All of lore.kernel.org
 help / color / mirror / Atom feed
From: John Fastabend <john.fastabend@gmail.com>
To: intel-wired-lan@osuosl.org
Subject: [Intel-wired-lan] [net-next PATCH v2 2/2] e1000: bundle xdp xmit routines
Date: Fri, 9 Sep 2016 16:37:06 -0700	[thread overview]
Message-ID: <57D347A2.4060909@gmail.com> (raw)
In-Reply-To: <20160909212938.4001.40540.stgit@john-Precision-Tower-5810>

On 16-09-09 02:29 PM, John Fastabend wrote:
> e1000 supports a single TX queue so it is being shared with the stack
> when XDP runs XDP_TX action. This requires taking the xmit lock to
> ensure we don't corrupt the tx ring. To avoid taking and dropping the
> lock per packet this patch adds a bundling implementation to submit
> a bundle of packets to the xmit routine.
> 
> I tested this patch running e1000 in a VM using KVM over a tap
> device using pktgen to generate traffic along with 'ping -f -l 100'.
> 
> Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
> ---

This patch is a bit bogus in a few spots as well...


> -
> -	if (E1000_DESC_UNUSED(tx_ring) < 2) {
> -		HARD_TX_UNLOCK(netdev, txq);
> -		return;
> +	for (; i < E1000_XDP_XMIT_BUNDLE_MAX && buffer_info[i].buffer; i++) {
> +		e1000_xmit_raw_frame(buffer_info[i].buffer,
> +				     buffer_info[i].length,
> +				     adapter, tx_ring);
> +		buffer_info[i].buffer->rxbuf.page = NULL;
> +		buffer_info[i].buffer = NULL;
> +		buffer_info[i].length = 0;
> +		i++;
              ^^^^^^^^


WARNING: multiple messages have this Message-ID (diff)
From: John Fastabend <john.fastabend@gmail.com>
To: bblanco@plumgrid.com, alexei.starovoitov@gmail.com,
	jeffrey.t.kirsher@intel.com, brouer@redhat.com,
	davem@davemloft.net
Cc: xiyou.wangcong@gmail.com, intel-wired-lan@lists.osuosl.org,
	u9012063@gmail.com, netdev@vger.kernel.org
Subject: Re: [net-next PATCH v2 2/2] e1000: bundle xdp xmit routines
Date: Fri, 9 Sep 2016 16:37:06 -0700	[thread overview]
Message-ID: <57D347A2.4060909@gmail.com> (raw)
In-Reply-To: <20160909212938.4001.40540.stgit@john-Precision-Tower-5810>

On 16-09-09 02:29 PM, John Fastabend wrote:
> e1000 supports a single TX queue so it is being shared with the stack
> when XDP runs XDP_TX action. This requires taking the xmit lock to
> ensure we don't corrupt the tx ring. To avoid taking and dropping the
> lock per packet this patch adds a bundling implementation to submit
> a bundle of packets to the xmit routine.
> 
> I tested this patch running e1000 in a VM using KVM over a tap
> device using pktgen to generate traffic along with 'ping -f -l 100'.
> 
> Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
> ---

This patch is a bit bogus in a few spots as well...


> -
> -	if (E1000_DESC_UNUSED(tx_ring) < 2) {
> -		HARD_TX_UNLOCK(netdev, txq);
> -		return;
> +	for (; i < E1000_XDP_XMIT_BUNDLE_MAX && buffer_info[i].buffer; i++) {
> +		e1000_xmit_raw_frame(buffer_info[i].buffer,
> +				     buffer_info[i].length,
> +				     adapter, tx_ring);
> +		buffer_info[i].buffer->rxbuf.page = NULL;
> +		buffer_info[i].buffer = NULL;
> +		buffer_info[i].length = 0;
> +		i++;
              ^^^^^^^^

  reply	other threads:[~2016-09-09 23:37 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-09 21:29 [Intel-wired-lan] [net-next PATCH v2 1/2] e1000: add initial XDP support John Fastabend
2016-09-09 21:29 ` John Fastabend
2016-09-09 21:29 ` [Intel-wired-lan] [net-next PATCH v2 2/2] e1000: bundle xdp xmit routines John Fastabend
2016-09-09 21:29   ` John Fastabend
2016-09-09 23:37   ` John Fastabend [this message]
2016-09-09 23:37     ` John Fastabend
2016-09-09 23:44   ` [Intel-wired-lan] " Tom Herbert
2016-09-09 23:44     ` Tom Herbert
2016-09-10  0:01     ` [Intel-wired-lan] " John Fastabend
2016-09-10  0:01       ` John Fastabend
2016-09-10  1:04       ` [Intel-wired-lan] " Tom Herbert
2016-09-10  1:04         ` Tom Herbert
2016-09-10  1:12         ` [Intel-wired-lan] " John Fastabend
2016-09-10  1:12           ` John Fastabend
2016-09-10  1:19           ` [Intel-wired-lan] " Tom Herbert
2016-09-10  1:19             ` Tom Herbert
2016-09-10  1:40             ` [Intel-wired-lan] " Alexei Starovoitov
2016-09-10  1:40               ` Alexei Starovoitov
2016-09-10  3:12               ` [Intel-wired-lan] " Tom Herbert
2016-09-10  3:12                 ` Tom Herbert
2016-09-10  3:26                 ` [Intel-wired-lan] " John Fastabend
2016-09-10  3:26                   ` John Fastabend
2016-09-10  4:13                   ` [Intel-wired-lan] " Tom Herbert
2016-09-10  4:13                     ` Tom Herbert
2016-09-12  3:15                     ` [Intel-wired-lan] " John Fastabend
2016-09-12  3:15                       ` John Fastabend
2016-09-12  4:12                       ` [Intel-wired-lan] " Alexei Starovoitov
2016-09-12  4:12                         ` Alexei Starovoitov
2016-09-10  3:56                 ` [Intel-wired-lan] " Alexei Starovoitov
2016-09-10  3:56                   ` Alexei Starovoitov
2016-09-12 11:56             ` [Intel-wired-lan] " Jesper Dangaard Brouer
2016-09-12 11:56               ` Jesper Dangaard Brouer
2016-09-10 15:36   ` [Intel-wired-lan] " Tom Herbert
2016-09-10 15:36     ` Tom Herbert
2016-09-12  3:07     ` [Intel-wired-lan] " John Fastabend
2016-09-12  3:07       ` John Fastabend
2016-09-12 12:17   ` [Intel-wired-lan] " Jesper Dangaard Brouer
2016-09-12 12:17     ` Jesper Dangaard Brouer
2016-09-12 18:11     ` [Intel-wired-lan] " John Fastabend
2016-09-12 18:11       ` John Fastabend
2016-09-09 22:04 ` [Intel-wired-lan] [net-next PATCH v2 1/2] e1000: add initial XDP support Eric Dumazet
2016-09-09 22:04   ` Eric Dumazet
2016-09-09 23:33   ` [Intel-wired-lan] " John Fastabend
2016-09-09 23:33     ` John Fastabend
2016-09-21  4:26 ` [Intel-wired-lan] " zhuyj
2016-09-21  4:26   ` zhuyj
2016-09-21  4:30   ` [Intel-wired-lan] " John Fastabend
2016-09-21  4:30     ` John Fastabend

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=57D347A2.4060909@gmail.com \
    --to=john.fastabend@gmail.com \
    --cc=intel-wired-lan@osuosl.org \
    /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.