From: Kiran Kumar K <kirankumark@marvell.com> Adding suuport to DBDF action in the RTE Flow. Application can specify the dbdf value using rte_flow_action_dbdf. Matched traffic will be sent to specified PCI DBDF device. Signed-off-by: Kiran Kumar K <kirankumark@marvell.com> --- app/test-pmd/cmdline_flow.c | 64 ++++++++++++++++++++++++++++++ doc/guides/prog_guide/rte_flow.rst | 19 +++++++++ lib/librte_ethdev/rte_flow.c | 1 + lib/librte_ethdev/rte_flow.h | 16 ++++++++ 4 files changed, 100 insertions(+) diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c index a78154502..c318b4a27 100644 --- a/app/test-pmd/cmdline_flow.c +++ b/app/test-pmd/cmdline_flow.c @@ -342,8 +342,17 @@ enum index { ACTION_SET_IPV4_DSCP_VALUE, ACTION_SET_IPV6_DSCP, ACTION_SET_IPV6_DSCP_VALUE, + ACTION_DBDF, }; +#define DBDF_KEY_LENGTH 20 + +struct action_dbdf_data { + struct rte_flow_action_dbdf conf; + uint8_t dbdf_value[DBDF_KEY_LENGTH]; +}; + + /** Maximum size for pattern in struct rte_flow_item_raw. */ #define ITEM_RAW_PATTERN_SIZE 40 @@ -1144,6 +1153,7 @@ static const enum index next_action[] = { ACTION_SET_META, ACTION_SET_IPV4_DSCP, ACTION_SET_IPV6_DSCP, + ACTION_DBDF, ZERO, }; @@ -1369,6 +1379,11 @@ static const enum index action_set_ipv6_dscp[] = { ZERO, }; +static const enum index action_dbdf[] = { + ACTION_NEXT, + ZERO, +}; + static int parse_set_raw_encap_decap(struct context *, const struct token *, const char *, unsigned int, void *, unsigned int); @@ -1421,6 +1436,9 @@ static int parse_vc_action_mplsoudp_encap(struct context *, static int parse_vc_action_mplsoudp_decap(struct context *, const struct token *, const char *, unsigned int, void *, unsigned int); +static int parse_vc_action_dbdf_value(struct context *, + const struct token *, const char *, + unsigned int, void *, unsigned int); static int parse_vc_action_raw_encap(struct context *, const struct token *, const char *, unsigned int, void *, unsigned int); @@ -3684,6 +3702,18 @@ static const struct token token_list[] = { (struct rte_flow_action_set_dscp, dscp)), .call = parse_vc_conf, }, + [ACTION_DBDF] = { + .name = "dbdf", + .help = "set DBDF value", + .next = NEXT(action_dbdf, NEXT_ENTRY(STRING)), + .priv = PRIV_ACTION(DBDF, sizeof(struct action_dbdf_data)), + .args = ARGS(ARGS_ENTRY_ARB(0, 0), + ARGS_ENTRY_ARB(0, sizeof(uint8_t)), + ARGS_ENTRY_ARB( + offsetof(struct action_dbdf_data, dbdf_value), + DBDF_KEY_LENGTH)), + .call = parse_vc_action_dbdf_value, + }, }; /** Remove and return last entry from argument stack. */ @@ -5064,6 +5094,40 @@ parse_vc_action_raw_encap_index(struct context *ctx, const struct token *token, return len; } +static int +parse_vc_action_dbdf_value(struct context *ctx, const struct token *token, + const char *str, unsigned int len, void *buf, + unsigned int size) +{ + struct buffer *out = buf; + struct rte_flow_action *action; + struct action_dbdf_data *action_dbdf_data = NULL; + int ret; + + ret = parse_vc(ctx, token, str, len, buf, size); + if (ret < 0) + return ret; + /* Nothing else to do if there is no buffer. */ + if (!out) + return ret; + if (!out->args.vc.actions_n) + return -1; + action = &out->args.vc.actions[out->args.vc.actions_n - 1]; + /* Point to selected object. */ + ctx->object = out->args.vc.data; + ctx->objmask = NULL; + /* Copy the headers to the buffer. */ + action_dbdf_data = ctx->object; + *action_dbdf_data = (struct action_dbdf_data) { + .conf = (struct rte_flow_action_dbdf){ + .dbdf_value = action_dbdf_data->dbdf_value, + }, + .dbdf_value = {}, + }; + action->conf = &action_dbdf_data->conf; + return ret; +} + static int parse_vc_action_raw_encap(struct context *ctx, const struct token *token, const char *str, unsigned int len, void *buf, diff --git a/doc/guides/prog_guide/rte_flow.rst b/doc/guides/prog_guide/rte_flow.rst index 41c147913..b900e283c 100644 --- a/doc/guides/prog_guide/rte_flow.rst +++ b/doc/guides/prog_guide/rte_flow.rst @@ -2616,6 +2616,25 @@ Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned. | ``dscp`` | DSCP in low 6 bits, rest ignore | +-----------+---------------------------------+ +Action: ``DBDF`` +^^^^^^^^^^^^^^^^^^^^^^^^^ + +Set DBDF value. + +Send traffic to specified PCI DBDF device. + +.. _table_rte_flow_action_dbdf: + +.. table:: DBDF + + +-----------------+----------------------------+ + | Field | Value | + +=================+============================+ + | ``length`` | DBDF length | + +-----------------+----------------------------+ + | ``dbdf_value`` | DBDF value | + +-----------------+----------------------------+ + Negative types ~~~~~~~~~~~~~~ diff --git a/lib/librte_ethdev/rte_flow.c b/lib/librte_ethdev/rte_flow.c index a5ac1c7fb..6eada7785 100644 --- a/lib/librte_ethdev/rte_flow.c +++ b/lib/librte_ethdev/rte_flow.c @@ -172,6 +172,7 @@ static const struct rte_flow_desc_data rte_flow_desc_action[] = { MK_FLOW_ACTION(SET_META, sizeof(struct rte_flow_action_set_meta)), MK_FLOW_ACTION(SET_IPV4_DSCP, sizeof(struct rte_flow_action_set_dscp)), MK_FLOW_ACTION(SET_IPV6_DSCP, sizeof(struct rte_flow_action_set_dscp)), + MK_FLOW_ACTION(DBDF, sizeof(struct rte_flow_action_dbdf)), }; int diff --git a/lib/librte_ethdev/rte_flow.h b/lib/librte_ethdev/rte_flow.h index b43238b45..b6029c282 100644 --- a/lib/librte_ethdev/rte_flow.h +++ b/lib/librte_ethdev/rte_flow.h @@ -2082,6 +2082,22 @@ enum rte_flow_action_type { * See struct rte_flow_action_set_dscp. */ RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP, + + /** + * Send packet to specified PCIe device + */ + RTE_FLOW_ACTION_TYPE_DBDF, +}; + + +/** + * RTE_FLOW_ACTION_TYPE_DBDF + * + * Send the packet to specified PCI DBDF device + */ +struct rte_flow_action_dbdf { + uint8_t length; + const uint8_t *dbdf_value; }; /** -- 2.17.1
