* [net-next V2 09/18] net/mlx5: DR, Expose steering matcher functionality
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Mark Bloch,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
Matcher defines which packets fields are matched when a packet arrives.
Matcher is a part of a table and can contain one or more rules. Where
rule defines specific values of the matcher's mask definition.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/dr_matcher.c | 770 ++++++++++++++++++
1 file changed, 770 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c
new file mode 100644
index 000000000000..01008cd66f75
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c
@@ -0,0 +1,770 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#include "dr_types.h"
+
+static bool dr_mask_is_smac_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->smac_47_16 || spec->smac_15_0);
+}
+
+static bool dr_mask_is_dmac_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->dmac_47_16 || spec->dmac_15_0);
+}
+
+static bool dr_mask_is_src_addr_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->src_ip_127_96 || spec->src_ip_95_64 ||
+ spec->src_ip_63_32 || spec->src_ip_31_0);
+}
+
+static bool dr_mask_is_dst_addr_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->dst_ip_127_96 || spec->dst_ip_95_64 ||
+ spec->dst_ip_63_32 || spec->dst_ip_31_0);
+}
+
+static bool dr_mask_is_l3_base_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->ip_protocol || spec->frag || spec->tcp_flags ||
+ spec->ip_ecn || spec->ip_dscp);
+}
+
+static bool dr_mask_is_tcp_udp_base_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->tcp_sport || spec->tcp_dport ||
+ spec->udp_sport || spec->udp_dport);
+}
+
+static bool dr_mask_is_ipv4_set(struct mlx5dr_match_spec *spec)
+{
+ return (spec->dst_ip_31_0 || spec->src_ip_31_0);
+}
+
+static bool dr_mask_is_ipv4_5_tuple_set(struct mlx5dr_match_spec *spec)
+{
+ return (dr_mask_is_l3_base_set(spec) ||
+ dr_mask_is_tcp_udp_base_set(spec) ||
+ dr_mask_is_ipv4_set(spec));
+}
+
+static bool dr_mask_is_eth_l2_tnl_set(struct mlx5dr_match_misc *misc)
+{
+ return misc->vxlan_vni;
+}
+
+static bool dr_mask_is_ttl_set(struct mlx5dr_match_spec *spec)
+{
+ return spec->ttl_hoplimit;
+}
+
+#define DR_MASK_IS_L2_DST(_spec, _misc, _inner_outer) (_spec.first_vid || \
+ (_spec).first_cfi || (_spec).first_prio || (_spec).cvlan_tag || \
+ (_spec).svlan_tag || (_spec).dmac_47_16 || (_spec).dmac_15_0 || \
+ (_spec).ethertype || (_spec).ip_version || \
+ (_misc)._inner_outer##_second_vid || \
+ (_misc)._inner_outer##_second_cfi || \
+ (_misc)._inner_outer##_second_prio || \
+ (_misc)._inner_outer##_second_cvlan_tag || \
+ (_misc)._inner_outer##_second_svlan_tag)
+
+#define DR_MASK_IS_ETH_L4_SET(_spec, _misc, _inner_outer) ( \
+ dr_mask_is_l3_base_set(&(_spec)) || \
+ dr_mask_is_tcp_udp_base_set(&(_spec)) || \
+ dr_mask_is_ttl_set(&(_spec)) || \
+ (_misc)._inner_outer##_ipv6_flow_label)
+
+#define DR_MASK_IS_ETH_L4_MISC_SET(_misc3, _inner_outer) ( \
+ (_misc3)._inner_outer##_tcp_seq_num || \
+ (_misc3)._inner_outer##_tcp_ack_num)
+
+#define DR_MASK_IS_FIRST_MPLS_SET(_misc2, _inner_outer) ( \
+ (_misc2)._inner_outer##_first_mpls_label || \
+ (_misc2)._inner_outer##_first_mpls_exp || \
+ (_misc2)._inner_outer##_first_mpls_s_bos || \
+ (_misc2)._inner_outer##_first_mpls_ttl)
+
+static bool dr_mask_is_gre_set(struct mlx5dr_match_misc *misc)
+{
+ return (misc->gre_key_h || misc->gre_key_l ||
+ misc->gre_protocol || misc->gre_c_present ||
+ misc->gre_k_present || misc->gre_s_present);
+}
+
+#define DR_MASK_IS_OUTER_MPLS_OVER_GRE_UDP_SET(_misc2, gre_udp) ( \
+ (_misc2).outer_first_mpls_over_##gre_udp##_label || \
+ (_misc2).outer_first_mpls_over_##gre_udp##_exp || \
+ (_misc2).outer_first_mpls_over_##gre_udp##_s_bos || \
+ (_misc2).outer_first_mpls_over_##gre_udp##_ttl)
+
+#define DR_MASK_IS_FLEX_PARSER_0_SET(_misc2) ( \
+ DR_MASK_IS_OUTER_MPLS_OVER_GRE_UDP_SET((_misc2), gre) || \
+ DR_MASK_IS_OUTER_MPLS_OVER_GRE_UDP_SET((_misc2), udp))
+
+static bool dr_mask_is_flex_parser_tnl_set(struct mlx5dr_match_misc3 *misc3)
+{
+ return (misc3->outer_vxlan_gpe_vni ||
+ misc3->outer_vxlan_gpe_next_protocol ||
+ misc3->outer_vxlan_gpe_flags);
+}
+
+static bool dr_mask_is_flex_parser_icmpv6_set(struct mlx5dr_match_misc3 *misc3)
+{
+ return (misc3->icmpv6_type || misc3->icmpv6_code ||
+ misc3->icmpv6_header_data);
+}
+
+static bool dr_mask_is_wqe_metadata_set(struct mlx5dr_match_misc2 *misc2)
+{
+ return misc2->metadata_reg_a;
+}
+
+static bool dr_mask_is_reg_c_0_3_set(struct mlx5dr_match_misc2 *misc2)
+{
+ return (misc2->metadata_reg_c_0 || misc2->metadata_reg_c_1 ||
+ misc2->metadata_reg_c_2 || misc2->metadata_reg_c_3);
+}
+
+static bool dr_mask_is_reg_c_4_7_set(struct mlx5dr_match_misc2 *misc2)
+{
+ return (misc2->metadata_reg_c_4 || misc2->metadata_reg_c_5 ||
+ misc2->metadata_reg_c_6 || misc2->metadata_reg_c_7);
+}
+
+static bool dr_mask_is_gvmi_or_qpn_set(struct mlx5dr_match_misc *misc)
+{
+ return (misc->source_sqn || misc->source_port);
+}
+
+static bool
+dr_matcher_supp_flex_parser_vxlan_gpe(struct mlx5dr_domain *dmn)
+{
+ return dmn->info.caps.flex_protocols &
+ MLX5_FLEX_PARSER_VXLAN_GPE_ENABLED;
+}
+
+int mlx5dr_matcher_select_builders(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ bool ipv6)
+{
+ if (ipv6) {
+ nic_matcher->ste_builder = nic_matcher->ste_builder6;
+ nic_matcher->num_of_builders = nic_matcher->num_of_builders6;
+ } else {
+ nic_matcher->ste_builder = nic_matcher->ste_builder4;
+ nic_matcher->num_of_builders = nic_matcher->num_of_builders4;
+ }
+
+ if (!nic_matcher->num_of_builders) {
+ mlx5dr_dbg(matcher->tbl->dmn,
+ "Rule not supported on this matcher due to IP related fields\n");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ bool ipv6)
+{
+ struct mlx5dr_domain_rx_tx *nic_dmn = nic_matcher->nic_tbl->nic_dmn;
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_match_param mask = {};
+ struct mlx5dr_match_misc3 *misc3;
+ struct mlx5dr_ste_build *sb;
+ u8 *num_of_builders;
+ bool inner, rx;
+ int idx = 0;
+ int ret, i;
+
+ if (ipv6) {
+ sb = nic_matcher->ste_builder6;
+ num_of_builders = &nic_matcher->num_of_builders6;
+ } else {
+ sb = nic_matcher->ste_builder4;
+ num_of_builders = &nic_matcher->num_of_builders4;
+ }
+
+ rx = nic_dmn->ste_type == MLX5DR_STE_TYPE_RX;
+
+ /* Create a temporary mask to track and clear used mask fields */
+ if (matcher->match_criteria & DR_MATCHER_CRITERIA_OUTER)
+ mask.outer = matcher->mask.outer;
+
+ if (matcher->match_criteria & DR_MATCHER_CRITERIA_MISC)
+ mask.misc = matcher->mask.misc;
+
+ if (matcher->match_criteria & DR_MATCHER_CRITERIA_INNER)
+ mask.inner = matcher->mask.inner;
+
+ if (matcher->match_criteria & DR_MATCHER_CRITERIA_MISC2)
+ mask.misc2 = matcher->mask.misc2;
+
+ if (matcher->match_criteria & DR_MATCHER_CRITERIA_MISC3)
+ mask.misc3 = matcher->mask.misc3;
+
+ ret = mlx5dr_ste_build_pre_check(dmn, matcher->match_criteria,
+ &matcher->mask, NULL);
+ if (ret)
+ return ret;
+
+ /* Outer */
+ if (matcher->match_criteria & (DR_MATCHER_CRITERIA_OUTER |
+ DR_MATCHER_CRITERIA_MISC |
+ DR_MATCHER_CRITERIA_MISC2 |
+ DR_MATCHER_CRITERIA_MISC3)) {
+ inner = false;
+
+ if (dr_mask_is_wqe_metadata_set(&mask.misc2))
+ mlx5dr_ste_build_general_purpose(&sb[idx++], &mask, inner, rx);
+
+ if (dr_mask_is_reg_c_0_3_set(&mask.misc2))
+ mlx5dr_ste_build_register_0(&sb[idx++], &mask, inner, rx);
+
+ if (dr_mask_is_reg_c_4_7_set(&mask.misc2))
+ mlx5dr_ste_build_register_1(&sb[idx++], &mask, inner, rx);
+
+ if (dr_mask_is_gvmi_or_qpn_set(&mask.misc) &&
+ (dmn->type == MLX5DR_DOMAIN_TYPE_FDB ||
+ dmn->type == MLX5DR_DOMAIN_TYPE_NIC_RX)) {
+ ret = mlx5dr_ste_build_src_gvmi_qpn(&sb[idx++], &mask,
+ &dmn->info.caps,
+ inner, rx);
+ if (ret)
+ return ret;
+ }
+
+ if (dr_mask_is_smac_set(&mask.outer) &&
+ dr_mask_is_dmac_set(&mask.outer)) {
+ ret = mlx5dr_ste_build_eth_l2_src_des(&sb[idx++], &mask,
+ inner, rx);
+ if (ret)
+ return ret;
+ }
+
+ if (dr_mask_is_smac_set(&mask.outer))
+ mlx5dr_ste_build_eth_l2_src(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_L2_DST(mask.outer, mask.misc, outer))
+ mlx5dr_ste_build_eth_l2_dst(&sb[idx++], &mask, inner, rx);
+
+ if (ipv6) {
+ if (dr_mask_is_dst_addr_set(&mask.outer))
+ mlx5dr_ste_build_eth_l3_ipv6_dst(&sb[idx++], &mask,
+ inner, rx);
+
+ if (dr_mask_is_src_addr_set(&mask.outer))
+ mlx5dr_ste_build_eth_l3_ipv6_src(&sb[idx++], &mask,
+ inner, rx);
+
+ if (DR_MASK_IS_ETH_L4_SET(mask.outer, mask.misc, outer))
+ mlx5dr_ste_build_ipv6_l3_l4(&sb[idx++], &mask,
+ inner, rx);
+ } else {
+ if (dr_mask_is_ipv4_5_tuple_set(&mask.outer))
+ mlx5dr_ste_build_eth_l3_ipv4_5_tuple(&sb[idx++], &mask,
+ inner, rx);
+
+ if (dr_mask_is_ttl_set(&mask.outer))
+ mlx5dr_ste_build_eth_l3_ipv4_misc(&sb[idx++], &mask,
+ inner, rx);
+ }
+
+ if (dr_mask_is_flex_parser_tnl_set(&mask.misc3) &&
+ dr_matcher_supp_flex_parser_vxlan_gpe(dmn))
+ mlx5dr_ste_build_flex_parser_tnl(&sb[idx++], &mask,
+ inner, rx);
+
+ if (DR_MASK_IS_ETH_L4_MISC_SET(mask.misc3, outer))
+ mlx5dr_ste_build_eth_l4_misc(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_FIRST_MPLS_SET(mask.misc2, outer))
+ mlx5dr_ste_build_mpls(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_FLEX_PARSER_0_SET(mask.misc2))
+ mlx5dr_ste_build_flex_parser_0(&sb[idx++], &mask,
+ inner, rx);
+
+ misc3 = &mask.misc3;
+ if ((DR_MASK_IS_FLEX_PARSER_ICMPV4_SET(misc3) &&
+ mlx5dr_matcher_supp_flex_parser_icmp_v4(&dmn->info.caps)) ||
+ (dr_mask_is_flex_parser_icmpv6_set(&mask.misc3) &&
+ mlx5dr_matcher_supp_flex_parser_icmp_v6(&dmn->info.caps))) {
+ ret = mlx5dr_ste_build_flex_parser_1(&sb[idx++],
+ &mask, &dmn->info.caps,
+ inner, rx);
+ if (ret)
+ return ret;
+ }
+ if (dr_mask_is_gre_set(&mask.misc))
+ mlx5dr_ste_build_gre(&sb[idx++], &mask, inner, rx);
+ }
+
+ /* Inner */
+ if (matcher->match_criteria & (DR_MATCHER_CRITERIA_INNER |
+ DR_MATCHER_CRITERIA_MISC |
+ DR_MATCHER_CRITERIA_MISC2 |
+ DR_MATCHER_CRITERIA_MISC3)) {
+ inner = true;
+
+ if (dr_mask_is_eth_l2_tnl_set(&mask.misc))
+ mlx5dr_ste_build_eth_l2_tnl(&sb[idx++], &mask, inner, rx);
+
+ if (dr_mask_is_smac_set(&mask.inner) &&
+ dr_mask_is_dmac_set(&mask.inner)) {
+ ret = mlx5dr_ste_build_eth_l2_src_des(&sb[idx++],
+ &mask, inner, rx);
+ if (ret)
+ return ret;
+ }
+
+ if (dr_mask_is_smac_set(&mask.inner))
+ mlx5dr_ste_build_eth_l2_src(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_L2_DST(mask.inner, mask.misc, inner))
+ mlx5dr_ste_build_eth_l2_dst(&sb[idx++], &mask, inner, rx);
+
+ if (ipv6) {
+ if (dr_mask_is_dst_addr_set(&mask.inner))
+ mlx5dr_ste_build_eth_l3_ipv6_dst(&sb[idx++], &mask,
+ inner, rx);
+
+ if (dr_mask_is_src_addr_set(&mask.inner))
+ mlx5dr_ste_build_eth_l3_ipv6_src(&sb[idx++], &mask,
+ inner, rx);
+
+ if (DR_MASK_IS_ETH_L4_SET(mask.inner, mask.misc, inner))
+ mlx5dr_ste_build_ipv6_l3_l4(&sb[idx++], &mask,
+ inner, rx);
+ } else {
+ if (dr_mask_is_ipv4_5_tuple_set(&mask.inner))
+ mlx5dr_ste_build_eth_l3_ipv4_5_tuple(&sb[idx++], &mask,
+ inner, rx);
+
+ if (dr_mask_is_ttl_set(&mask.inner))
+ mlx5dr_ste_build_eth_l3_ipv4_misc(&sb[idx++], &mask,
+ inner, rx);
+ }
+
+ if (DR_MASK_IS_ETH_L4_MISC_SET(mask.misc3, inner))
+ mlx5dr_ste_build_eth_l4_misc(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_FIRST_MPLS_SET(mask.misc2, inner))
+ mlx5dr_ste_build_mpls(&sb[idx++], &mask, inner, rx);
+
+ if (DR_MASK_IS_FLEX_PARSER_0_SET(mask.misc2))
+ mlx5dr_ste_build_flex_parser_0(&sb[idx++], &mask, inner, rx);
+ }
+ /* Empty matcher, takes all */
+ if (matcher->match_criteria == DR_MATCHER_CRITERIA_EMPTY)
+ mlx5dr_ste_build_empty_always_hit(&sb[idx++], rx);
+
+ if (idx == 0) {
+ mlx5dr_dbg(dmn, "Cannot generate any valid rules from mask\n");
+ return -EINVAL;
+ }
+
+ /* Check that all mask fields were consumed */
+ for (i = 0; i < sizeof(struct mlx5dr_match_param); i++) {
+ if (((u8 *)&mask)[i] != 0) {
+ mlx5dr_info(dmn, "Mask contains unsupported parameters\n");
+ return -EOPNOTSUPP;
+ }
+ }
+
+ *num_of_builders = idx;
+
+ return 0;
+}
+
+static int dr_matcher_connect(struct mlx5dr_domain *dmn,
+ struct mlx5dr_matcher_rx_tx *curr_nic_matcher,
+ struct mlx5dr_matcher_rx_tx *next_nic_matcher,
+ struct mlx5dr_matcher_rx_tx *prev_nic_matcher)
+{
+ struct mlx5dr_table_rx_tx *nic_tbl = curr_nic_matcher->nic_tbl;
+ struct mlx5dr_domain_rx_tx *nic_dmn = nic_tbl->nic_dmn;
+ struct mlx5dr_htbl_connect_info info;
+ struct mlx5dr_ste_htbl *prev_htbl;
+ int ret;
+
+ /* Connect end anchor hash table to next_htbl or to the default address */
+ if (next_nic_matcher) {
+ info.type = CONNECT_HIT;
+ info.hit_next_htbl = next_nic_matcher->s_htbl;
+ } else {
+ info.type = CONNECT_MISS;
+ info.miss_icm_addr = nic_tbl->default_icm_addr;
+ }
+ ret = mlx5dr_ste_htbl_init_and_postsend(dmn, nic_dmn,
+ curr_nic_matcher->e_anchor,
+ &info, info.type == CONNECT_HIT);
+ if (ret)
+ return ret;
+
+ /* Connect start hash table to end anchor */
+ info.type = CONNECT_MISS;
+ info.miss_icm_addr = curr_nic_matcher->e_anchor->chunk->icm_addr;
+ ret = mlx5dr_ste_htbl_init_and_postsend(dmn, nic_dmn,
+ curr_nic_matcher->s_htbl,
+ &info, false);
+ if (ret)
+ return ret;
+
+ /* Connect previous hash table to matcher start hash table */
+ if (prev_nic_matcher)
+ prev_htbl = prev_nic_matcher->e_anchor;
+ else
+ prev_htbl = nic_tbl->s_anchor;
+
+ info.type = CONNECT_HIT;
+ info.hit_next_htbl = curr_nic_matcher->s_htbl;
+ ret = mlx5dr_ste_htbl_init_and_postsend(dmn, nic_dmn, prev_htbl,
+ &info, true);
+ if (ret)
+ return ret;
+
+ /* Update the pointing ste and next hash table */
+ curr_nic_matcher->s_htbl->pointing_ste = prev_htbl->ste_arr;
+ prev_htbl->ste_arr[0].next_htbl = curr_nic_matcher->s_htbl;
+
+ if (next_nic_matcher) {
+ next_nic_matcher->s_htbl->pointing_ste = curr_nic_matcher->e_anchor->ste_arr;
+ curr_nic_matcher->e_anchor->ste_arr[0].next_htbl = next_nic_matcher->s_htbl;
+ }
+
+ return 0;
+}
+
+static int dr_matcher_add_to_tbl(struct mlx5dr_matcher *matcher)
+{
+ struct mlx5dr_matcher *next_matcher, *prev_matcher, *tmp_matcher;
+ struct mlx5dr_table *tbl = matcher->tbl;
+ struct mlx5dr_domain *dmn = tbl->dmn;
+ bool first = true;
+ int ret;
+
+ next_matcher = NULL;
+ if (!list_empty(&tbl->matcher_list))
+ list_for_each_entry(tmp_matcher, &tbl->matcher_list, matcher_list) {
+ if (tmp_matcher->prio >= matcher->prio) {
+ next_matcher = tmp_matcher;
+ break;
+ }
+ first = false;
+ }
+
+ prev_matcher = NULL;
+ if (next_matcher && !first)
+ prev_matcher = list_entry(next_matcher->matcher_list.prev,
+ struct mlx5dr_matcher,
+ matcher_list);
+ else if (!first)
+ prev_matcher = list_entry(tbl->matcher_list.prev,
+ struct mlx5dr_matcher,
+ matcher_list);
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB ||
+ dmn->type == MLX5DR_DOMAIN_TYPE_NIC_RX) {
+ ret = dr_matcher_connect(dmn, &matcher->rx,
+ next_matcher ? &next_matcher->rx : NULL,
+ prev_matcher ? &prev_matcher->rx : NULL);
+ if (ret)
+ return ret;
+ }
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB ||
+ dmn->type == MLX5DR_DOMAIN_TYPE_NIC_TX) {
+ ret = dr_matcher_connect(dmn, &matcher->tx,
+ next_matcher ? &next_matcher->tx : NULL,
+ prev_matcher ? &prev_matcher->tx : NULL);
+ if (ret)
+ return ret;
+ }
+
+ if (prev_matcher)
+ list_add(&matcher->matcher_list, &prev_matcher->matcher_list);
+ else if (next_matcher)
+ list_add_tail(&matcher->matcher_list,
+ &next_matcher->matcher_list);
+ else
+ list_add(&matcher->matcher_list, &tbl->matcher_list);
+
+ return 0;
+}
+
+static void dr_matcher_uninit_nic(struct mlx5dr_matcher_rx_tx *nic_matcher)
+{
+ mlx5dr_htbl_put(nic_matcher->s_htbl);
+ mlx5dr_htbl_put(nic_matcher->e_anchor);
+}
+
+static void dr_matcher_uninit_fdb(struct mlx5dr_matcher *matcher)
+{
+ dr_matcher_uninit_nic(&matcher->rx);
+ dr_matcher_uninit_nic(&matcher->tx);
+}
+
+static void dr_matcher_uninit(struct mlx5dr_matcher *matcher)
+{
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+
+ switch (dmn->type) {
+ case MLX5DR_DOMAIN_TYPE_NIC_RX:
+ dr_matcher_uninit_nic(&matcher->rx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_NIC_TX:
+ dr_matcher_uninit_nic(&matcher->tx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_FDB:
+ dr_matcher_uninit_fdb(matcher);
+ break;
+ default:
+ WARN_ON(true);
+ break;
+ }
+}
+
+static int dr_matcher_init_nic(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher)
+{
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ int ret, ret_v4, ret_v6;
+
+ ret_v4 = dr_matcher_set_ste_builders(matcher, nic_matcher, false);
+ ret_v6 = dr_matcher_set_ste_builders(matcher, nic_matcher, true);
+
+ if (ret_v4 && ret_v6) {
+ mlx5dr_dbg(dmn, "Cannot generate IPv4 or IPv6 rules with given mask\n");
+ return -EINVAL;
+ }
+
+ if (!ret_v4)
+ nic_matcher->ste_builder = nic_matcher->ste_builder4;
+ else
+ nic_matcher->ste_builder = nic_matcher->ste_builder6;
+
+ nic_matcher->e_anchor = mlx5dr_ste_htbl_alloc(dmn->ste_icm_pool,
+ DR_CHUNK_SIZE_1,
+ MLX5DR_STE_LU_TYPE_DONT_CARE,
+ 0);
+ if (!nic_matcher->e_anchor)
+ return -ENOMEM;
+
+ nic_matcher->s_htbl = mlx5dr_ste_htbl_alloc(dmn->ste_icm_pool,
+ DR_CHUNK_SIZE_1,
+ nic_matcher->ste_builder[0].lu_type,
+ nic_matcher->ste_builder[0].byte_mask);
+ if (!nic_matcher->s_htbl) {
+ ret = -ENOMEM;
+ goto free_e_htbl;
+ }
+
+ /* make sure the tables exist while empty */
+ mlx5dr_htbl_get(nic_matcher->s_htbl);
+ mlx5dr_htbl_get(nic_matcher->e_anchor);
+
+ return 0;
+
+free_e_htbl:
+ mlx5dr_ste_htbl_free(nic_matcher->e_anchor);
+ return ret;
+}
+
+static int dr_matcher_init_fdb(struct mlx5dr_matcher *matcher)
+{
+ int ret;
+
+ ret = dr_matcher_init_nic(matcher, &matcher->rx);
+ if (ret)
+ return ret;
+
+ ret = dr_matcher_init_nic(matcher, &matcher->tx);
+ if (ret)
+ goto uninit_nic_rx;
+
+ return 0;
+
+uninit_nic_rx:
+ dr_matcher_uninit_nic(&matcher->rx);
+ return ret;
+}
+
+static int dr_matcher_init(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *mask)
+{
+ struct mlx5dr_table *tbl = matcher->tbl;
+ struct mlx5dr_domain *dmn = tbl->dmn;
+ int ret;
+
+ if (matcher->match_criteria >= DR_MATCHER_CRITERIA_MAX) {
+ mlx5dr_info(dmn, "Invalid match criteria attribute\n");
+ return -EINVAL;
+ }
+
+ if (mask) {
+ if (mask->match_sz > sizeof(struct mlx5dr_match_param)) {
+ mlx5dr_info(dmn, "Invalid match size attribute\n");
+ return -EINVAL;
+ }
+ mlx5dr_ste_copy_param(matcher->match_criteria,
+ &matcher->mask, mask);
+ }
+
+ switch (dmn->type) {
+ case MLX5DR_DOMAIN_TYPE_NIC_RX:
+ matcher->rx.nic_tbl = &tbl->rx;
+ ret = dr_matcher_init_nic(matcher, &matcher->rx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_NIC_TX:
+ matcher->tx.nic_tbl = &tbl->tx;
+ ret = dr_matcher_init_nic(matcher, &matcher->tx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_FDB:
+ matcher->rx.nic_tbl = &tbl->rx;
+ matcher->tx.nic_tbl = &tbl->tx;
+ ret = dr_matcher_init_fdb(matcher);
+ break;
+ default:
+ WARN_ON(true);
+ return -EINVAL;
+ }
+
+ return ret;
+}
+
+struct mlx5dr_matcher *
+mlx5dr_matcher_create(struct mlx5dr_table *tbl,
+ u16 priority,
+ u8 match_criteria_enable,
+ struct mlx5dr_match_parameters *mask)
+{
+ struct mlx5dr_matcher *matcher;
+ int ret;
+
+ refcount_inc(&tbl->refcount);
+
+ matcher = kzalloc(sizeof(*matcher), GFP_KERNEL);
+ if (!matcher)
+ goto dec_ref;
+
+ matcher->tbl = tbl;
+ matcher->prio = priority;
+ matcher->match_criteria = match_criteria_enable;
+ refcount_set(&matcher->refcount, 1);
+ INIT_LIST_HEAD(&matcher->matcher_list);
+
+ mutex_lock(&tbl->dmn->mutex);
+
+ ret = dr_matcher_init(matcher, mask);
+ if (ret)
+ goto free_matcher;
+
+ ret = dr_matcher_add_to_tbl(matcher);
+ if (ret)
+ goto matcher_uninit;
+
+ mutex_unlock(&tbl->dmn->mutex);
+
+ return matcher;
+
+matcher_uninit:
+ dr_matcher_uninit(matcher);
+free_matcher:
+ mutex_unlock(&tbl->dmn->mutex);
+ kfree(matcher);
+dec_ref:
+ refcount_dec(&tbl->refcount);
+ return NULL;
+}
+
+static int dr_matcher_disconnect(struct mlx5dr_domain *dmn,
+ struct mlx5dr_table_rx_tx *nic_tbl,
+ struct mlx5dr_matcher_rx_tx *next_nic_matcher,
+ struct mlx5dr_matcher_rx_tx *prev_nic_matcher)
+{
+ struct mlx5dr_domain_rx_tx *nic_dmn = nic_tbl->nic_dmn;
+ struct mlx5dr_htbl_connect_info info;
+ struct mlx5dr_ste_htbl *prev_anchor;
+
+ if (prev_nic_matcher)
+ prev_anchor = prev_nic_matcher->e_anchor;
+ else
+ prev_anchor = nic_tbl->s_anchor;
+
+ /* Connect previous anchor hash table to next matcher or to the default address */
+ if (next_nic_matcher) {
+ info.type = CONNECT_HIT;
+ info.hit_next_htbl = next_nic_matcher->s_htbl;
+ next_nic_matcher->s_htbl->pointing_ste = prev_anchor->ste_arr;
+ prev_anchor->ste_arr[0].next_htbl = next_nic_matcher->s_htbl;
+ } else {
+ info.type = CONNECT_MISS;
+ info.miss_icm_addr = nic_tbl->default_icm_addr;
+ prev_anchor->ste_arr[0].next_htbl = NULL;
+ }
+
+ return mlx5dr_ste_htbl_init_and_postsend(dmn, nic_dmn, prev_anchor,
+ &info, true);
+}
+
+static int dr_matcher_remove_from_tbl(struct mlx5dr_matcher *matcher)
+{
+ struct mlx5dr_matcher *prev_matcher, *next_matcher;
+ struct mlx5dr_table *tbl = matcher->tbl;
+ struct mlx5dr_domain *dmn = tbl->dmn;
+ int ret = 0;
+
+ if (list_is_last(&matcher->matcher_list, &tbl->matcher_list))
+ next_matcher = NULL;
+ else
+ next_matcher = list_next_entry(matcher, matcher_list);
+
+ if (matcher->matcher_list.prev == &tbl->matcher_list)
+ prev_matcher = NULL;
+ else
+ prev_matcher = list_prev_entry(matcher, matcher_list);
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB ||
+ dmn->type == MLX5DR_DOMAIN_TYPE_NIC_RX) {
+ ret = dr_matcher_disconnect(dmn, &tbl->rx,
+ next_matcher ? &next_matcher->rx : NULL,
+ prev_matcher ? &prev_matcher->rx : NULL);
+ if (ret)
+ return ret;
+ }
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB ||
+ dmn->type == MLX5DR_DOMAIN_TYPE_NIC_TX) {
+ ret = dr_matcher_disconnect(dmn, &tbl->tx,
+ next_matcher ? &next_matcher->tx : NULL,
+ prev_matcher ? &prev_matcher->tx : NULL);
+ if (ret)
+ return ret;
+ }
+
+ list_del(&matcher->matcher_list);
+
+ return 0;
+}
+
+int mlx5dr_matcher_destroy(struct mlx5dr_matcher *matcher)
+{
+ struct mlx5dr_table *tbl = matcher->tbl;
+
+ if (refcount_read(&matcher->refcount) > 1)
+ return -EBUSY;
+
+ mutex_lock(&tbl->dmn->mutex);
+
+ dr_matcher_remove_from_tbl(matcher);
+ dr_matcher_uninit(matcher);
+ refcount_dec(&matcher->tbl->refcount);
+
+ mutex_unlock(&tbl->dmn->mutex);
+ kfree(matcher);
+
+ return 0;
+}
--
2.21.0
^ permalink raw reply related
* [net-next V2 10/18] net/mlx5: DR, Expose steering action functionality
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Mark Bloch,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
On rule creation a set of actions can be provided, the actions describe
what to do with the packet in case of a match. It is possible to provide
a set of actions which will be done by order.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/dr_action.c | 1588 +++++++++++++++++
1 file changed, 1588 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
new file mode 100644
index 000000000000..a02f87f85c17
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -0,0 +1,1588 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#include "dr_types.h"
+
+enum dr_action_domain {
+ DR_ACTION_DOMAIN_NIC_INGRESS,
+ DR_ACTION_DOMAIN_NIC_EGRESS,
+ DR_ACTION_DOMAIN_FDB_INGRESS,
+ DR_ACTION_DOMAIN_FDB_EGRESS,
+ DR_ACTION_DOMAIN_MAX,
+};
+
+enum dr_action_valid_state {
+ DR_ACTION_STATE_ERR,
+ DR_ACTION_STATE_NO_ACTION,
+ DR_ACTION_STATE_REFORMAT,
+ DR_ACTION_STATE_MODIFY_HDR,
+ DR_ACTION_STATE_MODIFY_VLAN,
+ DR_ACTION_STATE_NON_TERM,
+ DR_ACTION_STATE_TERM,
+ DR_ACTION_STATE_MAX,
+};
+
+static const enum dr_action_valid_state
+next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] = {
+ [DR_ACTION_DOMAIN_NIC_INGRESS] = {
+ [DR_ACTION_STATE_NO_ACTION] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_TNL_L3_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_REFORMAT] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_MODIFY_HDR] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR,
+ },
+ [DR_ACTION_STATE_MODIFY_VLAN] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ },
+ [DR_ACTION_STATE_NON_TERM] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_TNL_L3_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_TERM] = {
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_TERM,
+ },
+ },
+ [DR_ACTION_DOMAIN_NIC_EGRESS] = {
+ [DR_ACTION_STATE_NO_ACTION] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_REFORMAT] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_REFORMAT,
+ },
+ [DR_ACTION_STATE_MODIFY_HDR] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_MODIFY_VLAN] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ },
+ [DR_ACTION_STATE_NON_TERM] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ },
+ [DR_ACTION_STATE_TERM] = {
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_TERM,
+ },
+ },
+ [DR_ACTION_DOMAIN_FDB_INGRESS] = {
+ [DR_ACTION_STATE_NO_ACTION] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_TNL_L3_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_REFORMAT] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_MODIFY_HDR] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_MODIFY_VLAN] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ },
+ [DR_ACTION_STATE_NON_TERM] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_TNL_L3_TO_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_TERM] = {
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_TERM,
+ },
+ },
+ [DR_ACTION_DOMAIN_FDB_EGRESS] = {
+ [DR_ACTION_STATE_NO_ACTION] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_REFORMAT] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_MODIFY_HDR] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_MODIFY_VLAN] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_NON_TERM] = {
+ [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM,
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM,
+ [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR,
+ [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_L2_TO_TNL_L3] = DR_ACTION_STATE_REFORMAT,
+ [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_MODIFY_VLAN,
+ [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM,
+ },
+ [DR_ACTION_STATE_TERM] = {
+ [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_TERM,
+ },
+ },
+};
+
+struct dr_action_modify_field_conv {
+ u16 hw_field;
+ u8 start;
+ u8 end;
+ u8 l3_type;
+ u8 l4_type;
+};
+
+static const struct dr_action_modify_field_conv dr_action_conv_arr[] = {
+ [MLX5_ACTION_IN_FIELD_OUT_SMAC_47_16] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_1, .start = 16, .end = 47,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SMAC_15_0] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_1, .start = 0, .end = 15,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_ETHERTYPE] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_2, .start = 32, .end = 47,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DMAC_47_16] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_0, .start = 16, .end = 47,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DMAC_15_0] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_0, .start = 0, .end = 15,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_IP_DSCP] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_1, .start = 0, .end = 5,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_TCP_FLAGS] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_0, .start = 48, .end = 56,
+ .l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_TCP,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_TCP_SPORT] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_0, .start = 0, .end = 15,
+ .l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_TCP,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_TCP_DPORT] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_0, .start = 16, .end = 31,
+ .l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_TCP,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_IP_TTL] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_1, .start = 8, .end = 15,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV4,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_IPV6_HOPLIMIT] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_1, .start = 8, .end = 15,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_UDP_SPORT] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_0, .start = 0, .end = 15,
+ .l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_UDP,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_UDP_DPORT] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_0, .start = 16, .end = 31,
+ .l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_UDP,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SIPV6_127_96] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_3, .start = 32, .end = 63,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SIPV6_95_64] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_3, .start = 0, .end = 31,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SIPV6_63_32] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_4, .start = 32, .end = 63,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SIPV6_31_0] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_4, .start = 0, .end = 31,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DIPV6_127_96] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_0, .start = 32, .end = 63,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DIPV6_95_64] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_0, .start = 0, .end = 31,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DIPV6_63_32] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_2, .start = 32, .end = 63,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DIPV6_31_0] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_2, .start = 0, .end = 31,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV6,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_SIPV4] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_0, .start = 0, .end = 31,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV4,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_DIPV4] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L3_0, .start = 32, .end = 63,
+ .l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_IPV4,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_A] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_METADATA, .start = 0, .end = 31,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_B] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_METADATA, .start = 32, .end = 63,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_0] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_0, .start = 32, .end = 63,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_1] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_0, .start = 0, .end = 31,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_2] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_1, .start = 32, .end = 63,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_3] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_1, .start = 0, .end = 31,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_4] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_2, .start = 32, .end = 63,
+ },
+ [MLX5_ACTION_IN_FIELD_METADATA_REG_C_5] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_REG_2, .start = 0, .end = 31,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_TCP_SEQ_NUM] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_1, .start = 32, .end = 63,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_TCP_ACK_NUM] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L4_1, .start = 0, .end = 31,
+ },
+ [MLX5_ACTION_IN_FIELD_OUT_FIRST_VID] = {
+ .hw_field = MLX5DR_ACTION_MDFY_HW_FLD_L2_2, .start = 0, .end = 15,
+ },
+};
+
+#define MAX_VLANS 2
+struct dr_action_vlan_info {
+ int count;
+ u32 headers[MAX_VLANS];
+};
+
+struct dr_action_apply_attr {
+ u32 modify_index;
+ u16 modify_actions;
+ u32 decap_index;
+ u16 decap_actions;
+ u8 decap_with_vlan:1;
+ u64 final_icm_addr;
+ u32 flow_tag;
+ u32 ctr_id;
+ u16 gvmi;
+ u16 hit_gvmi;
+ u32 reformat_id;
+ u32 reformat_size;
+ struct dr_action_vlan_info vlans;
+};
+
+static int
+dr_action_reformat_to_action_type(enum mlx5dr_action_reformat_type reformat_type,
+ enum mlx5dr_action_type *action_type)
+{
+ switch (reformat_type) {
+ case DR_ACTION_REFORMAT_TYP_TNL_L2_TO_L2:
+ *action_type = DR_ACTION_TYP_TNL_L2_TO_L2;
+ break;
+ case DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L2:
+ *action_type = DR_ACTION_TYP_L2_TO_TNL_L2;
+ break;
+ case DR_ACTION_REFORMAT_TYP_TNL_L3_TO_L2:
+ *action_type = DR_ACTION_TYP_TNL_L3_TO_L2;
+ break;
+ case DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L3:
+ *action_type = DR_ACTION_TYP_L2_TO_TNL_L3;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static void dr_actions_init_next_ste(u8 **last_ste,
+ u32 *added_stes,
+ enum mlx5dr_ste_entry_type entry_type,
+ u16 gvmi)
+{
+ (*added_stes)++;
+ *last_ste += DR_STE_SIZE;
+ mlx5dr_ste_init(*last_ste, MLX5DR_STE_LU_TYPE_DONT_CARE, entry_type, gvmi);
+}
+
+static void dr_actions_apply_tx(struct mlx5dr_domain *dmn,
+ u8 *action_type_set,
+ u8 *last_ste,
+ struct dr_action_apply_attr *attr,
+ u32 *added_stes)
+{
+ bool encap = action_type_set[DR_ACTION_TYP_L2_TO_TNL_L2] ||
+ action_type_set[DR_ACTION_TYP_L2_TO_TNL_L3];
+
+ /* We want to make sure the modify header comes before L2
+ * encapsulation. The reason for that is that we support
+ * modify headers for outer headers only
+ */
+ if (action_type_set[DR_ACTION_TYP_MODIFY_HDR]) {
+ mlx5dr_ste_set_entry_type(last_ste, MLX5DR_STE_TYPE_MODIFY_PKT);
+ mlx5dr_ste_set_rewrite_actions(last_ste,
+ attr->modify_actions,
+ attr->modify_index);
+ }
+
+ if (action_type_set[DR_ACTION_TYP_PUSH_VLAN]) {
+ int i;
+
+ for (i = 0; i < attr->vlans.count; i++) {
+ if (i || action_type_set[DR_ACTION_TYP_MODIFY_HDR])
+ dr_actions_init_next_ste(&last_ste,
+ added_stes,
+ MLX5DR_STE_TYPE_TX,
+ attr->gvmi);
+
+ mlx5dr_ste_set_tx_push_vlan(last_ste,
+ attr->vlans.headers[i],
+ encap);
+ }
+ }
+
+ if (encap) {
+ /* Modify header and encapsulation require a different STEs.
