[v3,1/8] ethdev: introduce shared Rx queue

  In current DPDK framework, each RX queue is pre-loaded with mbufs for
incoming packets. When number of representors scale out in a switch
domain, the memory consumption became significant. Most important,
polling all ports leads to high cache miss, high latency and low
throughput.

This patch introduces shared RX queue. Ports with same configuration in
a switch domain could share RX queue set by specifying sharing group.
Polling any queue using same shared RX queue receives packets from all
member ports. Source port is identified by mbuf->port.

Port queue number in a shared group should be identical. Queue index is
1:1 mapped in shared group.

Share RX queue must be polled on single thread or core.

Multiple groups is supported by group ID.

Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Cc: Jerin Jacob <jerinjacobk@gmail.com>
---
Rx queue object could be used as shared Rx queue object, it's important
to clear all queue control callback api that using queue object:
  https://mails.dpdk.org/archives/dev/2021-July/215574.html
---
 doc/guides/nics/features.rst                    | 11 +++++++++++
 doc/guides/nics/features/default.ini            |  1 +
 doc/guides/prog_guide/switch_representation.rst | 10 ++++++++++
 lib/ethdev/rte_ethdev.c                         |  1 +
 lib/ethdev/rte_ethdev.h                         |  7 +++++++
 5 files changed, 30 insertions(+)
  

On Fri, Sep 17, 2021 at 1:02 AM Xueming Li <xuemingl@nvidia.com> wrote:
>
> In current DPDK framework, each RX queue is pre-loaded with mbufs for
> incoming packets. When number of representors scale out in a switch
> domain, the memory consumption became significant. Most important,
> polling all ports leads to high cache miss, high latency and low
> throughput.
>
> This patch introduces shared RX queue. Ports with same configuration in
> a switch domain could share RX queue set by specifying sharing group.
> Polling any queue using same shared RX queue receives packets from all
> member ports. Source port is identified by mbuf->port.
>
> Port queue number in a shared group should be identical. Queue index is
> 1:1 mapped in shared group.
>
> Share RX queue must be polled on single thread or core.
>
> Multiple groups is supported by group ID.
Can you clarify this a little more?

Apologies if this was already covered:
* Can't we do this for Tx also?

Couple of nits inline. Thanks

>
> Signed-off-by: Xueming Li <xuemingl@nvidia.com>
> Cc: Jerin Jacob <jerinjacobk@gmail.com>
> ---
> Rx queue object could be used as shared Rx queue object, it's important
> to clear all queue control callback api that using queue object:
>   https://mails.dpdk.org/archives/dev/2021-July/215574.html
> ---
>  doc/guides/nics/features.rst                    | 11 +++++++++++
>  doc/guides/nics/features/default.ini            |  1 +
>  doc/guides/prog_guide/switch_representation.rst | 10 ++++++++++
>  lib/ethdev/rte_ethdev.c                         |  1 +
>  lib/ethdev/rte_ethdev.h                         |  7 +++++++
>  5 files changed, 30 insertions(+)
>
> diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
> index a96e12d155..2e2a9b1554 100644
> --- a/doc/guides/nics/features.rst
> +++ b/doc/guides/nics/features.rst
> @@ -624,6 +624,17 @@ Supports inner packet L4 checksum.
>    ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_OUTER_UDP_CKSUM``.
>
>
> +.. _nic_features_shared_rx_queue:
> +
> +Shared Rx queue
> +---------------
> +
> +Supports shared Rx queue for ports in same switch domain.
> +
> +* **[uses]     rte_eth_rxconf,rte_eth_rxmode**: ``offloads:RTE_ETH_RX_OFFLOAD_SHARED_RXQ``.
> +* **[provides] mbuf**: ``mbuf.port``.
> +
> +
>  .. _nic_features_packet_type_parsing:
>
>  Packet type parsing
> diff --git a/doc/guides/nics/features/default.ini b/doc/guides/nics/features/default.ini
> index 754184ddd4..ebeb4c1851 100644
> --- a/doc/guides/nics/features/default.ini
> +++ b/doc/guides/nics/features/default.ini
> @@ -19,6 +19,7 @@ Free Tx mbuf on demand =
>  Queue start/stop     =
>  Runtime Rx queue setup =
>  Runtime Tx queue setup =
> +Shared Rx queue      =
>  Burst mode info      =
>  Power mgmt address monitor =
>  MTU update           =
> diff --git a/doc/guides/prog_guide/switch_representation.rst b/doc/guides/prog_guide/switch_representation.rst
> index ff6aa91c80..45bf5a3a10 100644
> --- a/doc/guides/prog_guide/switch_representation.rst
> +++ b/doc/guides/prog_guide/switch_representation.rst
> @@ -123,6 +123,16 @@ thought as a software "patch panel" front-end for applications.
>  .. [1] `Ethernet switch device driver model (switchdev)
>         <https://www.kernel.org/doc/Documentation/networking/switchdev.txt>`_
>
> +- Memory usage of representors is huge when number of representor grows,
> +  because PMD always allocate mbuf for each descriptor of Rx queue.
> +  Polling the large number of ports brings more CPU load, cache miss and
> +  latency. Shared Rx queue can be used to share Rx queue between PF and
> +  representors in same switch domain. ``RTE_ETH_RX_OFFLOAD_SHARED_RXQ``

