netdev - Re: [PATCH net-next 8/8] net/mlx5e: XDP TX xmit more

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: <0e732cab-ed54-5e09-569b-f75258e98ad4@gmail.com>
Date:   Tue, 20 Sep 2016 11:19:46 +0300
From:   Tariq Toukan <ttoukan.linux@...il.com>
To:     Jesper Dangaard Brouer <brouer@...hat.com>,
        Tariq Toukan <tariqt@...lanox.com>
Cc:     "David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
        Eran Ben Elisha <eranbe@...lanox.com>,
        Saeed Mahameed <saeedm@...lanox.com>,
        Rana Shahout <ranas@...lanox.com>
Subject: Re: [PATCH net-next 8/8] net/mlx5e: XDP TX xmit more

Hi Jesper,

On 20/09/2016 10:46 AM, Jesper Dangaard Brouer wrote:
> On Mon, 19 Sep 2016 16:58:59 +0300
> Tariq Toukan <tariqt@...lanox.com> wrote:
>
>> From: Saeed Mahameed <saeedm@...lanox.com>
>>
>> Previously we rang XDP SQ doorbell on every forwarded XDP packet.
>>
>> Here we introduce a xmit more like mechanism that will queue up more
>> than one packet into SQ (up to RX napi budget) w/o notifying the hardware.
>>
>> Once RX napi budget is consumed and we exit napi RX loop, we will
>> flush (doorbell) all XDP looped packets in case there are such.
> I've already raised strong concerns with this approach on the RFC
> patchset.  Of not really taking advantage of RX bulking.
> Please do not ignore this!
Sure. Your approach can fit with our plans to split the RX completion 
poll loop into several stages.
I will try it when we get there.
>
> If you can promise, that we/you will also try to other approach I'm
> suggesting, then I'm fine with this patch.
>
Thanks.