings anymore,
> but it seems that in my installation it has broken offload capability.
Yes, this is expected, this specific flow can not be of
Hi Ilya, Eelco,
I’ve tried this patch against 3.1 and latest master branch. There are no
warnings anymore,
but it seems that in my installation it has broken offload capability.
In tcpdump I see packets, which I expect to be offloaded,
so not to appear in tcpdump output, also there is an empty ou
On 30 Oct 2023, at 15:00, Ilya Maximets wrote:
> There is no TCA_TUNNEL_KEY_ENC_SRC_PORT in the kernel, so the offload
> should not be attempted if OVS_TUNNEL_KEY_ATTR_TP_SRC is requested
> by OVS. Current code just ignores the attribute in the tunnel(set())
> action leading to a flow mismatch
There is no TCA_TUNNEL_KEY_ENC_SRC_PORT in the kernel, so the offload
should not be attempted if OVS_TUNNEL_KEY_ATTR_TP_SRC is requested
by OVS. Current code just ignores the attribute in the tunnel(set())
action leading to a flow mismatch and potential incorrect datapath
behavior:
|tc(handler2
15 matches
Mail list logo