Hi Kiran, > -----Original Message----- > From: kirankumark@marvell.com <kirankumark@marvell.com> > Sent: Tuesday, March 10, 2020 6:06 PM > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu <wenzhuo.lu@intel.com>; > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > <bernard.iremonger@intel.com>; John McNamara > <john.mcnamara@intel.com>; Marko Kovacevic > <marko.kovacevic@intel.com>; Thomas Monjalon <thomas@monjalon.net>; > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > <arybchenko@solarflare.com> > Cc: dev@dpdk.org; Kiran Kumar K <kirankumark@marvell.com> > Subject: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > From: Kiran Kumar K <kirankumark@marvell.com> > > Adding suuport to DBDF action in the RTE Flow. > Application can specify the dbdf value using rte_flow_action_dbdf. > Matched traffic will be sent to specified PCI DBDF device. > I would like to see more detail use case, for example to which device / device type will the traffic be routed to? > Signed-off-by: Kiran Kumar K <kirankumark@marvell.com> > --- > app/test-pmd/cmdline_flow.c | 64 ++++++++++++++++++++++++++++++ > doc/guides/prog_guide/rte_flow.rst | 19 +++++++++ > lib/librte_ethdev/rte_flow.c | 1 + > lib/librte_ethdev/rte_flow.h | 16 ++++++++ > 4 files changed, 100 insertions(+) > > diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c > index a78154502..c318b4a27 100644 > --- a/app/test-pmd/cmdline_flow.c > +++ b/app/test-pmd/cmdline_flow.c > @@ -342,8 +342,17 @@ enum index { > ACTION_SET_IPV4_DSCP_VALUE, > ACTION_SET_IPV6_DSCP, > ACTION_SET_IPV6_DSCP_VALUE, > + ACTION_DBDF, > }; > > +#define DBDF_KEY_LENGTH 20 > + > +struct action_dbdf_data { > + struct rte_flow_action_dbdf conf; > + uint8_t dbdf_value[DBDF_KEY_LENGTH]; > +}; > + > + > /** Maximum size for pattern in struct rte_flow_item_raw. */ > #define ITEM_RAW_PATTERN_SIZE 40 > > @@ -1144,6 +1153,7 @@ static const enum index next_action[] = { > ACTION_SET_META, > ACTION_SET_IPV4_DSCP, > ACTION_SET_IPV6_DSCP, > + ACTION_DBDF, > ZERO, > }; > > @@ -1369,6 +1379,11 @@ static const enum index action_set_ipv6_dscp[] = { > ZERO, > }; > > +static const enum index action_dbdf[] = { > + ACTION_NEXT, > + ZERO, > +}; > + > static int parse_set_raw_encap_decap(struct context *, const struct token *, > const char *, unsigned int, > void *, unsigned int); > @@ -1421,6 +1436,9 @@ static int parse_vc_action_mplsoudp_encap(struct > context *, > static int parse_vc_action_mplsoudp_decap(struct context *, > const struct token *, const char *, > unsigned int, void *, unsigned int); > +static int parse_vc_action_dbdf_value(struct context *, > + const struct token *, const char *, > + unsigned int, void *, unsigned int); > static int parse_vc_action_raw_encap(struct context *, > const struct token *, const char *, > unsigned int, void *, unsigned int); > @@ -3684,6 +3702,18 @@ static const struct token token_list[] = { > (struct rte_flow_action_set_dscp, dscp)), > .call = parse_vc_conf, > }, > + [ACTION_DBDF] = { > + .name = "dbdf", > + .help = "set DBDF value", > + .next = NEXT(action_dbdf, NEXT_ENTRY(STRING)), > + .priv = PRIV_ACTION(DBDF, sizeof(struct action_dbdf_data)), > + .args = ARGS(ARGS_ENTRY_ARB(0, 0), > + ARGS_ENTRY_ARB(0, sizeof(uint8_t)), > + ARGS_ENTRY_ARB( > + offsetof(struct action_dbdf_data, dbdf_value), > + DBDF_KEY_LENGTH)), > + .call = parse_vc_action_dbdf_value, > + }, > }; > > /** Remove and return last entry from argument stack. */ > @@ -5064,6 +5094,40 @@ parse_vc_action_raw_encap_index(struct context > *ctx, const struct token *token, > return len; > } > > +static int > +parse_vc_action_dbdf_value(struct context *ctx, const struct token *token, > + const char *str, unsigned int len, void *buf, > + unsigned int size) > +{ > + struct buffer *out = buf; > + struct rte_flow_action *action; > + struct action_dbdf_data *action_dbdf_data = NULL; > + int ret; > + > + ret = parse_vc(ctx, token, str, len, buf, size); > + if (ret < 0) > + return ret; > + /* Nothing else to do if there is no buffer. */ > + if (!out) > + return ret; > + if (!out->args.vc.actions_n) > + return -1; > + action = &out->args.vc.actions[out->args.vc.actions_n - 1]; > + /* Point to selected object. */ > + ctx->object = out->args.vc.data; > + ctx->objmask = NULL; > + /* Copy the headers to the buffer. */ > + action_dbdf_data = ctx->object; > + *action_dbdf_data = (struct action_dbdf_data) { > + .conf = (struct rte_flow_action_dbdf){ > + .dbdf_value = action_dbdf_data->dbdf_value, > + }, > + .dbdf_value = {}, > + }; > + action->conf = &action_dbdf_data->conf; I think you are missing the setting of the len value in the conf. > + return ret; > +} > + > static int > parse_vc_action_raw_encap(struct context *ctx, const struct token *token, > const char *str, unsigned int len, void *buf, > diff --git a/doc/guides/prog_guide/rte_flow.rst > b/doc/guides/prog_guide/rte_flow.rst > index 41c147913..b900e283c 100644 > --- a/doc/guides/prog_guide/rte_flow.rst > +++ b/doc/guides/prog_guide/rte_flow.rst > @@ -2616,6 +2616,25 @@ Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error > will be returned. > | ``dscp`` | DSCP in low 6 bits, rest ignore | > +-----------+---------------------------------+ > > +Action: ``DBDF`` > +^^^^^^^^^^^^^^^^^^^^^^^^^ > + > +Set DBDF value. > + > +Send traffic to specified PCI DBDF device. > + > +.. _table_rte_flow_action_dbdf: > + > +.. table:: DBDF > + > + +-----------------+----------------------------+ > + | Field | Value | > + +=================+============================+ > + | ``length`` | DBDF length | > + +-----------------+----------------------------+ > + | ``dbdf_value`` | DBDF value | > + +-----------------+----------------------------+ > + > Negative types > ~~~~~~~~~~~~~~ > > diff --git a/lib/librte_ethdev/rte_flow.c b/lib/librte_ethdev/rte_flow.c > index a5ac1c7fb..6eada7785 100644 > --- a/lib/librte_ethdev/rte_flow.c > +++ b/lib/librte_ethdev/rte_flow.c > @@ -172,6 +172,7 @@ static const struct rte_flow_desc_data > rte_flow_desc_action[] = { > MK_FLOW_ACTION(SET_META, sizeof(struct > rte_flow_action_set_meta)), > MK_FLOW_ACTION(SET_IPV4_DSCP, sizeof(struct > rte_flow_action_set_dscp)), > MK_FLOW_ACTION(SET_IPV6_DSCP, sizeof(struct > rte_flow_action_set_dscp)), > + MK_FLOW_ACTION(DBDF, sizeof(struct rte_flow_action_dbdf)), > }; > > int > diff --git a/lib/librte_ethdev/rte_flow.h b/lib/librte_ethdev/rte_flow.h > index b43238b45..b6029c282 100644 > --- a/lib/librte_ethdev/rte_flow.h > +++ b/lib/librte_ethdev/rte_flow.h > @@ -2082,6 +2082,22 @@ enum rte_flow_action_type { > * See struct rte_flow_action_set_dscp. > */ > RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP, > + > + /** > + * Send packet to specified PCIe device > + */ > + RTE_FLOW_ACTION_TYPE_DBDF, > +}; > + > + > +/** > + * RTE_FLOW_ACTION_TYPE_DBDF > + * > + * Send the packet to specified PCI DBDF device > + */ > +struct rte_flow_action_dbdf { > + uint8_t length; > + const uint8_t *dbdf_value; > }; > > /** > -- > 2.17.1 Thanks, Ori