+ * Since modify header STE format doesn't support encapsulation
+ * tunneling_action.
+ */
+ if (action_type_set[DR_ACTION_TYP_MODIFY_HDR] ||
+ action_type_set[DR_ACTION_TYP_PUSH_VLAN])
+ dr_actions_init_next_ste(&last_ste,
+ added_stes,
+ MLX5DR_STE_TYPE_TX,
+ attr->gvmi);
+
+ mlx5dr_ste_set_tx_encap(last_ste,
+ attr->reformat_id,
+ attr->reformat_size,
+ action_type_set[DR_ACTION_TYP_L2_TO_TNL_L3]);
+ /* Whenever prio_tag_required enabled, we can be sure that the
+ * previous table (ACL) already push vlan to our packet,
+ * And due to HW limitation we need to set this bit, otherwise
+ * push vlan + reformat will not work.
+ */
+ if (MLX5_CAP_GEN(dmn->mdev, prio_tag_required))
+ mlx5dr_ste_set_go_back_bit(last_ste);
+ }
+
+ if (action_type_set[DR_ACTION_TYP_CTR])
+ mlx5dr_ste_set_counter_id(last_ste, attr->ctr_id);
+}
+
+static void dr_actions_apply_rx(u8 *action_type_set,
+ u8 *last_ste,
+ struct dr_action_apply_attr *attr,
+ u32 *added_stes)
+{
+ if (action_type_set[DR_ACTION_TYP_CTR])
+ mlx5dr_ste_set_counter_id(last_ste, attr->ctr_id);
+
+ if (action_type_set[DR_ACTION_TYP_TNL_L3_TO_L2]) {
+ mlx5dr_ste_set_entry_type(last_ste, MLX5DR_STE_TYPE_MODIFY_PKT);
+ mlx5dr_ste_set_rx_decap_l3(last_ste, attr->decap_with_vlan);
+ mlx5dr_ste_set_rewrite_actions(last_ste,
+ attr->decap_actions,
+ attr->decap_index);
+ }
+
+ if (action_type_set[DR_ACTION_TYP_TNL_L2_TO_L2])
+ mlx5dr_ste_set_rx_decap(last_ste);
+
+ if (action_type_set[DR_ACTION_TYP_POP_VLAN]) {
+ int i;
+
+ for (i = 0; i < attr->vlans.count; i++) {
+ if (i ||
+ action_type_set[DR_ACTION_TYP_TNL_L2_TO_L2] ||
+ action_type_set[DR_ACTION_TYP_TNL_L3_TO_L2])
+ dr_actions_init_next_ste(&last_ste,
+ added_stes,
+ MLX5DR_STE_TYPE_RX,
+ attr->gvmi);
+
+ mlx5dr_ste_set_rx_pop_vlan(last_ste);
+ }
+ }
+
+ if (action_type_set[DR_ACTION_TYP_MODIFY_HDR]) {
+ if (mlx5dr_ste_get_entry_type(last_ste) == MLX5DR_STE_TYPE_MODIFY_PKT)
+ dr_actions_init_next_ste(&last_ste,
+ added_stes,
+ MLX5DR_STE_TYPE_MODIFY_PKT,
+ attr->gvmi);
+ else
+ mlx5dr_ste_set_entry_type(last_ste, MLX5DR_STE_TYPE_MODIFY_PKT);
+
+ mlx5dr_ste_set_rewrite_actions(last_ste,
+ attr->modify_actions,
+ attr->modify_index);
+ }
+
+ if (action_type_set[DR_ACTION_TYP_TAG]) {
+ if (mlx5dr_ste_get_entry_type(last_ste) == MLX5DR_STE_TYPE_MODIFY_PKT)
+ dr_actions_init_next_ste(&last_ste,
+ added_stes,
+ MLX5DR_STE_TYPE_RX,
+ attr->gvmi);
+
+ mlx5dr_ste_rx_set_flow_tag(last_ste, attr->flow_tag);
+ }
+}
+
+/* Apply the actions on the rule STE array starting from the last_ste.
+ * Actions might require more than one STE, new_num_stes will return
+ * the new size of the STEs array, rule with actions.
+ */
+static void dr_actions_apply(struct mlx5dr_domain *dmn,
+ enum mlx5dr_ste_entry_type ste_type,
+ u8 *action_type_set,
+ u8 *last_ste,
+ struct dr_action_apply_attr *attr,
+ u32 *new_num_stes)
+{
+ u32 added_stes = 0;
+
+ if (ste_type == MLX5DR_STE_TYPE_RX)
+ dr_actions_apply_rx(action_type_set, last_ste, attr, &added_stes);
+ else
+ dr_actions_apply_tx(dmn, action_type_set, last_ste, attr, &added_stes);
+
+ last_ste += added_stes * DR_STE_SIZE;
+ *new_num_stes += added_stes;
+
+ mlx5dr_ste_set_hit_gvmi(last_ste, attr->hit_gvmi);
+ mlx5dr_ste_set_hit_addr(last_ste, attr->final_icm_addr, 1);
+}
+
+static enum dr_action_domain
+dr_action_get_action_domain(enum mlx5dr_domain_type domain,
+ enum mlx5dr_ste_entry_type ste_type)
+{
+ switch (domain) {
+ case MLX5DR_DOMAIN_TYPE_NIC_RX:
+ return DR_ACTION_DOMAIN_NIC_INGRESS;
+ case MLX5DR_DOMAIN_TYPE_NIC_TX:
+ return DR_ACTION_DOMAIN_NIC_EGRESS;
+ case MLX5DR_DOMAIN_TYPE_FDB:
+ if (ste_type == MLX5DR_STE_TYPE_RX)
+ return DR_ACTION_DOMAIN_FDB_INGRESS;
+ return DR_ACTION_DOMAIN_FDB_EGRESS;
+ default:
+ WARN_ON(true);
+ return DR_ACTION_DOMAIN_MAX;
+ }
+}
+
+static
+int dr_action_validate_and_get_next_state(enum dr_action_domain action_domain,
+ u32 action_type,
+ u32 *state)
+{
+ u32 cur_state = *state;
+
+ /* Check action state machine is valid */
+ *state = next_action_state[action_domain][cur_state][action_type];
+
+ if (*state == DR_ACTION_STATE_ERR)
+ return -EOPNOTSUPP;
+
+ return 0;
+}
+
+static int dr_action_handle_cs_recalc(struct mlx5dr_domain *dmn,
+ struct mlx5dr_action *dest_action,
+ u64 *final_icm_addr)
+{
+ int ret;
+
+ switch (dest_action->action_type) {
+ case DR_ACTION_TYP_FT:
+ /* Allow destination flow table only if table is a terminating
+ * table, since there is an *assumption* that in such case FW
+ * will recalculate the CS.
+ */
+ if (dest_action->dest_tbl.is_fw_tbl) {
+ *final_icm_addr = dest_action->dest_tbl.fw_tbl.rx_icm_addr;
+ } else {
+ mlx5dr_dbg(dmn,
+ "Destination FT should be terminating when modify TTL is used\n");
+ return -EINVAL;
+ }
+ break;
+
+ case DR_ACTION_TYP_VPORT:
+ /* If destination is vport we will get the FW flow table
+ * that recalculates the CS and forwards to the vport.
+ */
+ ret = mlx5dr_domain_cache_get_recalc_cs_ft_addr(dest_action->vport.dmn,
+ dest_action->vport.num,
+ final_icm_addr);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed to get FW cs recalc flow table\n");
+ return ret;
+ }
+ break;
+
+ default:
+ break;
+ }
+
+ return 0;
+}
+
+#define WITH_VLAN_NUM_HW_ACTIONS 6
+
+int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_action *actions[],
+ u32 num_actions,
+ u8 *ste_arr,
+ u32 *new_hw_ste_arr_sz)
+{
+ struct mlx5dr_domain_rx_tx *nic_dmn = nic_matcher->nic_tbl->nic_dmn;
+ bool rx_rule = nic_dmn->ste_type == MLX5DR_STE_TYPE_RX;
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ u8 action_type_set[DR_ACTION_TYP_MAX] = {};
+ struct mlx5dr_action *dest_action = NULL;
+ u32 state = DR_ACTION_STATE_NO_ACTION;
+ struct dr_action_apply_attr attr = {};
+ enum dr_action_domain action_domain;
+ bool recalc_cs_required = false;
+ u8 *last_ste;
+ int i, ret;
+
+ attr.gvmi = dmn->info.caps.gvmi;
+ attr.hit_gvmi = dmn->info.caps.gvmi;
+ attr.final_icm_addr = nic_dmn->default_icm_addr;
+ action_domain = dr_action_get_action_domain(dmn->type, nic_dmn->ste_type);
+
+ for (i = 0; i < num_actions; i++) {
+ struct mlx5dr_action *action;
+ int max_actions_type = 1;
+ u32 action_type;
+
+ action = actions[i];
+ action_type = action->action_type;
+
+ switch (action_type) {
+ case DR_ACTION_TYP_DROP:
+ attr.final_icm_addr = nic_dmn->drop_icm_addr;
+ break;
+ case DR_ACTION_TYP_FT:
+ dest_action = action;
+ if (!action->dest_tbl.is_fw_tbl) {
+ if (action->dest_tbl.tbl->dmn != dmn) {
+ mlx5dr_dbg(dmn,
+ "Destination table belongs to a different domain\n");
+ goto out_invalid_arg;
+ }
+ if (action->dest_tbl.tbl->level <= matcher->tbl->level) {
+ mlx5dr_dbg(dmn,
+ "Destination table level should be higher than source table\n");
+ goto out_invalid_arg;
+ }
+ attr.final_icm_addr = rx_rule ?
+ action->dest_tbl.tbl->rx.s_anchor->chunk->icm_addr :
+ action->dest_tbl.tbl->tx.s_anchor->chunk->icm_addr;
+ } else {
+ struct mlx5dr_cmd_query_flow_table_details output;
+ int ret;
+
+ /* get the relevant addresses */
+ if (!action->dest_tbl.fw_tbl.rx_icm_addr) {
+ ret = mlx5dr_cmd_query_flow_table(action->dest_tbl.fw_tbl.mdev,
+ action->dest_tbl.fw_tbl.ft->type,
+ action->dest_tbl.fw_tbl.ft->id,
+ &output);
+ if (!ret) {
+ action->dest_tbl.fw_tbl.tx_icm_addr =
+ output.sw_owner_icm_root_1;
+ action->dest_tbl.fw_tbl.rx_icm_addr =
+ output.sw_owner_icm_root_0;
+ } else {
+ mlx5dr_dbg(dmn,
+ "Failed mlx5_cmd_query_flow_table ret: %d\n",
+ ret);
+ return ret;
+ }
+ }
+ attr.final_icm_addr = rx_rule ?
+ action->dest_tbl.fw_tbl.rx_icm_addr :
+ action->dest_tbl.fw_tbl.tx_icm_addr;
+ }
+ break;
+ case DR_ACTION_TYP_QP:
+ mlx5dr_info(dmn, "Domain doesn't support QP\n");
+ goto out_invalid_arg;
+ case DR_ACTION_TYP_CTR:
+ attr.ctr_id = action->ctr.ctr_id +
+ action->ctr.offeset;
+ break;
+ case DR_ACTION_TYP_TAG:
+ attr.flow_tag = action->flow_tag;
+ break;
+ case DR_ACTION_TYP_TNL_L2_TO_L2:
+ break;
+ case DR_ACTION_TYP_TNL_L3_TO_L2:
+ attr.decap_index = action->rewrite.index;
+ attr.decap_actions = action->rewrite.num_of_actions;
+ attr.decap_with_vlan =
+ attr.decap_actions == WITH_VLAN_NUM_HW_ACTIONS;
+ break;
+ case DR_ACTION_TYP_MODIFY_HDR:
+ attr.modify_index = action->rewrite.index;
+ attr.modify_actions = action->rewrite.num_of_actions;
+ recalc_cs_required = action->rewrite.modify_ttl;
+ break;
+ case DR_ACTION_TYP_L2_TO_TNL_L2:
+ case DR_ACTION_TYP_L2_TO_TNL_L3:
+ attr.reformat_size = action->reformat.reformat_size;
+ attr.reformat_id = action->reformat.reformat_id;
+ break;
+ case DR_ACTION_TYP_VPORT:
+ attr.hit_gvmi = action->vport.caps->vhca_gvmi;
+ dest_action = action;
+ if (rx_rule) {
+ /* Loopback on WIRE vport is not supported */
+ if (action->vport.num == WIRE_PORT)
+ goto out_invalid_arg;
+
+ attr.final_icm_addr = action->vport.caps->icm_address_rx;
+ } else {
+ attr.final_icm_addr = action->vport.caps->icm_address_tx;
+ }
+ break;
+ case DR_ACTION_TYP_POP_VLAN:
+ max_actions_type = MAX_VLANS;
+ attr.vlans.count++;
+ break;
+ case DR_ACTION_TYP_PUSH_VLAN:
+ max_actions_type = MAX_VLANS;
+ if (attr.vlans.count == MAX_VLANS)
+ return -EINVAL;
+
+ attr.vlans.headers[attr.vlans.count++] = action->push_vlan.vlan_hdr;
+ break;
+ default:
+ goto out_invalid_arg;
+ }
+
+ /* Check action duplication */
+ if (++action_type_set[action_type] > max_actions_type) {
+ mlx5dr_dbg(dmn, "Action type %d supports only max %d time(s)\n",
+ action_type, max_actions_type);
+ goto out_invalid_arg;
+ }
+
+ /* Check action state machine is valid */
+ if (dr_action_validate_and_get_next_state(action_domain,
+ action_type,
+ &state)) {
+ mlx5dr_dbg(dmn, "Invalid action sequence provided\n");
+ return -EOPNOTSUPP;
+ }
+ }
+
+ *new_hw_ste_arr_sz = nic_matcher->num_of_builders;
+ last_ste = ste_arr + DR_STE_SIZE * (nic_matcher->num_of_builders - 1);
+
+ /* Due to a HW bug, modifying TTL on RX flows will cause an incorrect
+ * checksum calculation. In this case we will use a FW table to
+ * recalculate.
