diff options
author | David S. Miller | 2017-10-20 14:06:53 +0200 |
---|---|---|
committer | David S. Miller | 2017-10-20 14:09:09 +0200 |
commit | 322d95f04a8f375ced4f704eb612287d0730a163 (patch) | |
tree | 7fbd4da791119e527ceef742169e6cf598fd1567 /drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | |
parent | tcp: Remove use of inet6_sk and add IPv6 checks to tracepoint (diff) | |
parent | cxgb4: add tc flower support for L3/L4 rewrite (diff) | |
download | kernel-qcow2-linux-322d95f04a8f375ced4f704eb612287d0730a163.tar.gz kernel-qcow2-linux-322d95f04a8f375ced4f704eb612287d0730a163.tar.xz kernel-qcow2-linux-322d95f04a8f375ced4f704eb612287d0730a163.zip |
Merge branch 'cxgb4-more-flower-offloads'
Rahul Lakkireddy says:
====================
cxgb4: enable more tc flower offload matches and actions
This patch series enable more matches and actions for TC Flower
Offload support on Chelsio adapters.
Patch 1 enables matching on IP TOS.
Patch 2 enables matching on VLAN TCI.
Patch 3 adds support for action PASS.
Patch 4 adds support for ETH-DMAC rewrite via TC-PEDIT action. Also,
adds a check to assert that vlan/eth-dmac rewrite actions are valid
only in combination with action egress redirect.
Patch 5 introduces SMT ops for adding/removing entries from SMAC Table
in HW in preparation for patch 6.
Patch 6 adds support for ETH-SMAC rewrite via TC-PEDIT action.
Patch 7 introduces fw_filter2_wr to support L3/L4 header rewrites
in preparation for patch 8.
Patch 8 adds support for rewrite on L3/L4 header fields via TC-PEDIT
action. Supported fields for rewrite are:
IPv4 src/dst address, IPv6 src/dst address, TCP/UDP sport/dport.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c')
-rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 385 |
1 files changed, 380 insertions, 5 deletions
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index 92a311767381..9b6aabe4f963 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -32,8 +32,9 @@ * SOFTWARE. */ -#include <net/tc_act/tc_gact.h> #include <net/tc_act/tc_mirred.h> +#include <net/tc_act/tc_pedit.h> +#include <net/tc_act/tc_gact.h> #include <net/tc_act/tc_vlan.h> #include "cxgb4.h" @@ -41,6 +42,27 @@ #define STATS_CHECK_PERIOD (HZ / 2) +struct ch_tc_pedit_fields pedits[] = { + PEDIT_FIELDS(ETH_, DMAC_31_0, 4, dmac, 0), + PEDIT_FIELDS(ETH_, DMAC_47_32, 2, dmac, 4), + PEDIT_FIELDS(ETH_, SMAC_15_0, 2, smac, 0), + PEDIT_FIELDS(ETH_, SMAC_47_16, 4, smac, 2), + PEDIT_FIELDS(IP4_, SRC, 4, nat_fip, 0), + PEDIT_FIELDS(IP4_, DST, 4, nat_lip, 0), + PEDIT_FIELDS(IP6_, SRC_31_0, 4, nat_fip, 0), + PEDIT_FIELDS(IP6_, SRC_63_32, 4, nat_fip, 4), + PEDIT_FIELDS(IP6_, SRC_95_64, 4, nat_fip, 8), + PEDIT_FIELDS(IP6_, SRC_127_96, 4, nat_fip, 12), + PEDIT_FIELDS(IP6_, DST_31_0, 4, nat_lip, 0), + PEDIT_FIELDS(IP6_, DST_63_32, 4, nat_lip, 4), + PEDIT_FIELDS(IP6_, DST_95_64, 4, nat_lip, 8), + PEDIT_FIELDS(IP6_, DST_127_96, 4, nat_lip, 12), + PEDIT_FIELDS(TCP_, SPORT, 2, nat_fport, 0), + PEDIT_FIELDS(TCP_, DPORT, 2, nat_lport, 0), + PEDIT_FIELDS(UDP_, SPORT, 2, nat_fport, 0), + PEDIT_FIELDS(UDP_, DPORT, 2, nat_lport, 