All of lore.kernel.org
 help / color / mirror / Atom feed
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Network Development <netdev@vger.kernel.org>,
	Jason Wang <jasowang@redhat.com>,
	virtualization@lists.linux-foundation.org,
	David Miller <davem@davemloft.net>,
	Willem de Bruijn <willemb@google.com>
Subject: Re: [PATCH net-next 2/3] virtio-net: transmit napi
Date: Mon, 3 Apr 2017 01:07:49 -0400	[thread overview]
Message-ID: <CAF=yD-+5+X6UBYaX=Dh3M3caty0fJ7cjamTfX0=_+Lwx6r_VQw@mail.gmail.com> (raw)
In-Reply-To: <20170403052938-mutt-send-email-mst@kernel.org>

On Sun, Apr 2, 2017 at 10:30 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Sun, Apr 02, 2017 at 04:10:11PM -0400, Willem de Bruijn wrote:
>> From: Willem de Bruijn <willemb@google.com>
>>
>> Convert virtio-net to a standard napi tx completion path. This enables
>> better TCP pacing using TCP small queues and increases single stream
>> throughput.
>>
>> The virtio-net driver currently cleans tx descriptors on transmission
>> of new packets in ndo_start_xmit. Latency depends on new traffic, so
>> is unbounded. To avoid deadlock when a socket reaches its snd limit,
>> packets are orphaned on tranmission. This breaks socket backpressure,
>> including TSQ.
>>
>> Napi increases the number of interrupts generated compared to the
>> current model, which keeps interrupts disabled as long as the ring
>> has enough free descriptors. Keep tx napi optional for now. Follow-on
>> patches will reduce the interrupt cost.
>>
>> Signed-off-by: Willem de Bruijn <willemb@google.com>
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>> ---
>>  drivers/net/virtio_net.c | 63 ++++++++++++++++++++++++++++++++++++++++--------
>>  1 file changed, 53 insertions(+), 10 deletions(-)
>>
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index 6aac0ad0d9b2..95d938e82080 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
>> @@ -33,9 +33,10 @@
>>  static int napi_weight = NAPI_POLL_WEIGHT;
>>  module_param(napi_weight, int, 0444);
>>
>> -static bool csum = true, gso = true;
>> +static bool csum = true, gso = true, napi_tx = true;
>>  module_param(csum, bool, 0444);
>>  module_param(gso, bool, 0444);
>> +module_param(napi_tx, bool, 0644);
>>
>>  /* FIXME: MTU in config. */
>>  #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
>
> Off by default seems safer until we can find better ways
> to reduce the overhead, esp for UDP.

Okay, I'll change that. I don't have an immediate idea for that
unidirectional case.

  parent reply	other threads:[~2017-04-03  5:08 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-02 20:10 [PATCH net-next 0/3] virtio-net tx napi Willem de Bruijn
2017-04-02 20:10 ` [PATCH net-next 1/3] virtio-net: napi helper functions Willem de Bruijn
2017-04-02 20:10 ` Willem de Bruijn
2017-04-02 20:10 ` [PATCH net-next 2/3] virtio-net: transmit napi Willem de Bruijn
2017-04-02 20:10 ` Willem de Bruijn
2017-04-03  2:30   ` Michael S. Tsirkin
2017-04-03  5:07     ` Willem de Bruijn
2017-04-03  5:07     ` Willem de Bruijn [this message]
2017-04-02 20:10 ` [PATCH net-next 3/3] virtio-net: clean tx descriptors from rx napi Willem de Bruijn
2017-04-02 20:10 ` Willem de Bruijn
2017-04-03  2:47   ` Michael S. Tsirkin
2017-04-03  5:02     ` Willem de Bruijn
2017-04-03  5:02     ` Willem de Bruijn
2017-04-07 19:28       ` Michael S. Tsirkin
2017-04-07 19:28       ` Michael S. Tsirkin
2017-04-07 20:59         ` Willem de Bruijn
2017-04-07 20:59         ` Willem de Bruijn
2017-04-07 21:10           ` Michael S. Tsirkin
2017-04-07 21:10           ` Michael S. Tsirkin
2017-04-07 21:14             ` Willem de Bruijn
2017-04-07 21:14             ` Willem de Bruijn
2017-04-03  2:47   ` Michael S. Tsirkin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAF=yD-+5+X6UBYaX=Dh3M3caty0fJ7cjamTfX0=_+Lwx6r_VQw@mail.gmail.com' \
    --to=willemdebruijn.kernel@gmail.com \
    --cc=davem@davemloft.net \
    --cc=jasowang@redhat.com \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=willemb@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.