"in the same switch"

> +  is present in Rx offloading capability of device info. Setting the
> +  offloading flag in device Rx mode or Rx queue configuration to enable
> +  shared Rx queue. Polling any member port of shared Rx queue can return
"of the shared Rx queue.."

> +  packets of all ports in group, port ID is saved in ``mbuf.port``.

"ports in the group, "

> +
>  Basic SR-IOV
>  ------------
>
> diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
> index a7c090ce79..b3a58d5e65 100644
> --- a/lib/ethdev/rte_ethdev.c
> +++ b/lib/ethdev/rte_ethdev.c
> @@ -127,6 +127,7 @@ static const struct {
>         RTE_RX_OFFLOAD_BIT2STR(OUTER_UDP_CKSUM),
>         RTE_RX_OFFLOAD_BIT2STR(RSS_HASH),
>         RTE_ETH_RX_OFFLOAD_BIT2STR(BUFFER_SPLIT),
> +       RTE_ETH_RX_OFFLOAD_BIT2STR(SHARED_RXQ),
>  };
>
>  #undef RTE_RX_OFFLOAD_BIT2STR
> diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
> index d2b27c351f..a578c9db9d 100644
> --- a/lib/ethdev/rte_ethdev.h
> +++ b/lib/ethdev/rte_ethdev.h
> @@ -1047,6 +1047,7 @@ struct rte_eth_rxconf {
>         uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
>         uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
>         uint16_t rx_nseg; /**< Number of descriptions in rx_seg array. */
> +       uint32_t shared_group; /**< Shared port group index in switch domain. */
>         /**
>          * Per-queue Rx offloads to be set using DEV_RX_OFFLOAD_* flags.
>          * Only offloads set on rx_queue_offload_capa or rx_offload_capa
> @@ -1373,6 +1374,12 @@ struct rte_eth_conf {
>  #define DEV_RX_OFFLOAD_OUTER_UDP_CKSUM  0x00040000
>  #define DEV_RX_OFFLOAD_RSS_HASH                0x00080000
>  #define RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT 0x00100000
> +/**
> + * Rx queue is shared among ports in same switch domain to save memory,
> + * avoid polling each port. Any port in group can be used to receive packets.

