net/mlx5: support the new VLAN matching fields
PATCH 1/1 net/mlx5: support the new VLAN matching fields
The fields ``has_vlan`` and ``has_more_vlan`` were added in rte_flow by patch [1].
Using these fields, the application can match all the VLAN options by single flow: any, VLAN only and non-VLAN only.
Add the support for the fields.
By the way, add the support for QinQ packets matching.
VLAN\QinQ limitations are listed in the driver document.
[1] https://patches.dpdk.org/patch/80965/
--- doc/guides/nics/mlx5.rst | 24 +++-- doc/guides/rel_notes/release_20_11.rst | 3 + drivers/net/mlx5/mlx5_flow.c | 5 +- drivers/net/mlx5/mlx5_flow.h | 2 +- drivers/net/mlx5/mlx5_flow_dv.c | 176 +++++++++++++++++---------------- drivers/net/mlx5/mlx5_flow_verbs.c | 2 +- 6 files changed, 117 insertions(+), 95 deletions(-)
doc/guides/nics/mlx5.rst +15 -9
@@ -122,23 +122,29 @@ Limitations Will match any ipv4 packet (VLAN included). -- When using DV flow engine (``dv_flow_en`` = 1), flow pattern without VLAN item - will match untagged packets only. +- When using Verbs flow engine (``dv_flow_en`` = 0), multi-tagged(QinQ) match is not supported. + +- When using DV flow engine (``dv_flow_en`` = 1), flow pattern with any VLAN specification will match only single-tagged packets unless the ETH item ``type`` field is 0x88A8 or the VLAN item ``has_more_vlan`` field is 1. The flow rule:: flow create 0 ingress pattern eth / ipv4 / end ... - Will match untagged packets only. - The flow rule:: + Will match any ipv4 packet. + The flow rules:: - flow create 0 ingress pattern eth / vlan / ipv4 / end ... + flow create 0 ingress pattern eth / vlan / end ... + flow create 0 ingress pattern eth has_vlan is 1 / end ... + flow create 0 ingress pattern eth type is 0x8100 / end ... - Will match tagged packets only, with any VLAN ID value. - The flow rule:: + Will match single-tagged packets only, with any VLAN ID value. + The flow rules:: - flow create 0 ingress pattern eth / vlan vid is 3 / ipv4 / end ... + flow create 0 ingress pattern eth type is 0x88A8 / end ... + flow create 0 ingress pattern eth / vlan has_more_vlan is 1 / end ... + + Will match multi-tagged packets only, with any VLAN ID value. - Will only match tagged packets with VLAN ID 3. +- A flow pattern with 2 sequential VLAN items is not supported. - VLAN pop offload command:
doc/guides/rel_notes/release_20_11.rst +3 -0
@@ -353,6 +353,9 @@ New Features * Updated the supported timeout for Age action to the maximal value supported by rte_flow API. * Added support of Age action query. + * Added support for QinQ packets matching. + * Added support for the new vlan fields ``has_vlan`` in the eth item and + ``has_more_vlan`` in the vlan item. Removed Items -------------
drivers/net/mlx5/mlx5_flow.c +4 -1
@@ -1794,6 +1794,8 @@ struct mlx5_flow_tunnel_info {* Item specification. * @param[in] item_flags * Bit-fields that holds the items detected until now. +* @param[in] ext_vlan_sup +* Whether extended VLAN features are supported or not. * @param[out] error * Pointer to error structure. *
@@ -1914,7 +1916,7 @@ struct mlx5_flow_tunnel_info {*/ int mlx5_flow_validate_item_eth (const struct rte_flow_item *item, -uint64_t item_flags, +uint64_t item_flags,bool ext_vlan_sup,struct rte_flow_error *error) {const struct rte_flow_item_eth *mask = item->mask;
@@ -1922,6 +1924,7 @@ struct mlx5_flow_tunnel_info { .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff", .src.addr_bytes = "\xff\xff\xff\xff\xff\xff", .type = RTE_BE16(0xffff), + .has_vlan = ext_vlan_sup ? 1 : 0, };int ret;int tunnel = !!(item_flags &MLX5_FLOW_LAYER_TUNNEL );
drivers/net/mlx5/mlx5_flow.h +1 -1
