netdev - Re: [PATCH iproute2-next] taprio: Add support for the SetAndHold and SetAndRelease commands

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

Date:   Fri, 3 Jan 2020 17:24:14 -0500
From:   Murali Karicheri <m-karicheri2@...com>
To:     Jose Abreu <Jose.Abreu@...opsys.com>,
        Vinicius Costa Gomes <vinicius.gomes@...el.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>
CC:     Joao Pinto <Joao.Pinto@...opsys.com>,
        David Ahern <dsahern@...il.com>
Subject: Re: [PATCH iproute2-next] taprio: Add support for the SetAndHold and
 SetAndRelease commands

Hi Jose,

On 12/18/2019 06:08 PM, Jose Abreu wrote:
> From: Vinicius Costa Gomes <vinicius.gomes@...el.com>
> Date: Dec/18/2019, 23:05:13 (UTC+00:00)
> 
>> Hi Jose,
>>
>> Jose Abreu <Jose.Abreu@...opsys.com> writes:
>>
>>> Although this is already in kernel, currently the tool does not support
>>> them. We need these commands for full TSN features which are currently
>>> supported in Synopsys IPs such as QoS and XGMAC3.
>>>
>>> Signed-off-by: Jose Abreu <Jose.Abreu@...opsys.com>
>>
>> This patch looks good in itself.
>>
>> However, I feel that this is incomplete. At least the way I understand
>> things, without specifying which traffic classes are going to be
>> preemptible (or it's dual concept, express), I don't see how this is
>> going to be used in practice. Or does the hardware have a default
>> configuration, that all traffic classes are preemptible, for example.
>>
>> What am I missing here?
> 
> On our IPs Queue 0 is by preemptible and all remaining ones are express
> by default.
> 
> The way I tested it is quite easy: send traffic from queue 0 and at same
> time configure EST with SetAndHold for remaining queues. Which means
> queue 0 traffic will be blocked while remaining ones are sending.
> 
So you have one sched entry that specify SetAndHold for all remaining
queues. So this means, queue 0 will never get sent. I guess you also
support SetAndRelease so that a mix of SetAndHold followed by 
SetAndRelease can be sent to enable sending from Queue 0. Is that
correct?

Something like
               sched-entry H 02 300000 \ <=== 300 usec tx from Q1
               sched-entry R 01 200000   <=== 300 usec tx from Q0

Just trying to understand how this is being used for real world
application.

Regards,
Murali
> ---
> Thanks,
> Jose Miguel Abreu
> 

-- 
Murali Karicheri
Texas Instruments