0), +}; + static struct ch_tc_flower_entry *allocate_flower_entry(void) { struct ch_tc_flower_entry *new = kzalloc(sizeof(*new), GFP_KERNEL); @@ -113,6 +135,11 @@ static void cxgb4_process_flow_match(struct net_device *dev, memcpy(&fs->val.fip[0], &key->src, sizeof(key->src)); memcpy(&fs->mask.lip[0], &mask->dst, sizeof(mask->dst)); memcpy(&fs->mask.fip[0], &mask->src, sizeof(mask->src)); + + /* also initialize nat_lip/fip to same values */ + memcpy(&fs->nat_lip[0], &key->dst, sizeof(key->dst)); + memcpy(&fs->nat_fip[0], &key->src, sizeof(key->src)); + } if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { @@ -130,6 +157,10 @@ static void cxgb4_process_flow_match(struct net_device *dev, memcpy(&fs->val.fip[0], key->src.s6_addr, sizeof(key->src)); memcpy(&fs->mask.lip[0], mask->dst.s6_addr, sizeof(mask->dst)); memcpy(&fs->mask.fip[0], mask->src.s6_addr, sizeof(mask->src)); + + /* also initialize nat_lip/fip to same values */ + memcpy(&fs->nat_lip[0], key->dst.s6_addr, sizeof(key->dst)); + memcpy(&fs->nat_fip[0], key->src.s6_addr, sizeof(key->src)); } if (dissector_uses_key(cls->dissector, FLOW_DISSECTOR_KEY_PORTS)) { @@ -145,6 +176,57 @@ static void cxgb4_process_flow_match(struct net_device *dev, fs->mask.lport = cpu_to_be16(mask->dst); fs->val.fport = cpu_to_be16(key->src); fs->mask.fport = cpu_to_be16(mask->src); + + /* also initialize nat_lport/fport to same values */ + fs->nat_lport = cpu_to_be16(key->dst); + fs->nat_fport = cpu_to_be16(key->src); + } + + if (dissector_uses_key(cls->dissector, FLOW_DISSECTOR_KEY_IP)) { + struct flow_dissector_key_ip *key, *mask; + + key = skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_IP, + cls->key); + mask = skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_IP, + cls->mask); + fs->val.tos = key->tos; + fs->mask.tos = mask->tos; + } + + if (dissector_uses_key(cls->dissector, FLOW_DISSECTOR_KEY_VLAN)) { + struct flow_dissector_key_vlan *key, *mask; + u16 vlan_tci, vlan_tci_mask; + + key = skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_VLAN, + cls->key); + mask = skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_VLAN, + cls->mask); + vlan_tci = key->vlan_id | (key->vlan_priority << + VLAN_PRIO_SHIFT); + vlan_tci_mask = mask->vlan_id | (mask->vlan_priority << + VLAN_PRIO_SHIFT); + fs->val.ivlan = cpu_to_be16(vlan_tci); + fs->mask.ivlan = cpu_to_be16(vlan_tci_mask); + + /* Chelsio adapters use ivlan_vld bit to match vlan packets + * as 802.1Q. Also, when vlan tag is present in packets, + * ethtype match is used then to match on ethtype of inner + * header ie. the header following the vlan header. + * So, set the ivlan_vld based on ethtype info supplied by + * TC for vlan packets if its 802.1Q. And then reset the + * ethtype value else, hw will try to match the supplied + * ethtype value with ethtype of inner header. + */ + if (fs->val.ethtype == ETH_P_8021Q) { + fs->val.ivlan_vld = 1; + fs->mask.ivlan_vld = 1; + fs->val.ethtype = 0; + fs->mask.ethtype = 0; + } } /* Match only packets coming from the ingress port where this @@ -157,19 +239,162 @@ static void