Hi Ori, > -----Original Message----- > From: Ori Kam <orika@mellanox.com> > Sent: Monday, March 16, 2020 7:04 PM > To: Kiran Kumar Kokkilagadda <kirankumark@marvell.com>; Wenzhuo Lu > <wenzhuo.lu@intel.com>; Jingjing Wu <jingjing.wu@intel.com>; Bernard > Iremonger <bernard.iremonger@intel.com>; John McNamara > <john.mcnamara@intel.com>; Marko Kovacevic <marko.kovacevic@intel.com>; > Thomas Monjalon <thomas@monjalon.net>; Ferruh Yigit > <ferruh.yigit@intel.com>; Andrew Rybchenko <arybchenko@solarflare.com> > Cc: dev@dpdk.org > Subject: [EXT] RE: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > External Email > > ---------------------------------------------------------------------- > Hi Kiran, > > > > -----Original Message----- > > From: kirankumark@marvell.com <kirankumark@marvell.com> > > Sent: Tuesday, March 10, 2020 6:06 PM > > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu <wenzhuo.lu@intel.com>; > > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > > <bernard.iremonger@intel.com>; John McNamara > > <john.mcnamara@intel.com>; Marko Kovacevic > > <marko.kovacevic@intel.com>; Thomas Monjalon <thomas@monjalon.net>; > > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > > <arybchenko@solarflare.com> > > Cc: dev@dpdk.org; Kiran Kumar K <kirankumark@marvell.com> > > Subject: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > > > From: Kiran Kumar K <kirankumark@marvell.com> > > > > Adding suuport to DBDF action in the RTE Flow. > > Application can specify the dbdf value using rte_flow_action_dbdf. > > Matched traffic will be sent to specified PCI DBDF device. > > > I would like to see more detail use case, for example to which device / device > type will the traffic be routed to? > We have the following use case. We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And we have 3 applications running. 1st application on pf0 and pf1 2nd application on pf0_vf0 3rd application on pf1_vf0. We want to direct the traffic matching condition1 from application 1 (traffic from both pf0 & pf1) needs to send to application 2 (pf0_vf0) And matching condition2 from application 1 (traffic from both pf0 & pf1) needs to send to application 3 (pf1_vf0). To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic from pf1 to pf0_vf0. In this case This DBDF action will be useful. > > Signed-off-by: Kiran Kumar K <kirankumark@marvell.com> > > --- > > app/test-pmd/cmdline_flow.c | 64 ++++++++++++++++++++++++++++++ > > doc/guides/prog_guide/rte_flow.rst | 19 +++++++++ > > lib/librte_ethdev/rte_flow.c | 1 + > > lib/librte_ethdev/rte_flow.h | 16 ++++++++ > > 4 files changed, 100 insertions(+) > > > > diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c > > index a78154502..c318b4a27 100644 > > --- a/app/test-pmd/cmdline_flow.c > > +++ b/app/test-pmd/cmdline_flow.c > > @@ -342,8 +342,17 @@ enum index { > > ACTION_SET_IPV4_DSCP_VALUE, > > ACTION_SET_IPV6_DSCP, > > ACTION_SET_IPV6_DSCP_VALUE, > > + ACTION_DBDF, > > }; > > > > +#define DBDF_KEY_LENGTH 20 > > + > > +struct action_dbdf_data { > > + struct rte_flow_action_dbdf conf; > > + uint8_t dbdf_value[DBDF_KEY_LENGTH]; }; > > + > > + > > /** Maximum size for pattern in struct rte_flow_item_raw. */ #define > > ITEM_RAW_PATTERN_SIZE 40 > > > > @@ -1144,6 +1153,7 @@ static const enum index next_action[] = { > > ACTION_SET_META, > > ACTION_SET_IPV4_DSCP, > > ACTION_SET_IPV6_DSCP, > > + ACTION_DBDF, > > ZERO, > > }; > > > > @@ -1369,6 +1379,11 @@ static const enum index action_set_ipv6_dscp[] = { > > ZERO, > > }; > > > > +static const enum index action_dbdf[] = { > > + ACTION_NEXT, > > + ZERO, > > +}; > > + > > static int parse_set_raw_encap_decap(struct context *, const struct token *, > > const char *, unsigned int, > > void *, unsigned int); > > @@ -1421,6 +1436,9 @@ static int parse_vc_action_mplsoudp_encap(struct > > context *, > > static int parse_vc_action_mplsoudp_decap(struct context *, > > const struct token *, const char *, > > unsigned int, void *, unsigned int); > > +static int parse_vc_action_dbdf_value(struct context *, > > + const struct token *, const char *, > > + unsigned int, void *, unsigned int); > > static int parse_vc_action_raw_encap(struct context *, > > const struct token *, const char *, > > unsigned int, void *, unsigned int); @@ - > 3684,6 +3702,18 @@ > > static const struct token token_list[] = { > > (struct rte_flow_action_set_dscp, dscp)), > > .call = parse_vc_conf, > > }, > > + [ACTION_DBDF] = { > > + .name = "dbdf", > > + .help = "set DBDF value", > > + .next = NEXT(action_dbdf, NEXT_ENTRY(STRING)), > > + .priv = PRIV_ACTION(DBDF, sizeof(struct action_dbdf_data)), > > + .args = ARGS(ARGS_ENTRY_ARB(0, 0), > > + ARGS_ENTRY_ARB(0, sizeof(uint8_t)), > > + ARGS_ENTRY_ARB( > > + offsetof(struct action_dbdf_data, dbdf_value), > > + DBDF_KEY_LENGTH)), > > + .call = parse_vc_action_dbdf_value, > > + }, > > }; > > > > /** Remove and return last entry from argument stack. */ @@ -5064,6 > > +5094,40 @@ parse_vc_action_raw_encap_index(struct context *ctx, const > > struct token *token, > > return len; > > } > > > > +static int > > +parse_vc_action_dbdf_value(struct context *ctx, const struct token *token, > > + const char *str, unsigned int len, void *buf, > > + unsigned int size) > > +{ > > + struct buffer *out = buf; > > + struct rte_flow_action *action; > > + struct action_dbdf_data *action_dbdf_data = NULL; > > + int ret; > > + > > + ret = parse_vc(ctx, token, str, len, buf, size); > > + if (ret < 0) > > + return ret; > > + /* Nothing else to do if there is no buffer. */ > > + if (!out) > > + return ret; > > + if (!out->args.vc.actions_n) > > + return -1; > > + action = &out->args.vc.actions[out->args.vc.actions_n - 1]; > > + /* Point to selected object. */ > > + ctx->object = out->args.vc.data; > > + ctx->objmask = NULL; > > + /* Copy the headers to the buffer. */ > > + action_dbdf_data = ctx->object; > > + *action_dbdf_data = (struct action_dbdf_data) { > > + .conf = (struct rte_flow_action_dbdf){ > > + .dbdf_value = action_dbdf_data->dbdf_value, > > + }, > > + .dbdf_value = {}, > > + }; > > + action->conf = &action_dbdf_data->conf; > > I think you are missing the setting of the len value in the conf. Length will be update while parsing the string (parse_string). > > > + return ret; > > +} > > + > > static int > > parse_vc_action_raw_encap(struct context *ctx, const struct token *token, > > const char *str, unsigned int len, void *buf, diff --git > > a/doc/guides/prog_guide/rte_flow.rst > > b/doc/guides/prog_guide/rte_flow.rst > > index 41c147913..b900e283c 100644 > > --- a/doc/guides/prog_guide/rte_flow.rst > > +++ b/doc/guides/prog_guide/rte_flow.rst > > @@ -2616,6 +2616,25 @@ Otherwise, RTE_FLOW_ERROR_TYPE_ACTION > error > > will be returned. > > | ``dscp`` | DSCP in low 6 bits, rest ignore | > > +-----------+---------------------------------+ > > > > +Action: ``DBDF`` > > +^^^^^^^^^^^^^^^^^^^^^^^^^ > > + > > +Set DBDF value. > > + > > +Send traffic to specified PCI DBDF device. > > + > > +.. _table_rte_flow_action_dbdf: > > + > > +.. table:: DBDF > > + > > + +-----------------+----------------------------+ > > + | Field | Value | > > + +=================+============================+ > > + | ``length`` | DBDF length | > > + +-----------------+----------------------------+ > > + | ``dbdf_value`` | DBDF value | > > + +-----------------+----------------------------+ > > + > > Negative types > > ~~~~~~~~~~~~~~ > > > > diff --git a/lib/librte_ethdev/rte_flow.c > > b/lib/librte_ethdev/rte_flow.c index a5ac1c7fb..6eada7785 100644 > > --- a/lib/librte_ethdev/rte_flow.c > > +++ b/lib/librte_ethdev/rte_flow.c > > @@ -172,6 +172,7 @@ static const struct rte_flow_desc_data > > rte_flow_desc_action[] = { > > MK_FLOW_ACTION(SET_META, sizeof(struct > rte_flow_action_set_meta)), > > MK_FLOW_ACTION(SET_IPV4_DSCP, sizeof(struct > > rte_flow_action_set_dscp)), > > MK_FLOW_ACTION(SET_IPV6_DSCP, sizeof(struct > > rte_flow_action_set_dscp)), > > + MK_FLOW_ACTION(DBDF, sizeof(struct rte_flow_action_dbdf)), > > }; > > > > int > > diff --git a/lib/librte_ethdev/rte_flow.h > > b/lib/librte_ethdev/rte_flow.h index b43238b45..b6029c282 100644 > > --- a/lib/librte_ethdev/rte_flow.h > > +++ b/lib/librte_ethdev/rte_flow.h > > @@ -2082,6 +2082,22 @@ enum rte_flow_action_type { > > * See struct rte_flow_action_set_dscp. > > */ > > RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP, > > + > > + /** > > + * Send packet to specified PCIe device > > + */ > > + RTE_FLOW_ACTION_TYPE_DBDF, > > +}; > > + > > + > > +/** > > + * RTE_FLOW_ACTION_TYPE_DBDF > > + * > > + * Send the packet to specified PCI DBDF device */ struct > > +rte_flow_action_dbdf { > > + uint8_t length; > > + const uint8_t *dbdf_value; > > }; > > > > /** > > -- > > 2.17.1 > > Thanks, > Ori
HI Kiran, > -----Original Message----- > From: Kiran Kumar Kokkilagadda <kirankumark@marvell.com> > Sent: Tuesday, March 17, 2020 12:34 PM > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu <wenzhuo.lu@intel.com>; > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > <bernard.iremonger@intel.com>; John McNamara > <john.mcnamara@intel.com>; Marko Kovacevic > <marko.kovacevic@intel.com>; Thomas Monjalon <thomas@monjalon.net>; > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > <arybchenko@solarflare.com> > Cc: dev@dpdk.org > Subject: RE: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > Hi Ori, > > > > > -----Original Message----- > > From: Ori Kam <orika@mellanox.com> > > Sent: Monday, March 16, 2020 7:04 PM > > To: Kiran Kumar Kokkilagadda <kirankumark@marvell.com>; Wenzhuo Lu > > <wenzhuo.lu@intel.com>; Jingjing Wu <jingjing.wu@intel.com>; Bernard > > Iremonger <bernard.iremonger@intel.com>; John McNamara > > <john.mcnamara@intel.com>; Marko Kovacevic > <marko.kovacevic@intel.com>; > > Thomas Monjalon <thomas@monjalon.net>; Ferruh Yigit > > <ferruh.yigit@intel.com>; Andrew Rybchenko <arybchenko@solarflare.com> > > Cc: dev@dpdk.org > > Subject: [EXT] RE: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > > > External Email > > > > ---------------------------------------------------------------------- > > Hi Kiran, > > > > > > > -----Original Message----- > > > From: kirankumark@marvell.com <kirankumark@marvell.com> > > > Sent: Tuesday, March 10, 2020 6:06 PM > > > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu > <wenzhuo.lu@intel.com>; > > > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > > > <bernard.iremonger@intel.com>; John McNamara > > > <john.mcnamara@intel.com>; Marko Kovacevic > > > <marko.kovacevic@intel.com>; Thomas Monjalon > <thomas@monjalon.net>; > > > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > > > <arybchenko@solarflare.com> > > > Cc: dev@dpdk.org; Kiran Kumar K <kirankumark@marvell.com> > > > Subject: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > > > > > From: Kiran Kumar K <kirankumark@marvell.com> > > > > > > Adding suuport to DBDF action in the RTE Flow. > > > Application can specify the dbdf value using rte_flow_action_dbdf. > > > Matched traffic will be sent to specified PCI DBDF device. > > > > > I would like to see more detail use case, for example to which device / device > > type will the traffic be routed to? > > > > We have the following use case. > We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And we have > 3 applications running. > 1st application on pf0 and pf1 > 2nd application on pf0_vf0 > 3rd application on pf1_vf0. > We want to direct the traffic matching condition1 from application 1 (traffic > from both pf0 & pf1) needs to send to application 2 (pf0_vf0) > And matching condition2 from application 1 (traffic from both pf0 & pf1) needs > to send to application 3 (pf1_vf0). > To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic from pf1 > to pf0_vf0. In this case This DBDF action will be useful. > It seems that what you are describing it the port action with representors, or any other way you wish to implement it. > > > > Signed-off-by: Kiran Kumar K <kirankumark@marvell.com> > > > --- > > > app/test-pmd/cmdline_flow.c | 64 > ++++++++++++++++++++++++++++++ > > > doc/guides/prog_guide/rte_flow.rst | 19 +++++++++ > > > lib/librte_ethdev/rte_flow.c | 1 + > > > lib/librte_ethdev/rte_flow.h | 16 ++++++++ > > > 4 files changed, 100 insertions(+) > > > > > > diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c > > > index a78154502..c318b4a27 100644 > > > --- a/app/test-pmd/cmdline_flow.c > > > +++ b/app/test-pmd/cmdline_flow.c > > > @@ -342,8 +342,17 @@ enum index { > > > ACTION_SET_IPV4_DSCP_VALUE, > > > ACTION_SET_IPV6_DSCP, > > > ACTION_SET_IPV6_DSCP_VALUE, > > > + ACTION_DBDF, > > > }; > > > > > > +#define DBDF_KEY_LENGTH 20 > > > + > > > +struct action_dbdf_data { > > > + struct rte_flow_action_dbdf conf; > > > + uint8_t dbdf_value[DBDF_KEY_LENGTH]; }; > > > + > > > + > > > /** Maximum size for pattern in struct rte_flow_item_raw. */ #define > > > ITEM_RAW_PATTERN_SIZE 40 > > > > > > @@ -1144,6 +1153,7 @@ static const enum index next_action[] = { > > > ACTION_SET_META, > > > ACTION_SET_IPV4_DSCP, > > > ACTION_SET_IPV6_DSCP, > > > + ACTION_DBDF, > > > ZERO, > > > }; > > > > > > @@ -1369,6 +1379,11 @@ static const enum index action_set_ipv6_dscp[] > = { > > > ZERO, > > > }; > > > > > > +static const enum index action_dbdf[] = { > > > + ACTION_NEXT, > > > + ZERO, > > > +}; > > > + > > > static int parse_set_raw_encap_decap(struct context *, const struct token > *, > > > const char *, unsigned int, > > > void *, unsigned int); > > > @@ -1421,6 +1436,9 @@ static int > parse_vc_action_mplsoudp_encap(struct > > > context *, > > > static int parse_vc_action_mplsoudp_decap(struct context *, > > > const struct token *, const char *, > > > unsigned int, void *, unsigned int); > > > +static int parse_vc_action_dbdf_value(struct context *, > > > + const struct token *, const char *, > > > + unsigned int, void *, unsigned int); > > > static int parse_vc_action_raw_encap(struct context *, > > > const struct token *, const char *, > > > unsigned int, void *, unsigned int); @@ - > > 3684,6 +3702,18 @@ > > > static const struct token token_list[] = { > > > (struct rte_flow_action_set_dscp, dscp)), > > > .call = parse_vc_conf, > > > }, > > > + [ACTION_DBDF] = { > > > + .name = "dbdf", > > > + .help = "set DBDF value", > > > + .next = NEXT(action_dbdf, NEXT_ENTRY(STRING)), > > > + .priv = PRIV_ACTION(DBDF, sizeof(struct action_dbdf_data)), > > > + .args = ARGS(ARGS_ENTRY_ARB(0, 0), > > > + ARGS_ENTRY_ARB(0, sizeof(uint8_t)), > > > + ARGS_ENTRY_ARB( > > > + offsetof(struct action_dbdf_data, dbdf_value), > > > + DBDF_KEY_LENGTH)), > > > + .call = parse_vc_action_dbdf_value, > > > + }, > > > }; > > > > > > /** Remove and return last entry from argument stack. */ @@ -5064,6 > > > +5094,40 @@ parse_vc_action_raw_encap_index(struct context *ctx, const > > > struct token *token, > > > return len; > > > } > > > > > > +static int > > > +parse_vc_action_dbdf_value(struct context *ctx, const struct token *token, > > > + const char *str, unsigned int len, void *buf, > > > + unsigned int size) > > > +{ > > > + struct buffer *out = buf; > > > + struct rte_flow_action *action; > > > + struct action_dbdf_data *action_dbdf_data = NULL; > > > + int ret; > > > + > > > + ret = parse_vc(ctx, token, str, len, buf, size); > > > + if (ret < 0) > > > + return ret; > > > + /* Nothing else to do if there is no buffer. */ > > > + if (!out) > > > + return ret; > > > + if (!out->args.vc.actions_n) > > > + return -1; > > > + action = &out->args.vc.actions[out->args.vc.actions_n - 1]; > > > + /* Point to selected object. */ > > > + ctx->object = out->args.vc.data; > > > + ctx->objmask = NULL; > > > + /* Copy the headers to the buffer. */ > > > + action_dbdf_data = ctx->object; > > > + *action_dbdf_data = (struct action_dbdf_data) { > > > + .conf = (struct rte_flow_action_dbdf){ > > > + .dbdf_value = action_dbdf_data->dbdf_value, > > > + }, > > > + .dbdf_value = {}, > > > + }; > > > + action->conf = &action_dbdf_data->conf; > > > > I think you are missing the setting of the len value in the conf. > > Length will be update while parsing the string (parse_string). You are right I missed it 😊 > > > > > > + return ret; > > > +} > > > + > > > static int > > > parse_vc_action_raw_encap(struct context *ctx, const struct token *token, > > > const char *str, unsigned int len, void *buf, diff --git > > > a/doc/guides/prog_guide/rte_flow.rst > > > b/doc/guides/prog_guide/rte_flow.rst > > > index 41c147913..b900e283c 100644 > > > --- a/doc/guides/prog_guide/rte_flow.rst > > > +++ b/doc/guides/prog_guide/rte_flow.rst > > > @@ -2616,6 +2616,25 @@ Otherwise, RTE_FLOW_ERROR_TYPE_ACTION > > error > > > will be returned. > > > | ``dscp`` | DSCP in low 6 bits, rest ignore | > > > +-----------+---------------------------------+ > > > > > > +Action: ``DBDF`` > > > +^^^^^^^^^^^^^^^^^^^^^^^^^ > > > + > > > +Set DBDF value. > > > + > > > +Send traffic to specified PCI DBDF device. > > > + > > > +.. _table_rte_flow_action_dbdf: > > > + > > > +.. table:: DBDF > > > + > > > + +-----------------+----------------------------+ > > > + | Field | Value | > > > + +=================+============================+ > > > + | ``length`` | DBDF length | > > > + +-----------------+----------------------------+ > > > + | ``dbdf_value`` | DBDF value | > > > + +-----------------+----------------------------+ > > > + > > > Negative types > > > ~~~~~~~~~~~~~~ > > > > > > diff --git a/lib/librte_ethdev/rte_flow.c > > > b/lib/librte_ethdev/rte_flow.c index a5ac1c7fb..6eada7785 100644 > > > --- a/lib/librte_ethdev/rte_flow.c > > > +++ b/lib/librte_ethdev/rte_flow.c > > > @@ -172,6 +172,7 @@ static const struct rte_flow_desc_data > > > rte_flow_desc_action[] = { > > > MK_FLOW_ACTION(SET_META, sizeof(struct > > rte_flow_action_set_meta)), > > > MK_FLOW_ACTION(SET_IPV4_DSCP, sizeof(struct > > > rte_flow_action_set_dscp)), > > > MK_FLOW_ACTION(SET_IPV6_DSCP, sizeof(struct > > > rte_flow_action_set_dscp)), > > > + MK_FLOW_ACTION(DBDF, sizeof(struct rte_flow_action_dbdf)), > > > }; > > > > > > int > > > diff --git a/lib/librte_ethdev/rte_flow.h > > > b/lib/librte_ethdev/rte_flow.h index b43238b45..b6029c282 100644 > > > --- a/lib/librte_ethdev/rte_flow.h > > > +++ b/lib/librte_ethdev/rte_flow.h > > > @@ -2082,6 +2082,22 @@ enum rte_flow_action_type { > > > * See struct rte_flow_action_set_dscp. > > > */ > > > RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP, > > > + > > > + /** > > > + * Send packet to specified PCIe device > > > + */ > > > + RTE_FLOW_ACTION_TYPE_DBDF, > > > +}; > > > + > > > + > > > +/** > > > + * RTE_FLOW_ACTION_TYPE_DBDF > > > + * > > > + * Send the packet to specified PCI DBDF device */ struct > > > +rte_flow_action_dbdf { > > > + uint8_t length; > > > + const uint8_t *dbdf_value; > > > }; > > > > > > /** > > > -- > > > 2.17.1 > > > > Thanks, > > Ori