+ */
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB &&
+ rx_rule && recalc_cs_required && dest_action) {
+ ret = dr_action_handle_cs_recalc(dmn, dest_action, &attr.final_icm_addr);
+ if (ret) {
+ mlx5dr_dbg(dmn,
+ "Failed to handle checksum recalculation err %d\n",
+ ret);
+ return ret;
+ }
+ }
+
+ dr_actions_apply(dmn,
+ nic_dmn->ste_type,
+ action_type_set,
+ last_ste,
+ &attr,
+ new_hw_ste_arr_sz);
+
+ return 0;
+
+out_invalid_arg:
+ return -EINVAL;
+}
+
+#define CVLAN_ETHERTYPE 0x8100
+#define SVLAN_ETHERTYPE 0x88a8
+#define HDR_LEN_L2_ONLY 14
+#define HDR_LEN_L2_VLAN 18
+#define REWRITE_HW_ACTION_NUM 6
+
+static int dr_actions_l2_rewrite(struct mlx5dr_domain *dmn,
+ struct mlx5dr_action *action,
+ void *data, size_t data_sz)
+{
+ struct mlx5_ifc_l2_hdr_bits *l2_hdr = data;
+ u64 ops[REWRITE_HW_ACTION_NUM] = {};
+ u32 hdr_fld_4b;
+ u16 hdr_fld_2b;
+ u16 vlan_type;
+ bool vlan;
+ int i = 0;
+ int ret;
+
+ vlan = (data_sz != HDR_LEN_L2_ONLY);
+
+ /* dmac_47_16 */
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_length, 0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 16);
+ hdr_fld_4b = MLX5_GET(l2_hdr, l2_hdr, dmac_47_16);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ inline_data, hdr_fld_4b);
+ i++;
+
+ /* smac_47_16 */
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_length, 0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_1);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 16);
+ hdr_fld_4b = (MLX5_GET(l2_hdr, l2_hdr, smac_31_0) >> 16 |
+ MLX5_GET(l2_hdr, l2_hdr, smac_47_32) << 16);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ inline_data, hdr_fld_4b);
+ i++;
+
+ /* dmac_15_0 */
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_length, 16);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 0);
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, dmac_15_0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ inline_data, hdr_fld_2b);
+ i++;
+
+ /* ethertype + (optional) vlan */
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_2);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 32);
+ if (!vlan) {
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, ethertype);
+ MLX5_SET(dr_action_hw_set, ops + i, inline_data, hdr_fld_2b);
+ MLX5_SET(dr_action_hw_set, ops + i, destination_length, 16);
+ } else {
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, ethertype);
+ vlan_type = hdr_fld_2b == SVLAN_ETHERTYPE ? DR_STE_SVLAN : DR_STE_CVLAN;
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, vlan);
+ hdr_fld_4b = (vlan_type << 16) | hdr_fld_2b;
+ MLX5_SET(dr_action_hw_set, ops + i, inline_data, hdr_fld_4b);
+ MLX5_SET(dr_action_hw_set, ops + i, destination_length, 18);
+ }
+ i++;
+
+ /* smac_15_0 */
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_length, 16);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_1);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 0);
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, smac_31_0);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ inline_data, hdr_fld_2b);
+ i++;
+
+ if (vlan) {
+ MLX5_SET(dr_action_hw_set, ops + i,
+ opcode, MLX5DR_ACTION_MDFY_HW_OP_SET);
+ hdr_fld_2b = MLX5_GET(l2_hdr, l2_hdr, vlan_type);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ inline_data, hdr_fld_2b);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_length, 16);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_field_code, MLX5DR_ACTION_MDFY_HW_FLD_L2_2);
+ MLX5_SET(dr_action_hw_set, ops + i,
+ destination_left_shifter, 0);
+ i++;
+ }
+
+ action->rewrite.data = (void *)ops;
+ action->rewrite.num_of_actions = i;
+ action->rewrite.chunk->byte_size = i * sizeof(*ops);
+
+ ret = mlx5dr_send_postsend_action(dmn, action);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Writing encapsulation action to ICM failed\n");
+ return ret;
+ }
+
+ return 0;
+}
+
+static struct mlx5dr_action *
+dr_action_create_generic(enum mlx5dr_action_type action_type)
+{
+ struct mlx5dr_action *action;
+
+ action = kzalloc(sizeof(*action), GFP_KERNEL);
+ if (!action)
+ return NULL;
+
+ action->action_type = action_type;
+ refcount_set(&action->refcount, 1);
+
+ return action;
+}
+
+struct mlx5dr_action *mlx5dr_action_create_drop(void)
+{
+ return dr_action_create_generic(DR_ACTION_TYP_DROP);
+}
+
+struct mlx5dr_action *
+mlx5dr_action_create_dest_table(struct mlx5dr_table *tbl)
+{
+ struct mlx5dr_action *action;
+
+ refcount_inc(&tbl->refcount);
+
+ action = dr_action_create_generic(DR_ACTION_TYP_FT);
+ if (!action)
+ goto dec_ref;
+
+ action->dest_tbl.tbl = tbl;
+
+ return action;
+
+dec_ref:
+ refcount_dec(&tbl->refcount);
+ return NULL;
+}
+
+struct mlx5dr_action *
+mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
+ struct mlx5_core_dev *mdev)
+{
+ struct mlx5dr_action *action;
+
+ action = dr_action_create_generic(DR_ACTION_TYP_FT);
+ if (!action)
+ return NULL;
+
+ action->dest_tbl.is_fw_tbl = 1;
+ action->dest_tbl.fw_tbl.ft = ft;
+ action->dest_tbl.fw_tbl.mdev = mdev;
+
+ return action;
+}
+
+struct mlx5dr_action *
+mlx5dr_action_create_flow_counter(u32 counter_id)
+{
+ struct mlx5dr_action *action;
+
+ action = dr_action_create_generic(DR_ACTION_TYP_CTR);
+ if (!action)
+ return NULL;
+
+ action->ctr.ctr_id = counter_id;
+
+ return action;
+}
+
+struct mlx5dr_action *mlx5dr_action_create_tag(u32 tag_value)
+{
+ struct mlx5dr_action *action;
+
+ action = dr_action_create_generic(DR_ACTION_TYP_TAG);
+ if (!action)
+ return NULL;
+
+ action->flow_tag = tag_value & 0xffffff;
+
+ return action;
+}
+
+static int
+dr_action_verify_reformat_params(enum mlx5dr_action_type reformat_type,
+ struct mlx5dr_domain *dmn,
+ size_t data_sz,
+ void *data)
+{
+ if ((!data && data_sz) || (data && !data_sz) || reformat_type >
+ DR_ACTION_TYP_L2_TO_TNL_L3) {
+ mlx5dr_dbg(dmn, "Invalid reformat parameter!\n");
+ goto out_err;
+ }
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB)
+ return 0;
+
+ if (dmn->type == MLX5DR_DOMAIN_TYPE_NIC_RX) {
+ if (reformat_type != DR_ACTION_TYP_TNL_L2_TO_L2 &&
+ reformat_type != DR_ACTION_TYP_TNL_L3_TO_L2) {
+ mlx5dr_dbg(dmn, "Action reformat type not support on RX domain\n");
+ goto out_err;
+ }
+ } else if (dmn->type == MLX5DR_DOMAIN_TYPE_NIC_TX) {
+ if (reformat_type != DR_ACTION_TYP_L2_TO_TNL_L2 &&
+ reformat_type != DR_ACTION_TYP_L2_TO_TNL_L3) {
+ mlx5dr_dbg(dmn, "Action reformat type not support on TX domain\n");
+ goto out_err;
+ }
+ }
+
+ return 0;
+
+out_err:
+ return -EINVAL;
+}
+
+#define ACTION_CACHE_LINE_SIZE 64
+
+static int
+dr_action_create_reformat_action(struct mlx5dr_domain *dmn,
+ size_t data_sz, void *data,
+ struct mlx5dr_action *action)
+{
+ u32 reformat_id;
+ int ret;
+
+ switch (action->action_type) {
+ case DR_ACTION_TYP_L2_TO_TNL_L2:
+ case DR_ACTION_TYP_L2_TO_TNL_L3:
+ {
+ enum mlx5dr_action_type rt;
+
+ if (action->action_type == DR_ACTION_TYP_L2_TO_TNL_L2)
+ rt = MLX5_REFORMAT_TYPE_L2_TO_L2_TUNNEL;
+ else
+ rt = MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL;
+
+ ret = mlx5dr_cmd_create_reformat_ctx(dmn->mdev, rt, data_sz, data,
+ &reformat_id);
+ if (ret)
+ return ret;
+
+ action->reformat.reformat_id = reformat_id;
+ action->reformat.reformat_size = data_sz;
+ return 0;
+ }
+ case DR_ACTION_TYP_TNL_L2_TO_L2:
+ {
+ return 0;
+ }
+ case DR_ACTION_TYP_TNL_L3_TO_L2:
+ {
+ /* Only Ethernet frame is supported, with VLAN (18) or without (14) */
+ if (data_sz != HDR_LEN_L2_ONLY && data_sz != HDR_LEN_L2_VLAN)
+ return -EINVAL;
+
+ action->rewrite.chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool,
+ DR_CHUNK_SIZE_8);
+ if (!action->rewrite.chunk)
+ return -ENOMEM;
+
+ action->rewrite.index = (action->rewrite.chunk->icm_addr -
+ dmn->info.caps.hdr_modify_icm_addr) /
+ ACTION_CACHE_LINE_SIZE;
+
+ ret = dr_actions_l2_rewrite(dmn, action, data, data_sz);
+ if (ret) {
+ mlx5dr_icm_free_chunk(action->rewrite.chunk);
+ return ret;
+ }
+ return 0;
+ }
+ default:
+ mlx5dr_info(dmn, "Reformat type is not supported %d\n", action->action_type);
+ return -EINVAL;
+ }
+}
+
+struct mlx5dr_action *mlx5dr_action_create_pop_vlan(void)
+{
+ return dr_action_create_generic(DR_ACTION_TYP_POP_VLAN);
+}
+
+struct mlx5dr_action *mlx5dr_action_create_push_vlan(struct mlx5dr_domain *dmn,
+ __be32 vlan_hdr)
+{
+ u32 vlan_hdr_h = ntohl(vlan_hdr);
+ u16 ethertype = vlan_hdr_h >> 16;
+ struct mlx5dr_action *action;
+
+ if (ethertype != SVLAN_ETHERTYPE && ethertype != CVLAN_ETHERTYPE) {
+ mlx5dr_dbg(dmn, "Invalid vlan ethertype\n");
+ return NULL;
+ }
+
+ action = dr_action_create_generic(DR_ACTION_TYP_PUSH_VLAN);
+ if (!action)
+ return NULL;
+
+ action->push_vlan.vlan_hdr = vlan_hdr_h;
+ return action;
+}
+
+struct mlx5dr_action *
+mlx5dr_action_create_packet_reformat(struct mlx5dr_domain *dmn,
+ enum mlx5dr_action_reformat_type reformat_type,
+ size_t data_sz,
+ void *data)
+{
+ enum mlx5dr_action_type action_type;
+ struct mlx5dr_action *action;
+ int ret;
+
+ refcount_inc(&dmn->refcount);
+
+ /* General checks */
+ ret = dr_action_reformat_to_action_type(reformat_type, &action_type);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Invalid reformat_type provided\n");
+ goto dec_ref;
+ }
+
+ ret = dr_action_verify_reformat_params(action_type, dmn, data_sz, data);
+ if (ret)
+ goto dec_ref;
+
+ action = dr_action_create_generic(action_type);
+ if (!action)
+ goto dec_ref;
+
+ action->reformat.dmn = dmn;
+
+ ret = dr_action_create_reformat_action(dmn,
+ data_sz,
+ data,
+ action);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Failed creating reformat action %d\n", ret);
+ goto free_action;
+ }
+
+ return action;
+
+free_action:
+ kfree(action);
+dec_ref:
+ refcount_dec(&dmn->refcount);
+ return NULL;
+}
+
+static const struct dr_action_modify_field_conv *
+dr_action_modify_get_hw_info(u16 sw_field)
+{
+ const struct dr_action_modify_field_conv *hw_action_info;
+
+ if (sw_field >= ARRAY_SIZE(dr_action_conv_arr))
+ goto not_found;
+
+ hw_action_info = &dr_action_conv_arr[sw_field];
+ if (!hw_action_info->end && !hw_action_info->start)
+ goto not_found;
+
+ return hw_action_info;
+
+not_found:
+ return NULL;
+}
+
+static int
+dr_action_modify_sw_to_hw(struct mlx5dr_domain *dmn,
+ __be64 *sw_action,
+ __be64 *hw_action,
+ const struct dr_action_modify_field_conv **ret_hw_info)
+{
+ const struct dr_action_modify_field_conv *hw_action_info;
+ u8 offset, length, max_length, action;
+ u16 sw_field;
+ u8 hw_opcode;
+ u32 data;
+
+ /* Get SW modify action data */
+ action = MLX5_GET(set_action_in, sw_action, action_type);
+ length = MLX5_GET(set_action_in, sw_action, length);
+ offset = MLX5_GET(set_action_in, sw_action, offset);
+ sw_field = MLX5_GET(set_action_in, sw_action, field);
+ data = MLX5_GET(set_action_in, sw_action, data);
+
+ /* Convert SW data to HW modify action format */
+ hw_action_info = dr_action_modify_get_hw_info(sw_field);
+ if (!hw_action_info) {
+ mlx5dr_dbg(dmn, "Modify action invalid field given\n");
+ return -EINVAL;
+ }
+
+ max_length = hw_action_info->end - hw_action_info->start + 1;
+
+ switch (action) {
+ case MLX5_ACTION_TYPE_SET:
+ hw_opcode = MLX5DR_ACTION_MDFY_HW_OP_SET;
+ /* PRM defines that length zero specific length of 32bits */
+ if (!length)
+ length = 32;
+
+ if (length + offset > max_length) {
+ mlx5dr_dbg(dmn, "Modify action length + offset exceeds limit\n");
+ return -EINVAL;
+ }
+ break;
+
+ case MLX5_ACTION_TYPE_ADD:
+ hw_opcode = MLX5DR_ACTION_MDFY_HW_OP_ADD;
+ offset = 0;
+ length = max_length;
+ break;
+
+ default:
+ mlx5dr_info(dmn, "Unsupported action_type for modify action\n");
+ return -EOPNOTSUPP;
+ }
+
+ MLX5_SET(dr_action_hw_set, hw_action, opcode, hw_opcode);
+
+ MLX5_SET(dr_action_hw_set, hw_action, destination_field_code,
+ hw_action_info->hw_field);
+
+ MLX5_SET(dr_action_hw_set, hw_action, destination_left_shifter,
+ hw_action_info->start + offset);
+
+ MLX5_SET(dr_action_hw_set, hw_action, destination_length,
+ length == 32 ? 0 : length);
+
+ MLX5_SET(dr_action_hw_set, hw_action, inline_data, data);
+
+ *ret_hw_info = hw_action_info;
+
+ return 0;
+}
+
+static int
+dr_action_modify_check_field_limitation(struct mlx5dr_domain *dmn,
+ const __be64 *sw_action)
+{
+ u16 sw_field;
+ u8 action;
+
+ sw_field = MLX5_GET(set_action_in, sw_action, field);
+ action = MLX5_GET(set_action_in, sw_action, action_type);
+
+ /* Check if SW field is supported in current domain (RX/TX) */
+ if (action == MLX5_ACTION_TYPE_SET) {
+ if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_A) {
+ if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_TX) {
+ mlx5dr_dbg(dmn, "Unsupported field %d for RX/FDB set action\n",
+ sw_field);
+ return -EINVAL;
+ }
+ }
+
+ if (sw_field == MLX5_ACTION_IN_FIELD_METADATA_REG_B) {
+ if (dmn->type != MLX5DR_DOMAIN_TYPE_NIC_RX) {
+ mlx5dr_dbg(dmn, "Unsupported field %d for TX/FDB set action\n",
+ sw_field);
+ return -EINVAL;
+ }
+ }
+ } else if (action == MLX5_ACTION_TYPE_ADD) {
+ if (sw_field != MLX5_ACTION_IN_FIELD_OUT_IP_TTL &&
+ sw_field != MLX5_ACTION_IN_FIELD_OUT_IPV6_HOPLIMIT &&
+ sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_SEQ_NUM &&
+ sw_field != MLX5_ACTION_IN_FIELD_OUT_TCP_ACK_NUM) {
+ mlx5dr_dbg(dmn, "Unsupported field %d for add action\n", sw_field);
+ return -EINVAL;
+ }
+ } else {
+ mlx5dr_info(dmn, "Unsupported action %d modify action\n", action);
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
+static bool
+dr_action_modify_check_is_ttl_modify(const u64 *sw_action)
+{
+ u16 sw_field = MLX5_GET(set_action_in, sw_action, field);
+
+ return sw_field == MLX5_ACTION_IN_FIELD_OUT_IP_TTL;
+}
+
+static int dr_actions_convert_modify_header(struct mlx5dr_domain *dmn,
+ u32 max_hw_actions,
+ u32 num_sw_actions,
+ __be64 sw_actions[],
+ __be64 hw_actions[],
+ u32 *num_hw_actions,
+ bool *modify_ttl)
+{
+ const struct dr_action_modify_field_conv *hw_action_info;
+ u16 hw_field = MLX5DR_ACTION_MDFY_HW_FLD_RESERVED;
+ u32 l3_type = MLX5DR_ACTION_MDFY_HW_HDR_L3_NONE;
+ u32 l4_type = MLX5DR_ACTION_MDFY_HW_HDR_L4_NONE;
+ int ret, i, hw_idx = 0;
+ __be64 *sw_action;
+ __be64 hw_action;
+
+ *modify_ttl = false;
+
+ for (i = 0; i < num_sw_actions; i++) {
+ sw_action = &sw_actions[i];
+
+ ret = dr_action_modify_check_field_limitation(dmn, sw_action);
+ if (ret)
+ return ret;
+
+ if (!(*modify_ttl))
+ *modify_ttl = dr_action_modify_check_is_ttl_modify(sw_action);
+
+ /* Convert SW action to HW action */
+ ret = dr_action_modify_sw_to_hw(dmn,
+ sw_action,
+ &hw_action,
+ &hw_action_info);
+ if (ret)
+ return ret;
+
+ /* Due to a HW limitation we cannot modify 2 different L3 types */
+ if (l3_type && hw_action_info->l3_type &&
+ hw_action_info->l3_type != l3_type) {
+ mlx5dr_dbg(dmn, "Action list can't support two different L3 types\n");
+ return -EINVAL;
+ }
+ if (hw_action_info->l3_type)
+ l3_type = hw_action_info->l3_type;
+
+ /* Due to a HW limitation we cannot modify two different L4 types */
+ if (l4_type && hw_action_info->l4_type &&
+ hw_action_info->l4_type != l4_type) {
+ mlx5dr_dbg(dmn, "Action list can't support two different L4 types\n");
+ return -EINVAL;
+ }
+ if (hw_action_info->l4_type)
+ l4_type = hw_action_info->l4_type;
+
+ /* HW reads and executes two actions at once this means we
+ * need to create a gap if two actions access the same field
+ */
+ if ((hw_idx % 2) && hw_field == hw_action_info->hw_field) {
+ /* Check if after gap insertion the total number of HW
+ * modify actions doesn't exceeds the limit
+ */
+ hw_idx++;
+ if ((num_sw_actions + hw_idx - i) >= max_hw_actions) {
+ mlx5dr_dbg(dmn, "Modify header action number exceeds HW limit\n");
+ return -EINVAL;
+ }
+ }
+ hw_field = hw_action_info->hw_field;
+
+ hw_actions[hw_idx] = hw_action;
+ hw_idx++;
+ }
+
+ *num_hw_actions = hw_idx;
+
+ return 0;
+}
+
+static int dr_action_create_modify_action(struct mlx5dr_domain *dmn,
+ size_t actions_sz,
+ __be64 actions[],
+ struct mlx5dr_action *action)
+{
+ struct mlx5dr_icm_chunk *chunk;
+ u32 max_hw_actions;
+ u32 num_hw_actions;
+ u32 num_sw_actions;
+ __be64 *hw_actions;
+ bool modify_ttl;
+ int ret;
+
+ num_sw_actions = actions_sz / DR_MODIFY_ACTION_SIZE;
+ max_hw_actions = mlx5dr_icm_pool_chunk_size_to_entries(DR_CHUNK_SIZE_16);
+
+ if (num_sw_actions > max_hw_actions) {
+ mlx5dr_dbg(dmn, "Max number of actions %d exceeds limit %d\n",
+ num_sw_actions, max_hw_actions);
+ return -EINVAL;
+ }
+
+ chunk = mlx5dr_icm_alloc_chunk(dmn->action_icm_pool, DR_CHUNK_SIZE_16);
+ if (!chunk)
+ return -ENOMEM;
+
+ hw_actions = kcalloc(1, max_hw_actions * DR_MODIFY_ACTION_SIZE, GFP_KERNEL);
+ if (!hw_actions) {
+ ret = -ENOMEM;
+ goto free_chunk;
+ }
+
+ ret = dr_actions_convert_modify_header(dmn,
+ max_hw_actions,
+ num_sw_actions,
+ actions,
+ hw_actions,
+ &num_hw_actions,
+ &modify_ttl);
+ if (ret)
+ goto free_hw_actions;
+
+ action->rewrite.chunk = chunk;
+ action->rewrite.modify_ttl = modify_ttl;
+ action->rewrite.data = (u8 *)hw_actions;
+ action->rewrite.num_of_actions = num_hw_actions;
+ action->rewrite.index = (chunk->icm_addr -
+ dmn->info.caps.hdr_modify_icm_addr) /
+ ACTION_CACHE_LINE_SIZE;
+
+ ret = mlx5dr_send_postsend_action(dmn, action);
+ if (ret)
+ goto free_hw_actions;
+
+ return 0;
+
+free_hw_actions:
+ kfree(hw_actions);
+free_chunk:
+ mlx5dr_icm_free_chunk(chunk);
+ return ret;
+}
+
+struct mlx5dr_action *
+mlx5dr_action_create_modify_header(struct mlx5dr_domain *dmn,
+ u32 flags,
+ size_t actions_sz,
+ __be64 actions[])
+{
+ struct mlx5dr_action *action;
+ int ret = 0;
+
+ refcount_inc(&dmn->refcount);
+
+ if (actions_sz % DR_MODIFY_ACTION_SIZE) {
+ mlx5dr_dbg(dmn, "Invalid modify actions size provided\n");
+ goto dec_ref;
+ }
+
+ action = dr_action_create_generic(DR_ACTION_TYP_MODIFY_HDR);
+ if (!action)
+ goto dec_ref;
+
+ action->rewrite.dmn = dmn;
+
+ ret = dr_action_create_modify_action(dmn,
+ actions_sz,
+ actions,
+ action);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Failed creating modify header action %d\n", ret);
+ goto free_action;
+ }
+
+ return action;
+
+free_action:
+ kfree(action);
+dec_ref:
+ refcount_dec(&dmn->refcount);
+ return NULL;
+}
+
+struct mlx5dr_action *
+mlx5dr_action_create_dest_vport(struct mlx5dr_domain *dmn,
+ u32 vport, u8 vhca_id_valid,
+ u16 vhca_id)
+{
+ struct mlx5dr_cmd_vport_cap *vport_cap;
+ struct mlx5dr_domain *vport_dmn;
+ struct mlx5dr_action *action;
+ u8 peer_vport;
+
+ peer_vport = vhca_id_valid && (vhca_id != dmn->info.caps.gvmi);
+ vport_dmn = peer_vport ? dmn->peer_dmn : dmn;
+ if (!vport_dmn) {
+ mlx5dr_dbg(dmn, "No peer vport domain for given vhca_id\n");
+ return NULL;
+ }
+
+ if (vport_dmn->type != MLX5DR_DOMAIN_TYPE_FDB) {
+ mlx5dr_dbg(dmn, "Domain doesn't support vport actions\n");
+ return NULL;
+ }
+
+ vport_cap = mlx5dr_get_vport_cap(&vport_dmn->info.caps, vport);
+ if (!vport_cap) {
+ mlx5dr_dbg(dmn, "Failed to get vport %d caps\n", vport);
+ return NULL;
+ }
+
+ action = dr_action_create_generic(DR_ACTION_TYP_VPORT);
+ if (!action)
+ return NULL;
+
+ action->vport.dmn = vport_dmn;
+ action->vport.caps = vport_cap;
+
+ return action;
+}
+
+int mlx5dr_action_destroy(struct mlx5dr_action *action)
+{
+ if (refcount_read(&action->refcount) > 1)
+ return -EBUSY;
+
+ switch (action->action_type) {
+ case DR_ACTION_TYP_FT:
+ if (!action->dest_tbl.is_fw_tbl)
+ refcount_dec(&action->dest_tbl.tbl->refcount);
+ break;
+ case DR_ACTION_TYP_TNL_L2_TO_L2:
+ refcount_dec(&action->reformat.dmn->refcount);
+ break;
+ case DR_ACTION_TYP_TNL_L3_TO_L2:
+ mlx5dr_icm_free_chunk(action->rewrite.chunk);
+ refcount_dec(&action->reformat.dmn->refcount);
+ break;
+ case DR_ACTION_TYP_L2_TO_TNL_L2:
+ case DR_ACTION_TYP_L2_TO_TNL_L3:
+ mlx5dr_cmd_destroy_reformat_ctx((action->reformat.dmn)->mdev,
+ action->reformat.reformat_id);
+ refcount_dec(&action->reformat.dmn->refcount);
+ break;
+ case DR_ACTION_TYP_MODIFY_HDR:
+ mlx5dr_icm_free_chunk(action->rewrite.chunk);
+ refcount_dec(&action->rewrite.dmn->refcount);
+ break;
+ default:
+ break;
+ }
+
+ kfree(action);
+ return 0;
+}
--
2.21.0
^ permalink raw reply related
* [net-next V2 12/18] net/mlx5: DR, Add required FW steering functionality
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Mark Bloch,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
SW steering is capable of doing many steering functionalities
but there are still some functionalities which are not exposed
to upper layers and therefore performed by the FW.
This is the support for recalculating checksum using a hairpin QP.
The recalculation is required after a modify TTL action which skips
the needed CS calculation in HW.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/dr_fw.c | 93 +++++++++++++++++++
1 file changed, 93 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
new file mode 100644
index 000000000000..60ef6e6171e3
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
@@ -0,0 +1,93 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#include <linux/types.h>
+#include "dr_types.h"
+
+struct mlx5dr_fw_recalc_cs_ft *
+mlx5dr_fw_create_recalc_cs_ft(struct mlx5dr_domain *dmn, u32 vport_num)
+{
+ struct mlx5dr_fw_recalc_cs_ft *recalc_cs_ft;
+ u32 table_id, group_id, modify_hdr_id;
+ u64 rx_icm_addr, modify_ttl_action;
+ int ret;
+
+ recalc_cs_ft = kzalloc(sizeof(*recalc_cs_ft), GFP_KERNEL);
+ if (!recalc_cs_ft)
+ return NULL;
+
+ ret = mlx5dr_cmd_create_flow_table(dmn->mdev, MLX5_FLOW_TABLE_TYPE_FDB,
+ 0, 0, dmn->info.caps.max_ft_level - 1,
+ false, true, &rx_icm_addr, &table_id);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed creating TTL W/A FW flow table %d\n", ret);
+ goto free_ttl_tbl;
+ }
+
+ ret = mlx5dr_cmd_create_empty_flow_group(dmn->mdev,
+ MLX5_FLOW_TABLE_TYPE_FDB,
+ table_id, &group_id);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed creating TTL W/A FW flow group %d\n", ret);
+ goto destroy_flow_table;
+ }
+
+ /* Modify TTL action by adding zero to trigger CS recalculation */
+ modify_ttl_action = 0;
+ MLX5_SET(set_action_in, &modify_ttl_action, action_type, MLX5_ACTION_TYPE_ADD);
+ MLX5_SET(set_action_in, &modify_ttl_action, field, MLX5_ACTION_IN_FIELD_OUT_IP_TTL);
+
+ ret = mlx5dr_cmd_alloc_modify_header(dmn->mdev, MLX5_FLOW_TABLE_TYPE_FDB, 1,
+ &modify_ttl_action,
+ &modify_hdr_id);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed modify header TTL %d\n", ret);
+ goto destroy_flow_group;
+ }
+
+ ret = mlx5dr_cmd_set_fte_modify_and_vport(dmn->mdev,
+ MLX5_FLOW_TABLE_TYPE_FDB,
+ table_id, group_id, modify_hdr_id,
+ vport_num);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed setting TTL W/A flow table entry %d\n", ret);
+ goto dealloc_modify_header;
+ }
+
+ recalc_cs_ft->modify_hdr_id = modify_hdr_id;
+ recalc_cs_ft->rx_icm_addr = rx_icm_addr;
+ recalc_cs_ft->table_id = table_id;
+ recalc_cs_ft->group_id = group_id;
+
+ return recalc_cs_ft;
+
+dealloc_modify_header:
+ mlx5dr_cmd_dealloc_modify_header(dmn->mdev, modify_hdr_id);
+destroy_flow_group:
+ mlx5dr_cmd_destroy_flow_group(dmn->mdev,
+ MLX5_FLOW_TABLE_TYPE_FDB,
+ table_id, group_id);
+destroy_flow_table:
+ mlx5dr_cmd_destroy_flow_table(dmn->mdev, table_id, MLX5_FLOW_TABLE_TYPE_FDB);
+free_ttl_tbl:
+ kfree(recalc_cs_ft);
+ return NULL;
+}
+
+void mlx5dr_fw_destroy_recalc_cs_ft(struct mlx5dr_domain *dmn,
+ struct mlx5dr_fw_recalc_cs_ft *recalc_cs_ft)
+{
+ mlx5dr_cmd_del_flow_table_entry(dmn->mdev,
+ MLX5_FLOW_TABLE_TYPE_FDB,
+ recalc_cs_ft->table_id);
+ mlx5dr_cmd_dealloc_modify_header(dmn->mdev, recalc_cs_ft->modify_hdr_id);
+ mlx5dr_cmd_destroy_flow_group(dmn->mdev,
+ MLX5_FLOW_TABLE_TYPE_FDB,
+ recalc_cs_ft->table_id,
+ recalc_cs_ft->group_id);
+ mlx5dr_cmd_destroy_flow_table(dmn->mdev,
+ recalc_cs_ft->table_id,
+ MLX5_FLOW_TABLE_TYPE_FDB);
+
+ kfree(recalc_cs_ft);
+}
--
2.21.0
^ permalink raw reply related
* [net-next V2 13/18] net/mlx5: DR, Expose APIs for direct rule managing
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Mark Bloch,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
Expose APIs for direct rule managing to increase insertion rate by
bypassing the firmware.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/mlx5dr.h | 212 ++++++++++++++++++
1 file changed, 212 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
new file mode 100644
index 000000000000..adda9cbfba45
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -0,0 +1,212 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2019, Mellanox Technologies */
+
+#ifndef _MLX5DR_H_
+#define _MLX5DR_H_
+
+struct mlx5dr_domain;
+struct mlx5dr_table;
+struct mlx5dr_matcher;
+struct mlx5dr_rule;
+struct mlx5dr_action;
+
+enum mlx5dr_domain_type {
+ MLX5DR_DOMAIN_TYPE_NIC_RX,
+ MLX5DR_DOMAIN_TYPE_NIC_TX,
+ MLX5DR_DOMAIN_TYPE_FDB,
+};
+
+enum mlx5dr_domain_sync_flags {
+ MLX5DR_DOMAIN_SYNC_FLAGS_SW = 1 << 0,
+ MLX5DR_DOMAIN_SYNC_FLAGS_HW = 1 << 1,
+};
+
+enum mlx5dr_action_reformat_type {
+ DR_ACTION_REFORMAT_TYP_TNL_L2_TO_L2,
+ DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L2,
+ DR_ACTION_REFORMAT_TYP_TNL_L3_TO_L2,
+ DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L3,
+};
+
+struct mlx5dr_match_parameters {
+ size_t match_sz;
+ u64 *match_buf; /* Device spec format */
+};
+
+#ifdef CONFIG_MLX5_SW_STEERING
+
+struct mlx5dr_domain *
+mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type);
+
+int mlx5dr_domain_destroy(struct mlx5dr_domain *domain);
+
+int mlx5dr_domain_sync(struct mlx5dr_domain *domain, u32 flags);
+
+void mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn,
+ struct mlx5dr_domain *peer_dmn);
+
+struct mlx5dr_table *
+mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level);
+
+int mlx5dr_table_destroy(struct mlx5dr_table *table);
+
+u32 mlx5dr_table_get_id(struct mlx5dr_table *table);
+
+struct mlx5dr_matcher *
+mlx5dr_matcher_create(struct mlx5dr_table *table,
+ u16 priority,
+ u8 match_criteria_enable,
+ struct mlx5dr_match_parameters *mask);
+
+int mlx5dr_matcher_destroy(struct mlx5dr_matcher *matcher);
+
+struct mlx5dr_rule *
+mlx5dr_rule_create(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *value,
+ size_t num_actions,
+ struct mlx5dr_action *actions[]);
+
+int mlx5dr_rule_destroy(struct mlx5dr_rule *rule);
+
+int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
+ struct mlx5dr_action *action);
+
+struct mlx5dr_action *
+mlx5dr_action_create_dest_table(struct mlx5dr_table *table);
+
+struct mlx5dr_action *
+mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
+ struct mlx5_core_dev *mdev);
+
+struct mlx5dr_action *
+mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
+ u32 vport, u8 vhca_id_valid,
+ u16 vhca_id);
+
+struct mlx5dr_action *mlx5dr_action_create_drop(void);
+
+struct mlx5dr_action *mlx5dr_action_create_tag(u32 tag_value);
+
+struct mlx5dr_action *
+mlx5dr_action_create_flow_counter(u32 counter_id);
+
+struct mlx5dr_action *
+mlx5dr_action_create_packet_reformat(struct mlx5dr_domain *dmn,
+ enum mlx5dr_action_reformat_type reformat_type,
+ size_t data_sz,
+ void *data);
+
+struct mlx5dr_action *
+mlx5dr_action_create_modify_header(struct mlx5dr_domain *domain,
+ u32 flags,
+ size_t actions_sz,
+ __be64 actions[]);
+
+struct mlx5dr_action *mlx5dr_action_create_pop_vlan(void);
+
+struct mlx5dr_action *
+mlx5dr_action_create_push_vlan(struct mlx5dr_domain *domain, __be32 vlan_hdr);
+
+int mlx5dr_action_destroy(struct mlx5dr_action *action);
+
+static inline bool
+mlx5dr_is_supported(struct mlx5_core_dev *dev)
+{
+ return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, sw_owner);
+}
+
+#else /* CONFIG_MLX5_SW_STEERING */
+
+static inline struct mlx5dr_domain *
+mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) { return NULL; }
+
+static inline int
+mlx5dr_domain_destroy(struct mlx5dr_domain *domain) { return 0; }
+
+static inline int
+mlx5dr_domain_sync(struct mlx5dr_domain *domain, u32 flags) { return 0; }
+
+static inline void
+mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn,
+ struct mlx5dr_domain *peer_dmn) { }
+
+static inline struct mlx5dr_table *
+mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level) { return NULL; }
+
+static inline int
+mlx5dr_table_destroy(struct mlx5dr_table *table) { return 0; }
+
+static inline u32
+mlx5dr_table_get_id(struct mlx5dr_table *table) { return 0; }
+
+static inline struct mlx5dr_matcher *
+mlx5dr_matcher_create(struct mlx5dr_table *table,
+ u16 priority,
+ u8 match_criteria_enable,
+ struct mlx5dr_match_parameters *mask) { return NULL; }
+
+static inline int
+mlx5dr_matcher_destroy(struct mlx5dr_matcher *matcher) { return 0; }
+
+static inline struct mlx5dr_rule *
+mlx5dr_rule_create(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *value,
+ size_t num_actions,
+ struct mlx5dr_action *actions[]) { return NULL; }
+
+static inline int
+mlx5dr_rule_destroy(struct mlx5dr_rule *rule) { return 0; }
+
+static inline int
+mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl,
+ struct mlx5dr_action *action) { return 0; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_dest_table(struct mlx5dr_table *table) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_create_action_dest_flow_fw_table(struct mlx5_flow_table *ft,
+ struct mlx5_core_dev *mdev) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_dest_vport(struct mlx5dr_domain *domain,
+ u32 vport, u8 vhca_id_valid,
+ u16 vhca_id) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_drop(void) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_tag(u32 tag_value) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_flow_counter(u32 counter_id) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_packet_reformat(struct mlx5dr_domain *dmn,
+ enum mlx5dr_action_reformat_type reformat_type,
+ size_t data_sz,
+ void *data) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_modify_header(struct mlx5dr_domain *domain,
+ u32 flags,
+ size_t actions_sz,
+ __be64 actions[]) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_pop_vlan(void) { return NULL; }
+
+static inline struct mlx5dr_action *
+mlx5dr_action_create_push_vlan(struct mlx5dr_domain *domain,
+ __be32 vlan_hdr) { return NULL; }
+
+static inline int
+mlx5dr_action_destroy(struct mlx5dr_action *action) { return 0; }
+
+static inline bool
+mlx5dr_is_supported(struct mlx5_core_dev *dev) { return false; }
+
+#endif /* CONFIG_MLX5_SW_STEERING */
+
+#endif /* _MLX5DR_H_ */
--
2.21.0
^ permalink raw reply related
* [net-next V2 14/18] net/mlx5: DR, Add CONFIG_MLX5_SW_STEERING for software steering support
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit,
Yevgeny Kliteynik, Mark Bloch, Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
Add new mlx5 Kconfig flag to allow selecting software steering
support and compile all the steering files only if the flag is
selected.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/Kconfig | 7 +++++++
drivers/net/ethernet/mellanox/mlx5/core/Makefile | 7 +++++++
drivers/net/ethernet/mellanox/mlx5/core/steering/Makefile | 2 ++
3 files changed, 16 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/Makefile
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index 37fef8cd25e3..0d8dd885b7d6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -154,3 +154,10 @@ config MLX5_EN_TLS
Build support for TLS cryptography-offload accelaration in the NIC.
Note: Support for hardware with this capability needs to be selected
for this option to become available.
+
+config MLX5_SW_STEERING
+ bool "Mellanox Technologies software-managed steering"
+ depends on MLX5_CORE_EN && MLX5_ESWITCH
+ default y
+ help
+ Build support for software-managed steering in the NIC.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index e9163875efd6..716558476eda 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -67,3 +67,10 @@ mlx5_core-$(CONFIG_MLX5_EN_IPSEC) += en_accel/ipsec.o en_accel/ipsec_rxtx.o \
mlx5_core-$(CONFIG_MLX5_EN_TLS) += en_accel/tls.o en_accel/tls_rxtx.o en_accel/tls_stats.o \
en_accel/ktls.o en_accel/ktls_tx.o
+
+mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o \
+ steering/dr_matcher.o steering/dr_rule.o \
+ steering/dr_icm_pool.o steering/dr_crc32.o \
+ steering/dr_ste.o steering/dr_send.o \
+ steering/dr_cmd.o steering/dr_fw.o \
+ steering/dr_action.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/steering/Makefile
new file mode 100644
index 000000000000..c78512eed8d7
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/Makefile
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+subdir-ccflags-y += -I$(src)/..