@@ -1306,7 +1306,7 @@ int mlx5_flow_item_acceptable(const struct rte_flow_item *item,bool range_accepted, struct rte_flow_error *error);int mlx5_flow_validate_item_eth (const struct rte_flow_item *item, -uint64_t item_flags, +uint64_t item_flags,bool ext_vlan_sup,struct rte_flow_error *error); int mlx5_flow_validate_item_gre(const struct rte_flow_item *item, uint64_t item_flags,
drivers/net/mlx5/mlx5_flow_dv.c +93 -83
298 lines not on this page · raw message
drivers/net/mlx5/mlx5_flow_verbs.c +1 -1
@@ -1263,7 +1263,7 @@break ;case RTE_FLOW_ITEM_TYPE_ETH : ret =mlx5_flow_validate_item_eth (items, item_flags, - error); + false, error);if (ret <0 )return ret; last_item = tunnel ? MLX5_FLOW_LAYER_INNER_L2 :
Signature · 1 line
1.8.3.1
Hi,
Quoted 534 lines
-----Original Message-----
From: dev <[email protected]> On Behalf Of Matan Azrad
Sent: Sunday, October 25, 2020 6:04 PM
To: Slava Ovsiienko <[email protected]>
Cc: [email protected]
Subject: [dpdk-dev] [PATCH] net/mlx5: support the new VLAN matching fieldsThe fields ``has_vlan`` and ``has_more_vlan`` were added in
rte_flowby patch [1].Using these fields, the application can match all the VLAN options by single flow: any, VLAN only and non-VLAN only.
Add the support for the fields.
By the way, add the support for QinQ packets matching.VLAN\QinQ limitations are listed in the driver document.
[1]
https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc
hes.dpdk.org%2Fpatch%2F80965%2F&data=04%7C01%7Crasland%40nvi dia.com%7C94645c5c01464724cccc08d878ff9f90%7C43083d15727340c1b7db39 efd9ccc17a%7C0%7C0%7C637392386576427312%7CUnknown%7CTWFpbGZs b3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn 0%3D%7C1000&sdata=V7AAqp8X8hOEdSq51wGtPSYVOucQi2gEcFExhua Sjls%3D&reserved=0Signed-off-by Matan Azrad <[email protected]>Acked-by Dekel Peled <[email protected]>Acked-by Ori Kam <[email protected]>--- doc/guides/nics/mlx5.rst | 24 +++-- doc/guides/rel_notes/release_20_11.rst | 3 + drivers/net/mlx5/mlx5_flow.c | 5 +- drivers/net/mlx5/mlx5_flow.h | 2 +- drivers/net/mlx5/mlx5_flow_dv.c | 176 +++++++++++++++++------------- --- drivers/net/mlx5/mlx5_flow_verbs.c | 2 +- 6 files changed, 117 insertions(+), 95 deletions(-)doc/guides/nics/mlx5.rst +0
-1@@ -122,23 +122,29 @@ Limitations Will match any ipv4 packet (VLAN included). -- When using DV flow engine (``dv_flow_en`` = 1), flow pattern withoutVLAN item
- will match untagged packets only.
+- When using Verbs flow engine (``dv_flow_en`` = 0), multi-tagged(QinQ) match is not supported.
+
+- When using DV flow engine (``dv_flow_en`` = 1), flow pattern with any VLAN specification will match only single-tagged packets unless the ETH item ``type`` field is0x88A8or the VLAN item ``has_more_vlan`` field is 1.
The flow rule::flow create 0 ingress pattern eth / ipv4 / end ...
doc/guides/nics/mlx5.rst +12
-7- Will match untagged packets only. - The flow rule:: + Will match any ipv4 packet. + The flow rules:: - flow create 0 ingress pattern eth / vlan / ipv4 / end ... + flow create 0 ingress pattern eth / vlan / end ... + flow create 0 ingress pattern eth has_vlan is 1 / end ... + flow create 0 ingress pattern eth type is 0x8100 / end ... - Will match tagged packets only, with any VLAN ID value. - The flow rule:: + Will match single-tagged packets only, with any VLAN ID value. + The flow rules:: - flow create 0 ingress pattern eth / vlan vid is 3 / ipv4 / end ... + flow create 0 ingress pattern eth type is 0x88A8 / end ... + flow create 0 ingress pattern eth / vlan has_more_vlan is 1 / end ... + + Will match multi-tagged packets only, with any VLAN ID value. - Will only match tagged packets with VLAN ID 3. +- A flow pattern with 2 sequential VLAN items is not supported. - VLAN pop offload command:doc/guides/rel_notes/release_20_11.rst +0
-0@@ -353,6 +353,9 @@ New Features * Updated the supported timeout for Age action to the maximal valuesupported
byrte_flowAPI.
