All of lore.kernel.org
 help / color / mirror / Atom feed
From: Amritha Nambiar <amritha.nambiar@intel.com>
To: intel-wired-lan@lists.osuosl.org
Cc: alexander.h.duyck@intel.com, kiran.patil@intel.com,
	amritha.nambiar@intel.com, sridhar.samudrala@intel.com,
	mitch.a.williams@intel.com, neerav.parikh@intel.com,
	jeffrey.t.kirsher@intel.com, netdev@vger.kernel.org
Subject: [PATCH 0/4] Configuring traffic classes via new hardware offload mechanism in tc/mqprio
Date: Fri, 19 May 2017 17:58:17 -0700	[thread overview]
Message-ID: <149524122523.11022.4541073724650541658.stgit@anamdev.jf.intel.com> (raw)

The following series introduces a new harware offload mode in tc/mqprio where the TCs, the queue configurations and bandwidth rate limits are offloaded to the hardware.
The i40e driver enables the new mqprio hardware offload mechanism factoring the TCs, queue configuration and bandwidth rates by creating HW channel VSIs. 

In this mode, the priority to traffic class mapping and the user specified queue ranges are used to configure the traffic class when the 'hw' option is set to 2. This is achieved by creating HW channels(VSI). A new channel is created for each of the traffic class configuration offloaded via mqprio framework except for the first TC (TC0) which is for the main VSI. TC0 for the main VSI is also reconfigured as per user provided queue parameters. Finally, bandwidth rate limits are set on these traffic classes through the mqprio offload framework by sending these rates in addition to the number of TCs and the queue configurations.

Example:
# tc qdisc add dev eth0 root mqprio num_tc 2  map 0 0 0 0 1 1 1 1\
  queues 4@0 4@4 min_rate 0Mbit 0Mbit max_rate 55Mbit 60Mbit hw 2

To dump the bandwidth rates:

# tc qdisc show dev eth0
  qdisc mqprio 804a: root  tc 2 map 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0
               queues:(0:3) (4:7)
               min rates:0bit 0bit
               max rates:55Mbit 60Mbit

---

Amritha Nambiar (4):
      [next-queue]net: mqprio: Introduce new hardware offload mode in mqprio for offloading full TC configurations
      [next-queue]net: i40e: Add infrastructure for queue channel support with the TCs and queue configurations offloaded via mqprio scheduler
      [next-queue]net: i40e: Enable mqprio full offload mode in the i40e driver for configuring TCs and queue mapping
      [next-queue]net: i40e: Add support to set max bandwidth rates for TCs offloaded via tc/mqprio


 drivers/net/ethernet/intel/i40e/i40e.h         |   42 +
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c |    6 
 drivers/net/ethernet/intel/i40e/i40e_main.c    | 1365 +++++++++++++++++++++---
 drivers/net/ethernet/intel/i40e/i40e_txrx.h    |    2 
 include/linux/netdevice.h                      |    2 
 include/net/pkt_cls.h                          |    7 
 include/uapi/linux/pkt_sched.h                 |   13 
 net/sched/sch_mqprio.c                         |  169 +++
 8 files changed, 1449 insertions(+), 157 deletions(-)

WARNING: multiple messages have this Message-ID (diff)
From: Amritha Nambiar <amritha.nambiar@intel.com>
To: intel-wired-lan@osuosl.org
Subject: [Intel-wired-lan] [PATCH 0/4] Configuring traffic classes via new hardware offload mechanism in tc/mqprio
Date: Fri, 19 May 2017 17:58:17 -0700	[thread overview]
Message-ID: <149524122523.11022.4541073724650541658.stgit@anamdev.jf.intel.com> (raw)

The following series introduces a new harware offload mode in tc/mqprio where the TCs, the queue configurations and bandwidth rate limits are offloaded to the hardware.
The i40e driver enables the new mqprio hardware offload mechanism factoring the TCs, queue configuration and bandwidth rates by creating HW channel VSIs. 

In this mode, the priority to traffic class mapping and the user specified queue ranges are used to configure the traffic class when the 'hw' option is set to 2. This is achieved by creating HW channels(VSI). A new channel is created for each of the traffic class configuration offloaded via mqprio framework except for the first TC (TC0) which is for the main VSI. TC0 for the main VSI is also reconfigured as per user provided queue parameters. Finally, bandwidth rate limits are set on these traffic classes through the mqprio offload framework by sending these rates in addition to the number of TCs and the queue configurations.