--
2.21.0
^ permalink raw reply related
* [net-next V2 02/18] net/mlx5: DR, Add the internal direct rule types definitions
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit,
Yevgeny Kliteynik, Mark Bloch, Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
Add the internal header file that contains various types
definition that will be used in coming patches as well as
the internal functions decelerations.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/dr_types.h | 1060 +++++++++++++++++
1 file changed, 1060 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
new file mode 100644
index 000000000000..a37ee6359be2
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -0,0 +1,1060 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2019, Mellanox Technologies */
+
+#ifndef _DR_TYPES_
+#define _DR_TYPES_
+
+#include <linux/mlx5/driver.h>
+#include <linux/refcount.h>
+#include "fs_core.h"
+#include "wq.h"
+#include "lib/mlx5.h"
+#include "mlx5_ifc_dr.h"
+#include "mlx5dr.h"
+
+#define DR_RULE_MAX_STES 17
+#define DR_ACTION_MAX_STES 5
+#define WIRE_PORT 0xFFFF
+#define DR_STE_SVLAN 0x1
+#define DR_STE_CVLAN 0x2
+
+#define mlx5dr_err(dmn, arg...) mlx5_core_err((dmn)->mdev, ##arg)
+#define mlx5dr_info(dmn, arg...) mlx5_core_info((dmn)->mdev, ##arg)
+#define mlx5dr_dbg(dmn, arg...) mlx5_core_dbg((dmn)->mdev, ##arg)
+
+enum mlx5dr_icm_chunk_size {
+ DR_CHUNK_SIZE_1,
+ DR_CHUNK_SIZE_MIN = DR_CHUNK_SIZE_1, /* keep updated when changing */
+ DR_CHUNK_SIZE_2,
+ DR_CHUNK_SIZE_4,
+ DR_CHUNK_SIZE_8,
+ DR_CHUNK_SIZE_16,
+ DR_CHUNK_SIZE_32,
+ DR_CHUNK_SIZE_64,
+ DR_CHUNK_SIZE_128,
+ DR_CHUNK_SIZE_256,
+ DR_CHUNK_SIZE_512,
+ DR_CHUNK_SIZE_1K,
+ DR_CHUNK_SIZE_2K,
+ DR_CHUNK_SIZE_4K,
+ DR_CHUNK_SIZE_8K,
+ DR_CHUNK_SIZE_16K,
+ DR_CHUNK_SIZE_32K,
+ DR_CHUNK_SIZE_64K,
+ DR_CHUNK_SIZE_128K,
+ DR_CHUNK_SIZE_256K,
+ DR_CHUNK_SIZE_512K,
+ DR_CHUNK_SIZE_1024K,
+ DR_CHUNK_SIZE_2048K,
+ DR_CHUNK_SIZE_MAX,
+};
+
+enum mlx5dr_icm_type {
+ DR_ICM_TYPE_STE,
+ DR_ICM_TYPE_MODIFY_ACTION,
+};
+
+static inline enum mlx5dr_icm_chunk_size
+mlx5dr_icm_next_higher_chunk(enum mlx5dr_icm_chunk_size chunk)
+{
+ chunk += 2;
+ if (chunk < DR_CHUNK_SIZE_MAX)
+ return chunk;
+
+ return DR_CHUNK_SIZE_MAX;
+}
+
+enum {
+ DR_STE_SIZE = 64,
+ DR_STE_SIZE_CTRL = 32,
+ DR_STE_SIZE_TAG = 16,
+ DR_STE_SIZE_MASK = 16,
+};
+
+enum {
+ DR_STE_SIZE_REDUCED = DR_STE_SIZE - DR_STE_SIZE_MASK,
+};
+
+enum {
+ DR_MODIFY_ACTION_SIZE = 8,
+};
+
+enum mlx5dr_matcher_criteria {
+ DR_MATCHER_CRITERIA_EMPTY = 0,
+ DR_MATCHER_CRITERIA_OUTER = 1 << 0,
+ DR_MATCHER_CRITERIA_MISC = 1 << 1,
+ DR_MATCHER_CRITERIA_INNER = 1 << 2,
+ DR_MATCHER_CRITERIA_MISC2 = 1 << 3,
+ DR_MATCHER_CRITERIA_MISC3 = 1 << 4,
+ DR_MATCHER_CRITERIA_MAX = 1 << 5,
+};
+
+enum mlx5dr_action_type {
+ DR_ACTION_TYP_TNL_L2_TO_L2,
+ DR_ACTION_TYP_L2_TO_TNL_L2,
+ DR_ACTION_TYP_TNL_L3_TO_L2,
+ DR_ACTION_TYP_L2_TO_TNL_L3,
+ DR_ACTION_TYP_DROP,
+ DR_ACTION_TYP_QP,
+ DR_ACTION_TYP_FT,
+ DR_ACTION_TYP_CTR,
+ DR_ACTION_TYP_TAG,
+ DR_ACTION_TYP_MODIFY_HDR,
+ DR_ACTION_TYP_VPORT,
+ DR_ACTION_TYP_POP_VLAN,
+ DR_ACTION_TYP_PUSH_VLAN,
+ DR_ACTION_TYP_MAX,
+};
+
+struct mlx5dr_icm_pool;
+struct mlx5dr_icm_chunk;
+struct mlx5dr_icm_bucket;
+struct mlx5dr_ste_htbl;
+struct mlx5dr_match_param;
+struct mlx5dr_cmd_caps;
+struct mlx5dr_matcher_rx_tx;
+
+struct mlx5dr_ste {
+ u8 *hw_ste;
+ /* refcount: indicates the num of rules that using this ste */
+ refcount_t refcount;
+
+ /* attached to the miss_list head at each htbl entry */
+ struct list_head miss_list_node;
+
+ /* each rule member that uses this ste attached here */
+ struct list_head rule_list;
+
+ /* this ste is member of htbl */
+ struct mlx5dr_ste_htbl *htbl;
+
+ struct mlx5dr_ste_htbl *next_htbl;
+
+ /* this ste is part of a rule, located in ste's chain */
+ u8 ste_chain_location;
+};
+
+struct mlx5dr_ste_htbl_ctrl {
+ /* total number of valid entries belonging to this hash table. This
+ * includes the non collision and collision entries
+ */
+ unsigned int num_of_valid_entries;
+
+ /* total number of collisions entries attached to this table */
+ unsigned int num_of_collisions;
+ unsigned int increase_threshold;
+ u8 may_grow:1;
+};
+
+struct mlx5dr_ste_htbl {
+ u8 lu_type;
+ u16 byte_mask;
+ refcount_t refcount;
+ struct mlx5dr_icm_chunk *chunk;
+ struct mlx5dr_ste *ste_arr;
+ u8 *hw_ste_arr;
+
+ struct list_head *miss_list;
+
+ enum mlx5dr_icm_chunk_size chunk_size;
+ struct mlx5dr_ste *pointing_ste;
+
+ struct mlx5dr_ste_htbl_ctrl ctrl;
+};
+
+struct mlx5dr_ste_send_info {
+ struct mlx5dr_ste *ste;
+ struct list_head send_list;
+ u16 size;
+ u16 offset;
+ u8 data_cont[DR_STE_SIZE];
+ u8 *data;
+};
+
+void mlx5dr_send_fill_and_append_ste_send_info(struct mlx5dr_ste *ste, u16 size,
+ u16 offset, u8 *data,
+ struct mlx5dr_ste_send_info *ste_info,
+ struct list_head *send_list,
+ bool copy_data);
+
+struct mlx5dr_ste_build {
+ u8 inner:1;
+ u8 rx:1;
+ struct mlx5dr_cmd_caps *caps;
+ u8 lu_type;
+ u16 byte_mask;
+ u8 bit_mask[DR_STE_SIZE_MASK];
+ int (*ste_build_tag_func)(struct mlx5dr_match_param *spec,
+ struct mlx5dr_ste_build *sb,
+ u8 *hw_ste_p);
+};
+
+struct mlx5dr_ste_htbl *
+mlx5dr_ste_htbl_alloc(struct mlx5dr_icm_pool *pool,
+ enum mlx5dr_icm_chunk_size chunk_size,
+ u8 lu_type, u16 byte_mask);
+
+int mlx5dr_ste_htbl_free(struct mlx5dr_ste_htbl *htbl);
+
+static inline void mlx5dr_htbl_put(struct mlx5dr_ste_htbl *htbl)
+{
+ if (refcount_dec_and_test(&htbl->refcount))
+ mlx5dr_ste_htbl_free(htbl);
+}
+
+static inline void mlx5dr_htbl_get(struct mlx5dr_ste_htbl *htbl)
+{
+ refcount_inc(&htbl->refcount);
+}
+
+/* STE utils */
+u32 mlx5dr_ste_calc_hash_index(u8 *hw_ste_p, struct mlx5dr_ste_htbl *htbl);
+void mlx5dr_ste_init(u8 *hw_ste_p, u8 lu_type, u8 entry_type, u16 gvmi);
+void mlx5dr_ste_always_hit_htbl(struct mlx5dr_ste *ste,
+ struct mlx5dr_ste_htbl *next_htbl);
+void mlx5dr_ste_set_miss_addr(u8 *hw_ste, u64 miss_addr);
+u64 mlx5dr_ste_get_miss_addr(u8 *hw_ste);
+void mlx5dr_ste_set_hit_gvmi(u8 *hw_ste_p, u16 gvmi);
+void mlx5dr_ste_set_hit_addr(u8 *hw_ste, u64 icm_addr, u32 ht_size);
+void mlx5dr_ste_always_miss_addr(struct mlx5dr_ste *ste, u64 miss_addr);
+void mlx5dr_ste_set_bit_mask(u8 *hw_ste_p, u8 *bit_mask);
+bool mlx5dr_ste_not_used_ste(struct mlx5dr_ste *ste);
+bool mlx5dr_ste_is_last_in_rule(struct mlx5dr_matcher_rx_tx *nic_matcher,
+ u8 ste_location);
+void mlx5dr_ste_rx_set_flow_tag(u8 *hw_ste_p, u32 flow_tag);
+void mlx5dr_ste_set_counter_id(u8 *hw_ste_p, u32 ctr_id);
+void mlx5dr_ste_set_tx_encap(void *hw_ste_p, u32 reformat_id,
+ int size, bool encap_l3);
+void mlx5dr_ste_set_rx_decap(u8 *hw_ste_p);
+void mlx5dr_ste_set_rx_decap_l3(u8 *hw_ste_p, bool vlan);
+void mlx5dr_ste_set_rx_pop_vlan(u8 *hw_ste_p);
+void mlx5dr_ste_set_tx_push_vlan(u8 *hw_ste_p, u32 vlan_tpid_pcp_dei_vid,
+ bool go_back);
+void mlx5dr_ste_set_entry_type(u8 *hw_ste_p, u8 entry_type);
+u8 mlx5dr_ste_get_entry_type(u8 *hw_ste_p);
+void mlx5dr_ste_set_rewrite_actions(u8 *hw_ste_p, u16 num_of_actions,
+ u32 re_write_index);
+void mlx5dr_ste_set_go_back_bit(u8 *hw_ste_p);
+u64 mlx5dr_ste_get_icm_addr(struct mlx5dr_ste *ste);
+u64 mlx5dr_ste_get_mr_addr(struct mlx5dr_ste *ste);
+struct list_head *mlx5dr_ste_get_miss_list(struct mlx5dr_ste *ste);
+
+void mlx5dr_ste_free(struct mlx5dr_ste *ste,
+ struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher);
+static inline void mlx5dr_ste_put(struct mlx5dr_ste *ste,
+ struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher)
+{
+ if (refcount_dec_and_test(&ste->refcount))
+ mlx5dr_ste_free(ste, matcher, nic_matcher);
+}
+
+/* initial as 0, increased only when ste appears in a new rule */
+static inline void mlx5dr_ste_get(struct mlx5dr_ste *ste)
+{
+ refcount_inc(&ste->refcount);
+}
+
+void mlx5dr_ste_set_hit_addr_by_next_htbl(u8 *hw_ste,
+ struct mlx5dr_ste_htbl *next_htbl);
+bool mlx5dr_ste_equal_tag(void *src, void *dst);
+int mlx5dr_ste_create_next_htbl(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste *ste,
+ u8 *cur_hw_ste,
+ enum mlx5dr_icm_chunk_size log_table_size);
+
+/* STE build functions */
+int mlx5dr_ste_build_pre_check(struct mlx5dr_domain *dmn,
+ u8 match_criteria,
+ struct mlx5dr_match_param *mask,
+ struct mlx5dr_match_param *value);
+int mlx5dr_ste_build_ste_arr(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_match_param *value,
+ u8 *ste_arr);
+int mlx5dr_ste_build_eth_l2_src_des(struct mlx5dr_ste_build *builder,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l3_ipv4_5_tuple(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l3_ipv4_misc(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l3_ipv6_dst(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l3_ipv6_src(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l2_src(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l2_dst(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l2_tnl(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_ipv6_l3_l4(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_eth_l4_misc(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_gre(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_mpls(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_flex_parser_0(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+int mlx5dr_ste_build_flex_parser_1(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ struct mlx5dr_cmd_caps *caps,
+ bool inner, bool rx);
+void mlx5dr_ste_build_flex_parser_tnl(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_general_purpose(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_register_0(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+void mlx5dr_ste_build_register_1(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ bool inner, bool rx);
+int mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb,
+ struct mlx5dr_match_param *mask,
+ struct mlx5dr_cmd_caps *caps,
+ bool inner, bool rx);
+void mlx5dr_ste_build_empty_always_hit(struct mlx5dr_ste_build *sb, bool rx);
+
+/* Actions utils */
+int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_action *actions[],
+ u32 num_actions,
+ u8 *ste_arr,
+ u32 *new_hw_ste_arr_sz);
+
+struct mlx5dr_match_spec {
+ u32 smac_47_16; /* Source MAC address of incoming packet */
+ /* Incoming packet Ethertype - this is the Ethertype
+ * following the last VLAN tag of the packet
+ */
+ u32 ethertype:16;
+ u32 smac_15_0:16; /* Source MAC address of incoming packet */
+ u32 dmac_47_16; /* Destination MAC address of incoming packet */
+ /* VLAN ID of first VLAN tag in the incoming packet.
+ * Valid only when cvlan_tag==1 or svlan_tag==1
+ */
+ u32 first_vid:12;
+ /* CFI bit of first VLAN tag in the incoming packet.
+ * Valid only when cvlan_tag==1 or svlan_tag==1
+ */
+ u32 first_cfi:1;
+ /* Priority of first VLAN tag in the incoming packet.
+ * Valid only when cvlan_tag==1 or svlan_tag==1
+ */
+ u32 first_prio:3;
+ u32 dmac_15_0:16; /* Destination MAC address of incoming packet */
+ /* TCP flags. ;Bit 0: FIN;Bit 1: SYN;Bit 2: RST;Bit 3: PSH;Bit 4: ACK;
+ * Bit 5: URG;Bit 6: ECE;Bit 7: CWR;Bit 8: NS
+ */
+ u32 tcp_flags:9;
+ u32 ip_version:4; /* IP version */
+ u32 frag:1; /* Packet is an IP fragment */
+ /* The first vlan in the packet is s-vlan (0x8a88).
+ * cvlan_tag and svlan_tag cannot be set together
+ */
+ u32 svlan_tag:1;
+ /* The first vlan in the packet is c-vlan (0x8100).
+ * cvlan_tag and svlan_tag cannot be set together
+ */
+ u32 cvlan_tag:1;
+ /* Explicit Congestion Notification derived from
+ * Traffic Class/TOS field of IPv6/v4
+ */
+ u32 ip_ecn:2;
+ /* Differentiated Services Code Point derived from
+ * Traffic Class/TOS field of IPv6/v4
+ */
+ u32 ip_dscp:6;
+ u32 ip_protocol:8; /* IP protocol */
+ /* TCP destination port.
+ * tcp and udp sport/dport are mutually exclusive
+ */
+ u32 tcp_dport:16;
+ /* TCP source port.;tcp and udp sport/dport are mutually exclusive */
+ u32 tcp_sport:16;
+ u32 ttl_hoplimit:8;
+ u32 reserved:24;
+ /* UDP destination port.;tcp and udp sport/dport are mutually exclusive */
+ u32 udp_dport:16;
+ /* UDP source port.;tcp and udp sport/dport are mutually exclusive */
+ u32 udp_sport:16;
+ /* IPv6 source address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 src_ip_127_96;
+ /* IPv6 source address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 src_ip_95_64;
+ /* IPv6 source address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 src_ip_63_32;
+ /* IPv6 source address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 src_ip_31_0;
+ /* IPv6 destination address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 dst_ip_127_96;
+ /* IPv6 destination address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 dst_ip_95_64;
+ /* IPv6 destination address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 dst_ip_63_32;
+ /* IPv6 destination address of incoming packets
+ * For IPv4 address use bits 31:0 (rest of the bits are reserved)
+ * This field should be qualified by an appropriate ethertype
+ */
+ u32 dst_ip_31_0;
+};
+
+struct mlx5dr_match_misc {
+ u32 source_sqn:24; /* Source SQN */
+ u32 source_vhca_port:4;
+ /* used with GRE, sequence number exist when gre_s_present == 1 */
+ u32 gre_s_present:1;
+ /* used with GRE, key exist when gre_k_present == 1 */
+ u32 gre_k_present:1;
+ u32 reserved_auto1:1;
+ /* used with GRE, checksum exist when gre_c_present == 1 */
+ u32 gre_c_present:1;
+ /* Source port.;0xffff determines wire port */
+ u32 source_port:16;
+ u32 reserved_auto2:16;
+ /* VLAN ID of first VLAN tag the inner header of the incoming packet.
+ * Valid only when inner_second_cvlan_tag ==1 or inner_second_svlan_tag ==1
+ */
+ u32 inner_second_vid:12;
+ /* CFI bit of first VLAN tag in the inner header of the incoming packet.
+ * Valid only when inner_second_cvlan_tag ==1 or inner_second_svlan_tag ==1
+ */
+ u32 inner_second_cfi:1;
+ /* Priority of second VLAN tag in the inner header of the incoming packet.
+ * Valid only when inner_second_cvlan_tag ==1 or inner_second_svlan_tag ==1
+ */
+ u32 inner_second_prio:3;
+ /* VLAN ID of first VLAN tag the outer header of the incoming packet.
+ * Valid only when outer_second_cvlan_tag ==1 or outer_second_svlan_tag ==1
+ */
+ u32 outer_second_vid:12;
+ /* CFI bit of first VLAN tag in the outer header of the incoming packet.
+ * Valid only when outer_second_cvlan_tag ==1 or outer_second_svlan_tag ==1
+ */
+ u32 outer_second_cfi:1;
+ /* Priority of second VLAN tag in the outer header of the incoming packet.
+ * Valid only when outer_second_cvlan_tag ==1 or outer_second_svlan_tag ==1
+ */
+ u32 outer_second_prio:3;
+ u32 gre_protocol:16; /* GRE Protocol (outer) */
+ u32 reserved_auto3:12;
+ /* The second vlan in the inner header of the packet is s-vlan (0x8a88).
+ * inner_second_cvlan_tag and inner_second_svlan_tag cannot be set together
+ */
+ u32 inner_second_svlan_tag:1;
+ /* The second vlan in the outer header of the packet is s-vlan (0x8a88).
+ * outer_second_cvlan_tag and outer_second_svlan_tag cannot be set together
+ */
+ u32 outer_second_svlan_tag:1;
+ /* The second vlan in the inner header of the packet is c-vlan (0x8100).
+ * inner_second_cvlan_tag and inner_second_svlan_tag cannot be set together
+ */
+ u32 inner_second_cvlan_tag:1;
+ /* The second vlan in the outer header of the packet is c-vlan (0x8100).
+ * outer_second_cvlan_tag and outer_second_svlan_tag cannot be set together
+ */
+ u32 outer_second_cvlan_tag:1;
+ u32 gre_key_l:8; /* GRE Key [7:0] (outer) */
+ u32 gre_key_h:24; /* GRE Key[31:8] (outer) */
+ u32 reserved_auto4:8;
+ u32 vxlan_vni:24; /* VXLAN VNI (outer) */
+ u32 geneve_oam:1; /* GENEVE OAM field (outer) */
+ u32 reserved_auto5:7;
+ u32 geneve_vni:24; /* GENEVE VNI field (outer) */
+ u32 outer_ipv6_flow_label:20; /* Flow label of incoming IPv6 packet (outer) */
+ u32 reserved_auto6:12;
+ u32 inner_ipv6_flow_label:20; /* Flow label of incoming IPv6 packet (inner) */
+ u32 reserved_auto7:12;
+ u32 geneve_protocol_type:16; /* GENEVE protocol type (outer) */
+ u32 geneve_opt_len:6; /* GENEVE OptLen (outer) */
+ u32 reserved_auto8:10;
+ u32 bth_dst_qp:24; /* Destination QP in BTH header */
+ u32 reserved_auto9:8;
+ u8 reserved_auto10[20];
+};
+
+struct mlx5dr_match_misc2 {
+ u32 outer_first_mpls_ttl:8; /* First MPLS TTL (outer) */
+ u32 outer_first_mpls_s_bos:1; /* First MPLS S_BOS (outer) */
+ u32 outer_first_mpls_exp:3; /* First MPLS EXP (outer) */
+ u32 outer_first_mpls_label:20; /* First MPLS LABEL (outer) */
+ u32 inner_first_mpls_ttl:8; /* First MPLS TTL (inner) */
+ u32 inner_first_mpls_s_bos:1; /* First MPLS S_BOS (inner) */
+ u32 inner_first_mpls_exp:3; /* First MPLS EXP (inner) */
+ u32 inner_first_mpls_label:20; /* First MPLS LABEL (inner) */
+ u32 outer_first_mpls_over_gre_ttl:8; /* last MPLS TTL (outer) */
+ u32 outer_first_mpls_over_gre_s_bos:1; /* last MPLS S_BOS (outer) */
+ u32 outer_first_mpls_over_gre_exp:3; /* last MPLS EXP (outer) */
+ u32 outer_first_mpls_over_gre_label:20; /* last MPLS LABEL (outer) */
+ u32 outer_first_mpls_over_udp_ttl:8; /* last MPLS TTL (outer) */
+ u32 outer_first_mpls_over_udp_s_bos:1; /* last MPLS S_BOS (outer) */
+ u32 outer_first_mpls_over_udp_exp:3; /* last MPLS EXP (outer) */
+ u32 outer_first_mpls_over_udp_label:20; /* last MPLS LABEL (outer) */
+ u32 metadata_reg_c_7; /* metadata_reg_c_7 */
+ u32 metadata_reg_c_6; /* metadata_reg_c_6 */
+ u32 metadata_reg_c_5; /* metadata_reg_c_5 */
+ u32 metadata_reg_c_4; /* metadata_reg_c_4 */
+ u32 metadata_reg_c_3; /* metadata_reg_c_3 */
+ u32 metadata_reg_c_2; /* metadata_reg_c_2 */
+ u32 metadata_reg_c_1; /* metadata_reg_c_1 */
+ u32 metadata_reg_c_0; /* metadata_reg_c_0 */
+ u32 metadata_reg_a; /* metadata_reg_a */
+ u32 metadata_reg_b; /* metadata_reg_b */
+ u8 reserved_auto2[8];
+};
+
+struct mlx5dr_match_misc3 {
+ u32 inner_tcp_seq_num;
+ u32 outer_tcp_seq_num;
+ u32 inner_tcp_ack_num;
+ u32 outer_tcp_ack_num;
+ u32 outer_vxlan_gpe_vni:24;
+ u32 reserved_auto1:8;
+ u32 reserved_auto2:16;
+ u32 outer_vxlan_gpe_flags:8;
+ u32 outer_vxlan_gpe_next_protocol:8;
+ u32 icmpv4_header_data;
+ u32 icmpv6_header_data;
+ u32 icmpv6_code:8;
+ u32 icmpv6_type:8;
+ u32 icmpv4_code:8;
+ u32 icmpv4_type:8;
+ u8 reserved_auto3[0x1c];
+};
+
+struct mlx5dr_match_param {
+ struct mlx5dr_match_spec outer;
+ struct mlx5dr_match_misc misc;
+ struct mlx5dr_match_spec inner;
+ struct mlx5dr_match_misc2 misc2;
+ struct mlx5dr_match_misc3 misc3;
+};
+
+#define DR_MASK_IS_FLEX_PARSER_ICMPV4_SET(_misc3) ((_misc3)->icmpv4_type || \
+ (_misc3)->icmpv4_code || \
+ (_misc3)->icmpv4_header_data)
+
+struct mlx5dr_esw_caps {
+ u64 drop_icm_address_rx;
+ u64 drop_icm_address_tx;
+ u64 uplink_icm_address_rx;
+ u64 uplink_icm_address_tx;
+ bool sw_owner;
+};
+
+struct mlx5dr_cmd_vport_cap {
+ u16 vport_gvmi;
+ u16 vhca_gvmi;
+ u64 icm_address_rx;
+ u64 icm_address_tx;
+ u32 num;
+};
+
+struct mlx5dr_cmd_caps {
+ u16 gvmi;
+ u64 nic_rx_drop_address;
+ u64 nic_tx_drop_address;
+ u64 nic_tx_allow_address;
+ u64 esw_rx_drop_address;
+ u64 esw_tx_drop_address;
+ u32 log_icm_size;
+ u64 hdr_modify_icm_addr;
+ u32 flex_protocols;
+ u8 flex_parser_id_icmp_dw0;
+ u8 flex_parser_id_icmp_dw1;
+ u8 flex_parser_id_icmpv6_dw0;
+ u8 flex_parser_id_icmpv6_dw1;
+ u8 max_ft_level;
+ u16 roce_min_src_udp;
+ u8 num_esw_ports;
+ bool eswitch_manager;
+ bool rx_sw_owner;
+ bool tx_sw_owner;
+ bool fdb_sw_owner;
+ u32 num_vports;
+ struct mlx5dr_esw_caps esw_caps;
+ struct mlx5dr_cmd_vport_cap *vports_caps;
+ bool prio_tag_required;
+};
+
+struct mlx5dr_domain_rx_tx {
+ u64 drop_icm_addr;
+ u64 default_icm_addr;
+ enum mlx5dr_ste_entry_type ste_type;
+};
+
+struct mlx5dr_domain_info {
+ bool supp_sw_steering;
+ u32 max_inline_size;
+ u32 max_send_wr;
+ u32 max_log_sw_icm_sz;
+ u32 max_log_action_icm_sz;
+ struct mlx5dr_domain_rx_tx rx;
+ struct mlx5dr_domain_rx_tx tx;
+ struct mlx5dr_cmd_caps caps;
+};
+
+struct mlx5dr_domain_cache {
+ struct mlx5dr_fw_recalc_cs_ft **recalc_cs_ft;
+};
+
+struct mlx5dr_domain {
+ struct mlx5dr_domain *peer_dmn;
+ struct mlx5_core_dev *mdev;
+ u32 pdn;
+ struct mlx5_uars_page *uar;
+ enum mlx5dr_domain_type type;
+ refcount_t refcount;
+ struct mutex mutex; /* protect domain */
+ struct mlx5dr_icm_pool *ste_icm_pool;
+ struct mlx5dr_icm_pool *action_icm_pool;
+ struct mlx5dr_send_ring *send_ring;
+ struct mlx5dr_domain_info info;
+ struct mlx5dr_domain_cache cache;
+};
+
+struct mlx5dr_table_rx_tx {
+ struct mlx5dr_ste_htbl *s_anchor;
+ struct mlx5dr_domain_rx_tx *nic_dmn;
+ u64 default_icm_addr;
+};
+
+struct mlx5dr_table {
+ struct mlx5dr_domain *dmn;
+ struct mlx5dr_table_rx_tx rx;
+ struct mlx5dr_table_rx_tx tx;
+ u32 level;
+ u32 table_type;
+ u32 table_id;
+ struct list_head matcher_list;
+ struct mlx5dr_action *miss_action;
+ refcount_t refcount;
+};
+
+struct mlx5dr_matcher_rx_tx {
+ struct mlx5dr_ste_htbl *s_htbl;
+ struct mlx5dr_ste_htbl *e_anchor;
+ struct mlx5dr_ste_build *ste_builder;
+ struct mlx5dr_ste_build ste_builder4[DR_RULE_MAX_STES];
+ struct mlx5dr_ste_build ste_builder6[DR_RULE_MAX_STES];
+ u8 num_of_builders;
+ u8 num_of_builders4;
+ u8 num_of_builders6;
+ u64 default_icm_addr;
+ struct mlx5dr_table_rx_tx *nic_tbl;
+};
+
+struct mlx5dr_matcher {
+ struct mlx5dr_table *tbl;
+ struct mlx5dr_matcher_rx_tx rx;
+ struct mlx5dr_matcher_rx_tx tx;
+ struct list_head matcher_list;
+ u16 prio;
+ struct mlx5dr_match_param mask;
+ u8 match_criteria;
+ refcount_t refcount;
+ struct mlx5dv_flow_matcher *dv_matcher;
+};
+
+struct mlx5dr_rule_member {
+ struct mlx5dr_ste *ste;
+ /* attached to mlx5dr_rule via this */
+ struct list_head list;
+ /* attached to mlx5dr_ste via this */
+ struct list_head use_ste_list;
+};
+
+struct mlx5dr_action {
+ enum mlx5dr_action_type action_type;
+ refcount_t refcount;
+ union {
+ struct {
+ struct mlx5dr_domain *dmn;
+ struct mlx5dr_icm_chunk *chunk;
+ u8 *data;
+ u32 data_size;
+ u16 num_of_actions;
+ u32 index;
+ u8 allow_rx:1;
+ u8 allow_tx:1;
+ u8 modify_ttl:1;
+ } rewrite;
+ struct {
+ struct mlx5dr_domain *dmn;
+ u32 reformat_id;
+ u32 reformat_size;
+ } reformat;
+ struct {
+ u8 is_fw_tbl:1;
+ union {
+ struct mlx5dr_table *tbl;
+ struct {
+ struct mlx5_flow_table *ft;
+ u64 rx_icm_addr;
+ u64 tx_icm_addr;
+ struct mlx5_core_dev *mdev;
+ } fw_tbl;
+ };
+ } dest_tbl;
+ struct {
+ u32 ctr_id;
+ u32 offeset;
+ } ctr;
+ struct {
+ struct mlx5dr_domain *dmn;
+ struct mlx5dr_cmd_vport_cap *caps;
+ u32 num;
+ } vport;
+ struct {
+ u32 vlan_hdr; /* tpid_pcp_dei_vid */
+ } push_vlan;
+ u32 flow_tag;
+ };
+};
+
+enum mlx5dr_connect_type {
+ CONNECT_HIT = 1,
+ CONNECT_MISS = 2,
+};
+
+struct mlx5dr_htbl_connect_info {
+ enum mlx5dr_connect_type type;
+ union {
+ struct mlx5dr_ste_htbl *hit_next_htbl;
+ u64 miss_icm_addr;
+ };
+};
+
+struct mlx5dr_rule_rx_tx {
+ struct list_head rule_members_list;
+ struct mlx5dr_matcher_rx_tx *nic_matcher;
+};
+
+struct mlx5dr_rule {
+ struct mlx5dr_matcher *matcher;
+ struct mlx5dr_rule_rx_tx rx;
+ struct mlx5dr_rule_rx_tx tx;
+ struct list_head rule_actions_list;
+};
+
+void mlx5dr_rule_update_rule_member(struct mlx5dr_ste *new_ste,
+ struct mlx5dr_ste *ste);
+
+struct mlx5dr_icm_chunk {
+ struct mlx5dr_icm_bucket *bucket;
+ struct list_head chunk_list;
+ u32 rkey;
+ u32 num_of_entries;
+ u32 byte_size;
+ u64 icm_addr;
+ u64 mr_addr;
+
+ /* Memory optimisation */
+ struct mlx5dr_ste *ste_arr;
+ u8 *hw_ste_arr;
+ struct list_head *miss_list;
+};
+
+static inline int
+mlx5dr_matcher_supp_flex_parser_icmp_v4(struct mlx5dr_cmd_caps *caps)
+{
+ return caps->flex_protocols & MLX5_FLEX_PARSER_ICMP_V4_ENABLED;
+}
+
+static inline int
+mlx5dr_matcher_supp_flex_parser_icmp_v6(struct mlx5dr_cmd_caps *caps)
+{
+ return caps->flex_protocols & MLX5_FLEX_PARSER_ICMP_V6_ENABLED;
+}
+
+int mlx5dr_matcher_select_builders(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ bool ipv6);
+
+static inline u32
+mlx5dr_icm_pool_chunk_size_to_entries(enum mlx5dr_icm_chunk_size chunk_size)
+{
+ return 1 << chunk_size;
+}
+
+static inline int
+mlx5dr_icm_pool_chunk_size_to_byte(enum mlx5dr_icm_chunk_size chunk_size,
+ enum mlx5dr_icm_type icm_type)
+{
+ int num_of_entries;
+ int entry_size;
+
+ if (icm_type == DR_ICM_TYPE_STE)
+ entry_size = DR_STE_SIZE;
+ else
+ entry_size = DR_MODIFY_ACTION_SIZE;
+
+ num_of_entries = mlx5dr_icm_pool_chunk_size_to_entries(chunk_size);
+
+ return entry_size * num_of_entries;
+}
+
+static inline struct mlx5dr_cmd_vport_cap *
+mlx5dr_get_vport_cap(struct mlx5dr_cmd_caps *caps, u32 vport)
+{
+ if (!caps->vports_caps ||
+ (vport >= caps->num_vports && vport != WIRE_PORT))
+ return NULL;
+
+ if (vport == WIRE_PORT)
+ vport = caps->num_vports;
+
+ return &caps->vports_caps[vport];
+}
+
+struct mlx5dr_cmd_query_flow_table_details {
+ u8 status;
+ u8 level;
+ u64 sw_owner_icm_root_1;
+ u64 sw_owner_icm_root_0;
+};
+
+/* internal API functions */
+int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
+ struct mlx5dr_cmd_caps *caps);
+int mlx5dr_cmd_query_esw_vport_context(struct mlx5_core_dev *mdev,
+ bool other_vport, u16 vport_number,
+ u64 *icm_address_rx,
+ u64 *icm_address_tx);
+int mlx5dr_cmd_query_gvmi(struct mlx5_core_dev *mdev,
+ bool other_vport, u16 vport_number, u16 *gvmi);
+int mlx5dr_cmd_query_esw_caps(struct mlx5_core_dev *mdev,
+ struct mlx5dr_esw_caps *caps);
+int mlx5dr_cmd_sync_steering(struct mlx5_core_dev *mdev);
+int mlx5dr_cmd_set_fte_modify_and_vport(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u32 table_id,
+ u32 group_id,
+ u32 modify_header_id,
+ u32 vport_id);
+int mlx5dr_cmd_del_flow_table_entry(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u32 table_id);
+int mlx5dr_cmd_alloc_modify_header(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u8 num_of_actions,
+ u64 *actions,
+ u32 *modify_header_id);
+int mlx5dr_cmd_dealloc_modify_header(struct mlx5_core_dev *mdev,
+ u32 modify_header_id);
+int mlx5dr_cmd_create_empty_flow_group(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u32 table_id,
+ u32 *group_id);
+int mlx5dr_cmd_destroy_flow_group(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u32 table_id,
+ u32 group_id);
+int mlx5dr_cmd_create_flow_table(struct mlx5_core_dev *mdev,
+ u32 table_type,
+ u64 icm_addr_rx,
+ u64 icm_addr_tx,
+ u8 level,
+ bool sw_owner,
+ bool term_tbl,
+ u64 *fdb_rx_icm_addr,
+ u32 *table_id);
+int mlx5dr_cmd_destroy_flow_table(struct mlx5_core_dev *mdev,
+ u32 table_id,
+ u32 table_type);
+int mlx5dr_cmd_query_flow_table(struct mlx5_core_dev *dev,
+ enum fs_flow_table_type type,
+ u32 table_id,
+ struct mlx5dr_cmd_query_flow_table_details *output);
+int mlx5dr_cmd_create_reformat_ctx(struct mlx5_core_dev *mdev,
+ enum mlx5_reformat_ctx_type rt,
+ size_t reformat_size,
+ void *reformat_data,
+ u32 *reformat_id);
+void mlx5dr_cmd_destroy_reformat_ctx(struct mlx5_core_dev *mdev,
+ u32 reformat_id);
+