Hi Ori, > -----Original Message----- > From: dev <dev-bounces@dpdk.org> On Behalf Of Ori Kam > Sent: Tuesday, March 17, 2020 6:56 PM > To: Kiran Kumar Kokkilagadda <kirankumark@marvell.com>; Wenzhuo Lu > <wenzhuo.lu@intel.com>; Jingjing Wu <jingjing.wu@intel.com>; Bernard > Iremonger <bernard.iremonger@intel.com>; John McNamara > <john.mcnamara@intel.com>; Marko Kovacevic <marko.kovacevic@intel.com>; > Thomas Monjalon <thomas@monjalon.net>; Ferruh Yigit > <ferruh.yigit@intel.com>; Andrew Rybchenko <arybchenko@solarflare.com> > Cc: dev@dpdk.org > Subject: Re: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > HI Kiran, > > > -----Original Message----- > > From: Kiran Kumar Kokkilagadda <kirankumark@marvell.com> > > Sent: Tuesday, March 17, 2020 12:34 PM > > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu <wenzhuo.lu@intel.com>; > > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > > <bernard.iremonger@intel.com>; John McNamara > > <john.mcnamara@intel.com>; Marko Kovacevic > > <marko.kovacevic@intel.com>; Thomas Monjalon <thomas@monjalon.net>; > > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > > <arybchenko@solarflare.com> > > Cc: dev@dpdk.org > > Subject: RE: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > > > Hi Ori, > > > > > > > > > -----Original Message----- > > > From: Ori Kam <orika@mellanox.com> > > > Sent: Monday, March 16, 2020 7:04 PM > > > To: Kiran Kumar Kokkilagadda <kirankumark@marvell.com>; Wenzhuo Lu > > > <wenzhuo.lu@intel.com>; Jingjing Wu <jingjing.wu@intel.com>; Bernard > > > Iremonger <bernard.iremonger@intel.com>; John McNamara > > > <john.mcnamara@intel.com>; Marko Kovacevic > > <marko.kovacevic@intel.com>; > > > Thomas Monjalon <thomas@monjalon.net>; Ferruh Yigit > > > <ferruh.yigit@intel.com>; Andrew Rybchenko > > > <arybchenko@solarflare.com> > > > Cc: dev@dpdk.org > > > Subject: [EXT] RE: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE > > > Flow > > > > > > External Email > > > > > > -------------------------------------------------------------------- > > > -- > > > Hi Kiran, > > > > > > > > > > -----Original Message----- > > > > From: kirankumark@marvell.com <kirankumark@marvell.com> > > > > Sent: Tuesday, March 10, 2020 6:06 PM > > > > To: Ori Kam <orika@mellanox.com>; Wenzhuo Lu > > <wenzhuo.lu@intel.com>; > > > > Jingjing Wu <jingjing.wu@intel.com>; Bernard Iremonger > > > > <bernard.iremonger@intel.com>; John McNamara > > > > <john.mcnamara@intel.com>; Marko Kovacevic > > > > <marko.kovacevic@intel.com>; Thomas Monjalon > > <thomas@monjalon.net>; > > > > Ferruh Yigit <ferruh.yigit@intel.com>; Andrew Rybchenko > > > > <arybchenko@solarflare.com> > > > > Cc: dev@dpdk.org; Kiran Kumar K <kirankumark@marvell.com> > > > > Subject: [dpdk-dev] [PATCH] ethdev: add DBDF action to RTE Flow > > > > > > > > From: Kiran Kumar K <kirankumark@marvell.com> > > > > > > > > Adding suuport to DBDF action in the RTE Flow. > > > > Application can specify the dbdf value using rte_flow_action_dbdf. > > > > Matched traffic will be sent to specified PCI DBDF device. > > > > > > > I would like to see more detail use case, for example to which > > > device / device type will the traffic be routed to? > > > > > > > We have the following use case. > > We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And > > we have > > 3 applications running. > > 1st application on pf0 and pf1 > > 2nd application on pf0_vf0 > > 3rd application on pf1_vf0. > > We want to direct the traffic matching condition1 from application 1 > > (traffic from both pf0 & pf1) needs to send to application 2 > > (pf0_vf0) And matching condition2 from application 1 (traffic from > > both pf0 & pf1) needs to send to application 3 (pf1_vf0). > > To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic > > from pf1 to pf0_vf0. In this case This DBDF action will be useful. > > > > It seems that what you are describing it the port action with representors, or any > other way you wish to implement it. Let's say we have a VF with kernel and we want to send the traffic to that VF, then we can't Use port action. This will be useful in those scenarios. > > > > > > > Signed-off-by: Kiran Kumar K <kirankumark@marvell.com> > > > > --- > > > > app/test-pmd/cmdline_flow.c | 64 > > ++++++++++++++++++++++++++++++ > > > > doc/guides/prog_guide/rte_flow.rst | 19 +++++++++ > > > > lib/librte_ethdev/rte_flow.c | 1 + > > > > lib/librte_ethdev/rte_flow.h | 16 ++++++++ > > > > 4 files changed, 100 insertions(+) > > > > > > > > diff --git a/app/test-pmd/cmdline_flow.c > > > > b/app/test-pmd/cmdline_flow.c index a78154502..c318b4a27 100644 > > > > --- a/app/test-pmd/cmdline_flow.c > > > > +++ b/app/test-pmd/cmdline_flow.c > > > > @@ -342,8 +342,17 @@ enum index { > > > > ACTION_SET_IPV4_DSCP_VALUE, > > > > ACTION_SET_IPV6_DSCP, > > > > ACTION_SET_IPV6_DSCP_VALUE, > > > > + ACTION_DBDF, > > > > }; > > > > > > > > +#define DBDF_KEY_LENGTH 20 > > > > + > > > > +struct action_dbdf_data { > > > > + struct rte_flow_action_dbdf conf; > > > > + uint8_t dbdf_value[DBDF_KEY_LENGTH]; }; > > > > + > > > > + > > > > /** Maximum size for pattern in struct rte_flow_item_raw. */ > > > > #define ITEM_RAW_PATTERN_SIZE 40 > > > > > > > > @@ -1144,6 +1153,7 @@ static const enum index next_action[] = { > > > > ACTION_SET_META, > > > > ACTION_SET_IPV4_DSCP, > > > > ACTION_SET_IPV6_DSCP, > > > > + ACTION_DBDF, > > > > ZERO, > > > > }; > > > > > > > > @@ -1369,6 +1379,11 @@ static const enum index > > > > action_set_ipv6_dscp[] > > = { > > > > ZERO, > > > > }; > > > > > > > > +static const enum index action_dbdf[] = { > > > > + ACTION_NEXT, > > > > + ZERO, > > > > +}; > > > > + > > > > static int