cxgb4_process_flow_match(struct net_device *dev, static int cxgb4_validate_flow_match(struct net_device *dev, struct tc_cls_flower_offload *cls) { + u16 ethtype_mask = 0; + u16 ethtype_key = 0; + if (cls->dissector->used_keys & ~(BIT(FLOW_DISSECTOR_KEY_CONTROL) | BIT(FLOW_DISSECTOR_KEY_BASIC) | BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) | BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) | - BIT(FLOW_DISSECTOR_KEY_PORTS))) { + BIT(FLOW_DISSECTOR_KEY_PORTS) | + BIT(FLOW_DISSECTOR_KEY_VLAN) | + BIT(FLOW_DISSECTOR_KEY_IP))) { netdev_warn(dev, "Unsupported key used: 0x%x\n", cls->dissector->used_keys); return -EOPNOTSUPP; } + + if (dissector_uses_key(cls->dissector, FLOW_DISSECTOR_KEY_BASIC)) { + struct flow_dissector_key_basic *key = + skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_BASIC, + cls->key); + struct flow_dissector_key_basic *mask = + skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_BASIC, + cls->mask); + ethtype_key = ntohs(key->n_proto); + ethtype_mask = ntohs(mask->n_proto); + } + + if (dissector_uses_key(cls->dissector, FLOW_DISSECTOR_KEY_IP)) { + u16 eth_ip_type = ethtype_key & ethtype_mask; + struct flow_dissector_key_ip *mask; + + if (eth_ip_type != ETH_P_IP && eth_ip_type != ETH_P_IPV6) { + netdev_err(dev, "IP Key supported only with IPv4/v6"); + return -EINVAL; + } + + mask = skb_flow_dissector_target(cls->dissector, + FLOW_DISSECTOR_KEY_IP, + cls->mask); + if (mask->ttl) { + netdev_warn(dev, "ttl match unsupported for offload"); + return -EOPNOTSUPP; + } + } + return 0; } +static void offload_pedit(struct ch_filter_specification *fs, u32 val, u32 mask, + u8 field) +{ + u32 set_val = val & ~mask; + u32 offset = 0; + u8 size = 1; + int i; + + for (i = 0; i < ARRAY_SIZE(pedits); i++) { + if (pedits[i].field == field) { + offset = pedits[i].offset; + size = pedits[i].size; + break; + } + } + memcpy((u8 *)fs + offset, &set_val, size); +} + +static void process_pedit_field(struct ch_filter_specification *fs, u32 val, + u32 mask, u32 offset, u8 htype) +{ + switch (htype) { + case TCA_PEDIT_KEY_EX_HDR_TYPE_ETH: + switch (offset) { + case PEDIT_ETH_DMAC_31_0: + fs->newdmac = 1; + offload_pedit(fs, val, mask, ETH_DMAC_31_0); + break; + case PEDIT_ETH_DMAC_47_32_SMAC_15_0: + if (~mask & PEDIT_ETH_DMAC_MASK) + offload_pedit(fs, val, mask, ETH_DMAC_47_32); + else + offload_pedit(fs, val >> 16, mask >> 16, + ETH_SMAC_15_0); + break; + case PEDIT_ETH_SMAC_47_16: + fs->newsmac = 1; + offload_pedit(fs, val, mask, ETH_SMAC_47_16); + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP4: + switch (offset) { + case PEDIT_IP4_SRC: + offload_pedit(fs, val, mask, IP4_SRC); + break; + case PEDIT_IP4_DST: + offload_pedit(fs, val, mask, IP4_DST); + } + fs->nat_mode = NAT_MODE_ALL; + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP6: + switch (offset) { + case PEDIT_IP6_SRC_31_0: + offload_pedit(fs, val, mask, IP6_SRC_31_0); + break; + case PEDIT_IP6_SRC_63_32: + offload_pedit(fs, val, mask, IP6_SRC_63_32); + break; + case PEDIT_IP6_SRC_95_64: + offload_pedit(fs, val, mask, IP6_SRC_95_64); + break; + case