netdev - Re: [PATCH net-next v6 1/4] devlink: Extend devlink-rate api with export functions and new params

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] [day] [month] [year] [list]

Message-ID: <9f73fcab-5374-81d8-1069-6a7ad0140a2f@intel.com>
Date:   Tue, 18 Oct 2022 17:13:30 +0200
From:   "Wilczynski, Michal" <michal.wilczynski@...el.com>
To:     Jiri Pirko <jiri@...nulli.us>
CC:     <netdev@...r.kernel.org>, <alexandr.lobakin@...el.com>,
        <jacob.e.keller@...el.com>, <jesse.brandeburg@...el.com>,
        <przemyslaw.kitszel@...el.com>, <anthony.l.nguyen@...el.com>,
        <kuba@...nel.org>, <ecree.xilinx@...il.com>
Subject: Re: [PATCH net-next v6 1/4] devlink: Extend devlink-rate api with
 export functions and new params



On 10/18/2022 4:08 PM, Jiri Pirko wrote:
> Tue, Oct 18, 2022 at 02:35:39PM CEST, michal.wilczynski@...el.com wrote:
>> ice driver needs an ability to create devlink-rate nodes from inside the
>> driver. We have default Tx-scheduler tree that we would like to
>> export as devlink-rate objects.
>>
>> There is also a need to support additional parameters, besides two that
>> are supported currently:
>> tx_priority - priority among siblings (0-7)
>> tx_weight - weights for the WFQ algorithm (1-200)
>>
>> Allow creation of nodes from the driver, and introduce new argument
>> to devl_rate_leaf_create, so the parent can be set during the creation
>> of the leaf node.
>>
>> Implement new parameters - tx_priority, tx_weight.
>>
>> Allow modification of the priv field in the devlink_rate from parent_set
>> callbacks. This is needed because creating nodes without parents doesn't
>> make any sense in ice driver case. It's much more elegant to actually
>> create a node when the parent is assigned.
> This should be split into like 3-4 patches.

Sure,
Will try to split and re-send,
Thanks,
Michał