DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jerin Jacob <jerinjacobk@gmail.com>
To: "Shijith Thotton" <sthotton@marvell.com>,
	"McDaniel, Timothy" <timothy.mcdaniel@intel.com>,
	"Gujjar, Abhinandan S" <abhinandan.gujjar@intel.com>,
	"Van Haaren, Harry" <harry.van.haaren@intel.com>,
	"Hemant Agrawal" <hemant.agrawal@nxp.com>,
	"Liang Ma" <liangma@liangbit.com>,
	"Mattias Rönnblom" <mattias.ronnblom@ericsson.com>,
	"Erik Gabriel Carrillo" <erik.g.carrillo@intel.com>,
	"Nipun Gupta" <nipun.gupta@nxp.com>,
	"Jayatheerthan, Jay" <jay.jayatheerthan@intel.com>,
	"Peter Mccarthy" <peter.mccarthy@intel.com>
Cc: jerinj@marvell.com, dev@dpdk.org, Ray Kinsella <mdr@ashroe.eu>,
	 Pavan Nikhilesh <pbhagavatula@marvell.com>
Subject: Re: [PATCH] eventdev: add weight and affinity attributes to queue conf
Date: Wed, 14 Sep 2022 18:04:33 +0530	[thread overview]
Message-ID: <CALBAE1NTo11WvecJR0c-NXMCdrmKwz4zfXZ0bkCByD92EViPGg@mail.gmail.com> (raw)
In-Reply-To: <dcd3cf0ec034632f97223bb9df389f9cedf9753c.1660116951.git.sthotton@marvell.com>

On Wed, Aug 10, 2022 at 1:13 PM Shijith Thotton <sthotton@marvell.com> wrote:
>
> Added new fields to represent event queue weight and affinity in
> rte_event_queue_conf structure. Internal op to get queue attribute is
> removed as it is no longer needed. Updated driver to use the new field.
>
> Signed-off-by: Shijith Thotton <sthotton@marvell.com>

Acked-by: Jerin Jacob <jerinj@marvell.com>

+ eventdev PMD maintainers to raise any feedback before merging to next-eventdev


> ---
>  doc/guides/rel_notes/deprecation.rst   |  3 --
>  doc/guides/rel_notes/release_22_11.rst |  3 ++
>  drivers/event/cnxk/cn10k_eventdev.c    |  1 -
>  drivers/event/cnxk/cn9k_eventdev.c     |  1 -
>  drivers/event/cnxk/cnxk_eventdev.c     | 42 ++++++--------------------
>  drivers/event/cnxk/cnxk_eventdev.h     |  9 ------
>  lib/eventdev/eventdev_pmd.h            | 22 --------------
>  lib/eventdev/rte_eventdev.c            | 10 +++---
>  lib/eventdev/rte_eventdev.h            | 16 ++++++++++
>  9 files changed, 33 insertions(+), 74 deletions(-)
>
> diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst
> index e7583cae4c..13e7c6370e 100644
> --- a/doc/guides/rel_notes/deprecation.rst
> +++ b/doc/guides/rel_notes/deprecation.rst
> @@ -206,9 +206,6 @@ Deprecation Notices
>    ``rte_event_vector::elem_offset`` gives the number of valid elements left
>    to process from the ``rte_event_vector::elem_offset``.
>
> -* eventdev: New fields to represent event queue weight and affinity
> -  will be added to ``rte_event_queue_conf`` structure in DPDK 22.11.
> -
>  * metrics: The function ``rte_metrics_init`` will have a non-void return
>    in order to notify errors instead of calling ``rte_exit``.
>
> diff --git a/doc/guides/rel_notes/release_22_11.rst b/doc/guides/rel_notes/release_22_11.rst
> index 8c021cf050..8ffd71e650 100644
> --- a/doc/guides/rel_notes/release_22_11.rst
> +++ b/doc/guides/rel_notes/release_22_11.rst
> @@ -88,6 +88,9 @@ API Changes
>  ABI Changes
>  -----------
>
> +* eventdev: Added ``weight`` and ``affinity`` fields to ``rte_event_queue_conf``
> +  structure.