"Any port in the group can"


> + * Real source port number saved in mbuf->port field.
> + */
> +#define RTE_ETH_RX_OFFLOAD_SHARED_RXQ   0x00200000
>
>  #define DEV_RX_OFFLOAD_CHECKSUM (DEV_RX_OFFLOAD_IPV4_CKSUM | \
>                                  DEV_RX_OFFLOAD_UDP_CKSUM | \
> --
> 2.33.0
>
  

On Mon, 2021-09-27 at 16:53 -0700, Ajit Khaparde wrote:
> On Fri, Sep 17, 2021 at 1:02 AM Xueming Li <xuemingl@nvidia.com> wrote:
> > 
> > In current DPDK framework, each RX queue is pre-loaded with mbufs for
> > incoming packets. When number of representors scale out in a switch
> > domain, the memory consumption became significant. Most important,
> > polling all ports leads to high cache miss, high latency and low
> > throughput.
> > 
> > This patch introduces shared RX queue. Ports with same configuration in
> > a switch domain could share RX queue set by specifying sharing group.
> > Polling any queue using same shared RX queue receives packets from all
> > member ports. Source port is identified by mbuf->port.
> > 
> > Port queue number in a shared group should be identical. Queue index is
> > 1:1 mapped in shared group.
> > 
> > Share RX queue must be polled on single thread or core.
> > 
> > Multiple groups is supported by group ID.
> Can you clarify this a little more?

Thanks for the review!

By using group ID, user can specify for example:
 group 0: port 0-3, 2 queues per port, poll on core 0 and 1
 group 1: port 4-127, 1 queue per port, poll on core 1.
Normally used for QoS and load balance.

> 
> Apologies if this was already covered:
> * Can't we do this for Tx also?

Rx queue pre-fill mbufs for each queue, which consuming huge mbufs by
default, most of them are less active, saving memory is the primary
motatation for this feature.
Tx queue doesn't consume any mbuf by default until starting tx, no
strong reason so far. 

