Re: [dpdk-dev] [PATCH v6 1/2] mbuf: introduce accurate packet Tx scheduling

[Ferruh Yigit <ferruh.yigit@intel.com>]

On 7/9/2020 1:36 PM, Viacheslav Ovsiienko wrote:
> There is the requirement on some networks for precise traffic timing
> management. The ability to send (and, generally speaking, receive)
> the packets at the very precisely specified moment of time provides
> the opportunity to support the connections with Time Division
> Multiplexing using the contemporary general purpose NIC without involving
> an auxiliary hardware. For example, the supporting of O-RAN Fronthaul
> interface is one of the promising features for potentially usage of the
> precise time management for the egress packets.

Is this a HW support, or is the scheduling planned to be done in the driver?

> 
> The main objective of this RFC is to specify the way how applications

It is no more RFC.

> can provide the moment of time at what the packet transmission must be
> started and to describe in preliminary the supporting this feature from
> mlx5 PMD side.

I was about the ask this, will there be a PMD counterpart implementation of the
feature? It would be better to have it as part of this set.
What is the plan for the PMD implementation?

> 
> The new dynamic timestamp field is proposed, it provides some timing
> information, the units and time references (initial phase) are not
> explicitly defined but are maintained always the same for a given port.
> Some devices allow to query rte_eth_read_clock() that will return
> the current device timestamp. The dynamic timestamp flag tells whether
> the field contains actual timestamp value. For the packets being sent
> this value can be used by PMD to schedule packet sending.
> 
> The device clock is opaque entity, the units and frequency are
> vendor specific and might depend on hardware capabilities and
> configurations. If might (or not) be synchronized with real time
> via PTP, might (or not) be synchronous with CPU clock (for example
> if NIC and CPU share the same clock source there might be no
> any drift between the NIC and CPU clocks), etc.
> 
> After PKT_RX_TIMESTAMP flag and fixed timestamp field deprecation
> and obsoleting, these dynamic flag and field will be used to manage
> the timestamps on receiving datapath as well. Having the dedicated
> flags for Rx/Tx timestamps allows applications not to perform explicit
> flags reset on forwarding and not to promote received timestamps
> to the transmitting datapath by default. The static PKT_RX_TIMESTAMP
> is considered as candidate to become the dynamic flag.

Is there a deprecation notice for 'PKT_RX_TIMESTAMP'? Is this decided?

> 
> When PMD sees the "rte_dynfield_timestamp" set on the packet being sent
> it tries to synchronize the time of packet appearing on the wire with
> the specified packet timestamp. If the specified one is in the past it
> should be ignored, if one is in the distant future it should be capped
> with some reasonable value (in range of seconds). These specific cases
> ("too late" and "distant future") can be optionally reported via
> device xstats to assist applications to detect the time-related
> problems.
> 
> There is no any packet reordering according timestamps is supposed,
> neither within packet burst, nor between packets, it is an entirely
> application responsibility to generate packets and its timestamps
> in desired order. The timestamps can be put only in the first packet
> in the burst providing the entire burst scheduling.
> 
> PMD reports the ability to synchronize packet sending on timestamp
> with new offload flag:
> 
> This is palliative and is going to be replaced with new eth_dev API
> about reporting/managing the supported dynamic flags and its related
> features. This API would break ABI compatibility and can't be introduced
> at the moment, so is postponed to 20.11.

Good to hear that there will be a generic API to get supported dynamic flags. I
was concerned about adding 'DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP' flag, since not
sure if there will be any other PMD that will want to use it.
The trouble is it is hard to remove a public macro after it is introduced, in
this release I think only single PMD (mlx) will support this feature, and in
next release the plan is to remove the macro. In this case what do you think to
not introduce the flag at all?

> 
> For testing purposes it is proposed to update testpmd "txonly"
> forwarding mode routine. With this update testpmd application generates
> the packets and sets the dynamic timestamps according to specified time
> pattern if it sees the "rte_dynfield_timestamp" is registered.
> 
> The new testpmd command is proposed to configure sending pattern:
> 
> set tx_times <burst_gap>,<intra_gap>
> 
> <intra_gap> - the delay between the packets within the burst
>               specified in the device clock units. The number
>               of packets in the burst is defined by txburst parameter
> 
> <burst_gap> - the delay between the bursts in the device clock units
> 
> As the result the bursts of packet will be transmitted with specific
> delays between the packets within the burst and specific delay between
> the bursts. The rte_eth_get_clock is supposed to be engaged to get the

'rte_eth_read_clock()'?

> current device clock value and provide the reference for the timestamps.
> 
> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
> Acked-by: Olivier Matz <olivier.matz@6wind.com>
> 
> ---
>   v1->v4:
>      - dedicated dynamic Tx timestamp flag instead of shared with Rx
>   v4->v5:
>      - elaborated commit message
>      - more words about device clocks added,
>      - note about dedicated Rx/Tx timestamp flags added
>   v5->v6:
>      - release notes are updated
> ---
>  doc/guides/rel_notes/release_20_08.rst |  6 ++++++
>  lib/librte_ethdev/rte_ethdev.c         |  1 +
>  lib/librte_ethdev/rte_ethdev.h         |  4 ++++
>  lib/librte_mbuf/rte_mbuf_dyn.h         | 31 +++++++++++++++++++++++++++++++
>  4 files changed, 42 insertions(+)
> 
> diff --git a/doc/guides/rel_notes/release_20_08.rst b/doc/guides/rel_notes/release_20_08.rst
> index 988474c..5527bab 100644
> --- a/doc/guides/rel_notes/release_20_08.rst
> +++ b/doc/guides/rel_notes/release_20_08.rst
> @@ -200,6 +200,12 @@ New Features
>    See the :doc:`../sample_app_ug/l2_forward_real_virtual` for more
>    details of this parameter usage.
>  
> +* **Introduced send packet scheduling on the timestamps.**
> +
> +  Added the new mbuf dynamic field and flag to provide timestamp on what packet
> +  transmitting can be synchronized. The device Tx offload flag is added to
> +  indicate the PMD supports send scheduling.
> +

This is a core library change, can go up in the section, please check the
section comment for the ordering details.

>  
>  Removed Items
>  -------------
> diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
> index 7022bd7..c48ca2a 100644
> --- a/lib/librte_ethdev/rte_ethdev.c
> +++ b/lib/librte_ethdev/rte_ethdev.c
> @@ -160,6 +160,7 @@ struct rte_eth_xstats_name_off {
>  	RTE_TX_OFFLOAD_BIT2STR(UDP_TNL_TSO),
>  	RTE_TX_OFFLOAD_BIT2STR(IP_TNL_TSO),
>  	RTE_TX_OFFLOAD_BIT2STR(OUTER_UDP_CKSUM),
> +	RTE_TX_OFFLOAD_BIT2STR(SEND_ON_TIMESTAMP),
>  };
>  
>  #undef RTE_TX_OFFLOAD_BIT2STR
> diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
> index 631b146..97313a0 100644
> --- a/lib/librte_ethdev/rte_ethdev.h
> +++ b/lib/librte_ethdev/rte_ethdev.h
> @@ -1178,6 +1178,10 @@ struct rte_eth_conf {
>  /** Device supports outer UDP checksum */
>  #define DEV_TX_OFFLOAD_OUTER_UDP_CKSUM  0x00100000
>  
> +/** Device supports send on timestamp */
> +#define DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP 0x00200000

Please cc the ethdev maintainers.

As mentioned above my concern is if this is generic enough or are we adding a
flag to a specific PMD? And since commit log says this is temporary solution for
just this release, I repeat my question if we can remove the flag completely?