> +
>  .. This section should contain ABI changes. Sample format:
>
>     * sample: Add a short 1-2 sentence description of the ABI change
> diff --git a/drivers/event/cnxk/cn10k_eventdev.c b/drivers/event/cnxk/cn10k_eventdev.c
> index 5a0cab40a9..aa8ae394bc 100644
> --- a/drivers/event/cnxk/cn10k_eventdev.c
> +++ b/drivers/event/cnxk/cn10k_eventdev.c
> @@ -922,7 +922,6 @@ static struct eventdev_ops cn10k_sso_dev_ops = {
>         .queue_def_conf = cnxk_sso_queue_def_conf,
>         .queue_setup = cnxk_sso_queue_setup,
>         .queue_release = cnxk_sso_queue_release,
> -       .queue_attr_get = cnxk_sso_queue_attribute_get,
>         .queue_attr_set = cnxk_sso_queue_attribute_set,
>
>         .port_def_conf = cnxk_sso_port_def_conf,
> diff --git a/drivers/event/cnxk/cn9k_eventdev.c b/drivers/event/cnxk/cn9k_eventdev.c
> index 2e27030049..58c72a580a 100644
> --- a/drivers/event/cnxk/cn9k_eventdev.c
> +++ b/drivers/event/cnxk/cn9k_eventdev.c
> @@ -1153,7 +1153,6 @@ static struct eventdev_ops cn9k_sso_dev_ops = {
>         .queue_def_conf = cnxk_sso_queue_def_conf,
>         .queue_setup = cnxk_sso_queue_setup,
>         .queue_release = cnxk_sso_queue_release,
> -       .queue_attr_get = cnxk_sso_queue_attribute_get,
>         .queue_attr_set = cnxk_sso_queue_attribute_set,
>
>         .port_def_conf = cnxk_sso_port_def_conf,
> diff --git a/drivers/event/cnxk/cnxk_eventdev.c b/drivers/event/cnxk/cnxk_eventdev.c
> index 97dcf7b66e..45c53ffb4e 100644
> --- a/drivers/event/cnxk/cnxk_eventdev.c
> +++ b/drivers/event/cnxk/cnxk_eventdev.c
> @@ -321,6 +321,8 @@ cnxk_sso_queue_def_conf(struct rte_eventdev *event_dev, uint8_t queue_id,
>         queue_conf->nb_atomic_order_sequences = (1ULL << 20);
>         queue_conf->event_queue_cfg = RTE_EVENT_QUEUE_CFG_ALL_TYPES;
>         queue_conf->priority = RTE_EVENT_DEV_PRIORITY_NORMAL;
> +       queue_conf->weight = RTE_EVENT_QUEUE_WEIGHT_LOWEST;
> +       queue_conf->affinity = RTE_EVENT_QUEUE_AFFINITY_HIGHEST;
>  }
>
>  int
> @@ -330,18 +332,12 @@ cnxk_sso_queue_setup(struct rte_eventdev *event_dev, uint8_t queue_id,
>         struct cnxk_sso_evdev *dev = cnxk_sso_pmd_priv(event_dev);
>         uint8_t priority, weight, affinity;
>
> -       /* Default weight and affinity */
> -       dev->mlt_prio[queue_id].weight = RTE_EVENT_QUEUE_WEIGHT_LOWEST;
> -       dev->mlt_prio[queue_id].affinity = RTE_EVENT_QUEUE_AFFINITY_HIGHEST;
> -
>         priority = CNXK_QOS_NORMALIZE(queue_conf->priority, 0,
>                                       RTE_EVENT_DEV_PRIORITY_LOWEST,
>                                       CNXK_SSO_PRIORITY_CNT);
> -       weight = CNXK_QOS_NORMALIZE(
> -               dev->mlt_prio[queue_id].weight, CNXK_SSO_WEIGHT_MIN,
> -               RTE_EVENT_QUEUE_WEIGHT_HIGHEST, CNXK_SSO_WEIGHT_CNT);
