netdev - Re: [PATCH net-next] virtio

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: <d73dbd67-cc60-366c-7ec5-1f9818d70bd5@lab.ntt.co.jp>
Date:   Thu, 21 Dec 2017 11:12:39 +0900
From:   Toshiaki Makita <makita.toshiaki@....ntt.co.jp>
To:     Jason Wang <jasowang@...hat.com>,
        "David S . Miller" <davem@...emloft.net>,
        "Michael S . Tsirkin" <mst@...hat.com>
Cc:     netdev@...r.kernel.org, virtualization@...ts.linux-foundation.org
Subject: Re: [PATCH net-next] virtio_net: Add ethtool stats

On 2017/12/20 17:13, Jason Wang wrote:
> On 2017年12月20日 12:40, Toshiaki Makita wrote:
>> The main purpose of this patch is adding a way of checking per-queue
>> stats.
>> It's useful to debug performance problems on multiqueue environment.
>>
>> $ ethtool -S ens10
>> NIC statistics:
>>       rx_packets: 4172939
>>       tx_packets: 5855538
>>       rx_bytes: 6317757408
>>       tx_bytes: 8865151846
>>       rx_dropped: 0
>>       rx_length_errors: 0
>>       rx_frame_errors: 0
>>       tx_dropped: 0
>>       tx_fifo_errors: 0
>>       rx_queue_0_packets: 2090408
>>       rx_queue_0_bytes: 3164825094
>>       rx_queue_1_packets: 2082531
>>       rx_queue_1_bytes: 3152932314
>>       tx_queue_0_packets: 2770841
>>       tx_queue_0_bytes: 4194955474
>>       tx_queue_1_packets: 3084697
>>       tx_queue_1_bytes: 4670196372
>>
>> Signed-off-by: Toshiaki Makita<makita.toshiaki@....ntt.co.jp>
> 
> Thanks for the patch. This is not the first patch that wants to do this.
> Maybe you can go through the previous comments (E.g there's one in
> https://patchwork.ozlabs.org/patch/244413/).

Thanks for pointing it. I was not aware of it.
I checked the previous comments but not sure if there is anything to do.
The main logic is coincidentally the same - splitting percpu stats data
structure into rx and tx ones in receive_queue and send_queue, which has
been acked.
Comments from Ben Hutchings is about stats ordering but it seems no
consensus has been formed since then.
Feedback from Michael is kind of nit-picking ones and does not apply to
mine. He also requested other stats items like vm-exit, but not in the
same patch.
You requested performance numbers. Would you like to have it for this
patch as well?

> For me, I'd expect to add some XDP counters like other device did.

Nice to have it. I guess it would be another patch.

-- 
Toshiaki Makita