[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f759d33f-860c-454b-8553-3b840ed6da8d@gmail.com>
Date: Wed, 24 Apr 2024 17:47:15 +0700
From: Bagas Sanjaya <bagasdotme@...il.com>
To: Mateusz Polchlopek <mateusz.polchlopek@...el.com>,
Tony Nguyen <anthony.l.nguyen@...el.com>, davem@...emloft.net,
kuba@...nel.org, pabeni@...hat.com, edumazet@...gle.com,
netdev@...r.kernel.org
Cc: Michal Wilczynski <michal.wilczynski@...el.com>, corbet@....net,
linux-doc@...r.kernel.org, Jiri Pirko <jiri@...dia.com>,
Przemek Kitszel <przemyslaw.kitszel@...el.com>
Subject: Re: [PATCH net-next 6/6] ice: Document tx_scheduling_layers parameter
On 4/24/24 16:54, Mateusz Polchlopek wrote:
>
>
> On 4/23/2024 2:37 PM, Bagas Sanjaya wrote:
>> On Mon, Apr 22, 2024 at 01:39:11PM -0700, Tony Nguyen wrote:
>>> + The default 9-layer tree topology was deemed best for most workloads,
>>> + as it gives an optimal ratio of performance to configurability. However,
>>> + for some specific cases, this 9-layer topology might not be desired.
>>> + One example would be sending traffic to queues that are not a multiple
>>> + of 8. Because the maximum radix is limited to 8 in 9-layer topology,
>>> + the 9th queue has a different parent than the rest, and it's given
>>> + more bandwidth credits. This causes a problem when the system is
>>> + sending traffic to 9 queues:
>>> +
>>> + | tx_queue_0_packets: 24163396
>>> + | tx_queue_1_packets: 24164623
>>> + | tx_queue_2_packets: 24163188
>>> + | tx_queue_3_packets: 24163701
>>> + | tx_queue_4_packets: 24163683
>>> + | tx_queue_5_packets: 24164668
>>> + | tx_queue_6_packets: 23327200
>>> + | tx_queue_7_packets: 24163853
>>> + | tx_queue_8_packets: 91101417 < Too much traffic is sent from 9th
>>> +
>>> <snipped>...
>>> + To verify that value has been set:
>>> + $ devlink dev param show pci/0000:16:00.0 name tx_scheduling_layers
>>>
>>
>> For consistency with other code blocks, format above as such:
>>
>> ---- >8 ----
>> diff --git a/Documentation/networking/devlink/ice.rst b/Documentation/networking/devlink/ice.rst
>> index 830c04354222f8..0039ca45782400 100644
>> --- a/Documentation/networking/devlink/ice.rst
>> +++ b/Documentation/networking/devlink/ice.rst
>> @@ -41,15 +41,17 @@ Parameters
>> more bandwidth credits. This causes a problem when the system is
>> sending traffic to 9 queues:
>> - | tx_queue_0_packets: 24163396
>> - | tx_queue_1_packets: 24164623
>> - | tx_queue_2_packets: 24163188
>> - | tx_queue_3_packets: 24163701
>> - | tx_queue_4_packets: 24163683
>> - | tx_queue_5_packets: 24164668
>> - | tx_queue_6_packets: 23327200
>> - | tx_queue_7_packets: 24163853
>> - | tx_queue_8_packets: 91101417 < Too much traffic is sent from 9th
>> + .. code-block:: shell
>> +
>> + tx_queue_0_packets: 24163396
>> + tx_queue_1_packets: 24164623
>> + tx_queue_2_packets: 24163188
>> + tx_queue_3_packets: 24163701
>> + tx_queue_4_packets: 24163683
>> + tx_queue_5_packets: 24164668
>> + tx_queue_6_packets: 23327200
>> + tx_queue_7_packets: 24163853
>> + tx_queue_8_packets: 91101417 < Too much traffic is sent from 9th
>> To address this need, you can switch to a 5-layer topology, which
>> changes the maximum topology radix to 512. With this enhancement,
>> @@ -67,7 +69,10 @@ Parameters
>> You must do PCI slot powercycle for the selected topology to take effect.
>> To verify that value has been set:
>> - $ devlink dev param show pci/0000:16:00.0 name tx_scheduling_layers
>> +
>> + .. code-block:: shell
>> +
>> + $ devlink dev param show pci/0000:16:00.0 name tx_scheduling_layers
>> Info versions
>> =============
>>
>> Thanks.
>>
>
> Thank You for reporting that. I will verify this issue soon.
OK, thanks!
--
An old man doll... just what I always wanted! - Clara
Powered by blists - more mailing lists