> -       affinity = CNXK_QOS_NORMALIZE(dev->mlt_prio[queue_id].affinity, 0,
> -                                     RTE_EVENT_QUEUE_AFFINITY_HIGHEST,
> +       weight = CNXK_QOS_NORMALIZE(queue_conf->weight, CNXK_SSO_WEIGHT_MIN,
> +                                   RTE_EVENT_QUEUE_WEIGHT_HIGHEST, CNXK_SSO_WEIGHT_CNT);
> +       affinity = CNXK_QOS_NORMALIZE(queue_conf->affinity, 0, RTE_EVENT_QUEUE_AFFINITY_HIGHEST,
>                                       CNXK_SSO_AFFINITY_CNT);
>
>         plt_sso_dbg("Queue=%u prio=%u weight=%u affinity=%u", queue_id,
> @@ -358,22 +354,6 @@ cnxk_sso_queue_release(struct rte_eventdev *event_dev, uint8_t queue_id)
>         RTE_SET_USED(queue_id);
>  }
>
> -int
> -cnxk_sso_queue_attribute_get(struct rte_eventdev *event_dev, uint8_t queue_id,
> -                            uint32_t attr_id, uint32_t *attr_value)
> -{
> -       struct cnxk_sso_evdev *dev = cnxk_sso_pmd_priv(event_dev);
> -
> -       if (attr_id == RTE_EVENT_QUEUE_ATTR_WEIGHT)
> -               *attr_value = dev->mlt_prio[queue_id].weight;
> -       else if (attr_id == RTE_EVENT_QUEUE_ATTR_AFFINITY)
> -               *attr_value = dev->mlt_prio[queue_id].affinity;
> -       else
> -               return -EINVAL;
> -
> -       return 0;
> -}
> -
>  int
>  cnxk_sso_queue_attribute_set(struct rte_eventdev *event_dev, uint8_t queue_id,
>                              uint32_t attr_id, uint64_t attr_value)
> @@ -389,10 +369,10 @@ cnxk_sso_queue_attribute_set(struct rte_eventdev *event_dev, uint8_t queue_id,
>                 conf->priority = attr_value;
>                 break;
>         case RTE_EVENT_QUEUE_ATTR_WEIGHT:
> -               dev->mlt_prio[queue_id].weight = attr_value;
> +               conf->weight = attr_value;
>                 break;
>         case RTE_EVENT_QUEUE_ATTR_AFFINITY:
> -               dev->mlt_prio[queue_id].affinity = attr_value;
> +               conf->affinity = attr_value;
>                 break;
>         case RTE_EVENT_QUEUE_ATTR_NB_ATOMIC_FLOWS:
>         case RTE_EVENT_QUEUE_ATTR_NB_ATOMIC_ORDER_SEQUENCES:
> @@ -409,11 +389,9 @@ cnxk_sso_queue_attribute_set(struct rte_eventdev *event_dev, uint8_t queue_id,
>         priority = CNXK_QOS_NORMALIZE(conf->priority, 0,
>                                       RTE_EVENT_DEV_PRIORITY_LOWEST,
>                                       CNXK_SSO_PRIORITY_CNT);
> -       weight = CNXK_QOS_NORMALIZE(
> -               dev->mlt_prio[queue_id].weight, CNXK_SSO_WEIGHT_MIN,
> -               RTE_EVENT_QUEUE_WEIGHT_HIGHEST, CNXK_SSO_WEIGHT_CNT);
> -       affinity = CNXK_QOS_NORMALIZE(dev->mlt_prio[queue_id].affinity, 0,
> -                                     RTE_EVENT_QUEUE_AFFINITY_HIGHEST,
> +       weight = CNXK_QOS_NORMALIZE(conf->weight, CNXK_SSO_WEIGHT_MIN,
> +                                   RTE_EVENT_QUEUE_WEIGHT_HIGHEST, CNXK_SSO_WEIGHT_CNT);