parse_set_raw_encap_decap(struct context *, const > > > > struct token > > *, > > > > const char *, unsigned int, > > > > void *, unsigned int); > > > > @@ -1421,6 +1436,9 @@ static int > > parse_vc_action_mplsoudp_encap(struct > > > > context *, > > > > static int parse_vc_action_mplsoudp_decap(struct context *, > > > > const struct token *, const char *, > > > > unsigned int, void *, unsigned int); > > > > +static int parse_vc_action_dbdf_value(struct context *, > > > > + const struct token *, const char *, > > > > + unsigned int, void *, unsigned int); > > > > static int parse_vc_action_raw_encap(struct context *, > > > > const struct token *, const char *, > > > > unsigned int, void *, unsigned int); @@ - > > > 3684,6 +3702,18 @@ > > > > static const struct token token_list[] = { > > > > (struct rte_flow_action_set_dscp, dscp)), > > > > .call = parse_vc_conf, > > > > }, > > > > + [ACTION_DBDF] = { > > > > + .name = "dbdf", > > > > + .help = "set DBDF value", > > > > + .next = NEXT(action_dbdf, NEXT_ENTRY(STRING)), > > > > + .priv = PRIV_ACTION(DBDF, sizeof(struct action_dbdf_data)), > > > > + .args = ARGS(ARGS_ENTRY_ARB(0, 0), > > > > + ARGS_ENTRY_ARB(0, sizeof(uint8_t)), > > > > + ARGS_ENTRY_ARB( > > > > + offsetof(struct action_dbdf_data, dbdf_value), > > > > + DBDF_KEY_LENGTH)), > > > > + .call = parse_vc_action_dbdf_value, > > > > + }, > > > > }; > > > > > > > > /** Remove and return last entry from argument stack. */ @@ > > > > -5064,6 > > > > +5094,40 @@ parse_vc_action_raw_encap_index(struct context *ctx, > > > > +const > > > > struct token *token, > > > > return len; > > > > } > > > > > > > > +static int > > > > +parse_vc_action_dbdf_value(struct context *ctx, const struct token > *token, > > > > + const char *str, unsigned int len, void *buf, > > > > + unsigned int size) > > > > +{ > > > > + struct buffer *out = buf; > > > > + struct rte_flow_action *action; > > > > + struct action_dbdf_data *action_dbdf_data = NULL; > > > > + int ret; > > > > + > > > > + ret = parse_vc(ctx, token, str, len, buf, size); > > > > + if (ret < 0) > > > > + return ret; > > > > + /* Nothing else to do if there is no buffer. */ > > > > + if (!out) > > > > + return ret; > > > > + if (!out->args.vc.actions_n) > > > > + return -1; > > > > + action = &out->args.vc.actions[out->args.vc.actions_n - 1]; > > > > + /* Point to selected object. */ > > > > + ctx->object = out->args.vc.data; > > > > + ctx->objmask = NULL; > > > > + /* Copy the headers to the buffer. */ > > > > + action_dbdf_data = ctx->object; > > > > + *action_dbdf_data = (struct action_dbdf_data) { > > > > + .conf = (struct rte_flow_action_dbdf){ > > > > + .dbdf_value = action_dbdf_data->dbdf_value, > > > > + }, > > > > + .dbdf_value = {}, > > > > + }; > > > > + action->conf = &action_dbdf_data->conf; > > > > > > I think you are missing the setting of the len value in the conf. > > > > Length will be update while parsing the string (parse_string). > > You are right I missed it 😊 > > > > > > > > > > + return ret; > > > > +} > > > > + > > > > static int > > > > parse_vc_action_raw_encap(struct context *ctx, const struct token > *token, > > > > const char *str, unsigned int len, void *buf, diff --git > > > > a/doc/guides/prog_guide/rte_flow.rst > > > > b/doc/guides/prog_guide/rte_flow.rst > > > > index 41c147913..b900e283c 100644 > > > > --- a/doc/guides/prog_guide/rte_flow.rst > > > > +++ b/doc/guides/prog_guide/rte_flow.rst > > > > @@ -2616,6 +2616,25 @@ Otherwise, RTE_FLOW_ERROR_TYPE_ACTION > > > error > > > > will be returned. > > > > | ``dscp`` | DSCP in low 6 bits, rest ignore | > > > > +-----------+---------------------------------+ > > > > > > > > +Action: ``DBDF`` > > > > +^^^^^^^^^^^^^^^^^^^^^^^^^ > > > > + > > > > +Set DBDF value. > > > > + > > > > +Send traffic to specified PCI DBDF device. > > > > + > > > > +.. _table_rte_flow_action_dbdf: > > > > + > > > > +.. table:: DBDF > > > > + > > > > + +-----------------+----------------------------+ > > > > + | Field | Value | > > > > + +=================+============================+ > > > > + | ``length`` | DBDF length | > > > > + +-----------------+----------------------------+ > > > > + | ``dbdf_value`` | DBDF value | > > > > + +-----------------+----------------------------+ > > > > + > > > > Negative types > > > > ~~~~~~~~~~~~~~ > > > > > > > > diff --git a/lib/librte_ethdev/rte_flow.c > > > > b/lib/librte_ethdev/rte_flow.c index a5ac1c7fb..6eada7785 100644 > > > > --- a/lib/librte_ethdev/rte_flow.c > > > > +++ b/lib/librte_ethdev/rte_flow.c > > > > @@ -172,6 +172,7 @@ static const struct rte_flow_desc_data > > > > rte_flow_desc_action[] = { > > > > MK_FLOW_ACTION(SET_META, sizeof(struct > > > rte_flow_action_set_meta)), > > > > MK_FLOW_ACTION(SET_IPV4_DSCP, sizeof(struct > > > > rte_flow_action_set_dscp)), > > > > MK_FLOW_ACTION(SET_IPV6_DSCP, sizeof(struct > > > > rte_flow_action_set_dscp)), > > > > + MK_FLOW_ACTION(DBDF, sizeof(struct rte_flow_action_dbdf)), > > > > }; > > > > > > > > int > > > > diff --git a/lib/librte_ethdev/rte_flow.h > > > > b/lib/librte_ethdev/rte_flow.h index b43238b45..b6029c282 100644 > > > > --- a/lib/librte_ethdev/rte_flow.h > > > > +++ b/lib/librte_ethdev/rte_flow.h > > > > @@ -2082,6 +2082,22 @@ enum rte_flow_action_type { > > > > * See struct rte_flow_action_set_dscp. > > > > */ > > > > RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP, > > > > + > > > > + /** > > > > + * Send packet to specified PCIe device > > > > + */ > > > > + RTE_FLOW_ACTION_TYPE_DBDF, > > > > +}; > > > > + > > > > + > > > > +/** > > > > + * RTE_FLOW_ACTION_TYPE_DBDF > > > > + * > > > > + * Send the packet to specified PCI DBDF device */ struct > > > > +rte_flow_action_dbdf { > > > > + uint8_t length; > > > > + const uint8_t *dbdf_value; > > > > }; > > > > > > > > /** > > > > -- > > > > 2.17.1 > > > > > > Thanks, > > > Ori
19/03/2020 10:17, Kiran Kumar Kokkilagadda:
> From: Ori Kam
> > From: Kiran Kumar Kokkilagadda <kirankumark@marvell.com>
> > > From: Ori Kam <orika@mellanox.com>
> > > > From: kirankumark@marvell.com <kirankumark@marvell.com>
> > > > > From: Kiran Kumar K <kirankumark@marvell.com>
> > > > >
> > > > > Adding suuport to DBDF action in the RTE Flow.
