tc mqprio offload command error

tc mqprio offload command error

[Chopra, Manish]

Hello Folks,

I am trying to set below command to try mqprio offload on 4.18 kernel. It is throwing the flowing error.

# tc qdisc add dev eth0 root mqprio num_tc 2 map 1 1 1 1 0 0 0 0
RTNETLINK answers: Numerical result out of range

I can't really make out what's wrong with the above command, since this works fine with other OS kernels.
Any thoughts if it is something broken on upstream kernel ?

Thanks,
Manish

Re: tc mqprio offload command error

[Alexander Duyck]

On Sun, Jul 15, 2018 at 6:30 PM, Chopra, Manish
<Manish.Chopra@cavium.com> wrote:
> Hello Folks,
>
> I am trying to set below command to try mqprio offload on 4.18 kernel. It is throwing the flowing error.
>
> # tc qdisc add dev eth0 root mqprio num_tc 2 map 1 1 1 1 0 0 0 0
> RTNETLINK answers: Numerical result out of range
>
> I can't really make out what's wrong with the above command, since this works fine with other OS kernels.
> Any thoughts if it is something broken on upstream kernel ?
>
> Thanks,
> Manish

You might need to specify the traffic class for the 8 remaining
priorities. The full map size is 16 entries, not just 8. The default
value for the last 4 mapping entries is TC 3 which would be out of
range if you only have 2 TCs specified.

- Alex

Re: tc mqprio offload command error

[Jesus Sanchez-Palencia]

Hi,


On 07/16/2018 10:20 AM, Alexander Duyck wrote:
> On Sun, Jul 15, 2018 at 6:30 PM, Chopra, Manish
> <Manish.Chopra@cavium.com> wrote:
>> Hello Folks,
>>
>> I am trying to set below command to try mqprio offload on 4.18 kernel. It is throwing the flowing error.
>>
>> # tc qdisc add dev eth0 root mqprio num_tc 2 map 1 1 1 1 0 0 0 0
>> RTNETLINK answers: Numerical result out of range
>>
>> I can't really make out what's wrong with the above command, since this works fine with other OS kernels.
>> Any thoughts if it is something broken on upstream kernel ?
>>
>> Thanks,
>> Manish
> 
> You might need to specify the traffic class for the 8 remaining
> priorities. The full map size is 16 entries, not just 8. The default
> value for the last 4 mapping entries is TC 3 which would be out of
> range if you only have 2 TCs specified.


In addition to that, you might hit the same bug we brought up [1] a while ago.
If that is the case, a fix was just proposed here [2]. Note that other qdiscs
might be broken as well, but we could only spot the issue with mqprio and netem
so far.

[1] https://patchwork.ozlabs.org/patch/867860/#1893405
[2] https://patchwork.ozlabs.org/patch/944565/


Regards,
Jesus


> 
> - Alex
> 

RE: tc mqprio offload command error

[Chopra, Manish]

> -----Original Message-----
> From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
> Sent: Monday, July 16, 2018 11:28 PM
> To: Alexander Duyck <alexander.duyck@gmail.com>; Chopra, Manish
> <Manish.Chopra@cavium.com>
> Cc: Stephen Hemminger <stephen@networkplumber.org>; David Miller
> <davem@davemloft.net>; Jiri Pirko <jiri@resnulli.us>;
> netdev@vger.kernel.org
> Subject: Re: tc mqprio offload command error
> 
> External Email
> 
> Hi,
> 
> 
> On 07/16/2018 10:20 AM, Alexander Duyck wrote:
> > On Sun, Jul 15, 2018 at 6:30 PM, Chopra, Manish
> > <Manish.Chopra@cavium.com> wrote:
> >> Hello Folks,
> >>
> >> I am trying to set below command to try mqprio offload on 4.18 kernel. It
> is throwing the flowing error.
> >>
> >> # tc qdisc add dev eth0 root mqprio num_tc 2 map 1 1 1 1 0 0 0 0
> >> RTNETLINK answers: Numerical result out of range
> >>
> >> I can't really make out what's wrong with the above command, since this
> works fine with other OS kernels.
> >> Any thoughts if it is something broken on upstream kernel ?
> >>
> >> Thanks,
> >> Manish
> >
> > You might need to specify the traffic class for the 8 remaining
> > priorities. The full map size is 16 entries, not just 8. The default
> > value for the last 4 mapping entries is TC 3 which would be out of
> > range if you only have 2 TCs specified.
> 
> 
> In addition to that, you might hit the same bug we brought up [1] a while
> ago.
> If that is the case, a fix was just proposed here [2]. Note that other qdiscs
> might be broken as well, but we could only spot the issue with mqprio and
> netem so far.
> 
> [1] https://patchwork.ozlabs.org/patch/867860/#1893405
> [2] https://patchwork.ozlabs.org/patch/944565/
> 
> 

Issue is same with all of 16 prio-tc map supplied -

# tc qdisc add dev eth0 root mqprio num_tc 4 map 1 1 1 1 0 0 0 0 2 2 2 2 3 3 3 3
RTNETLINK answers: Numerical result out of range

Thanks Jesus, I will try the fix[2] and see.

Regards
-Manish