> +       affinity = CNXK_QOS_NORMALIZE(conf->affinity, 0, RTE_EVENT_QUEUE_AFFINITY_HIGHEST,
>                                       CNXK_SSO_AFFINITY_CNT);
>
>         return roc_sso_hwgrp_set_priority(&dev->sso, queue_id, weight, affinity,
> diff --git a/drivers/event/cnxk/cnxk_eventdev.h b/drivers/event/cnxk/cnxk_eventdev.h
> index bfd0c5627e..d78fb4ea2f 100644
> --- a/drivers/event/cnxk/cnxk_eventdev.h
> +++ b/drivers/event/cnxk/cnxk_eventdev.h
> @@ -88,11 +88,6 @@ struct cnxk_sso_qos {
>         uint16_t iaq_prcnt;
>  };
>
> -struct cnxk_sso_mlt_prio {
> -       uint8_t weight;
> -       uint8_t affinity;
> -};
> -
>  struct cnxk_sso_evdev {
>         struct roc_sso sso;
>         uint8_t max_event_queues;
> @@ -123,7 +118,6 @@ struct cnxk_sso_evdev {
>         uint64_t *timer_adptr_sz;
>         uint16_t vec_pool_cnt;
>         uint64_t *vec_pools;
> -       struct cnxk_sso_mlt_prio mlt_prio[RTE_EVENT_MAX_QUEUES_PER_DEV];
>         /* Dev args */
>         uint32_t xae_cnt;
>         uint8_t qos_queue_cnt;
> @@ -250,9 +244,6 @@ void cnxk_sso_queue_def_conf(struct rte_eventdev *event_dev, uint8_t queue_id,
>  int cnxk_sso_queue_setup(struct rte_eventdev *event_dev, uint8_t queue_id,
>                          const struct rte_event_queue_conf *queue_conf);
>  void cnxk_sso_queue_release(struct rte_eventdev *event_dev, uint8_t queue_id);
> -int cnxk_sso_queue_attribute_get(struct rte_eventdev *event_dev,
> -                                uint8_t queue_id, uint32_t attr_id,
> -                                uint32_t *attr_value);
>  int cnxk_sso_queue_attribute_set(struct rte_eventdev *event_dev,
>                                  uint8_t queue_id, uint32_t attr_id,
>                                  uint64_t attr_value);
> diff --git a/lib/eventdev/eventdev_pmd.h b/lib/eventdev/eventdev_pmd.h
> index 69402668d8..8879e43feb 100644
> --- a/lib/eventdev/eventdev_pmd.h
> +++ b/lib/eventdev/eventdev_pmd.h
> @@ -341,26 +341,6 @@ typedef int (*eventdev_queue_setup_t)(struct rte_eventdev *dev,
>  typedef void (*eventdev_queue_release_t)(struct rte_eventdev *dev,
>                 uint8_t queue_id);
>
> -/**
> - * Get an event queue attribute at runtime.
> - *
> - * @param dev
> - *   Event device pointer
> - * @param queue_id
> - *   Event queue index
> - * @param attr_id
> - *   Event queue attribute id
> - * @param[out] attr_value
> - *   Event queue attribute value
> - *
> - * @return
> - *  - 0: Success.
> - *  - <0: Error code on failure.
> - */
> -typedef int (*eventdev_queue_attr_get_t)(struct rte_eventdev *dev,
> -                                        uint8_t queue_id, uint32_t attr_id,
> -                                        uint32_t *attr_value);
> -
>  /**
>   * Set an event queue attribute at runtime.