* Added support of Age action query.doc/guides/rel_notes/release_20_11.rst +3
-0+ * Added support for QinQ packets matching. + * Added support for the new vlan fields ``has_vlan`` in the eth item and + ``has_more_vlan`` in the vlan item. Removed Items -------------drivers/net/mlx5/mlx5_flow.c +4
-1@@ -1794,6 +1794,8 @@ struct mlx5_flow_tunnel_info {* Item specification. * @param[in] item_flags * Bit-fields that holds the items detected until now. +* @param[in] ext_vlan_sup +* Whether extended VLAN features are supported or not. * @param[out] error * Pointer to error structure. * @@ -1914,7 +1916,7 @@ struct mlx5_flow_tunnel_info {*/ int mlx5_flow_validate_item_eth (const struct rte_flow_item *item, -uint64_t item_flags, +uint64_t item_flags,bool ext_vlan_sup,struct rte_flow_error *error) {const struct rte_flow_item_eth *mask = item->mask;@@ -1922,6 +1924,7 @@ struct mlx5_flow_tunnel_info { .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff", .src.addr_bytes = "\xff\xff\xff\xff\xff\xff", .type = RTE_BE16(0xffff), + .has_vlan = ext_vlan_sup ? 1 : 0, };int ret;int tunnel = !!(item_flags &MLX5_FLOW_LAYER_TUNNEL );drivers/net/mlx5/mlx5_flow.h +1
-1@@ -1306,7 +1306,7 @@ int mlx5_flow_item_acceptable(const structrte_flow_item *item, bool range_accepted, struct rte_flow_error *error);int mlx5_flow_validate_item_eth (const struct rte_flow_item *item, -uint64_t item_flags, +uint64_t item_flags,bool ext_vlan_sup,struct rte_flow_error *error); int mlx5_flow_validate_item_gre(const struct rte_flow_item *item,drivers/net/mlx5/mlx5_flow.h +0
-0uint64_t item_flags,drivers/net/mlx5/mlx5_flow_dv.c +2
-1@@ -1676,6 +1676,7 @@ struct field_modify_info modify_tcp[] = {const struct rte_flow_item_vlan nic_mask = { .tci =RTE_BE16 (UINT16_MAX ), .inner_type =RTE_BE16 (UINT16_MAX ), + .has_more_vlan =1 , };const int tunnel = !!(item_flags &MLX5_FLOW_LAYER_TUNNEL );int ret;@@ -5358,7 +5359,7 @@ struct field_modify_info modify_tcp[] = {break ;case RTE_FLOW_ITEM_TYPE_ETH : ret =mlx5_flow_validate_item_eth (items, item_flags, - error); + true, error);if (ret <0 )return ret;drivers/net/mlx5/mlx5_flow_dv.c +9
-9last_item = tunnel ? MLX5_FLOW_LAYER_INNER_L2 :@@ -6360,9 +6361,10 @@ struct field_modify_info modify_tcp[] = { .dst.addr_bytes = "\xff\xff\xff\xff\xff\xff", .src.addr_bytes = "\xff\xff\xff\xff\xff\xff", .type = RTE_BE16(0xffff), + .has_vlan = 0, }; -void *headers_m; -void *headers_v; +void *hdrs_m; +void *hdrs_v;char *l24_v;unsigned int i;@@ -6371,38 +6373,26 @@ struct field_modify_info modify_tcp[] = {if (!eth_m) eth_m = &nic_mask;if (inner) { - headers_m =MLX5_ADDR_OF (fte_match_param, matcher, + hdrs_m =MLX5_ADDR_OF (fte_match_param, matcher, inner_headers); - headers_v =MLX5_ADDR_OF (fte_match_param, key, inner_headers); + hdrs_v =MLX5_ADDR_OF (fte_match_param, key, inner_headers); }else { - headers_m =MLX5_ADDR_OF (fte_match_param, matcher, + hdrs_m =MLX5_ADDR_OF (fte_match_param, matcher, outer_headers); - headers_v =MLX5_ADDR_OF (fte_match_param, key, outer_headers); + hdrs_v =MLX5_ADDR_OF (fte_match_param, key, outer_headers); } -memcpy (MLX5_ADDR_OF (fte_match_set_lyr_2_4, headers_m, dmac_47_16), +memcpy (MLX5_ADDR_OF (fte_match_set_lyr_2_4, hdrs_m, dmac_47_16), ð_m->dst,sizeof (eth_m->dst));/* The value must be in the range of the mask. */ - l24_v =MLX5_ADDR_OF (fte_match_set_lyr_2_4, headers_v, dmac_47_16); + l24_v =MLX5_ADDR_OF (fte_match_set_lyr_2_4, hdrs_v, dmac_47_16);for (i =0 ; i <sizeof (eth_m->dst); ++i) l24_v[i] = eth_m->dst.addr_bytes[i] & eth_v- >dst.addr_bytes[i]; - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_m,
smac_47_16),drivers/net/mlx5/mlx5_flow_dv.c +1
-0+ memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, hdrs_m,
smac_47_16),drivers/net/mlx5/mlx5_flow_dv.c +0
-1ð_m->src, sizeof(eth_m->src)); - l24_v = MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
smac_47_16);drivers/net/mlx5/mlx5_flow_dv.c +1
-0+ l24_v = MLX5_ADDR_OF(fte_match_set_lyr_2_4, hdrs_v,
smac_47_16);drivers/net/mlx5/mlx5_flow_dv.c +0
-0/* The value must be in the range of the mask. */ for (i =0 ; i <sizeof (eth_m->dst); ++i) l24_v[i] = eth_m->src.addr_bytes[i] & eth_v-Quoted 1 line
src.addr_bytes[i];
drivers/net/mlx5/mlx5_flow_dv.c +0
-3-if (eth_v->type) { -/* When ethertype is present set mask for tagged VLAN. */ - MLX5_SET(fte_match_set_lyr_2_4, headers_m, cvlan_tag,1);
drivers/net/mlx5/mlx5_flow_dv.c +0
-4-/* Set value for tagged VLAN if ethertype is 802.1Q. */ -if (eth_v->type ==RTE_BE16 (RTE_ETHER_TYPE_VLAN ) || - eth_v->type ==RTE_BE16 (RTE_ETHER_TYPE_QINQ )) { - MLX5_SET(fte_match_set_lyr_2_4, headers_v,
cvlan_tag,drivers/net/mlx5/mlx5_flow_dv.c +0
-2- 1); -/* Return here to avoid setting match on ethertype. */
drivers/net/mlx5/mlx5_flow_dv.c +0
-3-return ; - } - }/* * HW supports match on one Ethertype, the Ethertype following the last
drivers/net/mlx5/mlx5_flow_dv.c +10
-12* VLAN tag of the packet (see PRM). @@ -6411,19 +6401,42 @@ struct field_modify_info modify_tcp[] = {* ethertype, and use ip_version field instead. * eCPRI over Ether layer will use type value 0xAEFE. */ -if (eth_v->type ==RTE_BE16 (RTE_ETHER_TYPE_IPV4 ) && - eth_m->type ==0xFFFF ) { -flow_dv_set_match_ip_version (group, headers_v, headers_m, 4); - }else if (eth_v->type ==RTE_BE16 (RTE_ETHER_TYPE_IPV6 ) && - eth_m->type ==0xFFFF ) { -flow_dv_set_match_ip_version (group, headers_v, headers_m, 6); - }else { -MLX5_SET (fte_match_set_lyr_2_4, headers_m, ethertype, -rte_be_to_cpu_16 (eth_m->type)); - l24_v =MLX5_ADDR_OF (fte_match_set_lyr_2_4, headers_v, - ethertype); - *(uint16_t *)(l24_v) = eth_m->type & eth_v->type; +if (eth_m->type ==0xFFFF ) { +/* Set cvlan_tag mask for any single\multi\un-tagged case. */ +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, cvlan_tag,1 ); +switch (eth_v->type) { +case RTE_BE16 (RTE_ETHER_TYPE_VLAN ): +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, cvlan_tag,1 ); +return ; +case RTE_BE16 (RTE_ETHER_TYPE_QINQ ): +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, svlan_tag,1 ); + MLX5_SET(fte_match_set_lyr_2_4, hdrs_v,