Example:
# tc qdisc add dev eth0 root mqprio num_tc 2  map 0 0 0 0 1 1 1 1\
  queues 4 at 0 4 at 4 min_rate 0Mbit 0Mbit max_rate 55Mbit 60Mbit hw 2

To dump the bandwidth rates:

# tc qdisc show dev eth0
  qdisc mqprio 804a: root  tc 2 map 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0
               queues:(0:3) (4:7)
               min rates:0bit 0bit
               max rates:55Mbit 60Mbit

---

Amritha Nambiar (4):
      [next-queue]net: mqprio: Introduce new hardware offload mode in mqprio for offloading full TC configurations
      [next-queue]net: i40e: Add infrastructure for queue channel support with the TCs and queue configurations offloaded via mqprio scheduler
      [next-queue]net: i40e: Enable mqprio full offload mode in the i40e driver for configuring TCs and queue mapping
      [next-queue]net: i40e: Add support to set max bandwidth rates for TCs offloaded via tc/mqprio


 drivers/net/ethernet/intel/i40e/i40e.h         |   42 +
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c |    6 
 drivers/net/ethernet/intel/i40e/i40e_main.c    | 1365 +++++++++++++++++++++---
 drivers/net/ethernet/intel/i40e/i40e_txrx.h    |    2 
 include/linux/netdevice.h                      |    2 
 include/net/pkt_cls.h                          |    7 
 include/uapi/linux/pkt_sched.h                 |   13 
 net/sched/sch_mqprio.c                         |  169 +++
 8 files changed, 1449 insertions(+), 157 deletions(-)

--

             reply	other threads:[~2017-05-19 20:28 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-20  0:58 Amritha Nambiar [this message]
2017-05-20  0:58 ` [Intel-wired-lan] [PATCH 0/4] Configuring traffic classes via new hardware offload mechanism in tc/mqprio Amritha Nambiar
2017-05-19 22:30 ` John Fastabend
2017-05-19 22:30   ` John Fastabend
2017-05-20  0:58 ` [PATCH 1/4] [next-queue]net: mqprio: Introduce new hardware offload mode in mqprio for offloading full TC configurations Amritha Nambiar
2017-05-20  0:58   ` [Intel-wired-lan] " Amritha Nambiar
2017-05-24 21:59   ` Alexander Duyck
2017-05-24 21:59     ` Alexander Duyck
2017-05-20  0:58 ` [PATCH 2/4] [next-queue]net: i40e: Add infrastructure for queue channel support with the TCs and queue configurations offloaded via mqprio scheduler Amritha Nambiar
2017-05-20  0:58   ` [Intel-wired-lan] " Amritha Nambiar
2017-05-24 21:45   ` Alexander Duyck
2017-05-24 21:45     ` Alexander Duyck
2017-05-24 22:03     ` Patil, Kiran
2017-05-20  0:58 ` [PATCH 3/4] [next-queue]net: i40e: Enable mqprio full offload mode in the i40e driver for configuring TCs and queue mapping Amritha Nambiar
2017-05-20  0:58   ` [Intel-wired-lan] " Amritha Nambiar
2017-05-24 22:05   ` Alexander Duyck
2017-05-24 22:05     ` Alexander Duyck
2017-05-20  0:58 ` [PATCH 4/4] [next-queue]net: i40e: Add support to set max bandwidth rates for TCs offloaded via tc/mqprio Amritha Nambiar
2017-05-20  0:58   ` [Intel-wired-lan] " Amritha Nambiar
2017-05-20 21:15 ` [PATCH 0/4] Configuring traffic classes via new hardware offload mechanism in tc/mqprio Or Gerlitz
2017-05-20 21:15   ` [Intel-wired-lan] " Or Gerlitz
2017-05-21 22:35   ` Alexander Duyck
2017-05-21 22:35     ` Alexander Duyck
2017-05-22  3:25     ` Or Gerlitz
2017-05-22  3:25       ` Or Gerlitz
2017-05-22 16:40       ` Duyck, Alexander H
2017-05-22 16:40         ` Duyck, Alexander H
2017-05-22 19:31 ` Jeff Kirsher
2017-05-22 19:31   ` [Intel-wired-lan] " Jeff Kirsher
2017-07-21  9:42   ` Richard Cochran
2017-07-21  9:42     ` [Intel-wired-lan] " Richard Cochran
2017-07-26 18:18     ` Nambiar, Amritha
2017-07-26 18:18       ` [Intel-wired-lan] " Nambiar, Amritha

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=149524122523.11022.4541073724650541658.stgit@anamdev.jf.intel.com \
    --to=amritha.nambiar@intel.com \
    --cc=alexander.h.duyck@intel.com \
    --cc=intel-wired-lan@lists.osuosl.org \
    --cc=jeffrey.t.kirsher@intel.com \
    --cc=kiran.patil@intel.com \
    --cc=mitch.a.williams@intel.com \
    --cc=neerav.parikh@intel.com \
    --cc=netdev@vger.kernel.org \
    --cc=sridhar.samudrala@intel.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.