Re: [dpdk-dev] [PATCH] doc: announce changes to ethdev rxconf structure

[Slava Ovsiienko <viacheslavo@mellanox.com>]

Hi, Andrew

Thanks for the comment, please, see below.

> -----Original Message-----
> From: Andrew Rybchenko <arybchenko@solarflare.com>
> Sent: Monday, August 3, 2020 17:31
> To: Slava Ovsiienko <viacheslavo@mellanox.com>; dev@dpdk.org
> Cc: Matan Azrad <matan@mellanox.com>; Raslan Darawsheh
> <rasland@mellanox.com>; Thomas Monjalon <thomas@monjalon.net>;
> ferruh.yigit@intel.com; jerinjacobk@gmail.com;
> stephen@networkplumber.org; ajit.khaparde@broadcom.com;
> maxime.coquelin@redhat.com; olivier.matz@6wind.com;
> david.marchand@redhat.com
> Subject: Re: ***Spam*** [PATCH] doc: announce changes to ethdev rxconf
> structure
> 
> On 8/3/20 1:58 PM, Viacheslav Ovsiienko wrote:
> > The DPDK datapath in the transmit direction is very flexible.
> > The applications can build multisegment packets and manages almost all
> > data aspects - the memory pools where segments are allocated from, the
> > segment lengths, the memory attributes like external, registered, etc.
> >
> > In the receiving direction, the datapath is much less flexible, the
> > applications can only specify the memory pool to configure the
> > receiving queue and nothing more. In order to extend the receiving
> > datapath capabilities it is proposed to add the new fields into
> > rte_eth_rxconf structure:
> >
> > struct rte_eth_rxconf {
> >     ...
> >     uint16_t rx_split_num; /* number of segments to split */
> >     uint16_t *rx_split_len; /* array of segment lengthes */
> >     struct rte_mempool **mp; /* array of segment memory pools */
> >     ...
> > };
> >
> > The non-zero value of rx_split_num field configures the receiving
> > queue to split ingress packets into multiple segments to the mbufs
> > allocated from various memory pools according to the specified
> > lengths. The zero value of rx_split_num field provides the backward
> > compatibility and queue should be configured in a regular way (with
> > single/multiple mbufs of the same data buffer length allocated from
> > the single memory pool).
> 
> From the above description it is not 100% clear how it will coexist with:
>  - existing mb_pool argument of the rte_eth_rx_queue_setup()
>  - DEV_RX_OFFLOAD_SCATTER

DEV_RX_OFFLOAD_SCATTER flag is required to be reported and configured
for the new feature to indicate the application is prepared for the 
multisegment packets.

But SCATTER it just tells that ingress packet length can exceed
the mbuf data buffer length and the chain of mbufs must be built to store
the entire packet. But there is the limitation - all mbufs are allocated
 from the same memory pool, and all data buffers have the same length.
The new feature provides an opportunity to allocated mbufs from the desired
pools and specifies the length of each buffer/part.

>  - DEV_RX_OFFLOAD_HEADER_SPLIT
The new feature (let's name it as "BUFFER_SPLIT") might be supported
in conjunction with HEADER_SPLIT (say, split the rest of the data after the header)
or rejected if HEADER_SPLIT is configured on the port, depending on PMD
implementation (return ENOTSUP if both features are requested on the same port).

> How will application know that the feature is supported? Limitations?
It is subject for further discussion, I see two options:
 - introduce the DEV_RX_OFFLOAD_BUFFER_SPLIT flag
- return ENOTSUP/EINVAL from rx_queue_setup() if feature is requested
  (mp parameter is supposed to be NULL for the case)

> Is it always split by specified/fixed length?
Yes, it is simple feature, it splits the data to the buffers with required
memory attributes provided by specified pools according to the fixed lengths.
It should be OK for protocols like eCPRI or some tunneling.

> What happens if header length is actually different?
It is per queue configuration, packets might be sorted with rte_flow engine between the queues.
The supposed use case is to filter out specific protocol packets (say eCPRI with fixed header length)
and split ones on specific Rx queue.


With best regards,
Slava

> 
> > The new approach would allow splitting the ingress packets into
> > multiple parts pushed to the memory with different attributes.
> > For example, the packet headers can be pushed to the embedded data
> > buffers within mbufs and the application data into the external
> > buffers attached to mbufs allocated from the different memory pools.
> > The memory attributes for the split parts may differ either - for
> > example the application data may be pushed into the external memory
> > located on the dedicated physical device, say GPU or NVMe. This would
> > improve the DPDK receiving datapath flexibility preserving
> > compatibility with existing API.
> >
> > Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
> > ---
> >  doc/guides/rel_notes/deprecation.rst | 5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/doc/guides/rel_notes/deprecation.rst
> > b/doc/guides/rel_notes/deprecation.rst
> > index ea4cfa7..cd700ae 100644
> > --- a/doc/guides/rel_notes/deprecation.rst
> > +++ b/doc/guides/rel_notes/deprecation.rst
> > @@ -99,6 +99,11 @@ Deprecation Notices
> >    In 19.11 PMDs will still update the field even when the offload is not
> >    enabled.
> >
> > +* ethdev: add new fields to ``rte_eth_rxconf`` to configure the
> > +receiving
> > +  queues to split ingress packets into multiple segments according to
> > +the
> > +  specified lengths into the buffers allocated from the specified
> > +  memory pools. The backward compatibility to existing API is preserved.
> > +
> >  * ethdev: ``rx_descriptor_done`` dev_ops and
> ``rte_eth_rx_descriptor_done``
> >    will be deprecated in 20.11 and will be removed in 21.11.
> >    Existing ``rte_eth_rx_descriptor_status`` and
> > ``rte_eth_tx_descriptor_status``