PEDIT_IP6_SRC_127_96: + offload_pedit(fs, val, mask, IP6_SRC_127_96); + break; + case PEDIT_IP6_DST_31_0: + offload_pedit(fs, val, mask, IP6_DST_31_0); + break; + case PEDIT_IP6_DST_63_32: + offload_pedit(fs, val, mask, IP6_DST_63_32); + break; + case PEDIT_IP6_DST_95_64: + offload_pedit(fs, val, mask, IP6_DST_95_64); + break; + case PEDIT_IP6_DST_127_96: + offload_pedit(fs, val, mask, IP6_DST_127_96); + } + fs->nat_mode = NAT_MODE_ALL; + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_TCP: + switch (offset) { + case PEDIT_TCP_SPORT_DPORT: + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + offload_pedit(fs, cpu_to_be32(val) >> 16, + cpu_to_be32(mask) >> 16, + TCP_SPORT); + else + offload_pedit(fs, cpu_to_be32(val), + cpu_to_be32(mask), TCP_DPORT); + } + fs->nat_mode = NAT_MODE_ALL; + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_UDP: + switch (offset) { + case PEDIT_UDP_SPORT_DPORT: + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + offload_pedit(fs, cpu_to_be32(val) >> 16, + cpu_to_be32(mask) >> 16, + UDP_SPORT); + else + offload_pedit(fs, cpu_to_be32(val), + cpu_to_be32(mask), UDP_DPORT); + } + fs->nat_mode = NAT_MODE_ALL; + } +} + static void cxgb4_process_flow_actions(struct net_device *in, struct tc_cls_flower_offload *cls, struct ch_filter_specification *fs) @@ -179,7 +404,9 @@ static void cxgb4_process_flow_actions(struct net_device *in, tcf_exts_to_list(cls->exts, &actions); list_for_each_entry(a, &actions, list) { - if (is_tcf_gact_shot(a)) { + if (is_tcf_gact_ok(a)) { + fs->action = FILTER_PASS; + } else if (is_tcf_gact_shot(a)) { fs->action = FILTER_DROP; } else if (is_tcf_mirred_egress_redirect(a)) { int ifindex = tcf_mirred_ifindex(a); @@ -210,19 +437,152 @@ static void cxgb4_process_flow_actions(struct net_device *in, default: break; } + } else if (is_tcf_pedit(a)) { + u32 mask, val, offset; + int nkeys, i; + u8 htype; + + nkeys = tcf_pedit_nkeys(a); + for (i = 0; i < nkeys; i++) { + htype = tcf_pedit_htype(a, i); + mask = tcf_pedit_mask(a, i); + val = tcf_pedit_val(a, i); + offset = tcf_pedit_offset(a, i); + + process_pedit_field(fs, val, mask, offset, + htype); + } + } + } +} + +static bool valid_l4_mask(u32 mask) +{ + u16 hi, lo; + + /* Either the upper 16-bits (SPORT) OR the lower + * 16-bits (DPORT) can be set, but NOT BOTH. + */ + hi = (mask >> 16) & 0xFFFF; + lo = mask & 0xFFFF; + + return hi && lo ? false : true; +} + +static bool valid_pedit_action(struct net_device *dev, + const struct tc_action *a) +{ + u32 mask, offset; + u8 cmd, htype; + int nkeys, i; + + nkeys = tcf_pedit_nkeys(a); + for (i = 0; i < nkeys; i++) { + htype = tcf_pedit_htype(a, i); + cmd = tcf_pedit_cmd(a, i); + mask = tcf_pedit_mask(a, i); + offset = tcf_pedit_offset(a, i); + + if (cmd != TCA_PEDIT_KEY_EX_CMD_SET) { + netdev_err(dev, "%s: Unsupported pedit cmd\n", + __func__); + return false; + } + + switch (htype) { + case TCA_PEDIT_KEY_EX_HDR_TYPE_ETH: + switch (offset) { + case PEDIT_ETH_DMAC_31_0: + case PEDIT_ETH_DMAC_47_32_SMAC_15_0: + case PEDIT_ETH_SMAC_47_16: + break; + default: + netdev_err(dev, "%s: Unsupported pedit