+struct mlx5dr_cmd_gid_attr {
+ u8 gid[16];
+ u8 mac[6];
+ u32 roce_ver;
+};
+
+struct mlx5dr_cmd_qp_create_attr {
+ u32 page_id;
+ u32 pdn;
+ u32 cqn;
+ u32 pm_state;
+ u32 service_type;
+ u32 buff_umem_id;
+ u32 db_umem_id;
+ u32 sq_wqe_cnt;
+ u32 rq_wqe_cnt;
+ u32 rq_wqe_shift;
+};
+
+int mlx5dr_cmd_query_gid(struct mlx5_core_dev *mdev, u8 vhca_port_num,
+ u16 index, struct mlx5dr_cmd_gid_attr *attr);
+
+struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
+ enum mlx5dr_icm_type icm_type);
+void mlx5dr_icm_pool_destroy(struct mlx5dr_icm_pool *pool);
+
+struct mlx5dr_icm_chunk *
+mlx5dr_icm_alloc_chunk(struct mlx5dr_icm_pool *pool,
+ enum mlx5dr_icm_chunk_size chunk_size);
+void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk);
+bool mlx5dr_ste_is_not_valid_entry(u8 *p_hw_ste);
+int mlx5dr_ste_htbl_init_and_postsend(struct mlx5dr_domain *dmn,
+ struct mlx5dr_domain_rx_tx *nic_dmn,
+ struct mlx5dr_ste_htbl *htbl,
+ struct mlx5dr_htbl_connect_info *connect_info,
+ bool update_hw_ste);
+void mlx5dr_ste_set_formatted_ste(u16 gvmi,
+ struct mlx5dr_domain_rx_tx *nic_dmn,
+ struct mlx5dr_ste_htbl *htbl,
+ u8 *formatted_ste,
+ struct mlx5dr_htbl_connect_info *connect_info);
+void mlx5dr_ste_copy_param(u8 match_criteria,
+ struct mlx5dr_match_param *set_param,
+ struct mlx5dr_match_parameters *mask);
+
+void mlx5dr_crc32_init_table(void);
+u32 mlx5dr_crc32_slice8_calc(const void *input_data, size_t length);
+
+struct mlx5dr_qp {
+ struct mlx5_core_dev *mdev;
+ struct mlx5_wq_qp wq;
+ struct mlx5_uars_page *uar;
+ struct mlx5_wq_ctrl wq_ctrl;
+ struct mlx5_core_qp mqp;
+ struct {
+ unsigned int pc;
+ unsigned int cc;
+ unsigned int size;
+ unsigned int *wqe_head;
+ unsigned int wqe_cnt;
+ } sq;
+ struct {
+ unsigned int pc;
+ unsigned int cc;
+ unsigned int size;
+ unsigned int wqe_cnt;
+ } rq;
+ int max_inline_data;
+};
+
+struct mlx5dr_cq {
+ struct mlx5_core_dev *mdev;
+ struct mlx5_cqwq wq;
+ struct mlx5_wq_ctrl wq_ctrl;
+ struct mlx5_core_cq mcq;
+ struct mlx5dr_qp *qp;
+};
+
+struct mlx5dr_mr {
+ struct mlx5_core_dev *mdev;
+ struct mlx5_core_mkey mkey;
+ dma_addr_t dma_addr;
+ void *addr;
+ size_t size;
+};
+
+#define MAX_SEND_CQE 64
+#define MIN_READ_SYNC 64
+
+struct mlx5dr_send_ring {
+ struct mlx5dr_cq *cq;
+ struct mlx5dr_qp *qp;
+ struct mlx5dr_mr *mr;
+ /* How much wqes are waiting for completion */
+ u32 pending_wqe;
+ /* Signal request per this trash hold value */
+ u16 signal_th;
+ /* Each post_send_size less than max_post_send_size */
+ u32 max_post_send_size;
+ /* manage the send queue */
+ u32 tx_head;
+ void *buf;
+ u32 buf_size;
+ struct ib_wc wc[MAX_SEND_CQE];
+ u8 sync_buff[MIN_READ_SYNC];
+ struct mlx5dr_mr *sync_mr;
+};
+
+int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn);
+void mlx5dr_send_ring_free(struct mlx5dr_domain *dmn,
+ struct mlx5dr_send_ring *send_ring);
+int mlx5dr_send_ring_force_drain(struct mlx5dr_domain *dmn);
+int mlx5dr_send_postsend_ste(struct mlx5dr_domain *dmn,
+ struct mlx5dr_ste *ste,
+ u8 *data,
+ u16 size,
+ u16 offset);
+int mlx5dr_send_postsend_htbl(struct mlx5dr_domain *dmn,
+ struct mlx5dr_ste_htbl *htbl,
+ u8 *formatted_ste, u8 *mask);
+int mlx5dr_send_postsend_formatted_htbl(struct mlx5dr_domain *dmn,
+ struct mlx5dr_ste_htbl *htbl,
+ u8 *ste_init_data,
+ bool update_hw_ste);
+int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn,
+ struct mlx5dr_action *action);
+
+struct mlx5dr_fw_recalc_cs_ft {
+ u64 rx_icm_addr;
+ u32 table_id;
+ u32 group_id;
+ u32 modify_hdr_id;
+};
+
+struct mlx5dr_fw_recalc_cs_ft *
+mlx5dr_fw_create_recalc_cs_ft(struct mlx5dr_domain *dmn, u32 vport_num);
+void mlx5dr_fw_destroy_recalc_cs_ft(struct mlx5dr_domain *dmn,
+ struct mlx5dr_fw_recalc_cs_ft *recalc_cs_ft);
+int mlx5dr_domain_cache_get_recalc_cs_ft_addr(struct mlx5dr_domain *dmn,
+ u32 vport_num,
+ u64 *rx_icm_addr);
+#endif /* _DR_TYPES_H_ */
--
2.21.0
^ permalink raw reply related
* [net-next V2 15/18] net/mlx5: Add direct rule fs_cmd implementation
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Maor Gottlieb,
Mark Bloch, Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Maor Gottlieb <maorg@mellanox.com>
Add support to create flow steering objects
via direct rule API (SW steering).
New layer is added - fs_dr, this layer translates the command that
fs_core sends to the FW into direct rule API. In case that direct
rule is not supported in some feature then -EOPNOTSUPP is
returned.
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
.../net/ethernet/mellanox/mlx5/core/fs_cmd.c | 28 +-
.../net/ethernet/mellanox/mlx5/core/fs_cmd.h | 7 +
.../net/ethernet/mellanox/mlx5/core/fs_core.c | 6 +
.../net/ethernet/mellanox/mlx5/core/fs_core.h | 20 +-
.../mellanox/mlx5/core/steering/fs_dr.c | 600 ++++++++++++++++++
.../mellanox/mlx5/core/steering/fs_dr.h | 60 ++
7 files changed, 717 insertions(+), 6 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 716558476eda..5708fcc079ca 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -73,4 +73,4 @@ mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o
steering/dr_icm_pool.o steering/dr_crc32.o \
steering/dr_ste.o steering/dr_send.o \
steering/dr_cmd.o steering/dr_fw.o \
- steering/dr_action.o
+ steering/dr_action.o steering/fs_dr.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 488f50dfb404..579c306caa7b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -135,6 +135,22 @@ static void mlx5_cmd_stub_modify_header_dealloc(struct mlx5_flow_root_namespace
{
}
+static int mlx5_cmd_stub_set_peer(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_root_namespace *peer_ns)
+{
+ return 0;
+}
+
+static int mlx5_cmd_stub_create_ns(struct mlx5_flow_root_namespace *ns)
+{
+ return 0;
+}
+
+static int mlx5_cmd_stub_destroy_ns(struct mlx5_flow_root_namespace *ns)
+{
+ return 0;
+}
+
static int mlx5_cmd_update_root_ft(struct mlx5_flow_root_namespace *ns,
struct mlx5_flow_table *ft, u32 underlay_qpn,
bool disconnect)
@@ -838,7 +854,10 @@ static const struct mlx5_flow_cmds mlx5_flow_cmds = {
.packet_reformat_alloc = mlx5_cmd_packet_reformat_alloc,
.packet_reformat_dealloc = mlx5_cmd_packet_reformat_dealloc,
.modify_header_alloc = mlx5_cmd_modify_header_alloc,
- .modify_header_dealloc = mlx5_cmd_modify_header_dealloc
+ .modify_header_dealloc = mlx5_cmd_modify_header_dealloc,
+ .set_peer = mlx5_cmd_stub_set_peer,
+ .create_ns = mlx5_cmd_stub_create_ns,
+ .destroy_ns = mlx5_cmd_stub_destroy_ns,
};
static const struct mlx5_flow_cmds mlx5_flow_cmd_stubs = {
@@ -854,10 +873,13 @@ static const struct mlx5_flow_cmds mlx5_flow_cmd_stubs = {
.packet_reformat_alloc = mlx5_cmd_stub_packet_reformat_alloc,
.packet_reformat_dealloc = mlx5_cmd_stub_packet_reformat_dealloc,
.modify_header_alloc = mlx5_cmd_stub_modify_header_alloc,
- .modify_header_dealloc = mlx5_cmd_stub_modify_header_dealloc
+ .modify_header_dealloc = mlx5_cmd_stub_modify_header_dealloc,
+ .set_peer = mlx5_cmd_stub_set_peer,
+ .create_ns = mlx5_cmd_stub_create_ns,
+ .destroy_ns = mlx5_cmd_stub_destroy_ns,
};
-static const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void)
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void)
{
return &mlx5_flow_cmds;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
index 3268654d6748..d62de642eca9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
@@ -93,6 +93,12 @@ struct mlx5_flow_cmds {
void (*modify_header_dealloc)(struct mlx5_flow_root_namespace *ns,
struct mlx5_modify_hdr *modify_hdr);
+
+ int (*set_peer)(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_root_namespace *peer_ns);
+
+ int (*create_ns)(struct mlx5_flow_root_namespace *ns);
+ int (*destroy_ns)(struct mlx5_flow_root_namespace *ns);
};
int mlx5_cmd_fc_alloc(struct mlx5_core_dev *dev, u32 *id);
@@ -108,5 +114,6 @@ int mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, u32 base_id, int bulk_len,
u32 *out);
const struct mlx5_flow_cmds *mlx5_fs_cmd_get_default(enum fs_flow_table_type type);
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void);
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 1d2333fd3080..c2d6e9f4cb90 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2991,3 +2991,9 @@ void mlx5_packet_reformat_dealloc(struct mlx5_core_dev *dev,
kfree(pkt_reformat);
}
EXPORT_SYMBOL(mlx5_packet_reformat_dealloc);
+
+int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_root_namespace *peer_ns)
+{
+ return ns->cmds->set_peer(ns, peer_ns);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index ea0f221685ab..a133ec5487ae 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -37,16 +37,23 @@
#include <linux/mlx5/fs.h>
#include <linux/rhashtable.h>
#include <linux/llist.h>
+#include <steering/fs_dr.h>
struct mlx5_modify_hdr {
enum mlx5_flow_namespace_type ns_type;
- u32 id;
+ union {
+ struct mlx5_fs_dr_action action;
+ u32 id;
+ };
};
struct mlx5_pkt_reformat {
enum mlx5_flow_namespace_type ns_type;
int reformat_type; /* from mlx5_ifc */
- u32 id;
+ union {
+ struct mlx5_fs_dr_action action;
+ u32 id;
+ };
};
/* FS_TYPE_PRIO_CHAINS is a PRIO that will have namespaces only,
@@ -139,6 +146,7 @@ struct mlx5_flow_handle {
/* Type of children is mlx5_flow_group */
struct mlx5_flow_table {
struct fs_node node;
+ struct mlx5_fs_dr_table fs_dr_table;
u32 id;
u16 vport;
unsigned int max_fte;
@@ -179,6 +187,7 @@ struct mlx5_ft_underlay_qp {
/* Type of children is mlx5_flow_rule */
struct fs_fte {
struct fs_node node;
+ struct mlx5_fs_dr_rule fs_dr_rule;
u32 val[MLX5_ST_SZ_DW_MATCH_PARAM];
u32 dests_size;
u32 index;
@@ -214,6 +223,7 @@ struct mlx5_flow_group_mask {
/* Type of children is fs_fte */
struct mlx5_flow_group {
struct fs_node node;
+ struct mlx5_fs_dr_matcher fs_dr_matcher;
struct mlx5_flow_group_mask mask;
u32 start_index;
u32 max_ftes;
@@ -225,6 +235,7 @@ struct mlx5_flow_group {
struct mlx5_flow_root_namespace {
struct mlx5_flow_namespace ns;
+ struct mlx5_fs_dr_domain fs_dr_domain;
enum fs_flow_table_type table_type;
struct mlx5_core_dev *dev;
struct mlx5_flow_table *root_ft;
@@ -242,6 +253,11 @@ void mlx5_fc_queue_stats_work(struct mlx5_core_dev *dev,
void mlx5_fc_update_sampling_interval(struct mlx5_core_dev *dev,
unsigned long interval);
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void);
+
+int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_root_namespace *peer_ns);
+
int mlx5_init_fs(struct mlx5_core_dev *dev);
void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
new file mode 100644
index 000000000000..3d587d0bdbbe
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -0,0 +1,600 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies */
+
+#include "mlx5_core.h"
+#include "fs_core.h"
+#include "fs_cmd.h"
+#include "mlx5dr.h"
+#include "fs_dr.h"
+
+static bool mlx5_dr_is_fw_table(u32 flags)
+{
+ if (flags & MLX5_FLOW_TABLE_TERMINATION)
+ return true;
+
+ return false;
+}
+
+static int mlx5_cmd_dr_update_root_ft(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ u32 underlay_qpn,
+ bool disconnect)
+{
+ return mlx5_fs_cmd_get_fw_cmds()->update_root_ft(ns, ft, underlay_qpn,
+ disconnect);
+}
+
+static int set_miss_action(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct mlx5_flow_table *next_ft)
+{
+ struct mlx5dr_action *old_miss_action;
+ struct mlx5dr_action *action = NULL;
+ struct mlx5dr_table *next_tbl;
+ int err;
+
+ next_tbl = next_ft ? next_ft->fs_dr_table.dr_table : NULL;
+ if (next_tbl) {
+ action = mlx5dr_action_create_dest_table(next_tbl);
+ if (!action)
+ return -EINVAL;
+ }
+ old_miss_action = ft->fs_dr_table.miss_action;
+ err = mlx5dr_table_set_miss_action(ft->fs_dr_table.dr_table, action);
+ if (err && action) {
+ err = mlx5dr_action_destroy(action);
+ if (err) {
+ action = NULL;
+ mlx5_core_err(ns->dev, "Failed to destroy action (%d)\n",
+ err);
+ }
+ }
+ ft->fs_dr_table.miss_action = action;
+ if (old_miss_action) {
+ err = mlx5dr_action_destroy(old_miss_action);
+ if (err)
+ mlx5_core_err(ns->dev, "Failed to destroy action (%d)\n",
+ err);
+ }
+
+ return err;
+}
+
+static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ unsigned int log_size,
+ struct mlx5_flow_table *next_ft)
+{
+ struct mlx5dr_table *tbl;
+ int err;
+
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->create_flow_table(ns, ft,
+ log_size,
+ next_ft);
+
+ tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain,
+ ft->level);
+ if (!tbl) {
+ mlx5_core_err(ns->dev, "Failed creating dr flow_table\n");
+ return -EINVAL;
+ }
+
+ ft->fs_dr_table.dr_table = tbl;
+ ft->id = mlx5dr_table_get_id(tbl);
+
+ if (next_ft) {
+ err = set_miss_action(ns, ft, next_ft);
+ if (err) {
+ mlx5dr_table_destroy(tbl);
+ ft->fs_dr_table.dr_table = NULL;
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+static int mlx5_cmd_dr_destroy_flow_table(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft)
+{
+ struct mlx5dr_action *action = ft->fs_dr_table.miss_action;
+ int err;
+
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_table(ns, ft);
+
+ err = mlx5dr_table_destroy(ft->fs_dr_table.dr_table);
+ if (err) {
+ mlx5_core_err(ns->dev, "Failed to destroy flow_table (%d)\n",
+ err);
+ return err;
+ }
+ if (action) {
+ err = mlx5dr_action_destroy(action);
+ if (err) {
+ mlx5_core_err(ns->dev, "Failed to destroy action(%d)\n",
+ err);
+ return err;
+ }
+ }
+
+ return err;
+}
+
+static int mlx5_cmd_dr_modify_flow_table(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct mlx5_flow_table *next_ft)
+{
+ return set_miss_action(ns, ft, next_ft);
+}
+
+static int mlx5_cmd_dr_create_flow_group(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ u32 *in,
+ struct mlx5_flow_group *fg)
+{
+ struct mlx5dr_matcher *matcher;
+ u16 priority = MLX5_GET(create_flow_group_in, in,
+ start_flow_index);
+ u8 match_criteria_enable = MLX5_GET(create_flow_group_in,
+ in,
+ match_criteria_enable);
+ struct mlx5dr_match_parameters mask;
+
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->create_flow_group(ns, ft, in,
+ fg);
+
+ mask.match_buf = MLX5_ADDR_OF(create_flow_group_in,
+ in, match_criteria);
+ mask.match_sz = sizeof(fg->mask.match_criteria);
+
+ matcher = mlx5dr_matcher_create(ft->fs_dr_table.dr_table,
+ priority,
+ match_criteria_enable,
+ &mask);
+ if (!matcher) {
+ mlx5_core_err(ns->dev, "Failed creating matcher\n");
+ return -EINVAL;
+ }
+
+ fg->fs_dr_matcher.dr_matcher = matcher;
+ return 0;
+}
+
+static int mlx5_cmd_dr_destroy_flow_group(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct mlx5_flow_group *fg)
+{
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_group(ns, ft, fg);
+
+ return mlx5dr_matcher_destroy(fg->fs_dr_matcher.dr_matcher);
+}
+
+static struct mlx5dr_action *create_vport_action(struct mlx5dr_domain *domain,
+ struct mlx5_flow_rule *dst)
+{
+ struct mlx5_flow_destination *dest_attr = &dst->dest_attr;
+
+ return mlx5dr_action_create_dest_vport(domain, dest_attr->vport.num,
+ dest_attr->vport.flags &
+ MLX5_FLOW_DEST_VPORT_VHCA_ID,
+ dest_attr->vport.vhca_id);
+}
+
+static struct mlx5dr_action *create_ft_action(struct mlx5_core_dev *dev,
+ struct mlx5_flow_rule *dst)
+{
+ struct mlx5_flow_table *dest_ft = dst->dest_attr.ft;
+
+ if (mlx5_dr_is_fw_table(dest_ft->flags))
+ return mlx5dr_create_action_dest_flow_fw_table(dest_ft, dev);
+ return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+}
+
+static struct mlx5dr_action *create_action_push_vlan(struct mlx5dr_domain *domain,
+ struct mlx5_fs_vlan *vlan)
+{
+ u16 n_ethtype = vlan->ethtype;
+ u8 prio = vlan->prio;
+ u16 vid = vlan->vid;
+ u32 vlan_hdr;
+
+ vlan_hdr = (u32)n_ethtype << 16 | (u32)(prio) << 12 | (u32)vid;
+ return mlx5dr_action_create_push_vlan(domain, htonl(vlan_hdr));
+}
+
+#define MLX5_FLOW_CONTEXT_ACTION_MAX 20
+static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct mlx5_flow_group *group,
+ struct fs_fte *fte)
+{
+ struct mlx5dr_domain *domain = ns->fs_dr_domain.dr_domain;
+ struct mlx5dr_action *term_action = NULL;
+ struct mlx5dr_match_parameters params;
+ struct mlx5_core_dev *dev = ns->dev;
+ struct mlx5dr_action **fs_dr_actions;
+ struct mlx5dr_action *tmp_action;
+ struct mlx5dr_action **actions;
+ bool delay_encap_set = false;
+ struct mlx5dr_rule *rule;
+ struct mlx5_flow_rule *dst;
+ int fs_dr_num_actions = 0;
+ int num_actions = 0;
+ size_t match_sz;
+ int err = 0;
+ int i;
+
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->create_fte(ns, ft, group, fte);
+
+ actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX, sizeof(*actions),
+ GFP_KERNEL);
+ if (!actions)
+ return -ENOMEM;
+
+ fs_dr_actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX,
+ sizeof(*fs_dr_actions), GFP_KERNEL);
+ if (!fs_dr_actions) {
+ kfree(actions);
+ return -ENOMEM;
+ }
+
+ match_sz = sizeof(fte->val);
+
+ /* The order of the actions are must to be keep, only the following
+ * order is supported by SW steering:
+ * TX: push vlan -> modify header -> encap
+ * RX: decap -> pop vlan -> modify header
+ */
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
+ tmp_action = create_action_push_vlan(domain, &fte->action.vlan[0]);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) {
+ tmp_action = create_action_push_vlan(domain, &fte->action.vlan[1]);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_DECAP) {
+ enum mlx5dr_action_reformat_type decap_type =
+ DR_ACTION_REFORMAT_TYP_TNL_L2_TO_L2;
+
+ tmp_action = mlx5dr_action_create_packet_reformat(domain,
+ decap_type, 0,
+ NULL);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT) {
+ bool is_decap = fte->action.pkt_reformat->reformat_type ==
+ MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2;
+
+ if (is_decap)
+ actions[num_actions++] =
+ fte->action.pkt_reformat->action.dr_action;
+ else
+ delay_encap_set = true;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) {
+ tmp_action =
+ mlx5dr_action_create_pop_vlan();
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP_2) {
+ tmp_action =
+ mlx5dr_action_create_pop_vlan();
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
+ actions[num_actions++] =
+ fte->action.modify_hdr->action.dr_action;
+
+ if (delay_encap_set)
+ actions[num_actions++] =
+ fte->action.pkt_reformat->action.dr_action;
+
+ /* The order of the actions below is not important */
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_DROP) {
+ tmp_action = mlx5dr_action_create_drop();
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ term_action = tmp_action;
+ }
+
+ if (fte->flow_context.flow_tag) {
+ tmp_action =
+ mlx5dr_action_create_tag(fte->flow_context.flow_tag);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ }
+
+ if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
+ list_for_each_entry(dst, &fte->node.children, node.list) {
+ enum mlx5_flow_destination_type type = dst->dest_attr.type;
+ u32 id;
+
+ if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX) {
+ err = -ENOSPC;
+ goto free_actions;
+ }
+
+ switch (type) {
+ case MLX5_FLOW_DESTINATION_TYPE_COUNTER:
+ id = dst->dest_attr.counter_id;
+
+ tmp_action =
+ mlx5dr_action_create_flow_counter(id);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ actions[num_actions++] = tmp_action;
+ break;
+ case MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE:
+ tmp_action = create_ft_action(dev, dst);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ term_action = tmp_action;
+ break;
+ case MLX5_FLOW_DESTINATION_TYPE_VPORT:
+ tmp_action = create_vport_action(domain, dst);
+ if (!tmp_action) {
+ err = -ENOMEM;
+ goto free_actions;
+ }
+ fs_dr_actions[fs_dr_num_actions++] = tmp_action;
+ term_action = tmp_action;
+ break;
+ default:
+ err = -EOPNOTSUPP;
+ goto free_actions;
+ }
+ }
+ }
+
+ params.match_sz = match_sz;
+ params.match_buf = (u64 *)fte->val;
+
+ if (term_action)
+ actions[num_actions++] = term_action;
+
+ rule = mlx5dr_rule_create(group->fs_dr_matcher.dr_matcher,
+ ¶ms,
+ num_actions,
+ actions);
+ if (!rule) {
+ err = -EINVAL;
+ goto free_actions;
+ }
+
+ kfree(actions);
+ fte->fs_dr_rule.dr_rule = rule;
+ fte->fs_dr_rule.num_actions = fs_dr_num_actions;
+ fte->fs_dr_rule.dr_actions = fs_dr_actions;
+
+ return 0;
+
+free_actions:
+ for (i = 0; i < fs_dr_num_actions; i++)
+ if (!IS_ERR_OR_NULL(fs_dr_actions[i]))
+ mlx5dr_action_destroy(fs_dr_actions[i]);
+
+ mlx5_core_err(dev, "Failed to create dr rule err(%d)\n", err);
+ kfree(actions);
+ kfree(fs_dr_actions);
+ return err;
+}
+
+static int mlx5_cmd_dr_packet_reformat_alloc(struct mlx5_flow_root_namespace *ns,
+ int reformat_type,
+ size_t size,
+ void *reformat_data,
+ enum mlx5_flow_namespace_type namespace,
+ struct mlx5_pkt_reformat *pkt_reformat)
+{
+ struct mlx5dr_domain *dr_domain = ns->fs_dr_domain.dr_domain;
+ struct mlx5dr_action *action;
+ int dr_reformat;
+
+ switch (reformat_type) {
+ case MLX5_REFORMAT_TYPE_L2_TO_VXLAN:
+ case MLX5_REFORMAT_TYPE_L2_TO_NVGRE:
+ case MLX5_REFORMAT_TYPE_L2_TO_L2_TUNNEL:
+ dr_reformat = DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L2;
+ break;
+ case MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2:
+ dr_reformat = DR_ACTION_REFORMAT_TYP_TNL_L3_TO_L2;
+ break;
+ case MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL:
+ dr_reformat = DR_ACTION_REFORMAT_TYP_L2_TO_TNL_L3;
+ break;
+ default:
+ mlx5_core_err(ns->dev, "Packet-reformat not supported(%d)\n",
+ reformat_type);
+ return -EOPNOTSUPP;
+ }
+
+ action = mlx5dr_action_create_packet_reformat(dr_domain,
+ dr_reformat,
+ size,
+ reformat_data);
+ if (!action) {
+ mlx5_core_err(ns->dev, "Failed allocating packet-reformat action\n");
+ return -EINVAL;
+ }
+
+ pkt_reformat->action.dr_action = action;
+
+ return 0;
+}
+
+static void mlx5_cmd_dr_packet_reformat_dealloc(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_pkt_reformat *pkt_reformat)
+{
+ mlx5dr_action_destroy(pkt_reformat->action.dr_action);
+}
+
+static int mlx5_cmd_dr_modify_header_alloc(struct mlx5_flow_root_namespace *ns,
+ u8 namespace, u8 num_actions,
+ void *modify_actions,
+ struct mlx5_modify_hdr *modify_hdr)
+{
+ struct mlx5dr_domain *dr_domain = ns->fs_dr_domain.dr_domain;
+ struct mlx5dr_action *action;
+ size_t actions_sz;
+
+ actions_sz = MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto) *
+ num_actions;
+ action = mlx5dr_action_create_modify_header(dr_domain, 0,
+ actions_sz,
+ modify_actions);
+ if (!action) {
+ mlx5_core_err(ns->dev, "Failed allocating modify-header action\n");
+ return -EINVAL;
+ }
+
+ modify_hdr->action.dr_action = action;
+
+ return 0;
+}
+
+static void mlx5_cmd_dr_modify_header_dealloc(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_modify_hdr *modify_hdr)
+{
+ mlx5dr_action_destroy(modify_hdr->action.dr_action);
+}
+
+static int mlx5_cmd_dr_update_fte(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct mlx5_flow_group *group,
+ int modify_mask,
+ struct fs_fte *fte)
+{
+ return -EOPNOTSUPP;
+}
+
+static int mlx5_cmd_dr_delete_fte(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_table *ft,
+ struct fs_fte *fte)
+{
+ struct mlx5_fs_dr_rule *rule = &fte->fs_dr_rule;
+ int err;
+ int i;
+
+ if (mlx5_dr_is_fw_table(ft->flags))
+ return mlx5_fs_cmd_get_fw_cmds()->delete_fte(ns, ft, fte);
+
+ err = mlx5dr_rule_destroy(rule->dr_rule);
+ if (err)
+ return err;
+
+ for (i = 0; i < rule->num_actions; i++)
+ if (!IS_ERR_OR_NULL(rule->dr_actions[i]))
+ mlx5dr_action_destroy(rule->dr_actions[i]);
+
+ kfree(rule->dr_actions);
+ return 0;
+}
+
+static int mlx5_cmd_dr_set_peer(struct mlx5_flow_root_namespace *ns,
+ struct mlx5_flow_root_namespace *peer_ns)
+{
+ struct mlx5dr_domain *peer_domain = NULL;
+
+ if (peer_ns)
+ peer_domain = peer_ns->fs_dr_domain.dr_domain;
+ mlx5dr_domain_set_peer(ns->fs_dr_domain.dr_domain,
+ peer_domain);
+ return 0;
+}
+
+static int mlx5_cmd_dr_create_ns(struct mlx5_flow_root_namespace *ns)
+{
+ ns->fs_dr_domain.dr_domain =
+ mlx5dr_domain_create(ns->dev,
+ MLX5DR_DOMAIN_TYPE_FDB);
+ if (!ns->fs_dr_domain.dr_domain) {
+ mlx5_core_err(ns->dev, "Failed to create dr flow namespace\n");
+ return -EOPNOTSUPP;
+ }
+ return 0;
+}
+
+static int mlx5_cmd_dr_destroy_ns(struct mlx5_flow_root_namespace *ns)
+{
+ return mlx5dr_domain_destroy(ns->fs_dr_domain.dr_domain);
+}
+
+bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev)
+{
+ return mlx5dr_is_supported(dev);
+}
+
+static const struct mlx5_flow_cmds mlx5_flow_cmds_dr = {
+ .create_flow_table = mlx5_cmd_dr_create_flow_table,
+ .destroy_flow_table = mlx5_cmd_dr_destroy_flow_table,
+ .modify_flow_table = mlx5_cmd_dr_modify_flow_table,
+ .create_flow_group = mlx5_cmd_dr_create_flow_group,
+ .destroy_flow_group = mlx5_cmd_dr_destroy_flow_group,
+ .create_fte = mlx5_cmd_dr_create_fte,
+ .update_fte = mlx5_cmd_dr_update_fte,
+ .delete_fte = mlx5_cmd_dr_delete_fte,
+ .update_root_ft = mlx5_cmd_dr_update_root_ft,
+ .packet_reformat_alloc = mlx5_cmd_dr_packet_reformat_alloc,
+ .packet_reformat_dealloc = mlx5_cmd_dr_packet_reformat_dealloc,
+ .modify_header_alloc = mlx5_cmd_dr_modify_header_alloc,
+ .modify_header_dealloc = mlx5_cmd_dr_modify_header_dealloc,
+ .set_peer = mlx5_cmd_dr_set_peer,
+ .create_ns = mlx5_cmd_dr_create_ns,
+ .destroy_ns = mlx5_cmd_dr_destroy_ns,
+};
+
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void)
+{
+ return &mlx5_flow_cmds_dr;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h
new file mode 100644
index 000000000000..1fb185d6ac7f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+ * Copyright (c) 2019 Mellanox Technologies
+ */
+
+#ifndef _MLX5_FS_DR_
+#define _MLX5_FS_DR_
+
+#include "mlx5dr.h"
+
+struct mlx5_flow_root_namespace;
+struct fs_fte;
+
+struct mlx5_fs_dr_action {
+ struct mlx5dr_action *dr_action;
+};
+
+struct mlx5_fs_dr_ns {
+ struct mlx5_dr_ns *dr_ns;
+};
+
+struct mlx5_fs_dr_rule {
+ struct mlx5dr_rule *dr_rule;
+ /* Only actions created by fs_dr */
+ struct mlx5dr_action **dr_actions;
+ int num_actions;
+};
+
+struct mlx5_fs_dr_domain {
+ struct mlx5dr_domain *dr_domain;
+};
+
+struct mlx5_fs_dr_matcher {
+ struct mlx5dr_matcher *dr_matcher;
+};
+
+struct mlx5_fs_dr_table {
+ struct mlx5dr_table *dr_table;
+ struct mlx5dr_action *miss_action;
+};
+
+#ifdef CONFIG_MLX5_SW_STEERING
+
+bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev);
+
+const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void);
+
+#else
+
+static inline const struct mlx5_flow_cmds *mlx5_fs_cmd_get_dr_cmds(void)
+{
+ return NULL;
+}
+
+static inline bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev)
+{
+ return false;
+}
+
+#endif /* CONFIG_MLX5_SW_STEERING */
+#endif
--
2.21.0
^ permalink raw reply related
* [net-next V2 16/18] net/mlx5: Add API to set the namespace steering mode
From: Saeed Mahameed @ 2019-09-03 20:05 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Maor Gottlieb,
Mark Bloch, Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Maor Gottlieb <maorg@mellanox.com>
Add API to set the flow steering root namesapce mode.