> > > > > Application can specify the dbdf value using rte_flow_action_dbdf.
> > > > > Matched traffic will be sent to specified PCI DBDF device.
> > > > >
> > > > I would like to see more detail use case, for example to which
> > > > device / device type will the traffic be routed to?
> > > >
> > >
> > > We have the following use case.
> > > We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And
> > > we have
> > > 3 applications running.
> > > 1st application on pf0 and pf1
> > > 2nd application on pf0_vf0
> > > 3rd application on pf1_vf0.
> > > We want to direct the traffic matching condition1 from application 1
> > > (traffic from both pf0 & pf1) needs to send to application 2
> > > (pf0_vf0) And matching condition2 from application 1 (traffic from
> > > both pf0 & pf1) needs to send to application 3 (pf1_vf0).
> > > To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic
> > > from pf1 to pf0_vf0. In this case This DBDF action will be useful.
> > >
> >
> > It seems that what you are describing it the port action with representors, or any
> > other way you wish to implement it.
>
> Let's say we have a VF with kernel and we want to send the traffic to that VF, then we can't
> Use port action. This will be useful in those scenarios.
Sorry I don't understand.
You mean the VF is managed by a kernel driver while the PF is managed by DPDK?
So what prevents having a VF representor?
Hello Jerin, Kiran,
On Thu, Mar 19, 2020 at 10:26 AM Thomas Monjalon <thomas@monjalon.net> wrote:
> > > > We have the following use case.
> > > > We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And
> > > > we have
> > > > 3 applications running.
> > > > 1st application on pf0 and pf1
> > > > 2nd application on pf0_vf0
> > > > 3rd application on pf1_vf0.
> > > > We want to direct the traffic matching condition1 from application 1
> > > > (traffic from both pf0 & pf1) needs to send to application 2
> > > > (pf0_vf0) And matching condition2 from application 1 (traffic from
> > > > both pf0 & pf1) needs to send to application 3 (pf1_vf0).
> > > > To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic
> > > > from pf1 to pf0_vf0. In this case This DBDF action will be useful.
> > > >
> > >
> > > It seems that what you are describing it the port action with representors, or any
> > > other way you wish to implement it.
> >
> > Let's say we have a VF with kernel and we want to send the traffic to that VF, then we can't
> > Use port action. This will be useful in those scenarios.
>
> Sorry I don't understand.
> You mean the VF is managed by a kernel driver while the PF is managed by DPDK?
> So what prevents having a VF representor?
The discussion did not reach a conclusion.
Looking at patchwork, I can see it set to "Not Applicable".
Do you still expect some work on this subject?
Thanks.
--
David Marchand
On Tue, May 26, 2020 at 10:25 PM David Marchand <david.marchand@redhat.com> wrote: > > Hello Jerin, Kiran, Hello David, > > On Thu, Mar 19, 2020 at 10:26 AM Thomas Monjalon <thomas@monjalon.net> wrote: > > > > > We have the following use case. > > > > > We have 2 PF's pf0, pf1 and corresponding VF's pf0_vf0 , pf1_vf0. And > > > > > we have > > > > > 3 applications running. > > > > > 1st application on pf0 and pf1 > > > > > 2nd application on pf0_vf0 > > > > > 3rd application on pf1_vf0. > > > > > We want to direct the traffic matching condition1 from application 1 > > > > > (traffic from both pf0 & pf1) needs to send to application 2 > > > > > (pf0_vf0) And matching condition2 from application 1 (traffic from > > > > > both pf0 & pf1) needs to send to application 3 (pf1_vf0). > > > > > To summarize, we need to send traffic from pf0 to pf1_vf0 and traffic > > > > > from pf1 to pf0_vf0. In this case This DBDF action will be useful. > > > > > > > > > > > > > It seems that what you are describing it the port action with representors, or any > > > > other way you wish to implement it. > > > > > > Let's say we have a VF with kernel and we want to send the traffic to that VF, then we can't > > > Use port action. This will be useful in those scenarios. > > > > Sorry I don't understand. > > You mean the VF is managed by a kernel driver while the PF is managed by DPDK? > > So what prevents having a VF representor? > > The discussion did not reach a conclusion. > Looking at patchwork, I can see it set to "Not Applicable". > > Do you still expect some work on this subject? No. We dont need API change, we can manage with VF representor. > > > Thanks. > > -- > David Marchand >
On Tue, May 26, 2020 at 6:58 PM Jerin Jacob <jerinjacobk@gmail.com> wrote:
> > > > > It seems that what you are describing it the port action with representors, or any
> > > > > other way you wish to implement it.
> > > >
> > > > Let's say we have a VF with kernel and we want to send the traffic to that VF, then we can't
> > > > Use port action. This will be useful in those scenarios.
> > >
> > > Sorry I don't understand.
> > > You mean the VF is managed by a kernel driver while the PF is managed by DPDK?
> > > So what prevents having a VF representor?
> >
> > The discussion did not reach a conclusion.
> > Looking at patchwork, I can see it set to "Not Applicable".
> >
> > Do you still expect some work on this subject?
>
> No. We dont need API change, we can manage with VF representor.
Ok, thanks for confirming.
I will mark this patch as Rejected.
"Not Applicable" is used for patches not for dpdk:
- patches for pktgen which uses dev@dpdk.org mailing list,
- incorrect patches sent to dev@dpdk.org,
- patches for stable branches,
--
David Marchand