mbox series

[v2,net-next,00/15] taprio automatic queueMaxSDU and new TXQ selection procedure

Message ID 20230207135440.1482856-1-vladimir.oltean@nxp.com (mailing list archive)
Headers show
Series taprio automatic queueMaxSDU and new TXQ selection procedure | expand

Message

Vladimir Oltean Feb. 7, 2023, 1:54 p.m. UTC
This patch set addresses 2 design limitations in the taprio software scheduler:

1. Software scheduling fundamentally prioritizes traffic incorrectly,
   in a way which was inspired from Intel igb/igc drivers and does not
   follow the inputs user space gives (traffic classes and TC to TXQ
   mapping). Patch 05/15 handles this, 01/15 - 04/15 are preparations
   for this work.

2. Software scheduling assumes that the gate for a traffic class closes
   as soon as the next interval begins. But this isn't true.
   If consecutive schedule entries have that traffic class gate open,
   there is no "gate close" event and taprio should keep dequeuing from
   that TC without interruptions. Patches 06/15 - 15/15 handle this.
   Patch 10/15 is a generic Qdisc change required for this to work.

Future development directions which depend on this patch set are:

- Propagating the automatic queueMaxSDU calculation down to offloading
  device drivers, instead of letting them calculate this, as
  vsc9959_tas_guard_bands_update() does today.

- A software data path for tc-taprio with preemptible traffic and
  Hold/Release events.

v1 at:
https://patchwork.kernel.org/project/netdevbpf/cover/20230128010719.2182346-1-vladimir.oltean@nxp.com/

Vladimir Oltean (15):
  net/sched: taprio: delete peek() implementation
  net/sched: taprio: continue with other TXQs if one dequeue() failed
  net/sched: taprio: refactor one skb dequeue from TXQ to separate
    function
  net/sched: taprio: avoid calling child->ops->dequeue(child) twice
  net/sched: taprio: give higher priority to higher TCs in software
    dequeue mode
  net/sched: taprio: calculate tc gate durations
  net/sched: taprio: rename close_time to end_time
  net/sched: taprio: calculate budgets per traffic class
  net/sched: taprio: calculate guard band against actual TC gate close
    time
  net/sched: make stab available before ops->init() call
  net/sched: taprio: warn about missing size table
  net/sched: keep the max_frm_len information inside struct
    sched_gate_list
  net/sched: taprio: automatically calculate queueMaxSDU based on TC
    gate durations
  net/sched: taprio: split segmentation logic from qdisc_enqueue()
  net/sched: taprio: don't segment unnecessarily

 drivers/net/ethernet/intel/igb/igb_main.c |  18 +
 drivers/net/ethernet/intel/igc/igc_main.c |   6 +-
 include/net/pkt_sched.h                   |   5 +
 net/sched/sch_api.c                       |  29 +-
 net/sched/sch_taprio.c                    | 639 ++++++++++++++++------
 5 files changed, 500 insertions(+), 197 deletions(-)

Comments

patchwork-bot+netdevbpf@kernel.org Feb. 8, 2023, 9:50 a.m. UTC | #1
Hello:

This series was applied to netdev/net-next.git (master)
by David S. Miller <davem@davemloft.net>:

On Tue,  7 Feb 2023 15:54:25 +0200 you wrote:
> This patch set addresses 2 design limitations in the taprio software scheduler:
> 
> 1. Software scheduling fundamentally prioritizes traffic incorrectly,
>    in a way which was inspired from Intel igb/igc drivers and does not
>    follow the inputs user space gives (traffic classes and TC to TXQ
>    mapping). Patch 05/15 handles this, 01/15 - 04/15 are preparations
>    for this work.
> 
> [...]

Here is the summary with links:
  - [v2,net-next,01/15] net/sched: taprio: delete peek() implementation
    https://git.kernel.org/netdev/net-next/c/ecc0cc98632a
  - [v2,net-next,02/15] net/sched: taprio: continue with other TXQs if one dequeue() failed
    https://git.kernel.org/netdev/net-next/c/1638bbbe4ece
  - [v2,net-next,03/15] net/sched: taprio: refactor one skb dequeue from TXQ to separate function
    https://git.kernel.org/netdev/net-next/c/92f966674f6a
  - [v2,net-next,04/15] net/sched: taprio: avoid calling child->ops->dequeue(child) twice
    https://git.kernel.org/netdev/net-next/c/4c22942734f0
  - [v2,net-next,05/15] net/sched: taprio: give higher priority to higher TCs in software dequeue mode
    https://git.kernel.org/netdev/net-next/c/2f530df76c8c
  - [v2,net-next,06/15] net/sched: taprio: calculate tc gate durations
    https://git.kernel.org/netdev/net-next/c/a306a90c8ffe
  - [v2,net-next,07/15] net/sched: taprio: rename close_time to end_time
    https://git.kernel.org/netdev/net-next/c/e5517551112f
  - [v2,net-next,08/15] net/sched: taprio: calculate budgets per traffic class
    https://git.kernel.org/netdev/net-next/c/d2ad689dec10
  - [v2,net-next,09/15] net/sched: taprio: calculate guard band against actual TC gate close time
    https://git.kernel.org/netdev/net-next/c/a1e6ad30fa19
  - [v2,net-next,10/15] net/sched: make stab available before ops->init() call
    https://git.kernel.org/netdev/net-next/c/1f62879e3632
  - [v2,net-next,11/15] net/sched: taprio: warn about missing size table
    https://git.kernel.org/netdev/net-next/c/a3d91b2c6f6b
  - [v2,net-next,12/15] net/sched: keep the max_frm_len information inside struct sched_gate_list
    https://git.kernel.org/netdev/net-next/c/a878fd46fe43
  - [v2,net-next,13/15] net/sched: taprio: automatically calculate queueMaxSDU based on TC gate durations
    https://git.kernel.org/netdev/net-next/c/fed87cc6718a
  - [v2,net-next,14/15] net/sched: taprio: split segmentation logic from qdisc_enqueue()
    https://git.kernel.org/netdev/net-next/c/2d5e8071c47a
  - [v2,net-next,15/15] net/sched: taprio: don't segment unnecessarily
    https://git.kernel.org/netdev/net-next/c/39b02d6d104a

You are awesome, thank you!