svlan_tag, 1);drivers/net/mlx5/mlx5_flow_dv.c +3
-0+return ; +case RTE_BE16 (RTE_ETHER_TYPE_IPV4 ): + flow_dv_set_match_ip_version(group, hdrs_v,
hdrs_m, 4);drivers/net/mlx5/mlx5_flow_dv.c +3
-0+return ; +case RTE_BE16 (RTE_ETHER_TYPE_IPV6 ): + flow_dv_set_match_ip_version(group, hdrs_v,
hdrs_m, 6);drivers/net/mlx5/mlx5_flow_dv.c +10
-0+return ; +default : +break ; + } + } +if (eth_m->has_vlan) { +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, cvlan_tag,1 ); +if (eth_v->has_vlan) { +/* +* Here, when also has_more_vlan field in VLAN item is
drivers/net/mlx5/mlx5_flow_dv.c +1
-0+* not set, only single-tagged packets will be matched.
drivers/net/mlx5/mlx5_flow_dv.c +2
-0+*/ + MLX5_SET(fte_match_set_lyr_2_4, hdrs_v,
cvlan_tag, 1);drivers/net/mlx5/mlx5_flow_dv.c +5
-0+return ; + } } +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, ethertype, +rte_be_to_cpu_16 (eth_m->type)); + l24_v = MLX5_ADDR_OF(fte_match_set_lyr_2_4, hdrs_v,ethertype);
drivers/net/mlx5/mlx5_flow_dv.c +7
-6+ *(uint16_t *)(l24_v) = eth_m->type & eth_v->type; }/** @@ -6448,19 +6461,19 @@ struct field_modify_info modify_tcp[] = { {const struct rte_flow_item_vlan *vlan_m = item->mask;const struct rte_flow_item_vlan *vlan_v = item->spec; -void *headers_m; -void *headers_v; +void *hdrs_m; +void *hdrs_v;uint16_t tci_m;uint16_t tci_v;if (inner) { - headers_m =MLX5_ADDR_OF (fte_match_param, matcher, + hdrs_m =MLX5_ADDR_OF (fte_match_param, matcher, inner_headers); - headers_v =MLX5_ADDR_OF (fte_match_param, key, inner_headers); + hdrs_v =MLX5_ADDR_OF (fte_match_param, key, inner_headers); }else { - headers_m =MLX5_ADDR_OF (fte_match_param, matcher, + hdrs_m =MLX5_ADDR_OF (fte_match_param, matcher, outer_headers); - headers_v =MLX5_ADDR_OF (fte_match_param, key, outer_headers); + hdrs_v = MLX5_ADDR_OF(fte_match_param, key,
outer_headers);drivers/net/mlx5/mlx5_flow_dv.c +10
-12/* * This is workaround, masks are not supported, * and pre-validated. @@ -6473,37 +6486,54 @@ struct field_modify_info modify_tcp[] = {* When VLAN item exists in flow, mark packet as tagged, * even if TCI is not specified. */ -MLX5_SET (fte_match_set_lyr_2_4, headers_m, cvlan_tag,1 ); -MLX5_SET (fte_match_set_lyr_2_4, headers_v, cvlan_tag,1 ); +if (!MLX5_GET (fte_match_set_lyr_2_4, hdrs_v, svlan_tag)) { +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, cvlan_tag,1 ); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, cvlan_tag,1 ); + }if (!vlan_v)return ;if (!vlan_m) vlan_m = &rte_flow_item_vlan_mask; tci_m =rte_be_to_cpu_16 (vlan_m->tci); tci_v =rte_be_to_cpu_16 (vlan_m->tci & vlan_v->tci); -MLX5_SET (fte_match_set_lyr_2_4, headers_m, first_vid, tci_m); -MLX5_SET (fte_match_set_lyr_2_4, headers_v, first_vid, tci_v); -MLX5_SET (fte_match_set_lyr_2_4, headers_m, first_cfi, tci_m >> 12); -MLX5_SET (fte_match_set_lyr_2_4, headers_v, first_cfi, tci_v >>12 ); -MLX5_SET (fte_match_set_lyr_2_4, headers_m, first_prio, tci_m >> 13); -MLX5_SET (fte_match_set_lyr_2_4, headers_v, first_prio, tci_v >> 13); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, first_vid, tci_m); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, first_vid, tci_v); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, first_cfi, tci_m >>12 ); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, first_cfi, tci_v >>12 ); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, first_prio, tci_m >>13 ); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, first_prio, tci_v >>13 );/* * HW is optimized for IPv4/IPv6. In such cases, avoid setting * ethertype, and use ip_version field instead. */ -if (vlan_v->inner_type ==RTE_BE16 (RTE_ETHER_TYPE_IPV4 ) && - vlan_m->inner_type ==0xFFFF ) { -flow_dv_set_match_ip_version (group, headers_v, headers_m, 4); - } else if (vlan_v->inner_type == RTE_BE16(RTE_ETHER_TYPE_IPV6)&&
drivers/net/mlx5/mlx5_flow_dv.c +0
-2- vlan_m->inner_type == 0xFFFF) { - flow_dv_set_match_ip_version(group, headers_v,
headers_m, 6);drivers/net/mlx5/mlx5_flow_dv.c +4
-6- }else { -MLX5_SET (fte_match_set_lyr_2_4, headers_m, ethertype, -rte_be_to_cpu_16 (vlan_m->inner_type)); -MLX5_SET (fte_match_set_lyr_2_4, headers_v, ethertype, -rte_be_to_cpu_16 (vlan_m->inner_type & - vlan_v->inner_type)); +if (vlan_m->inner_type ==0xFFFF ) { +switch (vlan_v->inner_type) { +case RTE_BE16 (RTE_ETHER_TYPE_VLAN ): + MLX5_SET(fte_match_set_lyr_2_4, hdrs_m,
svlan_tag, 1);drivers/net/mlx5/mlx5_flow_dv.c +1
-0+ MLX5_SET(fte_match_set_lyr_2_4, hdrs_v,
svlan_tag, 1);drivers/net/mlx5/mlx5_flow_dv.c +1
-0+ MLX5_SET(fte_match_set_lyr_2_4, hdrs_v,
cvlan_tag, 0);drivers/net/mlx5/mlx5_flow_dv.c +3
-0+return ; +case RTE_BE16 (RTE_ETHER_TYPE_IPV4 ): + flow_dv_set_match_ip_version(group, hdrs_v,
hdrs_m, 4);drivers/net/mlx5/mlx5_flow_dv.c +3
-0+return ; +case RTE_BE16 (RTE_ETHER_TYPE_IPV6 ): + flow_dv_set_match_ip_version(group, hdrs_v,
hdrs_m, 6);drivers/net/mlx5/mlx5_flow_dv.c +15
-0+return ; +default : +break ; + } } +if (vlan_m->has_more_vlan && vlan_v->has_more_vlan) { +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, svlan_tag,1 ); +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, svlan_tag,1 ); +/* Only one vlan_tag bit can be set. */ +MLX5_SET (fte_match_set_lyr_2_4, hdrs_v, cvlan_tag,0 ); +return ; + } +MLX5_SET (fte_match_set_lyr_2_4, hdrs_m, ethertype, +rte_be_to_cpu_16 (vlan_m->inner_type)); + MLX5_SET(fte_match_set_lyr_2_4, hdrs_v, ethertype, + rte_be_to_cpu_16(vlan_m->inner_type & vlan_v-Quoted 1 line
inner_type));}
/**
drivers/net/mlx5/mlx5_flow_dv.c +0
-10@@ -6515,8 +6545,6 @@ struct field_modify_info modify_tcp[] = {* Flow matcher value. * @param[in] item * Flow pattern to translate. -* @param[in] item_flags -* Bit-fields that holds the items detected until now. * @param[in] inner * Item is inner pattern. * @param[in] group @@ -6525,7 +6553,6 @@ struct field_modify_info modify_tcp[] = {static void flow_dv_translate_item_ipv4 (void *matcher,void *key,const struct rte_flow_item *item, -const uint64_t item_flags,int inner,uint32_t group) {const struct rte_flow_item_ipv4 *ipv4_m = item->mask;@@ -6555,13 +6582,6 @@ struct field_modify_info modify_tcp[] = { headers_v =MLX5_ADDR_OF (fte_match_param, key, outer_headers); }flow_dv_set_match_ip_version (group, headers_v, headers_m,4 ); -/* -* On outer header (which must contains L2), or inner header with L2, -* set cvlan_tag mask bit to mark this packet as untagged. -* This should be done even if item->spec is empty. -*/ -if (!inner || item_flags &MLX5_FLOW_LAYER_INNER_L2 ) -MLX5_SET (fte_match_set_lyr_2_4, headers_m, cvlan_tag, 1);if (!ipv4_v)drivers/net/mlx5/mlx5_flow_dv.c +0
-10return ;if (!ipv4_m)@@ -6608,8 +6628,6 @@ struct field_modify_info modify_tcp[] = {* Flow matcher value. * @param[in] item * Flow pattern to translate. -* @param[in] item_flags -* Bit-fields that holds the items detected until now. * @param[in] inner * Item is inner pattern. * @param[in] group @@ -6618,7 +6636,6 @@ struct field_modify_info modify_tcp[] = {static void flow_dv_translate_item_ipv6 (void *matcher,void *key,const struct rte_flow_item *item, -const uint64_t item_flags,int inner,uint32_t group) {const struct rte_flow_item_ipv6 *ipv6_m = item->mask;@@ -6657,13 +6674,6 @@ struct field_modify_info modify_tcp[] = { headers_v =MLX5_ADDR_OF (fte_match_param, key, outer_headers); }flow_dv_set_match_ip_version (group, headers_v, headers_m,6 ); -/* -* On outer header (which must contains L2), or inner header with L2, -* set cvlan_tag mask bit to mark this packet as untagged. -* This should be done even if item->spec is empty. -*/ -if (!inner || item_flags &MLX5_FLOW_LAYER_INNER_L2 ) -MLX5_SET (fte_match_set_lyr_2_4, headers_m, cvlan_tag, 1);if (!ipv6_v)drivers/net/mlx5/mlx5_flow_dv.c +1
-1return ;if (!ipv6_m)@@ -9915,7 +9925,7 @@ struct field_modify_info modify_tcp[] = {mlx5_flow_tunnel_ip_check (items, next_protocol, &item_flags, &tunnel);flow_dv_translate_item_ipv4 (match_mask, match_value, - items, item_flags, tunnel, + items, tunnel, dev_flow->dv.group); matcher.priority =MLX5_PRIORITY_MAP_L3 ;
last_item= tunnel ?MLX5_FLOW_LAYER_INNER_L3_IPV4:drivers/net/mlx5/mlx5_flow_dv.c +1
-1@@ -9938,7 +9948,7 @@ struct field_modify_info modify_tcp[] = {mlx5_flow_tunnel_ip_check (items, next_protocol, &item_flags, &tunnel);flow_dv_translate_item_ipv6 (match_mask, match_value, - items, item_flags, tunnel, + items, tunnel, dev_flow->dv.group); matcher.priority =MLX5_PRIORITY_MAP_L3 ;
last_item= tunnel ?MLX5_FLOW_LAYER_INNER_L3_IPV6:drivers/net/mlx5/mlx5_flow_verbs.c +1
-1@@ -1263,7 +1263,7 @@break ;case RTE_FLOW_ITEM_TYPE_ETH : ret =mlx5_flow_validate_item_eth (items, item_flags, - error); + false, error);if (ret <0 )return ret;
Quoted 3 lines
last_item= tunnel ?MLX5_FLOW_LAYER_INNER_L2:
--
1.8.3.1
Patch applied to next-net-mlx,
Kindest regards,
Raslan Darawsheh