netdev - Re: [PATCH net-next 00/13] Control action percpu counters allocation by netlink flag

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]

Message-ID: <1cd8ce66-cb27-b9e8-c8a6-da63c8226aae@mojatatu.com>
Date:   Wed, 23 Oct 2019 09:02:58 -0400
From:   Jamal Hadi Salim <jhs@...atatu.com>
To:     Vlad Buslov <vladbu@...lanox.com>, Roman Mashak <mrv@...atatu.com>
Cc:     Marcelo Ricardo Leitner <mleitner@...hat.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "xiyou.wangcong@...il.com" <xiyou.wangcong@...il.com>,
        "jiri@...nulli.us" <jiri@...nulli.us>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "dcaratti@...hat.com" <dcaratti@...hat.com>,
        "pabeni@...hat.com" <pabeni@...hat.com>
Subject: Re: [PATCH net-next 00/13] Control action percpu counters allocation
 by netlink flag


I shouldve read the thread backward. My earlier email was asking
similar question to Roman.

On 2019-10-23 2:38 a.m., Vlad Buslov wrote:
> 
> On Tue 22 Oct 2019 at 21:17, Roman Mashak <mrv@...atatu.com> wrote:

> Hi Roman,
> 
> I considered it, but didn't find good way to implement my change with
> TCA_ROOT_FLAGS. I needed some flags to be per-action for following
> reasons:
> 
> 1. Not all actions support the flag (only implemented for hw offloaded
>     actions).
> 
 >
> 2. TCA_ROOT_FLAGS is act API specific and I need this to work when
>     actions are created when actions are created with filters through cls
>     API. I guess I could have changed tcf_action_init_1() to require
>     having TCA_ROOT_FLAGS before actions attribute and then pass obtained
>     value to act_ops->init() as additional argument, but it sounds more
>     complex and ugly.

I really shouldve read the thread backwards;->
The question is if this uglier than introducing a new TLV for every
action.

cheers,
jamal