field\n", + __func__); + return false; + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP4: + switch (offset) { + case PEDIT_IP4_SRC: + case PEDIT_IP4_DST: + break; + default: + netdev_err(dev, "%s: Unsupported pedit field\n", + __func__); + return false; + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP6: + switch (offset) { + case PEDIT_IP6_SRC_31_0: + case PEDIT_IP6_SRC_63_32: + case PEDIT_IP6_SRC_95_64: + case PEDIT_IP6_SRC_127_96: + case PEDIT_IP6_DST_31_0: + case PEDIT_IP6_DST_63_32: + case PEDIT_IP6_DST_95_64: + case PEDIT_IP6_DST_127_96: + break; + default: + netdev_err(dev, "%s: Unsupported pedit field\n", + __func__); + return false; + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_TCP: + switch (offset) { + case PEDIT_TCP_SPORT_DPORT: + if (!valid_l4_mask(~mask)) { + netdev_err(dev, "%s: Unsupported mask for TCP L4 ports\n", + __func__); + return false; + } + break; + default: + netdev_err(dev, "%s: Unsupported pedit field\n", + __func__); + return false; + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_UDP: + switch (offset) { + case PEDIT_UDP_SPORT_DPORT: + if (!valid_l4_mask(~mask)) { + netdev_err(dev, "%s: Unsupported mask for UDP L4 ports\n", + __func__); + return false; + } + break; + default: + netdev_err(dev, "%s: Unsupported pedit field\n", + __func__); + return false; + } + break; + default: + netdev_err(dev, "%s: Unsupported pedit type\n", + __func__); + return false; } } + return true; } static int cxgb4_validate_flow_actions(struct net_device *dev, struct tc_cls_flower_offload *cls) { const struct tc_action *a; + bool act_redir = false; + bool act_pedit = false; + bool act_vlan = false; LIST_HEAD(actions); tcf_exts_to_list(cls->exts, &actions); list_for_each_entry(a, &actions, list) { - if (is_tcf_gact_shot(a)) { + if (is_tcf_gact_ok(a)) { + /* Do nothing */ + } else if (is_tcf_gact_shot(a)) { /* Do nothing */ } else if (is_tcf_mirred_egress_redirect(a)) { struct adapter *adap = netdev2adap(dev); @@ -247,6 +607,7 @@ static int cxgb4_validate_flow_actions(struct net_device *dev, __func__); return -EINVAL; } + act_redir = true; } else if (is_tcf_vlan(a)) { u16 proto = be16_to_cpu(tcf_vlan_push_proto(a)); u32 vlan_action = tcf_vlan_action(a); @@ -267,11 +628,25 @@ static int cxgb4_validate_flow_actions(struct net_device *dev, __func__); return -EOPNOTSUPP; } + act_vlan = true; + } else if (is_tcf_pedit(a)) { + bool pedit_valid = valid_pedit_action(dev, a); + + if (!pedit_valid) + return -EOPNOTSUPP; + act_pedit = true; } else { netdev_err(dev, "%s: Unsupported action\n", __func__); return -EOPNOTSUPP; } } + + if ((act_pedit || act_vlan) && !act_redir) { + netdev_err(dev, "%s: pedit/vlan rewrite invalid without egress redirect\n", + __func__); + return -EINVAL; + } + return 0; } @@ -299,8 +674,8 @@ int cxgb4_tc_flower_replace(struct net_device *dev, fs = &ch_flower->fs; fs->hitcnts = 1; - cxgb4_process_flow_actions(dev, cls, fs); cxgb4_process_flow_match(dev, cls, fs); + cxgb4_process_flow_actions(dev, cls, fs); fidx = cxgb4_get_free_ftid(dev, fs->type ? PF_INET6 : PF_INET); if (fidx < 0) { |