Setting new mode should be called before any steering operation
is executed on the namespace.
This API is going to be used by steering users such switchdev.
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/fs_core.c | 49 +++++++++++++++++++
.../net/ethernet/mellanox/mlx5/core/fs_core.h | 12 ++++-
2 files changed, 60 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index c2d6e9f4cb90..3bbb49354829 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2995,5 +2995,54 @@ EXPORT_SYMBOL(mlx5_packet_reformat_dealloc);
int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
struct mlx5_flow_root_namespace *peer_ns)
{
+ if (peer_ns && ns->mode != peer_ns->mode) {
+ mlx5_core_err(ns->dev,
+ "Can't peer namespace of different steering mode\n");
+ return -EINVAL;
+ }
+
return ns->cmds->set_peer(ns, peer_ns);
}
+
+/* This function should be called only at init stage of the namespace.
+ * It is not safe to call this function while steering operations
+ * are executed in the namespace.
+ */
+int mlx5_flow_namespace_set_mode(struct mlx5_flow_namespace *ns,
+ enum mlx5_flow_steering_mode mode)
+{
+ struct mlx5_flow_root_namespace *root;
+ const struct mlx5_flow_cmds *cmds;
+ int err;
+
+ root = find_root(&ns->node);
+ if (&root->ns != ns)
+ /* Can't set cmds to non root namespace */
+ return -EINVAL;
+
+ if (root->table_type != FS_FT_FDB)
+ return -EOPNOTSUPP;
+
+ if (root->mode == mode)
+ return 0;
+
+ if (mode == MLX5_FLOW_STEERING_MODE_SMFS)
+ cmds = mlx5_fs_cmd_get_dr_cmds();
+ else
+ cmds = mlx5_fs_cmd_get_fw_cmds();
+ if (!cmds)
+ return -EOPNOTSUPP;
+
+ err = cmds->create_ns(root);
+ if (err) {
+ mlx5_core_err(root->dev, "Failed to create flow namespace (%d)\n",
+ err);
+ return err;
+ }
+
+ root->cmds->destroy_ns(root);
+ root->cmds = cmds;
+ root->mode = mode;
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index a133ec5487ae..00717eba2256 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -98,9 +98,15 @@ enum fs_fte_status {
FS_FTE_STATUS_EXISTING = 1UL << 0,
};
+enum mlx5_flow_steering_mode {
+ MLX5_FLOW_STEERING_MODE_DMFS,
+ MLX5_FLOW_STEERING_MODE_SMFS
+};
+
struct mlx5_flow_steering {
struct mlx5_core_dev *dev;
- struct kmem_cache *fgs_cache;
+ enum mlx5_flow_steering_mode mode;
+ struct kmem_cache *fgs_cache;
struct kmem_cache *ftes_cache;
struct mlx5_flow_root_namespace *root_ns;
struct mlx5_flow_root_namespace *fdb_root_ns;
@@ -235,6 +241,7 @@ struct mlx5_flow_group {
struct mlx5_flow_root_namespace {
struct mlx5_flow_namespace ns;
+ enum mlx5_flow_steering_mode mode;
struct mlx5_fs_dr_domain fs_dr_domain;
enum fs_flow_table_type table_type;
struct mlx5_core_dev *dev;
@@ -258,6 +265,9 @@ const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void);
int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
struct mlx5_flow_root_namespace *peer_ns);
+int mlx5_flow_namespace_set_mode(struct mlx5_flow_namespace *ns,
+ enum mlx5_flow_steering_mode mode);
+
int mlx5_init_fs(struct mlx5_core_dev *dev);
void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
--
2.21.0
^ permalink raw reply related
* [net-next V2 18/18] net/mlx5: Add devlink flow_steering_mode parameter
From: Saeed Mahameed @ 2019-09-03 20:05 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Maor Gottlieb,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Maor Gottlieb <maorg@mellanox.com>
Add new parameter (flow_steering_mode) to control the flow steering
mode of the driver.
Two modes are supported:
1. DMFS - Device managed flow steering
2. SMFS - Software/Driver managed flow steering.
In the DMFS mode, the HW steering entities are created through the
FW. In the SMFS mode this entities are created though the driver
directly.
The driver will use the devlink steering mode only if the steering
domain supports it, for now SMFS will manages only the switchdev eswitch
steering domain.
User command examples:
- Set SMFS flow steering mode::
$ devlink dev param set pci/0000:06:00.0 name flow_steering_mode value "smfs" cmode runtime
- Read device flow steering mode::
$ devlink dev param show pci/0000:06:00.0 name flow_steering_mode
pci/0000:06:00.0:
name flow_steering_mode type driver-specific
values:
cmode runtime value smfs
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../device_drivers/mellanox/mlx5.rst | 33 ++++++
.../net/ethernet/mellanox/mlx5/core/devlink.c | 112 +++++++++++++++++-
2 files changed, 144 insertions(+), 1 deletion(-)
diff --git a/Documentation/networking/device_drivers/mellanox/mlx5.rst b/Documentation/networking/device_drivers/mellanox/mlx5.rst
index b30a63dbf4b7..d071c6b49e1f 100644
--- a/Documentation/networking/device_drivers/mellanox/mlx5.rst
+++ b/Documentation/networking/device_drivers/mellanox/mlx5.rst
@@ -11,6 +11,7 @@ Contents
- `Enabling the driver and kconfig options`_
- `Devlink info`_
+- `Devlink parameters`_
- `Devlink health reporters`_
- `mlx5 tracepoints`_
@@ -122,6 +123,38 @@ User command example::
stored:
fw.version 16.26.0100
+Devlink parameters
+==================
+
+flow_steering_mode: Device flow steering mode
+---------------------------------------------
+The flow steering mode parameter controls the flow steering mode of the driver.
+Two modes are supported:
+1. 'dmfs' - Device managed flow steering.
+2. 'smfs - Software/Driver managed flow steering.
+
+In DMFS mode, the HW steering entities are created and managed through the
+Firmware.
+In SMFS mode, the HW steering entities are created and managed though by
+the driver directly into Hardware without firmware intervention.
+
+SMFS mode is faster and provides better rule inserstion rate compared to default DMFS mode.
+
+User command examples:
+
+- Set SMFS flow steering mode::
+
+ $ devlink dev param set pci/0000:06:00.0 name flow_steering_mode value "smfs" cmode runtime
+
+- Read device flow steering mode::
+
+ $ devlink dev param show pci/0000:06:00.0 name flow_steering_mode
+ pci/0000:06:00.0:
+ name flow_steering_mode type driver-specific
+ values:
+ cmode runtime value smfs
+
+
Devlink health reporters
========================
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
index a400f4430c28..7bf7b6fbc776 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c
@@ -4,6 +4,7 @@
#include <devlink.h>
#include "mlx5_core.h"
+#include "fs_core.h"
#include "eswitch.h"
static int mlx5_devlink_flash_update(struct devlink *devlink,
@@ -107,12 +108,121 @@ void mlx5_devlink_free(struct devlink *devlink)
devlink_free(devlink);
}
+static int mlx5_devlink_fs_mode_validate(struct devlink *devlink, u32 id,
+ union devlink_param_value val,
+ struct netlink_ext_ack *extack)
+{
+ struct mlx5_core_dev *dev = devlink_priv(devlink);
+ char *value = val.vstr;
+ int err = 0;
+
+ if (!strcmp(value, "dmfs")) {
+ return 0;
+ } else if (!strcmp(value, "smfs")) {
+ u8 eswitch_mode;
+ bool smfs_cap;
+
+ eswitch_mode = mlx5_eswitch_mode(dev->priv.eswitch);
+ smfs_cap = mlx5_fs_dr_is_supported(dev);
+
+ if (!smfs_cap) {
+ err = -EOPNOTSUPP;
+ NL_SET_ERR_MSG_MOD(extack,
+ "Software managed steering is not supported by current device");
+ }
+
+ else if (eswitch_mode == MLX5_ESWITCH_OFFLOADS) {
+ NL_SET_ERR_MSG_MOD(extack,
+ "Software managed steering is not supported when eswitch offlaods enabled.");
+ err = -EOPNOTSUPP;
+ }
+ } else {
+ NL_SET_ERR_MSG_MOD(extack,
+ "Bad parameter: supported values are [\"dmfs\", \"smfs\"]");
+ err = -EINVAL;
+ }
+
+ return err;
+}
+
+static int mlx5_devlink_fs_mode_set(struct devlink *devlink, u32 id,
+ struct devlink_param_gset_ctx *ctx)
+{
+ struct mlx5_core_dev *dev = devlink_priv(devlink);
+ enum mlx5_flow_steering_mode mode;
+
+ if (!strcmp(ctx->val.vstr, "smfs"))
+ mode = MLX5_FLOW_STEERING_MODE_SMFS;
+ else
+ mode = MLX5_FLOW_STEERING_MODE_DMFS;
+ dev->priv.steering->mode = mode;
+
+ return 0;
+}
+
+static int mlx5_devlink_fs_mode_get(struct devlink *devlink, u32 id,
+ struct devlink_param_gset_ctx *ctx)
+{
+ struct mlx5_core_dev *dev = devlink_priv(devlink);
+
+ if (dev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_SMFS)
+ strcpy(ctx->val.vstr, "smfs");
+ else
+ strcpy(ctx->val.vstr, "dmfs");
+ return 0;
+}
+
+enum mlx5_devlink_param_id {
+ MLX5_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX,
+ MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
+};
+
+static const struct devlink_param mlx5_devlink_params[] = {
+ DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
+ "flow_steering_mode", DEVLINK_PARAM_TYPE_STRING,
+ BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+ mlx5_devlink_fs_mode_get, mlx5_devlink_fs_mode_set,
+ mlx5_devlink_fs_mode_validate),
+};
+
+static void mlx5_devlink_set_params_init_values(struct devlink *devlink)
+{
+ struct mlx5_core_dev *dev = devlink_priv(devlink);
+ union devlink_param_value value;
+
+ if (dev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_DMFS)
+ strcpy(value.vstr, "dmfs");
+ else
+ strcpy(value.vstr, "smfs");
+ devlink_param_driverinit_value_set(devlink,
+ MLX5_DEVLINK_PARAM_FLOW_STEERING_MODE,
+ value);
+}
+
int mlx5_devlink_register(struct devlink *devlink, struct device *dev)
{
- return devlink_register(devlink, dev);
+ int err;
+
+ err = devlink_register(devlink, dev);
+ if (err)
+ return err;
+
+ err = devlink_params_register(devlink, mlx5_devlink_params,
+ ARRAY_SIZE(mlx5_devlink_params));
+ if (err)
+ goto params_reg_err;
+ mlx5_devlink_set_params_init_values(devlink);
+ devlink_params_publish(devlink);
+ return 0;
+
+params_reg_err:
+ devlink_unregister(devlink);
+ return err;
}
void mlx5_devlink_unregister(struct devlink *devlink)
{
+ devlink_params_unregister(devlink, mlx5_devlink_params,
+ ARRAY_SIZE(mlx5_devlink_params));
devlink_unregister(devlink);
}
--
2.21.0
^ permalink raw reply related
* [net-next V2 17/18] net/mlx5: Add support to use SMFS in switchdev mode
From: Saeed Mahameed @ 2019-09-03 20:05 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Maor Gottlieb,
Mark Bloch, Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Maor Gottlieb <maorg@mellanox.com>
In case that flow steering mode of the driver is SMFS (Software Managed
Flow Steering), then use the DR (SW steering) API to create the steering
objects.
In addition, add a call to the set peer namespace when switchdev gets
devcom pair event. It is required to support VF LAG in SMFS.
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 1 +
.../mellanox/mlx5/core/eswitch_offloads.c | 61 ++++++++++++++++---
2 files changed, 55 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 4f70202db6af..6bd6f5895244 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -153,6 +153,7 @@ struct mlx5_eswitch_fdb {
} legacy;
struct offloads_fdb {
+ struct mlx5_flow_namespace *ns;
struct mlx5_flow_table *slow_fdb;
struct mlx5_flow_group *send_to_vport_grp;
struct mlx5_flow_group *peer_miss_grp;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index bee67ff58137..afa623b15a38 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1068,6 +1068,13 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
err = -EOPNOTSUPP;
goto ns_err;
}
+ esw->fdb_table.offloads.ns = root_ns;
+ err = mlx5_flow_namespace_set_mode(root_ns,
+ esw->dev->priv.steering->mode);
+ if (err) {
+ esw_warn(dev, "Failed to set FDB namespace steering mode\n");
+ goto ns_err;
+ }
max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) |
MLX5_CAP_GEN(dev, max_flow_counter_15_0);
@@ -1207,6 +1214,8 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports)
esw_destroy_offloads_fast_fdb_tables(esw);
mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb);
slow_fdb_err:
+ /* Holds true only as long as DMFS is the default */
+ mlx5_flow_namespace_set_mode(root_ns, MLX5_FLOW_STEERING_MODE_DMFS);
ns_err:
kvfree(flow_group_in);
return err;
@@ -1226,6 +1235,9 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw)
mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb);
esw_destroy_offloads_fast_fdb_tables(esw);
+ /* Holds true only as long as DMFS is the default */
+ mlx5_flow_namespace_set_mode(esw->fdb_table.offloads.ns,
+ MLX5_FLOW_STEERING_MODE_DMFS);
}
static int esw_create_offloads_table(struct mlx5_eswitch *esw, int nvports)
@@ -1623,13 +1635,42 @@ static void mlx5_esw_offloads_unpair(struct mlx5_eswitch *esw)
esw_del_fdb_peer_miss_rules(esw);
}
+static int mlx5_esw_offloads_set_ns_peer(struct mlx5_eswitch *esw,
+ struct mlx5_eswitch *peer_esw,
+ bool pair)
+{
+ struct mlx5_flow_root_namespace *peer_ns;
+ struct mlx5_flow_root_namespace *ns;
+ int err;
+
+ peer_ns = peer_esw->dev->priv.steering->fdb_root_ns;
+ ns = esw->dev->priv.steering->fdb_root_ns;
+
+ if (pair) {
+ err = mlx5_flow_namespace_set_peer(ns, peer_ns);
+ if (err)
+ return err;
+
+ mlx5_flow_namespace_set_peer(peer_ns, ns);
+ if (err) {
+ mlx5_flow_namespace_set_peer(ns, NULL);
+ return err;
+ }
+ } else {
+ mlx5_flow_namespace_set_peer(ns, NULL);
+ mlx5_flow_namespace_set_peer(peer_ns, NULL);
+ }
+
+ return 0;
+}
+
static int mlx5_esw_offloads_devcom_event(int event,
void *my_data,
void *event_data)
{
struct mlx5_eswitch *esw = my_data;
- struct mlx5_eswitch *peer_esw = event_data;
struct mlx5_devcom *devcom = esw->dev->priv.devcom;
+ struct mlx5_eswitch *peer_esw = event_data;
int err;
switch (event) {
@@ -1638,9 +1679,12 @@ static int mlx5_esw_offloads_devcom_event(int event,
mlx5_eswitch_vport_match_metadata_enabled(peer_esw))
break;
- err = mlx5_esw_offloads_pair(esw, peer_esw);
+ err = mlx5_esw_offloads_set_ns_peer(esw, peer_esw, true);
if (err)
goto err_out;
+ err = mlx5_esw_offloads_pair(esw, peer_esw);
+ if (err)
+ goto err_peer;
err = mlx5_esw_offloads_pair(peer_esw, esw);
if (err)
@@ -1656,6 +1700,7 @@ static int mlx5_esw_offloads_devcom_event(int event,
mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, false);
mlx5_esw_offloads_unpair(peer_esw);
mlx5_esw_offloads_unpair(esw);
+ mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
break;
}
@@ -1663,7 +1708,8 @@ static int mlx5_esw_offloads_devcom_event(int event,
err_pair:
mlx5_esw_offloads_unpair(esw);
-
+err_peer:
+ mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
err_out:
mlx5_core_err(esw->dev, "esw offloads devcom event failure, event %u err %d",
event, err);
@@ -2115,9 +2161,10 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
else
esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE;
+ mlx5_rdma_enable_roce(esw->dev);
err = esw_offloads_steering_init(esw);
if (err)
- return err;
+ goto err_steering_init;
err = esw_set_passing_vport_metadata(esw, true);
if (err)
@@ -2132,8 +2179,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
esw_offloads_devcom_init(esw);
mutex_init(&esw->offloads.termtbl_mutex);
- mlx5_rdma_enable_roce(esw->dev);
-
return 0;
err_reps:
@@ -2141,6 +2186,8 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
esw_set_passing_vport_metadata(esw, false);
err_vport_metadata:
esw_offloads_steering_cleanup(esw);
+err_steering_init:
+ mlx5_rdma_disable_roce(esw->dev);
return err;
}
@@ -2165,12 +2212,12 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw,
void esw_offloads_disable(struct mlx5_eswitch *esw)
{
- mlx5_rdma_disable_roce(esw->dev);
esw_offloads_devcom_cleanup(esw);
esw_offloads_unload_all_reps(esw);
mlx5_eswitch_disable_pf_vf_vports(esw);
esw_set_passing_vport_metadata(esw, false);
esw_offloads_steering_cleanup(esw);
+ mlx5_rdma_disable_roce(esw->dev);
esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE;
}
--
2.21.0
^ permalink raw reply related
* [net-next V2 11/18] net/mlx5: DR, Expose steering rule functionality
From: Saeed Mahameed @ 2019-09-03 20:04 UTC (permalink / raw)
To: David S. Miller
Cc: netdev@vger.kernel.org, Alex Vesker, Erez Shitrit, Mark Bloch,
Saeed Mahameed
In-Reply-To: <20190903200409.14406-1-saeedm@mellanox.com>
From: Alex Vesker <valex@mellanox.com>
Rules are the actual objects that tie matchers, header values and
actions. Each rule belongs to a matcher, which can hold multiple rules
sharing the same mask. Each rule is a specific set of values and
actions.
When a packet reaches a matcher it is being matched against the
matcher`s rules. In case of a match over a rule its actions will be
executed. Each rule object contains a set of STEs, where each STE is a
definition of match values and actions defined by the rule.
This file handles the rule operations and processing.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/steering/dr_rule.c | 1243 +++++++++++++++++
1 file changed, 1243 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c
new file mode 100644
index 000000000000..3bc3f66b8fa8
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c
@@ -0,0 +1,1243 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2019 Mellanox Technologies. */
+
+#include "dr_types.h"
+
+#define DR_RULE_MAX_STE_CHAIN (DR_RULE_MAX_STES + DR_ACTION_MAX_STES)
+
+struct mlx5dr_rule_action_member {
+ struct mlx5dr_action *action;
+ struct list_head list;
+};
+
+static int dr_rule_append_to_miss_list(struct mlx5dr_ste *new_last_ste,
+ struct list_head *miss_list,
+ struct list_head *send_list)
+{
+ struct mlx5dr_ste_send_info *ste_info_last;
+ struct mlx5dr_ste *last_ste;
+
+ /* The new entry will be inserted after the last */
+ last_ste = list_entry(miss_list->prev, struct mlx5dr_ste, miss_list_node);
+ WARN_ON(!last_ste);
+
+ ste_info_last = kzalloc(sizeof(*ste_info_last), GFP_KERNEL);
+ if (!ste_info_last)
+ return -ENOMEM;
+
+ mlx5dr_ste_set_miss_addr(last_ste->hw_ste,
+ mlx5dr_ste_get_icm_addr(new_last_ste));
+ list_add_tail(&new_last_ste->miss_list_node, miss_list);
+
+ mlx5dr_send_fill_and_append_ste_send_info(last_ste, DR_STE_SIZE_REDUCED,
+ 0, last_ste->hw_ste,
+ ste_info_last, send_list, true);
+
+ return 0;
+}
+
+static struct mlx5dr_ste *
+dr_rule_create_collision_htbl(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ u8 *hw_ste)
+{
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_ste_htbl *new_htbl;
+ struct mlx5dr_ste *ste;
+
+ /* Create new table for miss entry */
+ new_htbl = mlx5dr_ste_htbl_alloc(dmn->ste_icm_pool,
+ DR_CHUNK_SIZE_1,
+ MLX5DR_STE_LU_TYPE_DONT_CARE,
+ 0);
+ if (!new_htbl) {
+ mlx5dr_dbg(dmn, "Failed allocating collision table\n");
+ return NULL;
+ }
+
+ /* One and only entry, never grows */
+ ste = new_htbl->ste_arr;
+ mlx5dr_ste_set_miss_addr(hw_ste, nic_matcher->e_anchor->chunk->icm_addr);
+ mlx5dr_htbl_get(new_htbl);
+
+ return ste;
+}
+
+static struct mlx5dr_ste *
+dr_rule_create_collision_entry(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ u8 *hw_ste,
+ struct mlx5dr_ste *orig_ste)
+{
+ struct mlx5dr_ste *ste;
+
+ ste = dr_rule_create_collision_htbl(matcher, nic_matcher, hw_ste);
+ if (!ste) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed creating collision entry\n");
+ return NULL;
+ }
+
+ ste->ste_chain_location = orig_ste->ste_chain_location;
+
+ /* In collision entry, all members share the same miss_list_head */
+ ste->htbl->miss_list = mlx5dr_ste_get_miss_list(orig_ste);
+
+ /* Next table */
+ if (mlx5dr_ste_create_next_htbl(matcher, nic_matcher, ste, hw_ste,
+ DR_CHUNK_SIZE_1)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed allocating table\n");
+ goto free_tbl;
+ }
+
+ return ste;
+
+free_tbl:
+ mlx5dr_ste_free(ste, matcher, nic_matcher);
+ return NULL;
+}
+
+static int
+dr_rule_handle_one_ste_in_update_list(struct mlx5dr_ste_send_info *ste_info,
+ struct mlx5dr_domain *dmn)
+{
+ int ret;
+
+ list_del(&ste_info->send_list);
+ ret = mlx5dr_send_postsend_ste(dmn, ste_info->ste, ste_info->data,
+ ste_info->size, ste_info->offset);
+ if (ret)
+ goto out;
+ /* Copy data to ste, only reduced size, the last 16B (mask)
+ * is already written to the hw.
