[dpdk-dev] [PATCH v7 1/2] ethdev: add flow shared action API
Andrey Vesnovaty
andreyv at nvidia.com
Wed Oct 14 13:47:41 CEST 2020
Hi Ajit
All your suggestions applied in v8 series.
BW Thomas & Andrew pointed to testpmd example in RTE flow API.
Thanks,
Andrey
> -----Original Message-----
> From: Ajit Khaparde <ajit.khaparde at broadcom.com>
> Sent: Friday, October 9, 2020 1:31 AM
> To: Andrey Vesnovaty <andreyv at nvidia.com>
> Cc: dpdk-dev <dev at dpdk.org>; jer at marvell.com; Jerin Jacob
> <jerinjacobk at gmail.com>; NBU-Contact-Thomas Monjalon
> <thomas at monjalon.net>; Ferruh Yigit <ferruh.yigit at intel.com>; Stephen
> Hemminger <stephen at networkplumber.org>; Bruce Richardson
> <bruce.richardson at intel.com>; Ori Kam <orika at nvidia.com>; Slava Ovsiienko
> <viacheslavo at nvidia.com>; andrey.vesnovaty at gmail.com; Ray Kinsella
> <mdr at ashroe.eu>; Neil Horman <nhorman at tuxdriver.com>; Samik Gupta
> <samik.gupta at broadcom.com>; Andrew Rybchenko
> <arybchenko at solarflare.com>
> Subject: Re: [PATCH v7 1/2] ethdev: add flow shared action API
>
> On Thu, Oct 8, 2020 at 4:51 AM Andrey Vesnovaty <andreyv at nvidia.com>
> wrote:
> >
> > This commit introduces extension of DPDK flow action API enabling
> > sharing of single rte_flow_action in multiple flows. The API intended for
> > PMDs, where multiple HW offloaded flows can reuse the same HW
> > essence/object representing flow action and modification of such an
> > essence/object affects all the rules using it.
> >
> > Motivation and example
> > ===
> > Adding or removing one or more queues to RSS used by multiple flow rules
> > imposes per rule toll for current DPDK flow API; the scenario requires
> > for each flow sharing cloned RSS action:
> > - call `rte_flow_destroy()`
> > - call `rte_flow_create()` with modified RSS action
> >
> > API for sharing action and its in-place update benefits:
> > - reduce the overhead of multiple RSS flow rules reconfiguration
> > - optimize resource utilization by sharing action across multiple
> > flows
> >
> > Change description
> > ===
> >
> > Shared action
> > ===
> > In order to represent flow action shared by multiple flows new action
> > type RTE_FLOW_ACTION_TYPE_SHARED is introduced (see `enum
> > rte_flow_action_type`).
> > Actually the introduced API decouples action from any specific flow and
> > enables sharing of single action by its handle across multiple flows.
> >
> > Shared action create/use/destroy
> > ===
> > Shared action may be reused by some or none flow rules at any given
> > moment, i.e. shared action resides outside of the context of any flow.
> > Shared action represent HW resources/objects used for action offloading
> > implementation.
> > API for shared action create (see `rte_flow_shared_action_create()`):
> > - should allocate HW resources and make related initializations required
> > for shared action implementation.
> > - make necessary preparations to maintain shared access to
> > the action resources, configuration and state.
> > API for shared action destroy (see `rte_flow_shared_action_destroy()`)
> > should release HW resources and make related cleanups required for shared
> > action implementation.
> >
> > In order to share some flow action reuse the handle of type
> > `struct rte_flow_shared_action` returned by
> > rte_flow_shared_action_create() as a `conf` field of
> > `struct rte_flow_action` (see "example" section).
> >
> > If some shared action not used by any flow rule all resources allocated
> > by the shared action can be released by rte_flow_shared_action_destroy()
> > (see "example" section). The shared action handle passed as argument to
> > destroy API should not be used any further i.e. result of the usage is
> > undefined.
> >
> > Shared action re-configuration
> > ===
> > Shared action behavior defined by its configuration can be updated via
> > rte_flow_shared_action_update() (see "example" section). The shared
> > action update operation modifies HW related resources/objects allocated
> > on the action creation. The number of operations performed by the update
> > operation should not depend on the number of flows sharing the related
> > action. On return of shared action update API action behavior should be
> > according to updated configuration for all flows sharing the action.
> >
> > Shared action query
> > ===
> > Provide separate API to query shared action state (see
> > rte_flow_shared_action_update()). Taking a counter as an example: query
> > returns value aggregating all counter increments across all flow rules
> > sharing the counter. This API doesn't query shared action configuration
> > since it is controlled by rte_flow_shared_action_create() and
> > rte_flow_shared_action_update() APIs and no supposed to change by other
> > means.
> >
> > PMD support
> > ===
> > The support of introduced API is pure PMD specific design and
> > responsibility for each action type (see struct rte_flow_ops).