> 
> Couple of nits inline. Thanks
> 
> > 
> > Signed-off-by: Xueming Li <xuemingl@nvidia.com>
> > Cc: Jerin Jacob <jerinjacobk@gmail.com>
> > ---
> > Rx queue object could be used as shared Rx queue object, it's important
> > to clear all queue control callback api that using queue object:
> >   https://mails.dpdk.org/archives/dev/2021-July/215574.html
> > ---
> >  doc/guides/nics/features.rst                    | 11 +++++++++++
> >  doc/guides/nics/features/default.ini            |  1 +
> >  doc/guides/prog_guide/switch_representation.rst | 10 ++++++++++
> >  lib/ethdev/rte_ethdev.c                         |  1 +
> >  lib/ethdev/rte_ethdev.h                         |  7 +++++++
> >  5 files changed, 30 insertions(+)
> > 
> > diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
> > index a96e12d155..2e2a9b1554 100644
> > --- a/doc/guides/nics/features.rst
> > +++ b/doc/guides/nics/features.rst
> > @@ -624,6 +624,17 @@ Supports inner packet L4 checksum.
> >    ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_OUTER_UDP_CKSUM``.
> > 
> > 
> > +.. _nic_features_shared_rx_queue:
> > +
> > +Shared Rx queue
> > +---------------
> > +
> > +Supports shared Rx queue for ports in same switch domain.
> > +
> > +* **[uses]     rte_eth_rxconf,rte_eth_rxmode**: ``offloads:RTE_ETH_RX_OFFLOAD_SHARED_RXQ``.
> > +* **[provides] mbuf**: ``mbuf.port``.
> > +
> > +
> >  .. _nic_features_packet_type_parsing:
> > 
> >  Packet type parsing
> > diff --git a/doc/guides/nics/features/default.ini b/doc/guides/nics/features/default.ini
> > index 754184ddd4..ebeb4c1851 100644
> > --- a/doc/guides/nics/features/default.ini
> > +++ b/doc/guides/nics/features/default.ini
> > @@ -19,6 +19,7 @@ Free Tx mbuf on demand =
> >  Queue start/stop     =
> >  Runtime Rx queue setup =
> >  Runtime Tx queue setup =
> > +Shared Rx queue      =
> >  Burst mode info      =
> >  Power mgmt address monitor =
> >  MTU update           =
> > diff --git a/doc/guides/prog_guide/switch_representation.rst b/doc/guides/prog_guide/switch_representation.rst
> > index ff6aa91c80..45bf5a3a10 100644
> > --- a/doc/guides/prog_guide/switch_representation.rst
> > +++ b/doc/guides/prog_guide/switch_representation.rst
> > @@ -123,6 +123,16 @@ thought as a software "patch panel" front-end for applications.
> >  .. [1] `Ethernet switch device driver model (switchdev)
> >         <https://www.kernel.org/doc/Documentation/networking/switchdev.txt>`_
> > 
> > +- Memory usage of representors is huge when number of representor grows,
> > +  because PMD always allocate mbuf for each descriptor of Rx queue.
> > +  Polling the large number of ports brings more CPU load, cache miss and
> > +  latency. Shared Rx queue can be used to share Rx queue between PF and
> > +  representors in same switch domain. ``RTE_ETH_RX_OFFLOAD_SHARED_RXQ``
> 
> "in the same switch"
> 
> > +  is present in Rx offloading capability of device info. Setting the
> > +  offloading flag in device Rx mode or Rx queue configuration to enable
> > +  shared Rx queue. Polling any member port of shared Rx queue can return
> "of the shared Rx queue.."
> 
> > +  packets of all ports in group, port ID is saved in ``mbuf.port``.
> 
> "ports in the group, "
> 
> > +
> >  Basic SR-IOV
> >  ------------
> > 
> > diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
> > index a7c090ce79..b3a58d5e65 100644
> > --- a/lib/ethdev/rte_ethdev.c
> > +++ b/lib/ethdev/rte_ethdev.c
> > @@ -127,6 +127,7 @@ static const struct {
> >         RTE_RX_OFFLOAD_BIT2STR(OUTER_UDP_CKSUM),
> >         RTE_RX_OFFLOAD_BIT2STR(RSS_HASH),
> >         RTE_ETH_RX_OFFLOAD_BIT2STR(BUFFER_SPLIT),
> > +       RTE_ETH_RX_OFFLOAD_BIT2STR(SHARED_RXQ),
> >  };
> > 
> >  #undef RTE_RX_OFFLOAD_BIT2STR
> > diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
> > index d2b27c351f..a578c9db9d 100644
> > --- a/lib/ethdev/rte_ethdev.h
> > +++ b/lib/ethdev/rte_ethdev.h
> > @@ -1047,6 +1047,7 @@ struct rte_eth_rxconf {
> >         uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
> >         uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
> >         uint16_t rx_nseg; /**< Number of descriptions in rx_seg array. */
> > +       uint32_t shared_group; /**< Shared port group index in switch domain. */
> >         /**
> >          * Per-queue Rx offloads to be set using DEV_RX_OFFLOAD_* flags.
> >          * Only offloads set on rx_queue_offload_capa or rx_offload_capa
> > @@ -1373,6 +1374,12 @@ struct rte_eth_conf {
> >  #define DEV_RX_OFFLOAD_OUTER_UDP_CKSUM  0x00040000
> >  #define DEV_RX_OFFLOAD_RSS_HASH                0x00080000
> >  #define RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT 0x00100000
> > +/**
> > + * Rx queue is shared among ports in same switch domain to save memory,
> > + * avoid polling each port. Any port in group can be used to receive packets.
> 
> "Any port in the group can"
> 
> 
> > + * Real source port number saved in mbuf->port field.
> > + */
> > +#define RTE_ETH_RX_OFFLOAD_SHARED_RXQ   0x00200000
> > 
> >  #define DEV_RX_OFFLOAD_CHECKSUM (DEV_RX_OFFLOAD_IPV4_CKSUM | \
> >                                  DEV_RX_OFFLOAD_UDP_CKSUM | \
> > --
> > 2.33.0
> >