+ */
+ memcpy(ste_info->ste->hw_ste, ste_info->data, DR_STE_SIZE_REDUCED);
+
+out:
+ kfree(ste_info);
+ return ret;
+}
+
+static int dr_rule_send_update_list(struct list_head *send_ste_list,
+ struct mlx5dr_domain *dmn,
+ bool is_reverse)
+{
+ struct mlx5dr_ste_send_info *ste_info, *tmp_ste_info;
+ int ret;
+
+ if (is_reverse) {
+ list_for_each_entry_safe_reverse(ste_info, tmp_ste_info,
+ send_ste_list, send_list) {
+ ret = dr_rule_handle_one_ste_in_update_list(ste_info,
+ dmn);
+ if (ret)
+ return ret;
+ }
+ } else {
+ list_for_each_entry_safe(ste_info, tmp_ste_info,
+ send_ste_list, send_list) {
+ ret = dr_rule_handle_one_ste_in_update_list(ste_info,
+ dmn);
+ if (ret)
+ return ret;
+ }
+ }
+
+ return 0;
+}
+
+static struct mlx5dr_ste *
+dr_rule_find_ste_in_miss_list(struct list_head *miss_list, u8 *hw_ste)
+{
+ struct mlx5dr_ste *ste;
+
+ if (list_empty(miss_list))
+ return NULL;
+
+ /* Check if hw_ste is present in the list */
+ list_for_each_entry(ste, miss_list, miss_list_node) {
+ if (mlx5dr_ste_equal_tag(ste->hw_ste, hw_ste))
+ return ste;
+ }
+
+ return NULL;
+}
+
+static struct mlx5dr_ste *
+dr_rule_rehash_handle_collision(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct list_head *update_list,
+ struct mlx5dr_ste *col_ste,
+ u8 *hw_ste)
+{
+ struct mlx5dr_ste *new_ste;
+ int ret;
+
+ new_ste = dr_rule_create_collision_htbl(matcher, nic_matcher, hw_ste);
+ if (!new_ste)
+ return NULL;
+
+ /* In collision entry, all members share the same miss_list_head */
+ new_ste->htbl->miss_list = mlx5dr_ste_get_miss_list(col_ste);
+
+ /* Update the previous from the list */
+ ret = dr_rule_append_to_miss_list(new_ste,
+ mlx5dr_ste_get_miss_list(col_ste),
+ update_list);
+ if (ret) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed update dup entry\n");
+ goto err_exit;
+ }
+
+ return new_ste;
+
+err_exit:
+ mlx5dr_ste_free(new_ste, matcher, nic_matcher);
+ return NULL;
+}
+
+static void dr_rule_rehash_copy_ste_ctrl(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste *cur_ste,
+ struct mlx5dr_ste *new_ste)
+{
+ new_ste->next_htbl = cur_ste->next_htbl;
+ new_ste->ste_chain_location = cur_ste->ste_chain_location;
+
+ if (!mlx5dr_ste_is_last_in_rule(nic_matcher, new_ste->ste_chain_location))
+ new_ste->next_htbl->pointing_ste = new_ste;
+
+ /* We need to copy the refcount since this ste
+ * may have been traversed several times
+ */
+ refcount_set(&new_ste->refcount, refcount_read(&cur_ste->refcount));
+
+ /* Link old STEs rule_mem list to the new ste */
+ mlx5dr_rule_update_rule_member(cur_ste, new_ste);
+ INIT_LIST_HEAD(&new_ste->rule_list);
+ list_splice_tail_init(&cur_ste->rule_list, &new_ste->rule_list);
+}
+
+static struct mlx5dr_ste *
+dr_rule_rehash_copy_ste(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste *cur_ste,
+ struct mlx5dr_ste_htbl *new_htbl,
+ struct list_head *update_list)
+{
+ struct mlx5dr_ste_send_info *ste_info;
+ bool use_update_list = false;
+ u8 hw_ste[DR_STE_SIZE] = {};
+ struct mlx5dr_ste *new_ste;
+ int new_idx;
+ u8 sb_idx;
+
+ /* Copy STE mask from the matcher */
+ sb_idx = cur_ste->ste_chain_location - 1;
+ mlx5dr_ste_set_bit_mask(hw_ste, nic_matcher->ste_builder[sb_idx].bit_mask);
+
+ /* Copy STE control and tag */
+ memcpy(hw_ste, cur_ste->hw_ste, DR_STE_SIZE_REDUCED);
+ mlx5dr_ste_set_miss_addr(hw_ste, nic_matcher->e_anchor->chunk->icm_addr);
+
+ new_idx = mlx5dr_ste_calc_hash_index(hw_ste, new_htbl);
+ new_ste = &new_htbl->ste_arr[new_idx];
+
+ if (mlx5dr_ste_not_used_ste(new_ste)) {
+ mlx5dr_htbl_get(new_htbl);
+ list_add_tail(&new_ste->miss_list_node,
+ mlx5dr_ste_get_miss_list(new_ste));
+ } else {
+ new_ste = dr_rule_rehash_handle_collision(matcher,
+ nic_matcher,
+ update_list,
+ new_ste,
+ hw_ste);
+ if (!new_ste) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed adding collision entry, index: %d\n",
+ new_idx);
+ return NULL;
+ }
+ new_htbl->ctrl.num_of_collisions++;
+ use_update_list = true;
+ }
+
+ memcpy(new_ste->hw_ste, hw_ste, DR_STE_SIZE_REDUCED);
+
+ new_htbl->ctrl.num_of_valid_entries++;
+
+ if (use_update_list) {
+ ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL);
+ if (!ste_info)
+ goto err_exit;
+
+ mlx5dr_send_fill_and_append_ste_send_info(new_ste, DR_STE_SIZE, 0,
+ hw_ste, ste_info,
+ update_list, true);
+ }
+
+ dr_rule_rehash_copy_ste_ctrl(matcher, nic_matcher, cur_ste, new_ste);
+
+ return new_ste;
+
+err_exit:
+ mlx5dr_ste_free(new_ste, matcher, nic_matcher);
+ return NULL;
+}
+
+static int dr_rule_rehash_copy_miss_list(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct list_head *cur_miss_list,
+ struct mlx5dr_ste_htbl *new_htbl,
+ struct list_head *update_list)
+{
+ struct mlx5dr_ste *tmp_ste, *cur_ste, *new_ste;
+
+ if (list_empty(cur_miss_list))
+ return 0;
+
+ list_for_each_entry_safe(cur_ste, tmp_ste, cur_miss_list, miss_list_node) {
+ new_ste = dr_rule_rehash_copy_ste(matcher,
+ nic_matcher,
+ cur_ste,
+ new_htbl,
+ update_list);
+ if (!new_ste)
+ goto err_insert;
+
+ list_del(&cur_ste->miss_list_node);
+ mlx5dr_htbl_put(cur_ste->htbl);
+ }
+ return 0;
+
+err_insert:
+ mlx5dr_err(matcher->tbl->dmn, "Fatal error during resize\n");
+ WARN_ON(true);
+ return -EINVAL;
+}
+
+static int dr_rule_rehash_copy_htbl(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste_htbl *cur_htbl,
+ struct mlx5dr_ste_htbl *new_htbl,
+ struct list_head *update_list)
+{
+ struct mlx5dr_ste *cur_ste;
+ int cur_entries;
+ int err = 0;
+ int i;
+
+ cur_entries = mlx5dr_icm_pool_chunk_size_to_entries(cur_htbl->chunk_size);
+
+ if (cur_entries < 1) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Invalid number of entries\n");
+ return -EINVAL;
+ }
+
+ for (i = 0; i < cur_entries; i++) {
+ cur_ste = &cur_htbl->ste_arr[i];
+ if (mlx5dr_ste_not_used_ste(cur_ste)) /* Empty, nothing to copy */
+ continue;
+
+ err = dr_rule_rehash_copy_miss_list(matcher,
+ nic_matcher,
+ mlx5dr_ste_get_miss_list(cur_ste),
+ new_htbl,
+ update_list);
+ if (err)
+ goto clean_copy;
+ }
+
+clean_copy:
+ return err;
+}
+
+static struct mlx5dr_ste_htbl *
+dr_rule_rehash_htbl(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule,
+ struct mlx5dr_ste_htbl *cur_htbl,
+ u8 ste_location,
+ struct list_head *update_list,
+ enum mlx5dr_icm_chunk_size new_size)
+{
+ struct mlx5dr_ste_send_info *del_ste_info, *tmp_ste_info;
+ struct mlx5dr_matcher *matcher = rule->matcher;
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_matcher_rx_tx *nic_matcher;
+ struct mlx5dr_ste_send_info *ste_info;
+ struct mlx5dr_htbl_connect_info info;
+ struct mlx5dr_domain_rx_tx *nic_dmn;
+ u8 formatted_ste[DR_STE_SIZE] = {};
+ LIST_HEAD(rehash_table_send_list);
+ struct mlx5dr_ste *ste_to_update;
+ struct mlx5dr_ste_htbl *new_htbl;
+ int err;
+
+ nic_matcher = nic_rule->nic_matcher;
+ nic_dmn = nic_matcher->nic_tbl->nic_dmn;
+
+ ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL);
+ if (!ste_info)
+ return NULL;
+
+ new_htbl = mlx5dr_ste_htbl_alloc(dmn->ste_icm_pool,
+ new_size,
+ cur_htbl->lu_type,
+ cur_htbl->byte_mask);
+ if (!new_htbl) {
+ mlx5dr_err(dmn, "Failed to allocate new hash table\n");
+ goto free_ste_info;
+ }
+
+ /* Write new table to HW */
+ info.type = CONNECT_MISS;
+ info.miss_icm_addr = nic_matcher->e_anchor->chunk->icm_addr;
+ mlx5dr_ste_set_formatted_ste(dmn->info.caps.gvmi,
+ nic_dmn,
+ new_htbl,
+ formatted_ste,
+ &info);
+
+ new_htbl->pointing_ste = cur_htbl->pointing_ste;
+ new_htbl->pointing_ste->next_htbl = new_htbl;
+ err = dr_rule_rehash_copy_htbl(matcher,
+ nic_matcher,
+ cur_htbl,
+ new_htbl,
+ &rehash_table_send_list);
+ if (err)
+ goto free_new_htbl;
+
+ if (mlx5dr_send_postsend_htbl(dmn, new_htbl, formatted_ste,
+ nic_matcher->ste_builder[ste_location - 1].bit_mask)) {
+ mlx5dr_err(dmn, "Failed writing table to HW\n");
+ goto free_new_htbl;
+ }
+
+ /* Writing to the hw is done in regular order of rehash_table_send_list,
+ * in order to have the origin data written before the miss address of
+ * collision entries, if exists.
+ */
+ if (dr_rule_send_update_list(&rehash_table_send_list, dmn, false)) {
+ mlx5dr_err(dmn, "Failed updating table to HW\n");
+ goto free_ste_list;
+ }
+
+ /* Connect previous hash table to current */
+ if (ste_location == 1) {
+ /* The previous table is an anchor, anchors size is always one STE */
+ struct mlx5dr_ste_htbl *prev_htbl = cur_htbl->pointing_ste->htbl;
+
+ /* On matcher s_anchor we keep an extra refcount */
+ mlx5dr_htbl_get(new_htbl);
+ mlx5dr_htbl_put(cur_htbl);
+
+ nic_matcher->s_htbl = new_htbl;
+
+ /* It is safe to operate dr_ste_set_hit_addr on the hw_ste here
+ * (48B len) which works only on first 32B
+ */
+ mlx5dr_ste_set_hit_addr(prev_htbl->ste_arr[0].hw_ste,
+ new_htbl->chunk->icm_addr,
+ new_htbl->chunk->num_of_entries);
+
+ ste_to_update = &prev_htbl->ste_arr[0];
+ } else {
+ mlx5dr_ste_set_hit_addr_by_next_htbl(cur_htbl->pointing_ste->hw_ste,
+ new_htbl);
+ ste_to_update = cur_htbl->pointing_ste;
+ }
+
+ mlx5dr_send_fill_and_append_ste_send_info(ste_to_update, DR_STE_SIZE_REDUCED,
+ 0, ste_to_update->hw_ste, ste_info,
+ update_list, false);
+
+ return new_htbl;
+
+free_ste_list:
+ /* Clean all ste_info's from the new table */
+ list_for_each_entry_safe(del_ste_info, tmp_ste_info,
+ &rehash_table_send_list, send_list) {
+ list_del(&del_ste_info->send_list);
+ kfree(del_ste_info);
+ }
+
+free_new_htbl:
+ mlx5dr_ste_htbl_free(new_htbl);
+free_ste_info:
+ kfree(ste_info);
+ mlx5dr_info(dmn, "Failed creating rehash table\n");
+ return NULL;
+}
+
+static struct mlx5dr_ste_htbl *dr_rule_rehash(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule,
+ struct mlx5dr_ste_htbl *cur_htbl,
+ u8 ste_location,
+ struct list_head *update_list)
+{
+ struct mlx5dr_domain *dmn = rule->matcher->tbl->dmn;
+ enum mlx5dr_icm_chunk_size new_size;
+
+ new_size = mlx5dr_icm_next_higher_chunk(cur_htbl->chunk_size);
+ new_size = min_t(u32, new_size, dmn->info.max_log_sw_icm_sz);
+
+ if (new_size == cur_htbl->chunk_size)
+ return NULL; /* Skip rehash, we already at the max size */
+
+ return dr_rule_rehash_htbl(rule, nic_rule, cur_htbl, ste_location,
+ update_list, new_size);
+}
+
+static struct mlx5dr_ste *
+dr_rule_handle_collision(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste *ste,
+ u8 *hw_ste,
+ struct list_head *miss_list,
+ struct list_head *send_list)
+{
+ struct mlx5dr_ste_send_info *ste_info;
+ struct mlx5dr_ste *new_ste;
+
+ ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL);
+ if (!ste_info)
+ return NULL;
+
+ new_ste = dr_rule_create_collision_entry(matcher, nic_matcher, hw_ste, ste);
+ if (!new_ste)
+ goto free_send_info;
+
+ if (dr_rule_append_to_miss_list(new_ste, miss_list, send_list)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed to update prev miss_list\n");
+ goto err_exit;
+ }
+
+ mlx5dr_send_fill_and_append_ste_send_info(new_ste, DR_STE_SIZE, 0, hw_ste,
+ ste_info, send_list, false);
+
+ ste->htbl->ctrl.num_of_collisions++;
+ ste->htbl->ctrl.num_of_valid_entries++;
+
+ return new_ste;
+
+err_exit:
+ mlx5dr_ste_free(new_ste, matcher, nic_matcher);
+free_send_info:
+ kfree(ste_info);
+ return NULL;
+}
+
+static void dr_rule_remove_action_members(struct mlx5dr_rule *rule)
+{
+ struct mlx5dr_rule_action_member *action_mem;
+ struct mlx5dr_rule_action_member *tmp;
+
+ list_for_each_entry_safe(action_mem, tmp, &rule->rule_actions_list, list) {
+ list_del(&action_mem->list);
+ refcount_dec(&action_mem->action->refcount);
+ kvfree(action_mem);
+ }
+}
+
+static int dr_rule_add_action_members(struct mlx5dr_rule *rule,
+ size_t num_actions,
+ struct mlx5dr_action *actions[])
+{
+ struct mlx5dr_rule_action_member *action_mem;
+ int i;
+
+ for (i = 0; i < num_actions; i++) {
+ action_mem = kvzalloc(sizeof(*action_mem), GFP_KERNEL);
+ if (!action_mem)
+ goto free_action_members;
+
+ action_mem->action = actions[i];
+ INIT_LIST_HEAD(&action_mem->list);
+ list_add_tail(&action_mem->list, &rule->rule_actions_list);
+ refcount_inc(&action_mem->action->refcount);
+ }
+
+ return 0;
+
+free_action_members:
+ dr_rule_remove_action_members(rule);
+ return -ENOMEM;
+}
+
+/* While the pointer of ste is no longer valid, like while moving ste to be
+ * the first in the miss_list, and to be in the origin table,
+ * all rule-members that are attached to this ste should update their ste member
+ * to the new pointer
+ */
+void mlx5dr_rule_update_rule_member(struct mlx5dr_ste *ste,
+ struct mlx5dr_ste *new_ste)
+{
+ struct mlx5dr_rule_member *rule_mem;
+
+ if (!list_empty(&ste->rule_list))
+ list_for_each_entry(rule_mem, &ste->rule_list, use_ste_list)
+ rule_mem->ste = new_ste;
+}
+
+static void dr_rule_clean_rule_members(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule)
+{
+ struct mlx5dr_rule_member *rule_mem;
+ struct mlx5dr_rule_member *tmp_mem;
+
+ if (list_empty(&nic_rule->rule_members_list))
+ return;
+ list_for_each_entry_safe(rule_mem, tmp_mem, &nic_rule->rule_members_list, list) {
+ list_del(&rule_mem->list);
+ list_del(&rule_mem->use_ste_list);
+ mlx5dr_ste_put(rule_mem->ste, rule->matcher, nic_rule->nic_matcher);
+ kvfree(rule_mem);
+ }
+}
+
+static bool dr_rule_need_enlarge_hash(struct mlx5dr_ste_htbl *htbl,
+ struct mlx5dr_domain *dmn,
+ struct mlx5dr_domain_rx_tx *nic_dmn)
+{
+ struct mlx5dr_ste_htbl_ctrl *ctrl = &htbl->ctrl;
+
+ if (dmn->info.max_log_sw_icm_sz <= htbl->chunk_size)
+ return false;
+
+ if (!ctrl->may_grow)
+ return false;
+
+ if (ctrl->num_of_collisions >= ctrl->increase_threshold &&
+ (ctrl->num_of_valid_entries - ctrl->num_of_collisions) >= ctrl->increase_threshold)
+ return true;
+
+ return false;
+}
+
+static int dr_rule_add_member(struct mlx5dr_rule_rx_tx *nic_rule,
+ struct mlx5dr_ste *ste)
+{
+ struct mlx5dr_rule_member *rule_mem;
+
+ rule_mem = kvzalloc(sizeof(*rule_mem), GFP_KERNEL);
+ if (!rule_mem)
+ return -ENOMEM;
+
+ rule_mem->ste = ste;
+ list_add_tail(&rule_mem->list, &nic_rule->rule_members_list);
+
+ list_add_tail(&rule_mem->use_ste_list, &ste->rule_list);
+
+ return 0;
+}
+
+static int dr_rule_handle_action_stes(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule,
+ struct list_head *send_ste_list,
+ struct mlx5dr_ste *last_ste,
+ u8 *hw_ste_arr,
+ u32 new_hw_ste_arr_sz)
+{
+ struct mlx5dr_matcher_rx_tx *nic_matcher = nic_rule->nic_matcher;
+ struct mlx5dr_ste_send_info *ste_info_arr[DR_ACTION_MAX_STES];
+ u8 num_of_builders = nic_matcher->num_of_builders;
+ struct mlx5dr_matcher *matcher = rule->matcher;
+ u8 *curr_hw_ste, *prev_hw_ste;
+ struct mlx5dr_ste *action_ste;
+ int i, k, ret;
+
+ /* Two cases:
+ * 1. num_of_builders is equal to new_hw_ste_arr_sz, the action in the ste
+ * 2. num_of_builders is less then new_hw_ste_arr_sz, new ste was added
+ * to support the action.
+ */
+ if (num_of_builders == new_hw_ste_arr_sz)
+ return 0;
+
+ for (i = num_of_builders, k = 0; i < new_hw_ste_arr_sz; i++, k++) {
+ curr_hw_ste = hw_ste_arr + i * DR_STE_SIZE;
+ prev_hw_ste = (i == 0) ? curr_hw_ste : hw_ste_arr + ((i - 1) * DR_STE_SIZE);
+ action_ste = dr_rule_create_collision_htbl(matcher,
+ nic_matcher,
+ curr_hw_ste);
+ if (!action_ste)
+ return -ENOMEM;
+
+ mlx5dr_ste_get(action_ste);
+
+ /* While free ste we go over the miss list, so add this ste to the list */
+ list_add_tail(&action_ste->miss_list_node,
+ mlx5dr_ste_get_miss_list(action_ste));
+
+ ste_info_arr[k] = kzalloc(sizeof(*ste_info_arr[k]),
+ GFP_KERNEL);
+ if (!ste_info_arr[k])
+ goto err_exit;
+
+ /* Point current ste to the new action */
+ mlx5dr_ste_set_hit_addr_by_next_htbl(prev_hw_ste, action_ste->htbl);
+ ret = dr_rule_add_member(nic_rule, action_ste);
+ if (ret) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed adding rule member\n");
+ goto free_ste_info;
+ }
+ mlx5dr_send_fill_and_append_ste_send_info(action_ste, DR_STE_SIZE, 0,
+ curr_hw_ste,
+ ste_info_arr[k],
+ send_ste_list, false);
+ }
+
+ return 0;
+
+free_ste_info:
+ kfree(ste_info_arr[k]);
+err_exit:
+ mlx5dr_ste_put(action_ste, matcher, nic_matcher);
+ return -ENOMEM;
+}
+
+static int dr_rule_handle_empty_entry(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_matcher_rx_tx *nic_matcher,
+ struct mlx5dr_ste_htbl *cur_htbl,
+ struct mlx5dr_ste *ste,
+ u8 ste_location,
+ u8 *hw_ste,
+ struct list_head *miss_list,
+ struct list_head *send_list)
+{
+ struct mlx5dr_ste_send_info *ste_info;
+
+ /* Take ref on table, only on first time this ste is used */
+ mlx5dr_htbl_get(cur_htbl);
+
+ /* new entry -> new branch */
+ list_add_tail(&ste->miss_list_node, miss_list);
+
+ mlx5dr_ste_set_miss_addr(hw_ste, nic_matcher->e_anchor->chunk->icm_addr);
+
+ ste->ste_chain_location = ste_location;
+
+ ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL);
+ if (!ste_info)
+ goto clean_ste_setting;
+
+ if (mlx5dr_ste_create_next_htbl(matcher,
+ nic_matcher,
+ ste,
+ hw_ste,
+ DR_CHUNK_SIZE_1)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Failed allocating table\n");
+ goto clean_ste_info;
+ }
+
+ cur_htbl->ctrl.num_of_valid_entries++;
+
+ mlx5dr_send_fill_and_append_ste_send_info(ste, DR_STE_SIZE, 0, hw_ste,
+ ste_info, send_list, false);
+
+ return 0;
+
+clean_ste_info:
+ kfree(ste_info);
+clean_ste_setting:
+ list_del_init(&ste->miss_list_node);
+ mlx5dr_htbl_put(cur_htbl);
+
+ return -ENOMEM;
+}
+
+static struct mlx5dr_ste *
+dr_rule_handle_ste_branch(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule,
+ struct list_head *send_ste_list,
+ struct mlx5dr_ste_htbl *cur_htbl,
+ u8 *hw_ste,
+ u8 ste_location,
+ struct mlx5dr_ste_htbl **put_htbl)
+{
+ struct mlx5dr_matcher *matcher = rule->matcher;
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_matcher_rx_tx *nic_matcher;
+ struct mlx5dr_domain_rx_tx *nic_dmn;
+ struct mlx5dr_ste_htbl *new_htbl;
+ struct mlx5dr_ste *matched_ste;
+ struct list_head *miss_list;
+ bool skip_rehash = false;
+ struct mlx5dr_ste *ste;
+ int index;
+
+ nic_matcher = nic_rule->nic_matcher;
+ nic_dmn = nic_matcher->nic_tbl->nic_dmn;
+
+again:
+ index = mlx5dr_ste_calc_hash_index(hw_ste, cur_htbl);
+ miss_list = &cur_htbl->chunk->miss_list[index];
+ ste = &cur_htbl->ste_arr[index];
+
+ if (mlx5dr_ste_not_used_ste(ste)) {
+ if (dr_rule_handle_empty_entry(matcher, nic_matcher, cur_htbl,
+ ste, ste_location,
+ hw_ste, miss_list,
+ send_ste_list))
+ return NULL;
+ } else {
+ /* Hash table index in use, check if this ste is in the miss list */
+ matched_ste = dr_rule_find_ste_in_miss_list(miss_list, hw_ste);
+ if (matched_ste) {
+ /* If it is last STE in the chain, and has the same tag
+ * it means that all the previous stes are the same,
+ * if so, this rule is duplicated.
+ */
+ if (mlx5dr_ste_is_last_in_rule(nic_matcher,
+ matched_ste->ste_chain_location)) {
+ mlx5dr_info(dmn, "Duplicate rule inserted, aborting!!\n");
+ return NULL;
+ }
+ return matched_ste;
+ }
+
+ if (!skip_rehash && dr_rule_need_enlarge_hash(cur_htbl, dmn, nic_dmn)) {
+ /* Hash table index in use, try to resize of the hash */
+ skip_rehash = true;
+
+ /* Hold the table till we update.
+ * Release in dr_rule_create_rule()
+ */
+ *put_htbl = cur_htbl;
+ mlx5dr_htbl_get(cur_htbl);
+
+ new_htbl = dr_rule_rehash(rule, nic_rule, cur_htbl,
+ ste_location, send_ste_list);
+ if (!new_htbl) {
+ mlx5dr_htbl_put(cur_htbl);
+ mlx5dr_info(dmn, "failed creating rehash table, htbl-log_size: %d\n",
+ cur_htbl->chunk_size);
+ } else {
+ cur_htbl = new_htbl;
+ }
+ goto again;
+ } else {
+ /* Hash table index in use, add another collision (miss) */
+ ste = dr_rule_handle_collision(matcher,
+ nic_matcher,
+ ste,
+ hw_ste,
+ miss_list,
+ send_ste_list);
+ if (!ste) {
+ mlx5dr_dbg(dmn, "failed adding collision entry, index: %d\n",
+ index);
+ return NULL;
+ }
+ }
+ }
+ return ste;
+}
+
+static bool dr_rule_cmp_value_to_mask(u8 *mask, u8 *value,
+ u32 s_idx, u32 e_idx)
+{
+ u32 i;
+
+ for (i = s_idx; i < e_idx; i++) {
+ if (value[i] & ~mask[i]) {
+ pr_info("Rule parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+ return true;
+}
+
+static bool dr_rule_verify(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *value,
+ struct mlx5dr_match_param *param)
+{
+ u8 match_criteria = matcher->match_criteria;
+ size_t value_size = value->match_sz;
+ u8 *mask_p = (u8 *)&matcher->mask;
+ u8 *param_p = (u8 *)param;
+ u32 s_idx, e_idx;
+
+ if (!value_size ||
+ (value_size > sizeof(struct mlx5dr_match_param) ||
+ (value_size % sizeof(u32)))) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule parameters length is incorrect\n");
+ return false;
+ }
+
+ mlx5dr_ste_copy_param(matcher->match_criteria, param, value);
+
+ if (match_criteria & DR_MATCHER_CRITERIA_OUTER) {
+ s_idx = offsetof(struct mlx5dr_match_param, outer);
+ e_idx = min(s_idx + sizeof(param->outer), value_size);
+
+ if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule outer parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+
+ if (match_criteria & DR_MATCHER_CRITERIA_MISC) {
+ s_idx = offsetof(struct mlx5dr_match_param, misc);
+ e_idx = min(s_idx + sizeof(param->misc), value_size);
+
+ if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule misc parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+
+ if (match_criteria & DR_MATCHER_CRITERIA_INNER) {
+ s_idx = offsetof(struct mlx5dr_match_param, inner);
+ e_idx = min(s_idx + sizeof(param->inner), value_size);
+
+ if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule inner parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+
+ if (match_criteria & DR_MATCHER_CRITERIA_MISC2) {
+ s_idx = offsetof(struct mlx5dr_match_param, misc2);
+ e_idx = min(s_idx + sizeof(param->misc2), value_size);
+
+ if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule misc2 parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+
+ if (match_criteria & DR_MATCHER_CRITERIA_MISC3) {
+ s_idx = offsetof(struct mlx5dr_match_param, misc3);
+ e_idx = min(s_idx + sizeof(param->misc3), value_size);
+
+ if (!dr_rule_cmp_value_to_mask(mask_p, param_p, s_idx, e_idx)) {
+ mlx5dr_dbg(matcher->tbl->dmn, "Rule misc3 parameters contains a value not specified by mask\n");
+ return false;
+ }
+ }
+ return true;
+}
+
+static int dr_rule_destroy_rule_nic(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule)
+{
+ dr_rule_clean_rule_members(rule, nic_rule);
+ return 0;
+}
+
+static int dr_rule_destroy_rule_fdb(struct mlx5dr_rule *rule)
+{
+ dr_rule_destroy_rule_nic(rule, &rule->rx);
+ dr_rule_destroy_rule_nic(rule, &rule->tx);
+ return 0;
+}
+
+static int dr_rule_destroy_rule(struct mlx5dr_rule *rule)
+{
+ struct mlx5dr_domain *dmn = rule->matcher->tbl->dmn;
+
+ switch (dmn->type) {
+ case MLX5DR_DOMAIN_TYPE_NIC_RX:
+ dr_rule_destroy_rule_nic(rule, &rule->rx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_NIC_TX:
+ dr_rule_destroy_rule_nic(rule, &rule->tx);
+ break;
+ case MLX5DR_DOMAIN_TYPE_FDB:
+ dr_rule_destroy_rule_fdb(rule);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ dr_rule_remove_action_members(rule);
+ kfree(rule);
+ return 0;
+}
+
+static bool dr_rule_is_ipv6(struct mlx5dr_match_param *param)
+{
+ return (param->outer.ip_version == 6 ||
+ param->inner.ip_version == 6 ||
+ param->outer.ethertype == ETH_P_IPV6 ||
+ param->inner.ethertype == ETH_P_IPV6);
+}
+
+static bool dr_rule_skip(enum mlx5dr_domain_type domain,
+ enum mlx5dr_ste_entry_type ste_type,
+ struct mlx5dr_match_param *mask,
+ struct mlx5dr_match_param *value)
+{
+ if (domain != MLX5DR_DOMAIN_TYPE_FDB)
+ return false;
+
+ if (mask->misc.source_port) {
+ if (ste_type == MLX5DR_STE_TYPE_RX)
+ if (value->misc.source_port != WIRE_PORT)
+ return true;
+
+ if (ste_type == MLX5DR_STE_TYPE_TX)
+ if (value->misc.source_port == WIRE_PORT)
+ return true;
+ }
+
+ /* Metadata C can be used to describe the source vport */
+ if (mask->misc2.metadata_reg_c_0) {
+ if (ste_type == MLX5DR_STE_TYPE_RX)
+ if ((value->misc2.metadata_reg_c_0 & WIRE_PORT) != WIRE_PORT)
+ return true;
+
+ if (ste_type == MLX5DR_STE_TYPE_TX)
+ if ((value->misc2.metadata_reg_c_0 & WIRE_PORT) == WIRE_PORT)
+ return true;
+ }
+ return false;
+}
+
+static int
+dr_rule_create_rule_nic(struct mlx5dr_rule *rule,
+ struct mlx5dr_rule_rx_tx *nic_rule,
+ struct mlx5dr_match_param *param,
+ size_t num_actions,
+ struct mlx5dr_action *actions[])
+{
+ struct mlx5dr_ste_send_info *ste_info, *tmp_ste_info;
+ struct mlx5dr_matcher *matcher = rule->matcher;
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_matcher_rx_tx *nic_matcher;
+ struct mlx5dr_domain_rx_tx *nic_dmn;
+ struct mlx5dr_ste_htbl *htbl = NULL;
+ struct mlx5dr_ste_htbl *cur_htbl;
+ struct mlx5dr_ste *ste = NULL;
+ LIST_HEAD(send_ste_list);
+ u8 *hw_ste_arr = NULL;
+ u32 new_hw_ste_arr_sz;
+ int ret, i;
+
+ nic_matcher = nic_rule->nic_matcher;
+ nic_dmn = nic_matcher->nic_tbl->nic_dmn;
+
+ INIT_LIST_HEAD(&nic_rule->rule_members_list);
+
+ if (dr_rule_skip(dmn->type, nic_dmn->ste_type, &matcher->mask, param))
+ return 0;
+
+ ret = mlx5dr_matcher_select_builders(matcher,
+ nic_matcher,
+ dr_rule_is_ipv6(param));
+ if (ret)
+ goto out_err;
+
+ hw_ste_arr = kzalloc(DR_RULE_MAX_STE_CHAIN * DR_STE_SIZE, GFP_KERNEL);
+ if (!hw_ste_arr) {
+ ret = -ENOMEM;
+ goto out_err;
+ }
+
+ /* Set the tag values inside the ste array */
+ ret = mlx5dr_ste_build_ste_arr(matcher, nic_matcher, param, hw_ste_arr);
+ if (ret)
+ goto free_hw_ste;
+
+ /* Set the actions values/addresses inside the ste array */
+ ret = mlx5dr_actions_build_ste_arr(matcher, nic_matcher, actions,
+ num_actions, hw_ste_arr,
+ &new_hw_ste_arr_sz);
+ if (ret)
+ goto free_hw_ste;
+
+ cur_htbl = nic_matcher->s_htbl;
+
+ /* Go over the array of STEs, and build dr_ste accordingly.
+ * The loop is over only the builders which are equal or less to the
+ * number of stes, in case we have actions that lives in other stes.
+ */
+ for (i = 0; i < nic_matcher->num_of_builders; i++) {
+ /* Calculate CRC and keep new ste entry */
+ u8 *cur_hw_ste_ent = hw_ste_arr + (i * DR_STE_SIZE);
+
+ ste = dr_rule_handle_ste_branch(rule,
+ nic_rule,
+ &send_ste_list,
+ cur_htbl,
+ cur_hw_ste_ent,
+ i + 1,
+ &htbl);
+ if (!ste) {
+ mlx5dr_err(dmn, "Failed creating next branch\n");
+ ret = -ENOENT;
+ goto free_rule;
+ }
+
+ cur_htbl = ste->next_htbl;
+
+ /* Keep all STEs in the rule struct */
+ ret = dr_rule_add_member(nic_rule, ste);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Failed adding rule member index %d\n", i);
+ goto free_ste;
+ }
+
+ mlx5dr_ste_get(ste);
+ }
+
+ /* Connect actions */
+ ret = dr_rule_handle_action_stes(rule, nic_rule, &send_ste_list,
+ ste, hw_ste_arr, new_hw_ste_arr_sz);
+ if (ret) {
+ mlx5dr_dbg(dmn, "Failed apply actions\n");
+ goto free_rule;
+ }
+ ret = dr_rule_send_update_list(&send_ste_list, dmn, true);
+ if (ret) {
+ mlx5dr_err(dmn, "Failed sending ste!\n");
+ goto free_rule;
+ }
+
+ if (htbl)
+ mlx5dr_htbl_put(htbl);
+
+ return 0;
+
+free_ste:
+ mlx5dr_ste_put(ste, matcher, nic_matcher);
+free_rule:
+ dr_rule_clean_rule_members(rule, nic_rule);
+ /* Clean all ste_info's */
+ list_for_each_entry_safe(ste_info, tmp_ste_info, &send_ste_list, send_list) {
+ list_del(&ste_info->send_list);
+ kfree(ste_info);
+ }
+free_hw_ste:
+ kfree(hw_ste_arr);
+out_err:
+ return ret;
+}
+
+static int
+dr_rule_create_rule_fdb(struct mlx5dr_rule *rule,
+ struct mlx5dr_match_param *param,
+ size_t num_actions,
+ struct mlx5dr_action *actions[])
+{
+ struct mlx5dr_match_param copy_param = {};
+ int ret;
+
+ /* Copy match_param since they will be consumed during the first
+ * nic_rule insertion.