> >
> > testpmd
> > ===
> > In order to utilize introduced API testpmd cli may implement following
> > extension
> > create/update/destroy/query shared action accordingly
> >
> > flow shared_action (port) create {action_id (id)} (action) / end
> > flow shared_action (port) update (id) (action) / end
> > flow shared_action (port) destroy action_id (id) {action_id (id) [...]}
> > flow shared_action (port) query (id)
> >
> > testpmd example
> > ===
> >
> > configure rss to queues 1 & 2
> >
> > > flow shared_action 0 create action_id 100 rss queues 1 2 end / end
> >
> > create flow rule utilizing shared action
> >
> > > flow create 0 ingress \
> > pattern eth dst is 0c:42:a1:15:fd:ac / ipv6 / tcp / end \
> > actions shared 100 / end
> >
> > add 2 more queues
> >
> > > flow shared_action 0 modify 100 rss queues 1 2 3 4 end / end
> >
> > example
> > ===
> >
> > struct rte_flow_action actions[2];
> > struct rte_flow_shared_action_conf conf;
> > struct rte_flow_action action;
> > /* skipped: initialize conf and action */
> > struct rte_flow_shared_action *handle =
> > rte_flow_shared_action_create(port_id, &conf, &action, &error);
> > actions[0].type = RTE_FLOW_ACTION_TYPE_SHARED;
> > actions[0].conf = handle;
> > actions[1].type = RTE_FLOW_ACTION_TYPE_END;
> > /* skipped: init attr0 & pattern0 args */
> > struct rte_flow *flow0 = rte_flow_create(port_id, &attr0, pattern0,
> > actions, error);
> > /* create more rules reusing shared action */
> > struct rte_flow *flow1 = rte_flow_create(port_id, &attr1, pattern1,
> > actions, error);
> > /* skipped: for flows 2 till N */
> > struct rte_flow *flowN = rte_flow_create(port_id, &attrN, patternN,
> > actions, error);
> > /* update shared action */
> > struct rte_flow_action updated_action;
> > /*
> > * skipped: initialize updated_action according to desired action
> > * configuration change
> > */
> > rte_flow_shared_action_update(port_id, handle, &updated_action, error);
> > /*
> > * from now on all flows 1 till N will act according to configuration of
> > * updated_action
> > */
> > /* skipped: destroy all flows 1 till N */
> > rte_flow_shared_action_destroy(port_id, handle, error);
> >
> > Signed-off-by: Andrey Vesnovaty <andreyv at nvidia.com>
> > Acked-by: Ori Kam <orika at nvidia.com>
> Since this is an ethdev patch, the testpmd description is really not required.
> Moreover they are not in sync with the direction and other changes you made
> in the testpmd patch. Also there is a typo inline. Other than that..
>
> Acked-by: Ajit Khaparde <ajit.khaparde at broadcom.com>
>
> > ---
> > doc/guides/prog_guide/rte_flow.rst | 19 +++
> > doc/guides/rel_notes/release_20_11.rst | 9 ++
> > lib/librte_ethdev/rte_ethdev_version.map | 4 +
> > lib/librte_ethdev/rte_flow.c | 84 +++++++++++
> > lib/librte_ethdev/rte_flow.h | 169 ++++++++++++++++++++++-
> > lib/librte_ethdev/rte_flow_driver.h | 23 +++
> > 6 files changed, 307 insertions(+), 1 deletion(-)
> >
> [snip]
>
> > +
> > +/**
> > + * RTE_FLOW_ACTION_TYPE_SHARED
> > + *
> > + * Opaque type returned after successfully creating a shared action.
> > + *
> > + * This handle can be used to manage and query the related action:
> > + * - share it across multiple flow rules
> > + * - update action configuration
> > + * - query action data
> > + * - destroy action
> > + */
> > +struct rte_flow_shared_action;
> > +
> > /* Mbuf dynamic field offset for metadata. */
> > extern int32_t rte_flow_dynf_metadata_offs;
> >
> > @@ -3357,6 +3380,150 @@ int
> > rte_flow_get_aged_flows(uint16_t port_id, void **contexts,
> > uint32_t nb_contexts, struct rte_flow_error *error);
> >
> > +/**
> > + * Specify shared action configuration
> > + */
> > +struct rte_flow_shared_action_conf {
> > + /**
> > + * Flow direction for shared action configuration.
> > + *
> > + * Shred action should be valid at least for one flow direction,
> s/Shred/Shared
>
> > + * otherwise it is invalid for both ingress and egress rules.
> > + */
> > + uint32_t ingress:1;
> > + /**< Action valid for rules applied to ingress traffic. */
> > + uint32_t egress:1;
> > + /**< Action valid for rules applied to egress traffic. */
> > +};
> [snip]
More information about the dev
mailing list