>   *
> @@ -1268,8 +1248,6 @@ struct eventdev_ops {
>         /**< Set up an event queue. */
>         eventdev_queue_release_t queue_release;
>         /**< Release an event queue. */
> -       eventdev_queue_attr_get_t queue_attr_get;
> -       /**< Get an event queue attribute. */
>         eventdev_queue_attr_set_t queue_attr_set;
>         /**< Set an event queue attribute. */
>
> diff --git a/lib/eventdev/rte_eventdev.c b/lib/eventdev/rte_eventdev.c
> index 1dc4f966be..b96185b25d 100644
> --- a/lib/eventdev/rte_eventdev.c
> +++ b/lib/eventdev/rte_eventdev.c
> @@ -859,15 +859,13 @@ rte_event_queue_attr_get(uint8_t dev_id, uint8_t queue_id, uint32_t attr_id,
>                 break;
>         case RTE_EVENT_QUEUE_ATTR_WEIGHT:
>                 *attr_value = RTE_EVENT_QUEUE_WEIGHT_LOWEST;
> -               if (dev->dev_ops->queue_attr_get)
> -                       return (*dev->dev_ops->queue_attr_get)(
> -                               dev, queue_id, attr_id, attr_value);
> +               if (dev->data->event_dev_cap & RTE_EVENT_DEV_CAP_QUEUE_QOS)
> +                       *attr_value = conf->weight;
>                 break;
>         case RTE_EVENT_QUEUE_ATTR_AFFINITY:
>                 *attr_value = RTE_EVENT_QUEUE_AFFINITY_LOWEST;
> -               if (dev->dev_ops->queue_attr_get)
> -                       return (*dev->dev_ops->queue_attr_get)(
> -                               dev, queue_id, attr_id, attr_value);
> +               if (dev->data->event_dev_cap & RTE_EVENT_DEV_CAP_QUEUE_QOS)
> +                       *attr_value = conf->affinity;
>                 break;
>         default:
>                 return -EINVAL;
> diff --git a/lib/eventdev/rte_eventdev.h b/lib/eventdev/rte_eventdev.h
> index 6a6f6ea4c1..f1908b82b2 100644
> --- a/lib/eventdev/rte_eventdev.h
> +++ b/lib/eventdev/rte_eventdev.h
> @@ -640,6 +640,22 @@ struct rte_event_queue_conf {
>          * event device supported priority value.
>          * Valid when the device has RTE_EVENT_DEV_CAP_QUEUE_QOS capability
>          */
> +       uint8_t weight;
> +       /**< Weight of the event queue relative to other event queues.
> +        * The requested weight should be in the range of
> +        * [RTE_EVENT_DEV_WEIGHT_HIGHEST, RTE_EVENT_DEV_WEIGHT_LOWEST].
> +        * The implementation shall normalize the requested weight to event
> +        * device supported weight value.
> +        * Valid when the device has RTE_EVENT_DEV_CAP_QUEUE_QOS capability.
> +        */
> +       uint8_t affinity;
> +       /**< Affinity of the event queue relative to other event queues.
> +        * The requested affinity should be in the range of
> +        * [RTE_EVENT_DEV_AFFINITY_HIGHEST, RTE_EVENT_DEV_AFFINITY_LOWEST].
> +        * The implementation shall normalize the requested affinity to event
> +        * device supported affinity value.
> +        * Valid when the device has RTE_EVENT_DEV_CAP_QUEUE_QOS capability.
> +        */
>  };
>
>  /**
> --
> 2.25.1
>

  reply	other threads:[~2022-09-14 12:35 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-10  7:36 Shijith Thotton
2022-09-14 12:34 ` Jerin Jacob [this message]
2022-09-15 16:53   ` Jerin Jacob

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=CALBAE1NTo11WvecJR0c-NXMCdrmKwz4zfXZ0bkCByD92EViPGg@mail.gmail.com \
    --to=jerinjacobk@gmail.com \
    --cc=abhinandan.gujjar@intel.com \
    --cc=dev@dpdk.org \
    --cc=erik.g.carrillo@intel.com \
    --cc=harry.van.haaren@intel.com \
    --cc=hemant.agrawal@nxp.com \
    --cc=jay.jayatheerthan@intel.com \
    --cc=jerinj@marvell.com \
    --cc=liangma@liangbit.com \
    --cc=mattias.ronnblom@ericsson.com \
    --cc=mdr@ashroe.eu \
    --cc=nipun.gupta@nxp.com \
    --cc=pbhagavatula@marvell.com \
    --cc=peter.mccarthy@intel.com \
    --cc=sthotton@marvell.com \
    --cc=timothy.mcdaniel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).