+ */
+ memcpy(©_param, param, sizeof(struct mlx5dr_match_param));
+
+ ret = dr_rule_create_rule_nic(rule, &rule->rx, param,
+ num_actions, actions);
+ if (ret)
+ return ret;
+
+ ret = dr_rule_create_rule_nic(rule, &rule->tx, ©_param,
+ num_actions, actions);
+ if (ret)
+ goto destroy_rule_nic_rx;
+
+ return 0;
+
+destroy_rule_nic_rx:
+ dr_rule_destroy_rule_nic(rule, &rule->rx);
+ return ret;
+}
+
+static struct mlx5dr_rule *
+dr_rule_create_rule(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *value,
+ size_t num_actions,
+ struct mlx5dr_action *actions[])
+{
+ struct mlx5dr_domain *dmn = matcher->tbl->dmn;
+ struct mlx5dr_match_param param = {};
+ struct mlx5dr_rule *rule;
+ int ret;
+
+ if (!dr_rule_verify(matcher, value, ¶m))
+ return NULL;
+
+ rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+ if (!rule)
+ return NULL;
+
+ rule->matcher = matcher;
+ INIT_LIST_HEAD(&rule->rule_actions_list);
+
+ ret = dr_rule_add_action_members(rule, num_actions, actions);
+ if (ret)
+ goto free_rule;
+
+ switch (dmn->type) {
+ case MLX5DR_DOMAIN_TYPE_NIC_RX:
+ rule->rx.nic_matcher = &matcher->rx;
+ ret = dr_rule_create_rule_nic(rule, &rule->rx, ¶m,
+ num_actions, actions);
+ break;
+ case MLX5DR_DOMAIN_TYPE_NIC_TX:
+ rule->tx.nic_matcher = &matcher->tx;
+ ret = dr_rule_create_rule_nic(rule, &rule->tx, ¶m,
+ num_actions, actions);
+ break;
+ case MLX5DR_DOMAIN_TYPE_FDB:
+ rule->rx.nic_matcher = &matcher->rx;
+ rule->tx.nic_matcher = &matcher->tx;
+ ret = dr_rule_create_rule_fdb(rule, ¶m,
+ num_actions, actions);
+ break;
+ default:
+ ret = -EINVAL;
+ break;
+ }
+
+ if (ret)
+ goto remove_action_members;
+
+ return rule;
+
+remove_action_members:
+ dr_rule_remove_action_members(rule);
+free_rule:
+ kfree(rule);
+ mlx5dr_info(dmn, "Failed creating rule\n");
+ return NULL;
+}
+
+struct mlx5dr_rule *mlx5dr_rule_create(struct mlx5dr_matcher *matcher,
+ struct mlx5dr_match_parameters *value,
+ size_t num_actions,
+ struct mlx5dr_action *actions[])
+{
+ struct mlx5dr_rule *rule;
+
+ mutex_lock(&matcher->tbl->dmn->mutex);
+ refcount_inc(&matcher->refcount);
+
+ rule = dr_rule_create_rule(matcher, value, num_actions, actions);
+ if (!rule)
+ refcount_dec(&matcher->refcount);
+
+ mutex_unlock(&matcher->tbl->dmn->mutex);
+
+ return rule;
+}
+
+int mlx5dr_rule_destroy(struct mlx5dr_rule *rule)
+{
+ struct mlx5dr_matcher *matcher = rule->matcher;
+ struct mlx5dr_table *tbl = rule->matcher->tbl;
+ int ret;
+
+ mutex_lock(&tbl->dmn->mutex);
+
+ ret = dr_rule_destroy_rule(rule);
+
+ mutex_unlock(&tbl->dmn->mutex);
+
+ if (!ret)
+ refcount_dec(&matcher->refcount);
+ return ret;
+}
--
2.21.0
^ permalink raw reply related
* Re: [net-next 01/18] net/mlx5: Add flow steering actions to fs_cmd shim layer
From: Saeed Mahameed @ 2019-09-03 20:08 UTC (permalink / raw)
To: davem@davemloft.net
Cc: Maor Gottlieb, Mark Bloch, netdev@vger.kernel.org, Alex Vesker,
Erez Shitrit
In-Reply-To: <20190902.121012.1434735697208917415.davem@davemloft.net>
On Mon, 2019-09-02 at 12:10 -0700, David Miller wrote:
> From: Saeed Mahameed <saeedm@mellanox.com>
> Date: Mon, 2 Sep 2019 07:22:52 +0000
>
> > + maction->flow_action_raw.pkt_reformat =
> > + mlx5_packet_reformat_alloc(dev->mdev, prm_prt, len,
> > + in, namespace);
> > + if (IS_ERR(maction->flow_action_raw.pkt_reformat))
> > return ret;
>
> Don't you have to initialize 'ret' to the pointer error here?
>
> This transformation doesn't look correct.
Right! fixed in V2.
Thanks!
^ permalink raw reply
* Re: [PATCH] net/mlx5: Use PTR_ERR_OR_ZERO rather than its implementation
From: Saeed Mahameed @ 2019-09-03 20:08 UTC (permalink / raw)
To: zhongjiang@huawei.com, davem@davemloft.net
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
leon@kernel.org
In-Reply-To: <1567493770-20074-1-git-send-email-zhongjiang@huawei.com>
On Tue, 2019-09-03 at 14:56 +0800, zhong jiang wrote:
> PTR_ERR_OR_ZERO contains if(IS_ERR(...)) + PTR_ERR. It is better
> to use it directly. hence just replace it.
>
> Signed-off-by: zhong jiang <zhongjiang@huawei.com>
> ---
> drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 5 +----
> 1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> index 5581a80..2e0b467 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> @@ -989,10 +989,7 @@ static void mlx5e_hairpin_flow_del(struct
> mlx5e_priv *priv,
> &flow_act, dest, dest_ix);
> mutex_unlock(&priv->fs.tc.t_lock);
>
> - if (IS_ERR(flow->rule[0]))
> - return PTR_ERR(flow->rule[0]);
> -
> - return 0;
> + return PTR_ERR_OR_ZERO(flow->rule[0]);
> }
>
> static void mlx5e_tc_del_nic_flow(struct mlx5e_priv *priv,
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
^ permalink raw reply
* Re: [PATCH v4 1/2] netfilter: Terminate rule eval if protocol=IPv6 and ipv6 module is disabled
From: Pablo Neira Ayuso @ 2019-09-03 20:19 UTC (permalink / raw)
To: Florian Westphal
Cc: Leonardo Bras, netfilter-devel, coreteam, bridge, netdev,
linux-kernel, Jozsef Kadlecsik, Roopa Prabhu, Nikolay Aleksandrov,
David S. Miller
In-Reply-To: <20190903194809.GD13660@breakpoint.cc>
On Tue, Sep 03, 2019 at 09:48:09PM +0200, Florian Westphal wrote:
> Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> > > > I was expecting we could find a way to handle this from br_netfilter
> > > > alone itself.
> > >
> > > We can't because we support ipv6 fib lookups from the netdev family
> > > as well.
> > >
> > > Alternative is to auto-accept ipv6 packets from the nf_tables eval loop,
> > > but I think its worse.
> >
> > Could we add a restriction for nf_tables + br_netfilter + !ipv6. I
> > mean, if this is an IPv6 packet, nf_tables is on and IPv6 module if
> > off, then drop this packet?
>
> We could do that from nft_do_chain_netdev().
Indeed, this is all about the netdev case.
Probably add something similar to nf_ip6_route() to deal with
ip6_route_lookup() case? This is the one trigering the problem, right?
BTW, how does nft_fib_ipv6 module kicks in if ipv6 module is not
loaded? The symbol dependency would pull in the IPv6 module anyway.
^ permalink raw reply
* Re: net/mlx5e: bind() always returns EINVAL with XDP_ZEROCOPY
From: Saeed Mahameed @ 2019-09-03 20:19 UTC (permalink / raw)
To: kal.conley@dectris.com, brouer@redhat.com
Cc: Maxim Mikityanskiy, magnus.karlsson@intel.com,
toke.hoiland-jorgensen@kau.se, xdp-newbies@vger.kernel.org,
Tariq Toukan, gospo@broadcom.com, jakub.kicinski@netronome.com,
netdev@vger.kernel.org, bjorn.topel@intel.com
In-Reply-To: <20190902110818.2f6a8894@carbon>
On Mon, 2019-09-02 at 11:08 +0200, Jesper Dangaard Brouer wrote:
> On Sun, 1 Sep 2019 18:47:15 +0200
> Kal Cutter Conley <kal.conley@dectris.com> wrote:
>
> > Hi,
> > I figured out the problem. Let me document the issue here for
> > others
> > and hopefully start a discussion.
> >
> > The mlx5 driver uses special queue ids for ZC. If N is the number
> > of
> > configured queues, then for XDP_ZEROCOPY the queue ids start at N.
> > So
> > queue ids [0..N) can only be used with XDP_COPY and queue ids
> > [N..2N)
> > can only be used with XDP_ZEROCOPY.
>
> Thanks for the followup and explanation on how mlx5 AF_XDP queue
> implementation is different from other vendors.
>
>
> > sudo ethtool -L eth0 combined 16
> > sudo samples/bpf/xdpsock -r -i eth0 -c -q 0 # OK
> > sudo samples/bpf/xdpsock -r -i eth0 -z -q 0 # ERROR
> > sudo samples/bpf/xdpsock -r -i eth0 -c -q 16 # ERROR
> > sudo samples/bpf/xdpsock -r -i eth0 -z -q 16 # OK
> >
> > Why was this done? To use zerocopy if available and fallback on
> > copy
> > mode normally you would set sxdp_flags=0. However, here this is no
> > longer possible. To support this driver, you have to first try
> > binding
> > with XDP_ZEROCOPY and the special queue id, then if that fails, you
> > have to try binding again with a normal queue id. Peculiarities
> > like
> > this complicate the XDP user api. Maybe someone can explain the
> > benefits?
>
in mlx5 we like to keep full functional separation between different
queues. Unlike other implementations in mlx5 kernel standard rx rings
can still function while xsk queues are opened. from user perspective
this should be very simple and very usefull:
queues 0..(N-1): can't be used for XSK ZC since they are standard RX
queues managed by kernel and driver
queues N..(2N-1): Are XSK user app managed queues, they can't be used
for anything else.
benefits:
- RSS is not interrupted, Ongoing traffic and Current RX queues keeps
going normally when XSK apps are activated/deactivated on the fly.
- Well-defined full logical separation between different types of RX
queue.
as Jesper explained we understand the confusion, and we will come up
with a solution the fits all vendors.
> Thanks for complaining, it is actually valuable. It really illustrate
> the kernel need to improve in this area, which is what our talk[1] at
> LPC2019 (Sep 10) is about.
>
> Title: "Making Networking Queues a First Class Citizen in the Kernel"
> [1] https://linuxplumbersconf.org/event/4/contributions/462/
>
> As you can see, several vendors are actually involved. Kudos to
> Magnus
> for taking initiative here! It's unfortunately not solved
> "tomorrow",
> as first we have to agree this is needed (facility to register
> queues),
> then agree on API and get commitment from vendors, as this requires
> drivers changes. There is a long road ahead, but I think it will be
> worthwhile in the end, as effective use of dedicated hardware queues
> (both RX and TX) is key to performance.
>
^ permalink raw reply
* [net PATCH] net: sock_map, fix missing ulp check in sock hash case
From: John Fastabend @ 2019-09-03 20:24 UTC (permalink / raw)
To: hdanton, jakub.kicinski, davem; +Cc: netdev, john.fastabend
sock_map and ULP only work together when ULP is loaded after the sock
map is loaded. In the sock_map case we added a check for this to fail
the load if ULP is already set. However, we missed the check on the
sock_hash side.
Add a ULP check to the sock_hash update path.
Fixes: 604326b41a6fb ("bpf, sockmap: convert to generic sk_msg interface")
Reported-by: syzbot+7a6ee4d0078eac6bf782@syzkaller.appspotmail.com
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
net/core/sock_map.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index 1330a74..50916f9 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -656,6 +656,7 @@ static int sock_hash_update_common(struct bpf_map *map, void *key,
struct sock *sk, u64 flags)
{
struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+ struct inet_connection_sock *icsk = inet_csk(sk);
u32 key_size = map->key_size, hash;
struct bpf_htab_elem *elem, *elem_new;
struct bpf_htab_bucket *bucket;
@@ -666,6 +667,8 @@ static int sock_hash_update_common(struct bpf_map *map, void *key,
WARN_ON_ONCE(!rcu_read_lock_held());
if (unlikely(flags > BPF_EXIST))
return -EINVAL;
+ if (unlikely(icsk->icsk_ulp_data))
+ return -EINVAL;
link = sk_psock_init_link();
if (!link)
^ permalink raw reply related
* Re: [PATCH net-next] net: Fail explicit bind to local reserved ports
From: Subash Abhinov Kasiviswanathan @ 2019-09-03 20:34 UTC (permalink / raw)
To: David Miller; +Cc: netdev, stranche
In-Reply-To: <20190830.142202.1082989152863915040.davem@davemloft.net>
> I don't know how happy I am about this. Whatever sets up the
> transparent
> proxy business can block any attempt to communicate over these ports.
>
> Also, protocols like SCTP need the new handling too.
Hi David
The purpose of this patch was to allow the transparent proxy application
to block the specific socket ranges to prevent the communication on the
specific ports.
Dropping packets for this particular port using iptables could lead to
applications on the system getting stuck without getting a socket error.
If bind fails explicitly, the application can atleast retry for some
other
port.
Is there some alternate existing mechanism to achieve this already?
--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project
^ permalink raw reply
* Re: [PATCH v4 1/2] netfilter: Terminate rule eval if protocol=IPv6 and ipv6 module is disabled
From: Florian Westphal @ 2019-09-03 20:35 UTC (permalink / raw)
To: Pablo Neira Ayuso
Cc: Florian Westphal, Leonardo Bras, netfilter-devel, coreteam,
bridge, netdev, linux-kernel, Jozsef Kadlecsik, Roopa Prabhu,
Nikolay Aleksandrov, David S. Miller
In-Reply-To: <20190903201904.npna6dt25ug5gwvd@salvia>
Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> On Tue, Sep 03, 2019 at 09:48:09PM +0200, Florian Westphal wrote:
> > We could do that from nft_do_chain_netdev().
>
> Indeed, this is all about the netdev case.
>
> Probably add something similar to nf_ip6_route() to deal with
> ip6_route_lookup() case? This is the one trigering the problem, right?
Yes, this particular problem is caused by ipv6 fib not being
initialized due to ipv6.disable=1. I don't know if there are cases
other than FIB.
> BTW, how does nft_fib_ipv6 module kicks in if ipv6 module is not
> loaded? The symbol dependency would pull in the IPv6 module anyway.
ipv6.disabled=1 does load the ipv6 module, but its non-functional.
^ permalink raw reply
* Re: [PATCH bpf-next V9 1/3] bpf: new helper to obtain namespace data from current task
From: Yonghong Song @ 2019-09-03 20:36 UTC (permalink / raw)
To: Carlos Antonio Neira Bustos
Cc: netdev@vger.kernel.org, Eric Biederman, brouer@redhat.com,
bpf@vger.kernel.org
In-Reply-To: <20190903184502.2vpaqnoubbr7nnf6@ebpf-metal>
On 9/3/19 11:45 AM, Carlos Antonio Neira Bustos wrote:
> Hi Yonghong,
>
>>> Yes, the samples/bpf test case can be removed.
>>> Could you create a selftest with tracpoint net/netif_receive_skb, which
>>> also uses the proposed helper? net/netif_receive_skb will happen in
>>> interrupt context and it should catch the issue as well if
>>> filename_lookup still get called in interrupt context.
>>
> For this one scenario I just created another selftest with the only difference
> that the tracepoint is /net/netif_receive_skb so this fails with -EPERM.
> Is that enough?.
This should be fine.
>
> I have made this comment on include/uapi/linux/bpf.h, maybe is too terse?
>
> struct bpf_pidns_info {
> __u32 dev; /* dev_t from /proc/self/ns/pid inode */
> __u32 nsid;
> __u32 tgid;
> __u32 pid;
> };
Let us keep the above for now. I may have further comments based on
your test code which uses "dev".
>
> I'm only missing clearing out those questions to be ready to submit v11 of this patch.
Please go ahead to submit the new version.
Thanks.
^ permalink raw reply
* Re: [PATCH v4 1/2] netfilter: Terminate rule eval if protocol=IPv6 and ipv6 module is disabled
From: Pablo Neira Ayuso @ 2019-09-03 20:55 UTC (permalink / raw)
To: Florian Westphal
Cc: Leonardo Bras, netfilter-devel, coreteam, bridge, netdev,
linux-kernel, Jozsef Kadlecsik, Roopa Prabhu, Nikolay Aleksandrov,
David S. Miller
In-Reply-To: <20190903203531.GF13660@breakpoint.cc>
On Tue, Sep 03, 2019 at 10:35:31PM +0200, Florian Westphal wrote:
> Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> > On Tue, Sep 03, 2019 at 09:48:09PM +0200, Florian Westphal wrote:
> > > We could do that from nft_do_chain_netdev().
> >
> > Indeed, this is all about the netdev case.
> >
> > Probably add something similar to nf_ip6_route() to deal with
> > ip6_route_lookup() case? This is the one trigering the problem, right?
>
> Yes, this particular problem is caused by ipv6 fib not being
> initialized due to ipv6.disable=1. I don't know if there are cases
> other than FIB.
>
> > BTW, how does nft_fib_ipv6 module kicks in if ipv6 module is not
> > loaded? The symbol dependency would pull in the IPv6 module anyway.
>
> ipv6.disabled=1 does load the ipv6 module, but its non-functional.
I see, thanks for explaining.
^ permalink raw reply
* Re: [PATCH v4 1/2] netfilter: Terminate rule eval if protocol=IPv6 and ipv6 module is disabled
From: Pablo Neira Ayuso @ 2019-09-03 20:55 UTC (permalink / raw)
To: Leonardo Bras
Cc: netfilter-devel, coreteam, bridge, netdev, linux-kernel,
Jozsef Kadlecsik, Florian Westphal, Roopa Prabhu,
Nikolay Aleksandrov, David S. Miller
In-Reply-To: <20190830181354.26279-2-leonardo@linux.ibm.com>
On Fri, Aug 30, 2019 at 03:13:53PM -0300, Leonardo Bras wrote:
> If IPv6 is disabled on boot (ipv6.disable=1), but nft_fib_inet ends up
> dealing with a IPv6 packet, it causes a kernel panic in
> fib6_node_lookup_1(), crashing in bad_page_fault.
>
> The panic is caused by trying to deference a very low address (0x38
> in ppc64le), due to ipv6.fib6_main_tbl = NULL.
> BUG: Kernel NULL pointer dereference at 0x00000038
>
> The kernel panic was reproduced in a host that disabled IPv6 on boot and
> have to process guest packets (coming from a bridge) using it's ip6tables.
>
> Terminate rule evaluation when packet protocol is IPv6 but the ipv6 module
> is not loaded.
Patch is applied, thanks.
^ permalink raw reply
* Re: [PATCH bpf-next 00/13] bpf: adding map batch processing support
From: Alexei Starovoitov @ 2019-09-03 21:01 UTC (permalink / raw)
To: Stanislav Fomichev
Cc: Yonghong Song, Jakub Kicinski, Brian Vazquez, Alexei Starovoitov,
bpf@vger.kernel.org, netdev@vger.kernel.org, Daniel Borkmann,
Kernel Team
In-Reply-To: <20190830211809.GB2101@mini-arch>
On Fri, Aug 30, 2019 at 02:18:09PM -0700, Stanislav Fomichev wrote:
> > >
> > > I personally like Jakub's/Quentin's proposal more. So if I get to choose
> > > between this series and Jakub's filter+dump in BPF, I'd pick filter+dump
> > > (pending per-cpu issue which we actually care about).
> > >
> > > But if we can have both, I don't have any objections; this patch
I think we need to have both.
imo Jakub's and Yonghong's approach are solving slightly different cases.
filter+dump via program is better suited for LRU map walks where filter prog
would do some non-trivial logic.
Whereas plain 'delete all' or 'dump all' is much simpler to use without
loading yet another prog just to dump it.
bpf infra today isn't quite ready for this very short lived auxiliary progs.
At prog load pages get read-only mapping, tlbs across cpus flushed,
kallsyms populated, FDs allocated, etc.
Loading the prog is a heavy operation. There was a chatter before to have
built-in progs. This filter+dump could benefit from builtin 'allow all'
or 'delete all' progs, but imo that complicates design and asks even
more questions than it answers. Should this builtin progs show up
in 'bpftool prog show' ? When do they load/unload? Same safety requirements
as normal progs? etc.
imo it's fine to have little bit overlap between apis.
So I think we should proceed with both batching apis.
Having said that I think both are suffering from the important issue pointed out
by Brian: when kernel deletes an element get_next_key iterator over hash/lru
map will produce duplicates.
The amount of duplicates can be huge. When batched iterator is slow and
bpf prog is doing a lot of update/delete, there could be 10x worth of duplicates,
since walk will resume from the beginning.
User space cannot be tasked to deal with it.
I think this issue has to be solved in the kernel first and it may require
different batching api.
One idea is to use bucket spin_lock and batch process it bucket-at-a-time.
From api pov the user space will tell kernel:
- here is the buffer for N element. start dump from the beginning.
- kernel will return <= N elements and an iterator.
- user space will pass this opaque iterator back to get another batch
For well behaved hash/lru map there will be zero or one elements per bucket.
When there are 2+ the batching logic can process them together.
If 'lookup' is requested the kernel can check whether user space provided
enough space for these 2 elements. If not abort the batch earlier.
get_next_key won't be used. Instead some sort of opaque iterator
will be returned to user space, so next batch lookup can start from it.
This iterator could be the index of the last dumped bucket.
This idea won't work for pathological hash tables though.
A lot of elements in a single bucket may be more than room for single batch.
In such case iterator will get stuck, since num_of_elements_in_bucket > batch_buf_size.
May be special error code can be used to solve that?
I hope we can come up with other ideas to have a stable iterator over hash table.
Let's use email to describe the ideas and upcoming LPC conference to
sort out details and finalize the one to use.
^ permalink raw reply
* Re: [PATCH] net: sched: taprio: Fix potential integer overflow in taprio_set_picos_per_byte
From: Vinicius Costa Gomes @ 2019-09-03 21:26 UTC (permalink / raw)
To: Vladimir Oltean, Eric Dumazet
Cc: Gustavo A. R. Silva, Jamal Hadi Salim, Cong Wang, Jiri Pirko,
David S. Miller, netdev, lkml
In-Reply-To: <CA+h21hpCAJhE8xhsgDQ55_MUUiesV=uVY4tD=TzaCE6wynUPoQ@mail.gmail.com>
Hi,
Vladimir Oltean <olteanv@gmail.com> writes:
> Right. And while we're at it, there's still the potential
> division-by-zero problem which I still don't know how to solve without
> implementing a full-blown __ethtool_get_link_ksettings parser that
> checks against all the possible outputs it can have under the "no
> carrier" condition - see "[RFC PATCH 1/1] phylink: Set speed to
> SPEED_UNKNOWN when there is no PHY connected" for details.
> And there's also a third fix to be made: the netdev_dbg should be made
> to print "speed" instead of "ecmd.base.speed".
For the ksettings part I am thinking on adding something like this to
ethtool.c. Do you think anything is missing (apart from the
documentation)?
->
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 95991e43..d37c80b 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -177,6 +177,9 @@ void ethtool_convert_legacy_u32_to_link_mode(unsigned long *dst,
bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32,
const unsigned long *src);
+u32 ethtool_link_ksettings_to_speed(const struct ethtool_link_ksettings *settings,
+ u32 default_speed);
+
/**
* struct ethtool_ops - optional netdev operations
* @get_drvinfo: Report driver/device information. Should only set the
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 6288e69..80e3db3 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -539,6 +539,18 @@ struct ethtool_link_usettings {
} link_modes;
};
+u32 ethtool_link_ksettings_to_speed(const struct ethtool_link_ksettings *settings,
+ u32 default_speed)
+{
+ if (settings->base.speed == SPEED_UNKNOWN)
+ return default_speed;
+
+ if (settings->base.speed == 0)
+ return default_speed;
+
+ return settings->base.speed;
+}
+
/* Internal kernel helper to query a device ethtool_link_settings. */
int __ethtool_get_link_ksettings(struct net_device *dev,
struct ethtool_link_ksettings *link_ksettings)
^ permalink raw reply related
* Re: Proposal: r8152 firmware patching framework
From: Prashant Malani @ 2019-09-03 21:32 UTC (permalink / raw)
To: Bambi Yeh, David Miller
Cc: Hayes Wang, Amber Chen, netdev@vger.kernel.org, Ryankao, Jackc,
Albertk, marcochen@google.com, nic_swsd, Grant Grundler
In-Reply-To: <BAD4255E2724E442BCB37085A3D9C93AEEA087DF@RTITMBSVM03.realtek.com.tw>
Hi Bambi,
Thank you for your response. We'd be more than happy to assist in
working out a solution that would be acceptable by the upstream
maintainers.
I think having a maintainable and safe way to deploy firmware fixes
would be much appreciated by hardware users as well as upstream devs,
and certainly more manageable than big static byte-arrays in the
source code!
I've moved David to the TO list to hopefully get his suggestions and
guidance about how to design this in a upstream-compatible way.
I'd be happy to implement it too (I feel this can occur concurrent to
Hayes' upstreaming efforts).
David, could you kindly advise the best way to incorporate deploying
these firmware patches? This change link gives an idea of what we're
dealing with: https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/1417953
My original strawman is to just have a simple firmware format like so:
<section1><size_in_bytes><address1><data1><address2><data2>...<addressN><dataN><section2>
The driver code can have parts to deal with each section in an
appropriate fashion (e.g is each data entry a word or a byte? does
this section have a key which needs to be written to a certain
register etc.)
We'd be grateful if you can offer your advice about best practices (or
suggestions about who might be a good reviewer), so that we can have a
design in place before sending out any patches.
Thanks and best regards,
-Prashant
On Tue, Sep 3, 2019 at 2:01 AM Bambi Yeh <bambi.yeh@realtek.com> wrote:
>
> Hi Prashant:
>
> We will try to implement your requests.
> Based on our experience, upstream reviewer often reject our modification if they have any concern.
> Do you think you can talk to them about this idea and see if they will accept it or not?
> Or if you can help on this after we submit it?
>
> Also, Hayes is now updating our current upstream driver and it goes back and forth for a while.
> So we will need some time to finish it and the target schedule to have your request done is in the end of this month.
>
> Thank you very much.
>
> Best Regards,
> Bambi Yeh
>
> -----Original Message-----
> From: Hayes Wang <hayeswang@realtek.com>
> Sent: Monday, September 2, 2019 2:31 PM
> To: Amber Chen <amber.chen@realtek.com>; Prashant Malani <pmalani@chromium.org>
> Cc: David Miller <davem@davemloft.net>; netdev@vger.kernel.org; Bambi Yeh <bambi.yeh@realtek.com>; Ryankao <ryankao@realtek.com>; Jackc <jackc@realtek.com>; Albertk <albertk@realtek.com>; marcochen@google.com; nic_swsd <nic_swsd@realtek.com>; Grant Grundler <grundler@chromium.org>
> Subject: RE: Proposal: r8152 firmware patching framework
>
> Prashant Malani <pmalani@chromium.org>
> > >
> > > (Adding a few more Realtek folks)
> > >
> > > Friendly ping. Any thoughts / feedback, Realtek folks (and others) ?
> > >
> > >> On Thu, Aug 29, 2019 at 11:40 AM Prashant Malani
> > <pmalani@chromium.org> wrote:
> > >>
> > >> Hi,
> > >>
> > >> The r8152 driver source code distributed by Realtek (on
> > >> www.realtek.com) contains firmware patches. This involves binary
> > >> byte-arrays being written byte/word-wise to the hardware memory
> > >> Example: grundler@chromium.org (cc-ed) has an experimental patch
> > which
> > >> includes the firmware patching code which was distributed with the
> > >> Realtek source :
> > >>
> > https://chromium-review.googlesource.com/c/chromiumos/third_party/kern
> > el
> > /+/1417953
> > >>
> > >> It would be nice to have a way to incorporate these firmware fixes
> > >> into the upstream code. Since having indecipherable byte-arrays is
> > >> not possible upstream, I propose the following:
> > >> - We use the assistance of Realtek to come up with a format which
> > >> the firmware patch files can follow (this can be documented in the
> > >> comments).
> > >> - A real simple format could look like this:
> > >> +
> > >>
> > <section1><size_in_bytes><address1><data1><address2><data2>...<address
> > N
> > ><dataN><section2>...
> > >> + The driver would be able to understand how to
> > >> parse each section (e.g is each data entry a byte or a word?)
> > >>
> > >> - We use request_firmware() to load the firmware, parse it and
> > >> write the data to the relevant registers.
>
> I plan to finish the patches which I am going to submit, first. Then, I could focus on this. However, I don't think I would start this quickly. There are many preparations and they would take me a lot of time.
>
> Best Regards,
> Hayes
>
>
^ permalink raw reply
* Re: [PATCH] net/skbuff: silence warnings under memory pressure
From: Qian Cai @ 2019-09-03 21:42 UTC (permalink / raw)
To: Michal Hocko; +Cc: Eric Dumazet, davem, netdev, linux-mm, linux-kernel
In-Reply-To: <20190903185305.GA14028@dhcp22.suse.cz>
On Tue, 2019-09-03 at 20:53 +0200, Michal Hocko wrote:
> On Tue 03-09-19 11:42:22, Qian Cai wrote:
> > On Tue, 2019-09-03 at 15:22 +0200, Michal Hocko wrote:
> > > On Fri 30-08-19 18:15:22, Eric Dumazet wrote:
> > > > If there is a risk of flooding the syslog, we should fix this
> > > > generically
> > > > in mm layer, not adding hundred of __GFP_NOWARN all over the places.
> > >
> > > We do already ratelimit in warn_alloc. If it isn't sufficient then we
> > > can think of a different parameters. Or maybe it is the ratelimiting
> > > which doesn't work here. Hard to tell and something to explore.
> >
> > The time-based ratelimit won't work for skb_build() as when a system under
> > memory pressure, and the CPU is fast and IO is so slow, it could take a long
> > time to swap and trigger OOM.
>
> I really do not understand what does OOM and swapping have to do with
> the ratelimiting here. The sole purpose of the ratelimit is to reduce
> the amount of warnings to be printed. Slow IO might have an effect on
> when the OOM killer is invoked but atomic allocations are not directly
> dependent on IO.
When there is a heavy memory pressure, the system is trying hard to reclaim
memory to fill up the watermark. However, the IO is slow to page out, but the
memory pressure keep draining atomic reservoir, and some of those skb_build()
will fail eventually.
Only if there is a fast IO, it will finish swapping sooner and then invoke the
OOM to end the memory pressure.
>
> > I suppose what happens is those skb_build() allocations are from softirq,
> > and
> > once one of them failed, it calls printk() which generates more interrupts.
> > Hence, the infinite loop.
>
> Please elaborate more.
>
If you look at the original report, the failed allocation dump_stack() is,
<IRQ>
warn_alloc.cold.43+0x8a/0x148
__alloc_pages_nodemask+0x1a5c/0x1bb0
alloc_pages_current+0x9c/0x110
allocate_slab+0x34a/0x11f0
new_slab+0x46/0x70
___slab_alloc+0x604/0x950
__slab_alloc+0x12/0x20
kmem_cache_alloc+0x32a/0x400
__build_skb+0x23/0x60
build_skb+0x1a/0xb0
igb_clean_rx_irq+0xafc/0x1010 [igb]
igb_poll+0x4bb/0xe30 [igb]
net_rx_action+0x244/0x7a0
__do_softirq+0x1a0/0x60a
irq_exit+0xb5/0xd0
do_IRQ+0x81/0x170
common_interrupt+0xf/0xf
</IRQ>
Since it has no __GFP_NOWARN to begin with, it will call,
printk
vprintk_default
vprintk_emit
wake_up_klogd
irq_work_queue
__irq_work_queue_local
arch_irq_work_raise
apic->send_IPI_self(IRQ_WORK_VECTOR)
smp_irq_work_interrupt
exiting_irq
irq_exit
and end up processing pending net_rx_action softirqs again which are plenty due
to connected via ssh etc.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox