* [PATCH net-next 1/9] ice: support for indirect notification
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 2/9] ice: VXLAN and Geneve TC support Tony Nguyen
` (8 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Michal Swiatkowski, netdev, anthony.l.nguyen, Paul Menzel,
Sandeep Penigalapati
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Implement indirect notification mechanism to support offloading TC rules
on tunnel devices.
Keep indirect block list in netdev priv. Notification will call setting
tc cls flower function. For now we can offload only ingress type. Return
not supported for other flow block binder.
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice.h | 8 +
drivers/net/ethernet/intel/ice/ice_main.c | 188 +++++++++++++++++++-
drivers/net/ethernet/intel/ice/ice_tc_lib.h | 6 +
3 files changed, 200 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 967a90efcb11..763add09559c 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -34,6 +34,7 @@
#include <linux/if_bridge.h>
#include <linux/ctype.h>
#include <linux/bpf.h>
+#include <linux/btf.h>
#include <linux/auxiliary_bus.h>
#include <linux/avf/virtchnl.h>
#include <linux/cpu_rmap.h>
@@ -610,6 +611,13 @@ struct ice_pf {
struct ice_netdev_priv {
struct ice_vsi *vsi;
struct ice_repr *repr;
+ /* indirect block callbacks on registered higher level devices
+ * (e.g. tunnel devices)
+ *
+ * tc_indr_block_cb_priv_list is used to look up indirect callback
+ * private data
+ */
+ struct list_head tc_indr_block_priv_list;
};
/**
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 9ba22778011d..e6a8a07d30e5 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -58,6 +58,12 @@ static void ice_vsi_release_all(struct ice_pf *pf);
static int ice_rebuild_channels(struct ice_pf *pf);
static void ice_remove_q_channels(struct ice_vsi *vsi, bool rem_adv_fltr);
+static int
+ice_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch,
+ void *cb_priv, enum tc_setup_type type, void *type_data,
+ void *data,
+ void (*cleanup)(struct flow_block_cb *block_cb));
+
bool netif_is_ice(struct net_device *dev)
{
return dev && (dev->netdev_ops == &ice_netdev_ops);
@@ -3393,6 +3399,63 @@ ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto,
return ret;
}
+/**
+ * ice_rep_indr_tc_block_unbind
+ * @cb_priv: indirection block private data
+ */
+static void ice_rep_indr_tc_block_unbind(void *cb_priv)
+{
+ struct ice_indr_block_priv *indr_priv = cb_priv;
+
+ list_del(&indr_priv->list);
+ kfree(indr_priv);
+}
+
+/**
+ * ice_tc_indir_block_unregister - Unregister TC indirect block notifications
+ * @vsi: VSI struct which has the netdev
+ */
+static void ice_tc_indir_block_unregister(struct ice_vsi *vsi)
+{
+ struct ice_netdev_priv *np = netdev_priv(vsi->netdev);
+
+ flow_indr_dev_unregister(ice_indr_setup_tc_cb, np,
+ ice_rep_indr_tc_block_unbind);
+}
+
+/**
+ * ice_tc_indir_block_remove - clean indirect TC block notifications
+ * @pf: PF structure
+ */
+static void ice_tc_indir_block_remove(struct ice_pf *pf)
+{
+ struct ice_vsi *pf_vsi = ice_get_main_vsi(pf);
+
+ if (!pf_vsi)
+ return;
+
+ ice_tc_indir_block_unregister(pf_vsi);
+}
+
+/**
+ * ice_tc_indir_block_register - Register TC indirect block notifications
+ * @vsi: VSI struct which has the netdev
+ *
+ * Returns 0 on success, negative value on failure
+ */
+static int ice_tc_indir_block_register(struct ice_vsi *vsi)
+{
+ struct ice_netdev_priv *np;
+
+ if (!vsi || !vsi->netdev)
+ return -EINVAL;
+
+ np = netdev_priv(vsi->netdev);
+
+ INIT_LIST_HEAD(&np->tc_indr_block_priv_list);
+ return flow_indr_dev_register(ice_indr_setup_tc_cb, np);
+}
+
/**
* ice_setup_pf_sw - Setup the HW switch on startup or after reset
* @pf: board private structure
@@ -3401,6 +3464,7 @@ ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto,
*/
static int ice_setup_pf_sw(struct ice_pf *pf)
{
+ struct device *dev = ice_pf_to_dev(pf);
struct ice_vsi *vsi;
int status = 0;
@@ -3422,6 +3486,13 @@ static int ice_setup_pf_sw(struct ice_pf *pf)
/* netdev has to be configured before setting frame size */
ice_vsi_cfg_frame_size(vsi);
+ /* init indirect block notifications */
+ status = ice_tc_indir_block_register(vsi);
+ if (status) {
+ dev_err(dev, "Failed to register netdev notifier\n");
+ goto unroll_cfg_netdev;
+ }
+
/* Setup DCB netlink interface */
ice_dcbnl_setup(vsi);
@@ -3433,7 +3504,7 @@ static int ice_setup_pf_sw(struct ice_pf *pf)
status = ice_set_cpu_rx_rmap(vsi);
if (status) {
- dev_err(ice_pf_to_dev(pf), "Failed to set CPU Rx map VSI %d error %d\n",
+ dev_err(dev, "Failed to set CPU Rx map VSI %d error %d\n",
vsi->vsi_num, status);
status = -EINVAL;
goto unroll_napi_add;
@@ -3446,8 +3517,9 @@ static int ice_setup_pf_sw(struct ice_pf *pf)
free_cpu_rx_map:
ice_free_cpu_rx_rmap(vsi);
-
unroll_napi_add:
+ ice_tc_indir_block_unregister(vsi);
+unroll_cfg_netdev:
if (vsi) {
ice_napi_del(vsi);
if (vsi->netdev) {
@@ -4721,6 +4793,8 @@ static void ice_remove(struct pci_dev *pdev)
msleep(100);
}
+ ice_tc_indir_block_remove(pf);
+
if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) {
set_bit(ICE_VF_RESETS_DISABLED, pf->state);
ice_free_vfs(pf);
@@ -8155,6 +8229,116 @@ ice_setup_tc(struct net_device *netdev, enum tc_setup_type type,
return -EOPNOTSUPP;
}
+static struct ice_indr_block_priv *
+ice_indr_block_priv_lookup(struct ice_netdev_priv *np,
+ struct net_device *netdev)
+{
+ struct ice_indr_block_priv *cb_priv;
+
+ list_for_each_entry(cb_priv, &np->tc_indr_block_priv_list, list) {
+ if (!cb_priv->netdev)
+ return NULL;
+ if (cb_priv->netdev == netdev)
+ return cb_priv;
+ }
+ return NULL;
+}
+
+static int
+ice_indr_setup_block_cb(enum tc_setup_type type, void *type_data,
+ void *indr_priv)
+{
+ struct ice_indr_block_priv *priv = indr_priv;
+ struct ice_netdev_priv *np = priv->np;
+
+ switch (type) {
+ case TC_SETUP_CLSFLOWER:
+ return ice_setup_tc_cls_flower(np, priv->netdev,
+ (struct flow_cls_offload *)
+ type_data);
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static int
+ice_indr_setup_tc_block(struct net_device *netdev, struct Qdisc *sch,
+ struct ice_netdev_priv *np,
+ struct flow_block_offload *f, void *data,
+ void (*cleanup)(struct flow_block_cb *block_cb))
+{
+ struct ice_indr_block_priv *indr_priv;
+ struct flow_block_cb *block_cb;
+
+ if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
+ return -EOPNOTSUPP;
+
+ switch (f->command) {
+ case FLOW_BLOCK_BIND:
+ indr_priv = ice_indr_block_priv_lookup(np, netdev);
+ if (indr_priv)
+ return -EEXIST;
+
+ indr_priv = kzalloc(sizeof(*indr_priv), GFP_KERNEL);
+ if (!indr_priv)
+ return -ENOMEM;
+
+ indr_priv->netdev = netdev;
+ indr_priv->np = np;
+ list_add(&indr_priv->list, &np->tc_indr_block_priv_list);
+
+ block_cb =
+ flow_indr_block_cb_alloc(ice_indr_setup_block_cb,
+ indr_priv, indr_priv,
+ ice_rep_indr_tc_block_unbind,
+ f, netdev, sch, data, np,
+ cleanup);
+
+ if (IS_ERR(block_cb)) {
+ list_del(&indr_priv->list);
+ kfree(indr_priv);
+ return PTR_ERR(block_cb);
+ }
+ flow_block_cb_add(block_cb, f);
+ list_add_tail(&block_cb->driver_list, &ice_block_cb_list);
+ break;
+ case FLOW_BLOCK_UNBIND:
+ indr_priv = ice_indr_block_priv_lookup(np, netdev);
+ if (!indr_priv)
+ return -ENOENT;
+
+ block_cb = flow_block_cb_lookup(f->block,
+ ice_indr_setup_block_cb,
+ indr_priv);
+ if (!block_cb)
+ return -ENOENT;
+
+ flow_indr_block_cb_remove(block_cb, f);
+
+ list_del(&block_cb->driver_list);
+ break;
+ default:
+ return -EOPNOTSUPP;
+ }
+ return 0;
+}
+
+static int
+ice_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch,
+ void *cb_priv, enum tc_setup_type type, void *type_data,
+ void *data,
+ void (*cleanup)(struct flow_block_cb *block_cb))
+{
+ switch (type) {
+ case TC_SETUP_BLOCK:
+ return ice_indr_setup_tc_block(netdev, sch, cb_priv, type_data,
+ data, cleanup);
+
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
/**
* ice_open - Called when a network interface becomes active
* @netdev: network interface device structure
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.h b/drivers/net/ethernet/intel/ice/ice_tc_lib.h
index ee9b284fcc02..99ececeef445 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.h
@@ -23,6 +23,12 @@
#define ICE_TC_FLWR_FIELD_ENC_DST_MAC BIT(16)
#define ICE_TC_FLWR_FIELD_ETH_TYPE_ID BIT(17)
+struct ice_indr_block_priv {
+ struct net_device *netdev;
+ struct ice_netdev_priv *np;
+ struct list_head list;
+};
+
struct ice_tc_flower_action {
u32 tc_class;
enum ice_sw_fwd_act_type fltr_act;
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 2/9] ice: VXLAN and Geneve TC support
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 1/9] ice: support for indirect notification Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 3/9] ice: low level support for tunnels Tony Nguyen
` (7 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Michal Swiatkowski, netdev, anthony.l.nguyen, Sandeep Penigalapati
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Add definition for VXLAN and Geneve dummy packet. Define VXLAN and
Geneve type of fields to match on correct UDP tunnel header.
Parse tunnel specific fields from TC tool like outer MACs, outer IPs,
outer destination port and VNI. Save values and masks in outer header
struct and move header pointer to inner to simplify parsing inner
values.
There are two cases for redirect action:
- from uplink to VF - TC filter is added on tunnel device
- from VF to uplink - TC filter is added on PR, for this case check if
redirect device is tunnel device
VXLAN example:
- create tunnel device
ip l add $VXLAN_DEV type vxlan id $VXLAN_VNI dstport $VXLAN_PORT \
dev $PF
- add TC filter (in switchdev mode)
tc filter add dev $VXLAN_DEV protocol ip parent ffff: flower \
enc_dst_ip $VF1_IP enc_key_id $VXLAN_VNI action mirred egress \
redirect dev $VF1_PR
Geneve example:
- create tunnel device
ip l add $GENEVE_DEV type geneve id $GENEVE_VNI dstport $GENEVE_PORT \
remote $GENEVE_IP
- add TC filter (in switchdev mode)
tc filter add dev $GENEVE_DEV protocol ip parent ffff: flower \
enc_key_id $GENEVE_VNI dst_ip $GENEVE1_IP action mirred egress \
redirect dev $VF1_PR
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice_main.c | 5 +
.../ethernet/intel/ice/ice_protocol_type.h | 16 +
drivers/net/ethernet/intel/ice/ice_switch.h | 1 +
drivers/net/ethernet/intel/ice/ice_tc_lib.c | 380 ++++++++++++++++--
drivers/net/ethernet/intel/ice/ice_tc_lib.h | 4 +
5 files changed, 362 insertions(+), 44 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index e6a8a07d30e5..2ebbbe1edd82 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -8270,6 +8270,11 @@ ice_indr_setup_tc_block(struct net_device *netdev, struct Qdisc *sch,
struct ice_indr_block_priv *indr_priv;
struct flow_block_cb *block_cb;
+ if (!ice_is_tunnel_supported(netdev) &&
+ !(is_vlan_dev(netdev) &&
+ vlan_dev_real_dev(netdev) == np->vsi->netdev))
+ return -EOPNOTSUPP;
+
if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
return -EOPNOTSUPP;
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index 0b220dfa7457..d717d1158545 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -37,10 +37,19 @@ enum ice_protocol_type {
ICE_TCP_IL,
ICE_UDP_OF,
ICE_UDP_ILOS,
+ ICE_VXLAN,
+ ICE_GENEVE,
+ ICE_VXLAN_GPE,
ICE_SCTP_IL,
ICE_PROTOCOL_LAST
};
+enum ice_sw_tunnel_type {
+ ICE_NON_TUN = 0,
+ ICE_SW_TUN_VXLAN,
+ ICE_SW_TUN_GENEVE,
+};
+
/* Decoders for ice_prot_id:
* - F: First
* - I: Inner
@@ -152,6 +161,12 @@ struct ice_l4_hdr {
__be16 check;
};
+struct ice_udp_tnl_hdr {
+ __be16 field;
+ __be16 proto_type;
+ __be32 vni; /* only use lower 24-bits */
+};
+
union ice_prot_hdr {
struct ice_ether_hdr eth_hdr;
struct ice_ethtype_hdr ethertype;
@@ -160,6 +175,7 @@ union ice_prot_hdr {
struct ice_ipv6_hdr ipv6_hdr;
struct ice_l4_hdr l4_hdr;
struct ice_sctp_hdr sctp_hdr;
+ struct ice_udp_tnl_hdr tnl_hdr;
};
/* This is mapping table entry that maps every word within a given protocol
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index c4dd2062c469..7d661c9be81b 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -171,6 +171,7 @@ struct ice_adv_rule_flags_info {
};
struct ice_adv_rule_info {
+ enum ice_sw_tunnel_type tun_type;
struct ice_sw_act_ctrl sw_act;
u32 priority;
u8 rx; /* true means LOOKUP_RX otherwise LOOKUP_TX */
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
index 725caa160b13..920d9024a6c1 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
@@ -3,8 +3,9 @@
#include "ice.h"
#include "ice_tc_lib.h"
-#include "ice_lib.h"
#include "ice_fltr.h"
+#include "ice_lib.h"
+#include "ice_protocol_type.h"
/**
* ice_tc_count_lkups - determine lookup count for switch filter
@@ -20,7 +21,21 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
{
int lkups_cnt = 0;
- if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID)
+ if (flags & ICE_TC_FLWR_FIELD_TENANT_ID)
+ lkups_cnt++;
+
+ if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
+ ICE_TC_FLWR_FIELD_ENC_DEST_IPV4 |
+ ICE_TC_FLWR_FIELD_ENC_SRC_IPV6 |
+ ICE_TC_FLWR_FIELD_ENC_DEST_IPV6))
+ lkups_cnt++;
+
+ if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT)
+ lkups_cnt++;
+
+ /* currently inner etype filter isn't supported */
+ if ((flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) &&
+ fltr->tunnel_type == TNL_LAST)
lkups_cnt++;
/* are MAC fields specified? */
@@ -32,10 +47,8 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
lkups_cnt++;
/* are IPv[4|6] fields specified? */
- if (flags & (ICE_TC_FLWR_FIELD_DEST_IPV4 | ICE_TC_FLWR_FIELD_SRC_IPV4))
- lkups_cnt++;
- else if (flags & (ICE_TC_FLWR_FIELD_DEST_IPV6 |
- ICE_TC_FLWR_FIELD_SRC_IPV6))
+ if (flags & (ICE_TC_FLWR_FIELD_DEST_IPV4 | ICE_TC_FLWR_FIELD_SRC_IPV4 |
+ ICE_TC_FLWR_FIELD_DEST_IPV6 | ICE_TC_FLWR_FIELD_SRC_IPV6))
lkups_cnt++;
/* is L4 (TCP/UDP/any other L4 protocol fields) specified? */
@@ -46,6 +59,132 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
return lkups_cnt;
}
+static enum ice_protocol_type ice_proto_type_from_mac(bool inner)
+{
+ return inner ? ICE_MAC_IL : ICE_MAC_OFOS;
+}
+
+static enum ice_protocol_type ice_proto_type_from_ipv4(bool inner)
+{
+ return inner ? ICE_IPV4_IL : ICE_IPV4_OFOS;
+}
+
+static enum ice_protocol_type ice_proto_type_from_ipv6(bool inner)
+{
+ return inner ? ICE_IPV6_IL : ICE_IPV6_OFOS;
+}
+
+static enum ice_protocol_type
+ice_proto_type_from_l4_port(bool inner, u16 ip_proto)
+{
+ if (inner) {
+ switch (ip_proto) {
+ case IPPROTO_UDP:
+ return ICE_UDP_ILOS;
+ }
+ } else {
+ switch (ip_proto) {
+ case IPPROTO_TCP:
+ return ICE_TCP_IL;
+ case IPPROTO_UDP:
+ return ICE_UDP_OF;
+ }
+ }
+
+ return 0;
+}
+
+static enum ice_protocol_type
+ice_proto_type_from_tunnel(enum ice_tunnel_type type)
+{
+ switch (type) {
+ case TNL_VXLAN:
+ return ICE_VXLAN;
+ case TNL_GENEVE:
+ return ICE_GENEVE;
+ default:
+ return 0;
+ }
+}
+
+static enum ice_sw_tunnel_type
+ice_sw_type_from_tunnel(enum ice_tunnel_type type)
+{
+ switch (type) {
+ case TNL_VXLAN:
+ return ICE_SW_TUN_VXLAN;
+ case TNL_GENEVE:
+ return ICE_SW_TUN_GENEVE;
+ default:
+ return ICE_NON_TUN;
+ }
+}
+
+static int
+ice_tc_fill_tunnel_outer(u32 flags, struct ice_tc_flower_fltr *fltr,
+ struct ice_adv_lkup_elem *list)
+{
+ struct ice_tc_flower_lyr_2_4_hdrs *hdr = &fltr->outer_headers;
+ int i = 0;
+
+ if (flags & ICE_TC_FLWR_FIELD_TENANT_ID) {
+ u32 tenant_id;
+
+ list[i].type = ice_proto_type_from_tunnel(fltr->tunnel_type);
+ tenant_id = be32_to_cpu(fltr->tenant_id) << 8;
+ list[i].h_u.tnl_hdr.vni = cpu_to_be32(tenant_id);
+ memcpy(&list[i].m_u.tnl_hdr.vni, "\xff\xff\xff\x00", 4);
+ i++;
+ }
+
+ if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
+ ICE_TC_FLWR_FIELD_ENC_DEST_IPV4)) {
+ list[i].type = ice_proto_type_from_ipv4(false);
+
+ if (flags & ICE_TC_FLWR_FIELD_ENC_SRC_IPV4) {
+ list[i].h_u.ipv4_hdr.src_addr = hdr->l3_key.src_ipv4;
+ list[i].m_u.ipv4_hdr.src_addr = hdr->l3_mask.src_ipv4;
+ }
+ if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_IPV4) {
+ list[i].h_u.ipv4_hdr.dst_addr = hdr->l3_key.dst_ipv4;
+ list[i].m_u.ipv4_hdr.dst_addr = hdr->l3_mask.dst_ipv4;
+ }
+ i++;
+ }
+
+ if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV6 |
+ ICE_TC_FLWR_FIELD_ENC_DEST_IPV6)) {
+ list[i].type = ice_proto_type_from_ipv6(false);
+
+ if (flags & ICE_TC_FLWR_FIELD_ENC_SRC_IPV6) {
+ memcpy(&list[i].h_u.ipv6_hdr.src_addr,
+ &hdr->l3_key.src_ipv6_addr,
+ sizeof(hdr->l3_key.src_ipv6_addr));
+ memcpy(&list[i].m_u.ipv6_hdr.src_addr,
+ &hdr->l3_mask.src_ipv6_addr,
+ sizeof(hdr->l3_mask.src_ipv6_addr));
+ }
+ if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_IPV6) {
+ memcpy(&list[i].h_u.ipv6_hdr.dst_addr,
+ &hdr->l3_key.dst_ipv6_addr,
+ sizeof(hdr->l3_key.dst_ipv6_addr));
+ memcpy(&list[i].m_u.ipv6_hdr.dst_addr,
+ &hdr->l3_mask.dst_ipv6_addr,
+ sizeof(hdr->l3_mask.dst_ipv6_addr));
+ }
+ i++;
+ }
+
+ if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT) {
+ list[i].type = ice_proto_type_from_l4_port(false, hdr->l3_key.ip_proto);
+ list[i].h_u.l4_hdr.dst_port = hdr->l4_key.dst_port;
+ list[i].m_u.l4_hdr.dst_port = hdr->l4_mask.dst_port;
+ i++;
+ }
+
+ return i;
+}
+
/**
* ice_tc_fill_rules - fill filter rules based on TC fltr
* @hw: pointer to HW structure
@@ -67,9 +206,16 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
u16 *l4_proto)
{
struct ice_tc_flower_lyr_2_4_hdrs *headers = &tc_fltr->outer_headers;
+ bool inner = false;
int i = 0;
- if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) {
+ rule_info->tun_type = ice_sw_type_from_tunnel(tc_fltr->tunnel_type);
+ if (tc_fltr->tunnel_type != TNL_LAST) {
+ i = ice_tc_fill_tunnel_outer(flags, tc_fltr, list);
+
+ headers = &tc_fltr->inner_headers;
+ inner = true;
+ } else if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) {
list[i].type = ICE_ETYPE_OL;
list[i].h_u.ethertype.ethtype_id = headers->l2_key.n_proto;
list[i].m_u.ethertype.ethtype_id = headers->l2_mask.n_proto;
@@ -83,7 +229,7 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
l2_key = &headers->l2_key;
l2_mask = &headers->l2_mask;
- list[i].type = ICE_MAC_OFOS;
+ list[i].type = ice_proto_type_from_mac(inner);
if (flags & ICE_TC_FLWR_FIELD_DST_MAC) {
ether_addr_copy(list[i].h_u.eth_hdr.dst_addr,
l2_key->dst_mac);
@@ -112,7 +258,7 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
ICE_TC_FLWR_FIELD_SRC_IPV4)) {
struct ice_tc_l3_hdr *l3_key, *l3_mask;
- list[i].type = ICE_IPV4_OFOS;
+ list[i].type = ice_proto_type_from_ipv4(inner);
l3_key = &headers->l3_key;
l3_mask = &headers->l3_mask;
if (flags & ICE_TC_FLWR_FIELD_DEST_IPV4) {
@@ -129,7 +275,7 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
struct ice_ipv6_hdr *ipv6_hdr, *ipv6_mask;
struct ice_tc_l3_hdr *l3_key, *l3_mask;
- list[i].type = ICE_IPV6_OFOS;
+ list[i].type = ice_proto_type_from_ipv6(inner);
ipv6_hdr = &list[i].h_u.ipv6_hdr;
ipv6_mask = &list[i].m_u.ipv6_hdr;
l3_key = &headers->l3_key;
@@ -155,19 +301,10 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
ICE_TC_FLWR_FIELD_SRC_L4_PORT)) {
struct ice_tc_l4_hdr *l4_key, *l4_mask;
+ list[i].type = ice_proto_type_from_l4_port(inner, headers->l3_key.ip_proto);
l4_key = &headers->l4_key;
l4_mask = &headers->l4_mask;
- if (headers->l3_key.ip_proto == IPPROTO_TCP) {
- list[i].type = ICE_TCP_IL;
- /* detected L4 proto is TCP */
- if (l4_proto)
- *l4_proto = IPPROTO_TCP;
- } else if (headers->l3_key.ip_proto == IPPROTO_UDP) {
- list[i].type = ICE_UDP_ILOS;
- /* detected L4 proto is UDP */
- if (l4_proto)
- *l4_proto = IPPROTO_UDP;
- }
+
if (flags & ICE_TC_FLWR_FIELD_DEST_L4_PORT) {
list[i].h_u.l4_hdr.dst_port = l4_key->dst_port;
list[i].m_u.l4_hdr.dst_port = l4_mask->dst_port;
@@ -182,6 +319,27 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
return i;
}
+/**
+ * ice_tc_tun_get_type - get the tunnel type
+ * @tunnel_dev: ptr to tunnel device
+ *
+ * This function detects appropriate tunnel_type if specified device is
+ * tunnel device such as VXLAN/Geneve
+ */
+static int ice_tc_tun_get_type(struct net_device *tunnel_dev)
+{
+ if (netif_is_vxlan(tunnel_dev))
+ return TNL_VXLAN;
+ if (netif_is_geneve(tunnel_dev))
+ return TNL_GENEVE;
+ return TNL_LAST;
+}
+
+bool ice_is_tunnel_supported(struct net_device *dev)
+{
+ return ice_tc_tun_get_type(dev) != TNL_LAST;
+}
+
static int
ice_eswitch_tc_parse_action(struct ice_tc_flower_fltr *fltr,
struct flow_action_entry *act)
@@ -201,10 +359,8 @@ ice_eswitch_tc_parse_action(struct ice_tc_flower_fltr *fltr,
fltr->dest_vsi = repr->src_vsi;
fltr->direction = ICE_ESWITCH_FLTR_INGRESS;
- } else if (netif_is_ice(act->dev)) {
- struct ice_netdev_priv *np = netdev_priv(act->dev);
-
- fltr->dest_vsi = np->vsi;
+ } else if (netif_is_ice(act->dev) ||
+ ice_is_tunnel_supported(act->dev)) {
fltr->direction = ICE_ESWITCH_FLTR_EGRESS;
} else {
NL_SET_ERR_MSG_MOD(fltr->extack, "Unsupported netdevice in switchdev mode");
@@ -235,11 +391,7 @@ ice_eswitch_add_tc_fltr(struct ice_vsi *vsi, struct ice_tc_flower_fltr *fltr)
int ret = 0;
int i;
- if (!flags || (flags & (ICE_TC_FLWR_FIELD_ENC_DEST_IPV4 |
- ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
- ICE_TC_FLWR_FIELD_ENC_DEST_IPV6 |
- ICE_TC_FLWR_FIELD_ENC_SRC_IPV6 |
- ICE_TC_FLWR_FIELD_ENC_SRC_L4_PORT))) {
+ if (!flags || (flags & ICE_TC_FLWR_FIELD_ENC_SRC_L4_PORT)) {
NL_SET_ERR_MSG_MOD(fltr->extack, "Unsupported encap field(s)");
return -EOPNOTSUPP;
}
@@ -255,6 +407,10 @@ ice_eswitch_add_tc_fltr(struct ice_vsi *vsi, struct ice_tc_flower_fltr *fltr)
goto exit;
}
+ /* egress traffic is always redirect to uplink */
+ if (fltr->direction == ICE_ESWITCH_FLTR_EGRESS)
+ fltr->dest_vsi = vsi->back->switchdev.uplink_vsi;
+
rule_info.sw_act.fltr_act = fltr->action.fltr_act;
if (fltr->action.fltr_act != ICE_DROP_PACKET)
rule_info.sw_act.vsi_handle = fltr->dest_vsi->idx;
@@ -438,19 +594,26 @@ ice_add_tc_flower_adv_fltr(struct ice_vsi *vsi,
* @match: Pointer to flow match structure
* @fltr: Pointer to filter structure
* @headers: inner or outer header fields
+ * @is_encap: set true for tunnel IPv4 address
*/
static int
ice_tc_set_ipv4(struct flow_match_ipv4_addrs *match,
struct ice_tc_flower_fltr *fltr,
- struct ice_tc_flower_lyr_2_4_hdrs *headers)
+ struct ice_tc_flower_lyr_2_4_hdrs *headers, bool is_encap)
{
if (match->key->dst) {
- fltr->flags |= ICE_TC_FLWR_FIELD_DEST_IPV4;
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_DEST_IPV4;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_DEST_IPV4;
headers->l3_key.dst_ipv4 = match->key->dst;
headers->l3_mask.dst_ipv4 = match->mask->dst;
}
if (match->key->src) {
- fltr->flags |= ICE_TC_FLWR_FIELD_SRC_IPV4;
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_SRC_IPV4;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_SRC_IPV4;
headers->l3_key.src_ipv4 = match->key->src;
headers->l3_mask.src_ipv4 = match->mask->src;
}
@@ -462,11 +625,12 @@ ice_tc_set_ipv4(struct flow_match_ipv4_addrs *match,
* @match: Pointer to flow match structure
* @fltr: Pointer to filter structure
* @headers: inner or outer header fields
+ * @is_encap: set true for tunnel IPv6 address
*/
static int
ice_tc_set_ipv6(struct flow_match_ipv6_addrs *match,
struct ice_tc_flower_fltr *fltr,
- struct ice_tc_flower_lyr_2_4_hdrs *headers)
+ struct ice_tc_flower_lyr_2_4_hdrs *headers, bool is_encap)
{
struct ice_tc_l3_hdr *l3_key, *l3_mask;
@@ -484,21 +648,31 @@ ice_tc_set_ipv6(struct flow_match_ipv6_addrs *match,
NL_SET_ERR_MSG_MOD(fltr->extack, "Bad src/dest IPv6, addr is any");
return -EINVAL;
}
- if (!ipv6_addr_any(&match->mask->dst))
- fltr->flags |= ICE_TC_FLWR_FIELD_DEST_IPV6;
- if (!ipv6_addr_any(&match->mask->src))
- fltr->flags |= ICE_TC_FLWR_FIELD_SRC_IPV6;
+ if (!ipv6_addr_any(&match->mask->dst)) {
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_DEST_IPV6;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_DEST_IPV6;
+ }
+ if (!ipv6_addr_any(&match->mask->src)) {
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_SRC_IPV6;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_SRC_IPV6;
+ }
l3_key = &headers->l3_key;
l3_mask = &headers->l3_mask;
- if (fltr->flags & ICE_TC_FLWR_FIELD_SRC_IPV6) {
+ if (fltr->flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV6 |
+ ICE_TC_FLWR_FIELD_SRC_IPV6)) {
memcpy(&l3_key->src_ipv6_addr, &match->key->src.s6_addr,
sizeof(match->key->src.s6_addr));
memcpy(&l3_mask->src_ipv6_addr, &match->mask->src.s6_addr,
sizeof(match->mask->src.s6_addr));
}
- if (fltr->flags & ICE_TC_FLWR_FIELD_DEST_IPV6) {
+ if (fltr->flags & (ICE_TC_FLWR_FIELD_ENC_DEST_IPV6 |
+ ICE_TC_FLWR_FIELD_DEST_IPV6)) {
memcpy(&l3_key->dst_ipv6_addr, &match->key->dst.s6_addr,
sizeof(match->key->dst.s6_addr));
memcpy(&l3_mask->dst_ipv6_addr, &match->mask->dst.s6_addr,
@@ -513,18 +687,27 @@ ice_tc_set_ipv6(struct flow_match_ipv6_addrs *match,
* @match: Flow match structure
* @fltr: Pointer to filter structure
* @headers: inner or outer header fields
+ * @is_encap: set true for tunnel port
*/
static int
ice_tc_set_port(struct flow_match_ports match,
struct ice_tc_flower_fltr *fltr,
- struct ice_tc_flower_lyr_2_4_hdrs *headers)
+ struct ice_tc_flower_lyr_2_4_hdrs *headers, bool is_encap)
{
if (match.key->dst) {
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_DEST_L4_PORT;
fltr->flags |= ICE_TC_FLWR_FIELD_DEST_L4_PORT;
headers->l4_key.dst_port = match.key->dst;
headers->l4_mask.dst_port = match.mask->dst;
}
if (match.key->src) {
+ if (is_encap)
+ fltr->flags |= ICE_TC_FLWR_FIELD_ENC_SRC_L4_PORT;
+ else
+ fltr->flags |= ICE_TC_FLWR_FIELD_SRC_L4_PORT;
fltr->flags |= ICE_TC_FLWR_FIELD_SRC_L4_PORT;
headers->l4_key.src_port = match.key->src;
headers->l4_mask.src_port = match.mask->src;
@@ -532,6 +715,85 @@ ice_tc_set_port(struct flow_match_ports match,
return 0;
}
+static struct net_device *
+ice_get_tunnel_device(struct net_device *dev, struct flow_rule *rule)
+{
+ struct flow_action_entry *act;
+ int i;
+
+ if (ice_is_tunnel_supported(dev))
+ return dev;
+
+ flow_action_for_each(i, act, &rule->action) {
+ if (act->id == FLOW_ACTION_REDIRECT &&
+ ice_is_tunnel_supported(act->dev))
+ return act->dev;
+ }
+
+ return NULL;
+}
+
+static int
+ice_parse_tunnel_attr(struct net_device *dev, struct flow_rule *rule,
+ struct ice_tc_flower_fltr *fltr)
+{
+ struct ice_tc_flower_lyr_2_4_hdrs *headers = &fltr->outer_headers;
+ struct flow_match_control enc_control;
+
+ fltr->tunnel_type = ice_tc_tun_get_type(dev);
+ headers->l3_key.ip_proto = IPPROTO_UDP;
+
+ if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
+ struct flow_match_enc_keyid enc_keyid;
+
+ flow_rule_match_enc_keyid(rule, &enc_keyid);
+
+ if (!enc_keyid.mask->keyid ||
+ enc_keyid.mask->keyid != cpu_to_be32(ICE_TC_FLOWER_MASK_32))
+ return -EINVAL;
+
+ fltr->flags |= ICE_TC_FLWR_FIELD_TENANT_ID;
+ fltr->tenant_id = enc_keyid.key->keyid;
+ }
+
+ flow_rule_match_enc_control(rule, &enc_control);
+
+ if (enc_control.key->addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+ struct flow_match_ipv4_addrs match;
+
+ flow_rule_match_enc_ipv4_addrs(rule, &match);
+ if (ice_tc_set_ipv4(&match, fltr, headers, true))
+ return -EINVAL;
+ } else if (enc_control.key->addr_type ==
+ FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+ struct flow_match_ipv6_addrs match;
+
+ flow_rule_match_enc_ipv6_addrs(rule, &match);
+ if (ice_tc_set_ipv6(&match, fltr, headers, true))
+ return -EINVAL;
+ }
+
+ if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) {
+ struct flow_match_ip match;
+
+ flow_rule_match_enc_ip(rule, &match);
+ headers->l3_key.tos = match.key->tos;
+ headers->l3_key.ttl = match.key->ttl;
+ headers->l3_mask.tos = match.mask->tos;
+ headers->l3_mask.ttl = match.mask->ttl;
+ }
+
+ if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS)) {
+ struct flow_match_ports match;
+
+ flow_rule_match_enc_ports(rule, &match);
+ if (ice_tc_set_port(match, fltr, headers, true))
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
/**
* ice_parse_cls_flower - Parse TC flower filters provided by kernel
* @vsi: Pointer to the VSI
@@ -559,12 +821,42 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_ENC_CONTROL) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_KEYID) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_PORTS) |
BIT(FLOW_DISSECTOR_KEY_ENC_IP) |
BIT(FLOW_DISSECTOR_KEY_PORTS))) {
NL_SET_ERR_MSG_MOD(fltr->extack, "Unsupported key used");
return -EOPNOTSUPP;
}
+ if ((flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) ||
+ flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) ||
+ flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID) ||
+ flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS))) {
+ int err;
+
+ filter_dev = ice_get_tunnel_device(filter_dev, rule);
+ if (!filter_dev) {
+ NL_SET_ERR_MSG_MOD(fltr->extack, "Tunnel device not found");
+ return -EOPNOTSUPP;
+ }
+
+ err = ice_parse_tunnel_attr(filter_dev, rule, fltr);
+ if (err) {
+ NL_SET_ERR_MSG_MOD(fltr->extack, "Failed to parse TC flower tunnel attributes");
+ return err;
+ }
+
+ /* header pointers should point to the inner headers, outer
+ * header were already set by ice_parse_tunnel_attr
+ */
+ headers = &fltr->inner_headers;
+ } else {
+ fltr->tunnel_type = TNL_LAST;
+ }
+
if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) {
struct flow_match_basic match;
@@ -651,7 +943,7 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
struct flow_match_ipv4_addrs match;
flow_rule_match_ipv4_addrs(rule, &match);
- if (ice_tc_set_ipv4(&match, fltr, headers))
+ if (ice_tc_set_ipv4(&match, fltr, headers, false))
return -EINVAL;
}
@@ -659,7 +951,7 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
struct flow_match_ipv6_addrs match;
flow_rule_match_ipv6_addrs(rule, &match);
- if (ice_tc_set_ipv6(&match, fltr, headers))
+ if (ice_tc_set_ipv6(&match, fltr, headers, false))
return -EINVAL;
}
@@ -667,7 +959,7 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
struct flow_match_ports match;
flow_rule_match_ports(rule, &match);
- if (ice_tc_set_port(match, fltr, headers))
+ if (ice_tc_set_port(match, fltr, headers, false))
return -EINVAL;
switch (headers->l3_key.ip_proto) {
case IPPROTO_TCP:
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.h b/drivers/net/ethernet/intel/ice/ice_tc_lib.h
index 99ececeef445..319049477959 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.h
@@ -23,6 +23,8 @@
#define ICE_TC_FLWR_FIELD_ENC_DST_MAC BIT(16)
#define ICE_TC_FLWR_FIELD_ETH_TYPE_ID BIT(17)
+#define ICE_TC_FLOWER_MASK_32 0xFFFFFFFF
+
struct ice_indr_block_priv {
struct net_device *netdev;
struct ice_netdev_priv *np;
@@ -118,6 +120,7 @@ struct ice_tc_flower_fltr {
struct ice_vsi *src_vsi;
__be32 tenant_id;
u32 flags;
+ u8 tunnel_type;
struct ice_tc_flower_action action;
/* cache ptr which is used wherever needed to communicate netlink
@@ -154,5 +157,6 @@ ice_add_cls_flower(struct net_device *netdev, struct ice_vsi *vsi,
int
ice_del_cls_flower(struct ice_vsi *vsi, struct flow_cls_offload *cls_flower);
void ice_replay_tc_fltrs(struct ice_pf *pf);
+bool ice_is_tunnel_supported(struct net_device *dev);
#endif /* _ICE_TC_LIB_H_ */
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 3/9] ice: low level support for tunnels
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 1/9] ice: support for indirect notification Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 2/9] ice: VXLAN and Geneve TC support Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 4/9] ice: support for GRE in eswitch Tony Nguyen
` (6 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Michal Swiatkowski, netdev, anthony.l.nguyen, Sandeep Penigalapati
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Add definition of UDP tunnel dummy packets. Fill destination port value
in filter based on UDP tunnel port. Append tunnel flags to switch filter
definition in case of matching the tunnel.
Both VXLAN and Geneve are UDP tunnels, so only one new header is needed.
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
.../net/ethernet/intel/ice/ice_flex_pipe.c | 26 +-
.../net/ethernet/intel/ice/ice_flex_type.h | 2 +
.../ethernet/intel/ice/ice_protocol_type.h | 9 +
drivers/net/ethernet/intel/ice/ice_switch.c | 275 +++++++++++++++++-
drivers/net/ethernet/intel/ice/ice_switch.h | 2 +
5 files changed, 298 insertions(+), 16 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index e731b46270c3..8736d3ae230f 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -1565,6 +1565,26 @@ static struct ice_buf_build *ice_pkg_buf_alloc(struct ice_hw *hw)
return bld;
}
+/**
+ * ice_get_sw_prof_type - determine switch profile type
+ * @hw: pointer to the HW structure
+ * @fv: pointer to the switch field vector
+ */
+static enum ice_prof_type
+ice_get_sw_prof_type(struct ice_hw *hw, struct ice_fv *fv)
+{
+ u16 i;
+
+ for (i = 0; i < hw->blk[ICE_BLK_SW].es.fvw; i++) {
+ /* UDP tunnel will have UDP_OF protocol ID and VNI offset */
+ if (fv->ew[i].prot_id == (u8)ICE_PROT_UDP_OF &&
+ fv->ew[i].off == ICE_VNI_OFFSET)
+ return ICE_PROF_TUN_UDP;
+ }
+
+ return ICE_PROF_NON_TUN;
+}
+
/**
* ice_get_sw_fv_bitmap - Get switch field vector bitmap based on profile type
* @hw: pointer to hardware structure
@@ -1588,6 +1608,7 @@ ice_get_sw_fv_bitmap(struct ice_hw *hw, enum ice_prof_type req_profs,
bitmap_zero(bm, ICE_MAX_NUM_PROFILES);
ice_seg = hw->seg;
do {
+ enum ice_prof_type prof_type;
u32 offset;
fv = ice_pkg_enum_entry(ice_seg, &state, ICE_SID_FLD_VEC_SW,
@@ -1595,7 +1616,10 @@ ice_get_sw_fv_bitmap(struct ice_hw *hw, enum ice_prof_type req_profs,
ice_seg = NULL;
if (fv) {
- if (req_profs & ICE_PROF_NON_TUN)
+ /* Determine field vector type */
+ prof_type = ice_get_sw_prof_type(hw, fv);
+
+ if (req_profs & prof_type)
set_bit((u16)offset, bm);
}
} while (fv);
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h
index 120bcebaa080..07d3795d2b10 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h
@@ -614,6 +614,8 @@ struct ice_chs_chg {
enum ice_prof_type {
ICE_PROF_NON_TUN = 0x1,
+ ICE_PROF_TUN_UDP = 0x2,
+ ICE_PROF_TUN_ALL = 0x6,
ICE_PROF_ALL = 0xFF,
};
#endif /* _ICE_FLEX_TYPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index d717d1158545..8309ecaa771c 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -48,6 +48,7 @@ enum ice_sw_tunnel_type {
ICE_NON_TUN = 0,
ICE_SW_TUN_VXLAN,
ICE_SW_TUN_GENEVE,
+ ICE_ALL_TUNNELS /* All tunnel types */
};
/* Decoders for ice_prot_id:
@@ -83,6 +84,8 @@ enum ice_prot_id {
ICE_PROT_INVALID = 255 /* when offset == ICE_FV_OFFSET_INVAL */
};
+#define ICE_VNI_OFFSET 12 /* offset of VNI from ICE_PROT_UDP_OF */
+
#define ICE_MAC_OFOS_HW 1
#define ICE_MAC_IL_HW 4
#define ICE_ETYPE_OL_HW 9
@@ -96,6 +99,12 @@ enum ice_prot_id {
#define ICE_UDP_ILOS_HW 53
#define ICE_UDP_OF_HW 52 /* UDP Tunnels */
+#define ICE_META_DATA_ID_HW 255 /* this is used for tunnel type */
+
+#define ICE_MDID_SIZE 2
+#define ICE_TUN_FLAG_MDID 21
+#define ICE_TUN_FLAG_MDID_OFF (ICE_MDID_SIZE * ICE_TUN_FLAG_MDID)
+#define ICE_TUN_FLAG_MASK 0xFF
#define ICE_TUN_FLAG_FV_IND 2
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 2742e1c1e337..a2dfe8e3d3fa 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -35,6 +35,105 @@ struct ice_dummy_pkt_offsets {
u16 offset; /* ICE_PROTOCOL_LAST indicates end of list */
};
+static const struct ice_dummy_pkt_offsets dummy_udp_tun_tcp_packet_offsets[] = {
+ { ICE_MAC_OFOS, 0 },
+ { ICE_ETYPE_OL, 12 },
+ { ICE_IPV4_OFOS, 14 },
+ { ICE_UDP_OF, 34 },
+ { ICE_VXLAN, 42 },
+ { ICE_GENEVE, 42 },
+ { ICE_VXLAN_GPE, 42 },
+ { ICE_MAC_IL, 50 },
+ { ICE_IPV4_IL, 64 },
+ { ICE_TCP_IL, 84 },
+ { ICE_PROTOCOL_LAST, 0 },
+};
+
+static const u8 dummy_udp_tun_tcp_packet[] = {
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x08, 0x00, /* ICE_ETYPE_OL 12 */
+
+ 0x45, 0x00, 0x00, 0x5a, /* ICE_IPV4_OFOS 14 */
+ 0x00, 0x01, 0x00, 0x00,
+ 0x40, 0x11, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x12, 0xb5, /* ICE_UDP_OF 34 */
+ 0x00, 0x46, 0x00, 0x00,
+
+ 0x00, 0x00, 0x65, 0x58, /* ICE_VXLAN 42 */
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_IL 50 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x08, 0x00,
+
+ 0x45, 0x00, 0x00, 0x28, /* ICE_IPV4_IL 64 */
+ 0x00, 0x01, 0x00, 0x00,
+ 0x40, 0x06, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_TCP_IL 84 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x50, 0x02, 0x20, 0x00,
+ 0x00, 0x00, 0x00, 0x00
+};
+
+static const struct ice_dummy_pkt_offsets dummy_udp_tun_udp_packet_offsets[] = {
+ { ICE_MAC_OFOS, 0 },
+ { ICE_ETYPE_OL, 12 },
+ { ICE_IPV4_OFOS, 14 },
+ { ICE_UDP_OF, 34 },
+ { ICE_VXLAN, 42 },
+ { ICE_GENEVE, 42 },
+ { ICE_VXLAN_GPE, 42 },
+ { ICE_MAC_IL, 50 },
+ { ICE_IPV4_IL, 64 },
+ { ICE_UDP_ILOS, 84 },
+ { ICE_PROTOCOL_LAST, 0 },
+};
+
+static const u8 dummy_udp_tun_udp_packet[] = {
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x08, 0x00, /* ICE_ETYPE_OL 12 */
+
+ 0x45, 0x00, 0x00, 0x4e, /* ICE_IPV4_OFOS 14 */
+ 0x00, 0x01, 0x00, 0x00,
+ 0x00, 0x11, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x12, 0xb5, /* ICE_UDP_OF 34 */
+ 0x00, 0x3a, 0x00, 0x00,
+
+ 0x00, 0x00, 0x65, 0x58, /* ICE_VXLAN 42 */
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_IL 50 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x08, 0x00,
+
+ 0x45, 0x00, 0x00, 0x1c, /* ICE_IPV4_IL 64 */
+ 0x00, 0x01, 0x00, 0x00,
+ 0x00, 0x11, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_UDP_ILOS 84 */
+ 0x00, 0x08, 0x00, 0x00,
+};
+
/* offset info for MAC + IPv4 + UDP dummy packet */
static const struct ice_dummy_pkt_offsets dummy_udp_packet_offsets[] = {
{ ICE_MAC_OFOS, 0 },
@@ -3582,6 +3681,8 @@ static const struct ice_prot_ext_tbl_entry ice_prot_ext[ICE_PROTOCOL_LAST] = {
{ ICE_TCP_IL, { 0, 2 } },
{ ICE_UDP_OF, { 0, 2 } },
{ ICE_UDP_ILOS, { 0, 2 } },
+ { ICE_VXLAN, { 8, 10, 12, 14 } },
+ { ICE_GENEVE, { 8, 10, 12, 14 } },
};
static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = {
@@ -3596,6 +3697,8 @@ static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = {
{ ICE_TCP_IL, ICE_TCP_IL_HW },
{ ICE_UDP_OF, ICE_UDP_OF_HW },
{ ICE_UDP_ILOS, ICE_UDP_ILOS_HW },
+ { ICE_VXLAN, ICE_UDP_OF_HW },
+ { ICE_GENEVE, ICE_UDP_OF_HW },
};
/**
@@ -3915,12 +4018,11 @@ ice_find_free_recp_res_idx(struct ice_hw *hw, const unsigned long *profiles,
* ice_add_sw_recipe - function to call AQ calls to create switch recipe
* @hw: pointer to hardware structure
* @rm: recipe management list entry
- * @match_tun_mask: tunnel mask that needs to be programmed
* @profiles: bitmap of profiles that will be associated.
*/
static enum ice_status
ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
- u16 match_tun_mask, unsigned long *profiles)
+ unsigned long *profiles)
{
DECLARE_BITMAP(result_idx_bm, ICE_MAX_FV_WORDS);
struct ice_aqc_recipe_data_elem *tmp;
@@ -4128,15 +4230,6 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
}
buf[recps].content.act_ctrl_fwd_priority = rm->priority;
- /* To differentiate among different UDP tunnels, a meta data ID
- * flag is used.
- */
- if (match_tun_mask) {
- buf[recps].content.lkup_indx[i] = ICE_TUN_FLAG_FV_IND;
- buf[recps].content.mask[i] =
- cpu_to_le16(match_tun_mask);
- }
-
recps++;
rm->root_rid = (u8)rid;
}
@@ -4199,6 +4292,7 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
recp->chain_idx = entry->chain_idx;
recp->priority = buf[buf_idx].content.act_ctrl_fwd_priority;
recp->n_grp_count = rm->n_grp_count;
+ recp->tun_type = rm->tun_type;
recp->recp_created = true;
}
rm->root_buf = buf;
@@ -4279,6 +4373,54 @@ ice_get_fv(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
return status;
}
+/**
+ * ice_tun_type_match_word - determine if tun type needs a match mask
+ * @tun_type: tunnel type
+ * @mask: mask to be used for the tunnel
+ */
+static bool ice_tun_type_match_word(enum ice_sw_tunnel_type tun_type, u16 *mask)
+{
+ switch (tun_type) {
+ case ICE_SW_TUN_GENEVE:
+ case ICE_SW_TUN_VXLAN:
+ *mask = ICE_TUN_FLAG_MASK;
+ return true;
+
+ default:
+ *mask = 0;
+ return false;
+ }
+}
+
+/**
+ * ice_add_special_words - Add words that are not protocols, such as metadata
+ * @rinfo: other information regarding the rule e.g. priority and action info
+ * @lkup_exts: lookup word structure
+ */
+static enum ice_status
+ice_add_special_words(struct ice_adv_rule_info *rinfo,
+ struct ice_prot_lkup_ext *lkup_exts)
+{
+ u16 mask;
+
+ /* If this is a tunneled packet, then add recipe index to match the
+ * tunnel bit in the packet metadata flags.
+ */
+ if (ice_tun_type_match_word(rinfo->tun_type, &mask)) {
+ if (lkup_exts->n_val_words < ICE_MAX_CHAIN_WORDS) {
+ u8 word = lkup_exts->n_val_words++;
+
+ lkup_exts->fv_words[word].prot_id = ICE_META_DATA_ID_HW;
+ lkup_exts->fv_words[word].off = ICE_TUN_FLAG_MDID_OFF;
+ lkup_exts->field_mask[word] = mask;
+ } else {
+ return ICE_ERR_MAX_LIMIT;
+ }
+ }
+
+ return 0;
+}
+
/* ice_get_compat_fv_bitmap - Get compatible field vector bitmap for rule
* @hw: pointer to hardware structure
* @rinfo: other information regarding the rule e.g. priority and action info
@@ -4288,9 +4430,27 @@ static void
ice_get_compat_fv_bitmap(struct ice_hw *hw, struct ice_adv_rule_info *rinfo,
unsigned long *bm)
{
+ enum ice_prof_type prof_type;
+
bitmap_zero(bm, ICE_MAX_NUM_PROFILES);
- ice_get_sw_fv_bitmap(hw, ICE_PROF_NON_TUN, bm);
+ switch (rinfo->tun_type) {
+ case ICE_NON_TUN:
+ prof_type = ICE_PROF_NON_TUN;
+ break;
+ case ICE_ALL_TUNNELS:
+ prof_type = ICE_PROF_TUN_ALL;
+ break;
+ case ICE_SW_TUN_GENEVE:
+ case ICE_SW_TUN_VXLAN:
+ prof_type = ICE_PROF_TUN_UDP;
+ break;
+ default:
+ prof_type = ICE_PROF_ALL;
+ break;
+ }
+
+ ice_get_sw_fv_bitmap(hw, prof_type, bm);
}
/**
@@ -4315,7 +4475,6 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
struct ice_sw_fv_list_entry *tmp;
enum ice_status status = 0;
struct ice_sw_recipe *rm;
- u16 match_tun_mask = 0;
u8 i;
if (!lkups_cnt)
@@ -4365,6 +4524,13 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
if (status)
goto err_unroll;
+ /* Create any special protocol/offset pairs, such as looking at tunnel
+ * bits by extracting metadata
+ */
+ status = ice_add_special_words(rinfo, lkup_exts);
+ if (status)
+ goto err_free_lkup_exts;
+
/* Group match words into recipes using preferred recipe grouping
* criteria.
*/
@@ -4396,7 +4562,7 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
goto err_unroll;
/* Recipe we need does not exist, add a recipe */
- status = ice_add_sw_recipe(hw, rm, match_tun_mask, profiles);
+ status = ice_add_sw_recipe(hw, rm, profiles);
if (status)
goto err_unroll;
@@ -4466,12 +4632,14 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
* @lkups: lookup elements or match criteria for the advanced recipe, one
* structure per protocol header
* @lkups_cnt: number of protocols
+ * @tun_type: tunnel type
* @pkt: dummy packet to fill according to filter match criteria
* @pkt_len: packet length of dummy packet
* @offsets: pointer to receive the pointer to the offsets for the packet
*/
static void
ice_find_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
+ enum ice_sw_tunnel_type tun_type,
const u8 **pkt, u16 *pkt_len,
const struct ice_dummy_pkt_offsets **offsets)
{
@@ -4495,6 +4663,21 @@ ice_find_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
ipv6 = true;
}
+ if (tun_type == ICE_SW_TUN_VXLAN ||
+ tun_type == ICE_SW_TUN_GENEVE) {
+ if (tcp) {
+ *pkt = dummy_udp_tun_tcp_packet;
+ *pkt_len = sizeof(dummy_udp_tun_tcp_packet);
+ *offsets = dummy_udp_tun_tcp_packet_offsets;
+ return;
+ }
+
+ *pkt = dummy_udp_tun_udp_packet;
+ *pkt_len = sizeof(dummy_udp_tun_udp_packet);
+ *offsets = dummy_udp_tun_udp_packet_offsets;
+ return;
+ }
+
if (udp && !ipv6) {
if (vlan) {
*pkt = dummy_vlan_udp_packet;
@@ -4615,6 +4798,10 @@ ice_fill_adv_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
case ICE_SCTP_IL:
len = sizeof(struct ice_sctp_hdr);
break;
+ case ICE_VXLAN:
+ case ICE_GENEVE:
+ len = sizeof(struct ice_udp_tnl_hdr);
+ break;
default:
return ICE_ERR_PARAM;
}
@@ -4644,6 +4831,48 @@ ice_fill_adv_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
return 0;
}
+/**
+ * ice_fill_adv_packet_tun - fill dummy packet with udp tunnel port
+ * @hw: pointer to the hardware structure
+ * @tun_type: tunnel type
+ * @pkt: dummy packet to fill in
+ * @offsets: offset info for the dummy packet
+ */
+static enum ice_status
+ice_fill_adv_packet_tun(struct ice_hw *hw, enum ice_sw_tunnel_type tun_type,
+ u8 *pkt, const struct ice_dummy_pkt_offsets *offsets)
+{
+ u16 open_port, i;
+
+ switch (tun_type) {
+ case ICE_SW_TUN_VXLAN:
+ case ICE_SW_TUN_GENEVE:
+ if (!ice_get_open_tunnel_port(hw, &open_port))
+ return ICE_ERR_CFG;
+ break;
+
+ default:
+ /* Nothing needs to be done for this tunnel type */
+ return 0;
+ }
+
+ /* Find the outer UDP protocol header and insert the port number */
+ for (i = 0; offsets[i].type != ICE_PROTOCOL_LAST; i++) {
+ if (offsets[i].type == ICE_UDP_OF) {
+ struct ice_l4_hdr *hdr;
+ u16 offset;
+
+ offset = offsets[i].offset;
+ hdr = (struct ice_l4_hdr *)&pkt[offset];
+ hdr->dst_port = cpu_to_be16(open_port);
+
+ return 0;
+ }
+ }
+
+ return ICE_ERR_CFG;
+}
+
/**
* ice_find_adv_rule_entry - Search a rule entry
* @hw: pointer to the hardware structure
@@ -4678,6 +4907,7 @@ ice_find_adv_rule_entry(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
break;
}
if (rinfo->sw_act.flag == list_itr->rule_info.sw_act.flag &&
+ rinfo->tun_type == list_itr->rule_info.tun_type &&
lkups_matched)
return list_itr;
}
@@ -4852,7 +5082,7 @@ ice_add_adv_rule(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
return ICE_ERR_PARAM;
/* make sure that we can locate a dummy packet */
- ice_find_dummy_packet(lkups, lkups_cnt, &pkt, &pkt_len,
+ ice_find_dummy_packet(lkups, lkups_cnt, rinfo->tun_type, &pkt, &pkt_len,
&pkt_offsets);
if (!pkt) {
status = ICE_ERR_PARAM;
@@ -4963,6 +5193,14 @@ ice_add_adv_rule(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
if (status)
goto err_ice_add_adv_rule;
+ if (rinfo->tun_type != ICE_NON_TUN) {
+ status = ice_fill_adv_packet_tun(hw, rinfo->tun_type,
+ s_rule->pdata.lkup_tx_rx.hdr,
+ pkt_offsets);
+ if (status)
+ goto err_ice_add_adv_rule;
+ }
+
status = ice_aq_sw_rules(hw, (struct ice_aqc_sw_rules *)s_rule,
rule_buf_sz, 1, ice_aqc_opc_add_sw_rules,
NULL);
@@ -5198,6 +5436,13 @@ ice_rem_adv_rule(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
return ICE_ERR_CFG;
}
+ /* Create any special protocol/offset pairs, such as looking at tunnel
+ * bits by extracting metadata
+ */
+ status = ice_add_special_words(rinfo, &lkup_exts);
+ if (status)
+ return status;
+
rid = ice_find_recp(hw, &lkup_exts);
/* If did not find a recipe that match the existing criteria */
if (rid == ICE_MAX_NUM_RECIPES)
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index 7d661c9be81b..d8a38906f16f 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -212,6 +212,8 @@ struct ice_sw_recipe {
/* Bit map specifying the IDs associated with this group of recipe */
DECLARE_BITMAP(r_bitmap, ICE_MAX_NUM_RECIPES);
+ enum ice_sw_tunnel_type tun_type;
+
/* List of type ice_fltr_mgmt_list_entry or adv_rule */
u8 adv_rule;
struct list_head filt_rules;
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 4/9] ice: support for GRE in eswitch
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (2 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 3/9] ice: low level support for tunnels Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 5/9] ice: send correct vc status in switchdev Tony Nguyen
` (5 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Michal Swiatkowski, netdev, anthony.l.nguyen, Sandeep Penigalapati
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Mostly reuse code from Geneve and VXLAN in TC parsing code. Add new GRE
header to match on correct fields. Create new dummy packets with GRE
fields.
Instead of checking if any encap values are presented in TC flower,
check if device is tunnel type or redirect is to tunnel device. This
will allow adding all combination of rules. For example filters only
with inner fields.
Return error in case device isn't tunnel but encap values are presented.
gre example:
- create tunnel device
ip l add $NVGRE_DEV type gretap remote $NVGRE_REM_IP local $VF1_IP \
dev $PF
- add tc filter (in switchdev mode)
tc filter add dev $NVGRE_DEV protocol ip parent ffff: flower dst_ip \
$NVGRE1_IP action mirred egress redirect dev $VF1_PR
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
.../net/ethernet/intel/ice/ice_flex_pipe.c | 4 +
.../net/ethernet/intel/ice/ice_flex_type.h | 2 +
.../ethernet/intel/ice/ice_protocol_type.h | 12 +-
drivers/net/ethernet/intel/ice/ice_switch.c | 110 ++++++++++++++++++
drivers/net/ethernet/intel/ice/ice_tc_lib.c | 47 +++++---
5 files changed, 161 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
index 8736d3ae230f..23cfcceb1536 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
+++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c
@@ -1580,6 +1580,10 @@ ice_get_sw_prof_type(struct ice_hw *hw, struct ice_fv *fv)
if (fv->ew[i].prot_id == (u8)ICE_PROT_UDP_OF &&
fv->ew[i].off == ICE_VNI_OFFSET)
return ICE_PROF_TUN_UDP;
+
+ /* GRE tunnel will have GRE protocol */
+ if (fv->ew[i].prot_id == (u8)ICE_PROT_GRE_OF)
+ return ICE_PROF_TUN_GRE;
}
return ICE_PROF_NON_TUN;
diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h
index 07d3795d2b10..0f572a36d021 100644
--- a/drivers/net/ethernet/intel/ice/ice_flex_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h
@@ -373,6 +373,7 @@ struct ice_pkg_enum {
enum ice_tunnel_type {
TNL_VXLAN = 0,
TNL_GENEVE,
+ TNL_GRETAP,
__TNL_TYPE_CNT,
TNL_LAST = 0xFF,
TNL_ALL = 0xFF,
@@ -615,6 +616,7 @@ struct ice_chs_chg {
enum ice_prof_type {
ICE_PROF_NON_TUN = 0x1,
ICE_PROF_TUN_UDP = 0x2,
+ ICE_PROF_TUN_GRE = 0x4,
ICE_PROF_TUN_ALL = 0x6,
ICE_PROF_ALL = 0xFF,
};
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index 8309ecaa771c..dc1b0e9e6df5 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -39,6 +39,7 @@ enum ice_protocol_type {
ICE_UDP_ILOS,
ICE_VXLAN,
ICE_GENEVE,
+ ICE_NVGRE,
ICE_VXLAN_GPE,
ICE_SCTP_IL,
ICE_PROTOCOL_LAST
@@ -48,7 +49,8 @@ enum ice_sw_tunnel_type {
ICE_NON_TUN = 0,
ICE_SW_TUN_VXLAN,
ICE_SW_TUN_GENEVE,
- ICE_ALL_TUNNELS /* All tunnel types */
+ ICE_SW_TUN_NVGRE,
+ ICE_ALL_TUNNELS /* All tunnel types including NVGRE */
};
/* Decoders for ice_prot_id:
@@ -97,6 +99,7 @@ enum ice_prot_id {
#define ICE_IPV6_IL_HW 41
#define ICE_TCP_IL_HW 49
#define ICE_UDP_ILOS_HW 53
+#define ICE_GRE_OF_HW 64
#define ICE_UDP_OF_HW 52 /* UDP Tunnels */
#define ICE_META_DATA_ID_HW 255 /* this is used for tunnel type */
@@ -176,6 +179,12 @@ struct ice_udp_tnl_hdr {
__be32 vni; /* only use lower 24-bits */
};
+struct ice_nvgre_hdr {
+ __be16 flags;
+ __be16 protocol;
+ __be32 tni_flow;
+};
+
union ice_prot_hdr {
struct ice_ether_hdr eth_hdr;
struct ice_ethtype_hdr ethertype;
@@ -185,6 +194,7 @@ union ice_prot_hdr {
struct ice_l4_hdr l4_hdr;
struct ice_sctp_hdr sctp_hdr;
struct ice_udp_tnl_hdr tnl_hdr;
+ struct ice_nvgre_hdr nvgre_hdr;
};
/* This is mapping table entry that maps every word within a given protocol
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index a2dfe8e3d3fa..2af03b9845eb 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -35,6 +35,93 @@ struct ice_dummy_pkt_offsets {
u16 offset; /* ICE_PROTOCOL_LAST indicates end of list */
};
+static const struct ice_dummy_pkt_offsets dummy_gre_tcp_packet_offsets[] = {
+ { ICE_MAC_OFOS, 0 },
+ { ICE_ETYPE_OL, 12 },
+ { ICE_IPV4_OFOS, 14 },
+ { ICE_NVGRE, 34 },
+ { ICE_MAC_IL, 42 },
+ { ICE_IPV4_IL, 56 },
+ { ICE_TCP_IL, 76 },
+ { ICE_PROTOCOL_LAST, 0 },
+};
+
+static const u8 dummy_gre_tcp_packet[] = {
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x08, 0x00, /* ICE_ETYPE_OL 12 */
+
+ 0x45, 0x00, 0x00, 0x3E, /* ICE_IPV4_OFOS 14 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x2F, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x80, 0x00, 0x65, 0x58, /* ICE_NVGRE 34 */
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_IL 42 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x08, 0x00,
+
+ 0x45, 0x00, 0x00, 0x14, /* ICE_IPV4_IL 56 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x06, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_TCP_IL 76 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x50, 0x02, 0x20, 0x00,
+ 0x00, 0x00, 0x00, 0x00
+};
+
+static const struct ice_dummy_pkt_offsets dummy_gre_udp_packet_offsets[] = {
+ { ICE_MAC_OFOS, 0 },
+ { ICE_ETYPE_OL, 12 },
+ { ICE_IPV4_OFOS, 14 },
+ { ICE_NVGRE, 34 },
+ { ICE_MAC_IL, 42 },
+ { ICE_IPV4_IL, 56 },
+ { ICE_UDP_ILOS, 76 },
+ { ICE_PROTOCOL_LAST, 0 },
+};
+
+static const u8 dummy_gre_udp_packet[] = {
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_OFOS 0 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x08, 0x00, /* ICE_ETYPE_OL 12 */
+
+ 0x45, 0x00, 0x00, 0x3E, /* ICE_IPV4_OFOS 14 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x2F, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x80, 0x00, 0x65, 0x58, /* ICE_NVGRE 34 */
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_MAC_IL 42 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x08, 0x00,
+
+ 0x45, 0x00, 0x00, 0x14, /* ICE_IPV4_IL 56 */
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x11, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+ 0x00, 0x00, 0x00, 0x00,
+
+ 0x00, 0x00, 0x00, 0x00, /* ICE_UDP_ILOS 76 */
+ 0x00, 0x08, 0x00, 0x00,
+};
+
static const struct ice_dummy_pkt_offsets dummy_udp_tun_tcp_packet_offsets[] = {
{ ICE_MAC_OFOS, 0 },
{ ICE_ETYPE_OL, 12 },
@@ -3683,6 +3770,7 @@ static const struct ice_prot_ext_tbl_entry ice_prot_ext[ICE_PROTOCOL_LAST] = {
{ ICE_UDP_ILOS, { 0, 2 } },
{ ICE_VXLAN, { 8, 10, 12, 14 } },
{ ICE_GENEVE, { 8, 10, 12, 14 } },
+ { ICE_NVGRE, { 0, 2, 4, 6 } },
};
static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = {
@@ -3699,6 +3787,7 @@ static struct ice_protocol_entry ice_prot_id_tbl[ICE_PROTOCOL_LAST] = {
{ ICE_UDP_ILOS, ICE_UDP_ILOS_HW },
{ ICE_VXLAN, ICE_UDP_OF_HW },
{ ICE_GENEVE, ICE_UDP_OF_HW },
+ { ICE_NVGRE, ICE_GRE_OF_HW },
};
/**
@@ -4383,6 +4472,7 @@ static bool ice_tun_type_match_word(enum ice_sw_tunnel_type tun_type, u16 *mask)
switch (tun_type) {
case ICE_SW_TUN_GENEVE:
case ICE_SW_TUN_VXLAN:
+ case ICE_SW_TUN_NVGRE:
*mask = ICE_TUN_FLAG_MASK;
return true;
@@ -4445,6 +4535,9 @@ ice_get_compat_fv_bitmap(struct ice_hw *hw, struct ice_adv_rule_info *rinfo,
case ICE_SW_TUN_VXLAN:
prof_type = ICE_PROF_TUN_UDP;
break;
+ case ICE_SW_TUN_NVGRE:
+ prof_type = ICE_PROF_TUN_GRE;
+ break;
default:
prof_type = ICE_PROF_ALL;
break;
@@ -4663,6 +4756,20 @@ ice_find_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
ipv6 = true;
}
+ if (tun_type == ICE_SW_TUN_NVGRE) {
+ if (tcp) {
+ *pkt = dummy_gre_tcp_packet;
+ *pkt_len = sizeof(dummy_gre_tcp_packet);
+ *offsets = dummy_gre_tcp_packet_offsets;
+ return;
+ }
+
+ *pkt = dummy_gre_udp_packet;
+ *pkt_len = sizeof(dummy_gre_udp_packet);
+ *offsets = dummy_gre_udp_packet_offsets;
+ return;
+ }
+
if (tun_type == ICE_SW_TUN_VXLAN ||
tun_type == ICE_SW_TUN_GENEVE) {
if (tcp) {
@@ -4798,6 +4905,9 @@ ice_fill_adv_dummy_packet(struct ice_adv_lkup_elem *lkups, u16 lkups_cnt,
case ICE_SCTP_IL:
len = sizeof(struct ice_sctp_hdr);
break;
+ case ICE_NVGRE:
+ len = sizeof(struct ice_nvgre_hdr);
+ break;
case ICE_VXLAN:
case ICE_GENEVE:
len = sizeof(struct ice_udp_tnl_hdr);
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
index 920d9024a6c1..e5d23feb6701 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
@@ -102,6 +102,8 @@ ice_proto_type_from_tunnel(enum ice_tunnel_type type)
return ICE_VXLAN;
case TNL_GENEVE:
return ICE_GENEVE;
+ case TNL_GRETAP:
+ return ICE_NVGRE;
default:
return 0;
}
@@ -115,6 +117,8 @@ ice_sw_type_from_tunnel(enum ice_tunnel_type type)
return ICE_SW_TUN_VXLAN;
case TNL_GENEVE:
return ICE_SW_TUN_GENEVE;
+ case TNL_GRETAP:
+ return ICE_SW_TUN_NVGRE;
default:
return ICE_NON_TUN;
}
@@ -131,10 +135,22 @@ ice_tc_fill_tunnel_outer(u32 flags, struct ice_tc_flower_fltr *fltr,
u32 tenant_id;
list[i].type = ice_proto_type_from_tunnel(fltr->tunnel_type);
- tenant_id = be32_to_cpu(fltr->tenant_id) << 8;
- list[i].h_u.tnl_hdr.vni = cpu_to_be32(tenant_id);
- memcpy(&list[i].m_u.tnl_hdr.vni, "\xff\xff\xff\x00", 4);
- i++;
+ switch (fltr->tunnel_type) {
+ case TNL_VXLAN:
+ case TNL_GENEVE:
+ tenant_id = be32_to_cpu(fltr->tenant_id) << 8;
+ list[i].h_u.tnl_hdr.vni = cpu_to_be32(tenant_id);
+ memcpy(&list[i].m_u.tnl_hdr.vni, "\xff\xff\xff\x00", 4);
+ i++;
+ break;
+ case TNL_GRETAP:
+ list[i].h_u.nvgre_hdr.tni_flow = fltr->tenant_id;
+ memcpy(&list[i].m_u.nvgre_hdr.tni_flow, "\xff\xff\xff\xff", 4);
+ i++;
+ break;
+ default:
+ break;
+ }
}
if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
@@ -332,6 +348,9 @@ static int ice_tc_tun_get_type(struct net_device *tunnel_dev)
return TNL_VXLAN;
if (netif_is_geneve(tunnel_dev))
return TNL_GENEVE;
+ if (netif_is_gretap(tunnel_dev) ||
+ netif_is_ip6gretap(tunnel_dev))
+ return TNL_GRETAP;
return TNL_LAST;
}
@@ -810,6 +829,7 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
struct flow_rule *rule = flow_cls_offload_flow_rule(f);
u16 n_proto_mask = 0, n_proto_key = 0, addr_type = 0;
struct flow_dissector *dissector;
+ struct net_device *tunnel_dev;
dissector = rule->match.dissector;
@@ -831,17 +851,11 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
return -EOPNOTSUPP;
}
- if ((flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) ||
- flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) ||
- flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID) ||
- flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_PORTS))) {
+ tunnel_dev = ice_get_tunnel_device(filter_dev, rule);
+ if (tunnel_dev) {
int err;
- filter_dev = ice_get_tunnel_device(filter_dev, rule);
- if (!filter_dev) {
- NL_SET_ERR_MSG_MOD(fltr->extack, "Tunnel device not found");
- return -EOPNOTSUPP;
- }
+ filter_dev = tunnel_dev;
err = ice_parse_tunnel_attr(filter_dev, rule, fltr);
if (err) {
@@ -853,6 +867,13 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
* header were already set by ice_parse_tunnel_attr
*/
headers = &fltr->inner_headers;
+ } else if (dissector->used_keys &
+ (BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_KEYID) |
+ BIT(FLOW_DISSECTOR_KEY_ENC_PORTS))) {
+ NL_SET_ERR_MSG_MOD(fltr->extack, "Tunnel key used, but device isn't a tunnel");
+ return -EOPNOTSUPP;
} else {
fltr->tunnel_type = TNL_LAST;
}
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 5/9] ice: send correct vc status in switchdev
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (3 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 4/9] ice: support for GRE in eswitch Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 6/9] ice: Add support for changing MTU on PR in switchdev mode Tony Nguyen
` (4 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Michal Swiatkowski, netdev, anthony.l.nguyen, Sandeep Penigalapati
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Part of virtchannel messages are treated in different way in switchdev
mode to block configuring VFs from iavf driver side. This blocking was
done by doing nothing and returning success, event without sending
response.
Not sending response for opcodes that aren't supported in switchdev mode
leads to block iavf driver message handling. This happens for example
when vlan is configured at VF config time (VLAN module is already
loaded).
To get rid of it ice driver should answer for each VF message. In
switchdev mode:
- for adding/deleting VLAN driver should answer success without doing
anything to allow creating vlan device on VFs
- for enabling/disabling VLAN stripping and promiscuous mode driver
should answer not supported, this feature in switchdev can be only
set from host side
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
.../net/ethernet/intel/ice/ice_virtchnl_pf.c | 63 ++++++++++++++-----
1 file changed, 49 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index a42eaf6f942e..6a74344a3c21 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -4499,13 +4499,6 @@ void ice_vc_set_dflt_vf_ops(struct ice_vc_vf_ops *ops)
*ops = ice_vc_vf_dflt_ops;
}
-static int
-ice_vc_repr_no_action_msg(struct ice_vf __always_unused *vf,
- u8 __always_unused *msg)
-{
- return 0;
-}
-
/**
* ice_vc_repr_add_mac
* @vf: pointer to VF
@@ -4581,20 +4574,62 @@ ice_vc_repr_del_mac(struct ice_vf __always_unused *vf, u8 __always_unused *msg)
VIRTCHNL_STATUS_SUCCESS, NULL, 0);
}
-static int ice_vc_repr_no_action(struct ice_vf __always_unused *vf)
+static int ice_vc_repr_add_vlan(struct ice_vf *vf, u8 __always_unused *msg)
{
- return 0;
+ dev_dbg(ice_pf_to_dev(vf->pf),
+ "Can't add VLAN in switchdev mode for VF %d\n", vf->vf_id);
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_VLAN,
+ VIRTCHNL_STATUS_SUCCESS, NULL, 0);
+}
+
+static int ice_vc_repr_del_vlan(struct ice_vf *vf, u8 __always_unused *msg)
+{
+ dev_dbg(ice_pf_to_dev(vf->pf),
+ "Can't delete VLAN in switchdev mode for VF %d\n", vf->vf_id);
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DEL_VLAN,
+ VIRTCHNL_STATUS_SUCCESS, NULL, 0);
+}
+
+static int ice_vc_repr_ena_vlan_stripping(struct ice_vf *vf)
+{
+ dev_dbg(ice_pf_to_dev(vf->pf),
+ "Can't enable VLAN stripping in switchdev mode for VF %d\n",
+ vf->vf_id);
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_VLAN_STRIPPING,
+ VIRTCHNL_STATUS_ERR_NOT_SUPPORTED,
+ NULL, 0);
+}
+
+static int ice_vc_repr_dis_vlan_stripping(struct ice_vf *vf)
+{
+ dev_dbg(ice_pf_to_dev(vf->pf),
+ "Can't disable VLAN stripping in switchdev mode for VF %d\n",
+ vf->vf_id);
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_VLAN_STRIPPING,
+ VIRTCHNL_STATUS_ERR_NOT_SUPPORTED,
+ NULL, 0);
+}
+
+static int
+ice_vc_repr_cfg_promiscuous_mode(struct ice_vf *vf, u8 __always_unused *msg)
+{
+ dev_dbg(ice_pf_to_dev(vf->pf),
+ "Can't config promiscuous mode in switchdev mode for VF %d\n",
+ vf->vf_id);
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE,
+ VIRTCHNL_STATUS_ERR_NOT_SUPPORTED,
+ NULL, 0);
}
void ice_vc_change_ops_to_repr(struct ice_vc_vf_ops *ops)
{
ops->add_mac_addr_msg = ice_vc_repr_add_mac;
ops->del_mac_addr_msg = ice_vc_repr_del_mac;
- ops->add_vlan_msg = ice_vc_repr_no_action_msg;
- ops->remove_vlan_msg = ice_vc_repr_no_action_msg;
- ops->ena_vlan_stripping = ice_vc_repr_no_action;
- ops->dis_vlan_stripping = ice_vc_repr_no_action;
- ops->cfg_promiscuous_mode_msg = ice_vc_repr_no_action_msg;
+ ops->add_vlan_msg = ice_vc_repr_add_vlan;
+ ops->remove_vlan_msg = ice_vc_repr_del_vlan;
+ ops->ena_vlan_stripping = ice_vc_repr_ena_vlan_stripping;
+ ops->dis_vlan_stripping = ice_vc_repr_dis_vlan_stripping;
+ ops->cfg_promiscuous_mode_msg = ice_vc_repr_cfg_promiscuous_mode;
}
/**
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 6/9] ice: Add support for changing MTU on PR in switchdev mode
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (4 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 5/9] ice: send correct vc status in switchdev Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 7/9] ice: Add support to print error on PHY FW load failure Tony Nguyen
` (3 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba; +Cc: Marcin Szycik, netdev, anthony.l.nguyen, Sandeep Penigalapati
From: Marcin Szycik <marcin.szycik@intel.com>
This change adds support for changing MTU on port representor in
switchdev mode, by setting the min/max MTU values on port representor
netdev. Before it was possible to change the MTU only in a limited,
default range (68-1500).
Signed-off-by: Marcin Szycik <marcin.szycik@intel.com>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice_repr.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c
index c49eeea7cb67..af8e6ef5f571 100644
--- a/drivers/net/ethernet/intel/ice/ice_repr.c
+++ b/drivers/net/ethernet/intel/ice/ice_repr.c
@@ -267,6 +267,9 @@ static int ice_repr_add(struct ice_vf *vf)
if (err)
goto err_devlink;
+ repr->netdev->min_mtu = ETH_MIN_MTU;
+ repr->netdev->max_mtu = ICE_MAX_MTU;
+
err = ice_repr_reg_netdev(repr->netdev);
if (err)
goto err_netdev;
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 7/9] ice: Add support to print error on PHY FW load failure
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (5 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 6/9] ice: Add support for changing MTU on PR in switchdev mode Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 8/9] ice: Fix clang -Wimplicit-fallthrough in ice_pull_qvec_from_rc() Tony Nguyen
` (2 subsequent siblings)
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba; +Cc: Brett Creeley, netdev, anthony.l.nguyen, Sunitha Mekala
From: Brett Creeley <brett.creeley@intel.com>
Some devices have support for loading the PHY FW and in some cases this
can fail. When this fails, the FW will set the corresponding bit in the
link info structure. Also, the FW will send a link event if the correct
link event mask bit is set. Add support for printing an error message
when the PHY FW load fails during any link configuration flow and the
link event flow.
Since ice_check_module_power() is already doing something very similar
add a new function ice_check_link_cfg_err() so any failures reported in
the link info's link_cfg_err member can be printed in this one function.
Also, add the new ICE_FLAG_PHY_FW_LOAD_FAILED bit to the PF's flags so
we don't constantly print this error message during link polling if the
value never changed.
Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice.h | 1 +
.../net/ethernet/intel/ice/ice_adminq_cmd.h | 2 +
drivers/net/ethernet/intel/ice/ice_main.c | 49 +++++++++++++++++--
3 files changed, 47 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 763add09559c..bf4ecd9a517c 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -480,6 +480,7 @@ enum ice_pf_flags {
ICE_FLAG_NO_MEDIA,
ICE_FLAG_FW_LLDP_AGENT,
ICE_FLAG_MOD_POWER_UNSUPPORTED,
+ ICE_FLAG_PHY_FW_LOAD_FAILED,
ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */
ICE_FLAG_LEGACY_RX,
ICE_FLAG_VF_TRUE_PROMISC_ENA,
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index a5425f0dce3f..4eef3488d86f 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1185,6 +1185,7 @@ struct ice_aqc_get_link_status_data {
#define ICE_AQ_LINK_TOPO_UNSUPP_MEDIA BIT(7)
u8 link_cfg_err;
#define ICE_AQ_LINK_MODULE_POWER_UNSUPPORTED BIT(5)
+#define ICE_AQ_LINK_EXTERNAL_PHY_LOAD_FAILURE BIT(6)
#define ICE_AQ_LINK_INVAL_MAX_POWER_LIMIT BIT(7)
u8 link_info;
#define ICE_AQ_LINK_UP BIT(0) /* Link Status */
@@ -1268,6 +1269,7 @@ struct ice_aqc_set_event_mask {
#define ICE_AQ_LINK_EVENT_AN_COMPLETED BIT(7)
#define ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL BIT(8)
#define ICE_AQ_LINK_EVENT_PORT_TX_SUSPENDED BIT(9)
+#define ICE_AQ_LINK_EVENT_PHY_FW_LOAD_FAIL BIT(12)
u8 reserved1[6];
};
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 2ebbbe1edd82..66112addfb9a 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -936,6 +936,29 @@ static void ice_set_dflt_mib(struct ice_pf *pf)
kfree(lldpmib);
}
+/**
+ * ice_check_phy_fw_load - check if PHY FW load failed
+ * @pf: pointer to PF struct
+ * @link_cfg_err: bitmap from the link info structure
+ *
+ * check if external PHY FW load failed and print an error message if it did
+ */
+static void ice_check_phy_fw_load(struct ice_pf *pf, u8 link_cfg_err)
+{
+ if (!(link_cfg_err & ICE_AQ_LINK_EXTERNAL_PHY_LOAD_FAILURE)) {
+ clear_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags);
+ return;
+ }
+
+ if (test_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags))
+ return;
+
+ if (link_cfg_err & ICE_AQ_LINK_EXTERNAL_PHY_LOAD_FAILURE) {
+ dev_err(ice_pf_to_dev(pf), "Device failed to load the FW for the external PHY. Please download and install the latest NVM for your device and try again\n");
+ set_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags);
+ }
+}
+
/**
* ice_check_module_power
* @pf: pointer to PF struct
@@ -968,6 +991,20 @@ static void ice_check_module_power(struct ice_pf *pf, u8 link_cfg_err)
}
}
+/**
+ * ice_check_link_cfg_err - check if link configuration failed
+ * @pf: pointer to the PF struct
+ * @link_cfg_err: bitmap from the link info structure
+ *
+ * print if any link configuration failure happens due to the value in the
+ * link_cfg_err parameter in the link info structure
+ */
+static void ice_check_link_cfg_err(struct ice_pf *pf, u8 link_cfg_err)
+{
+ ice_check_module_power(pf, link_cfg_err);
+ ice_check_phy_fw_load(pf, link_cfg_err);
+}
+
/**
* ice_link_event - process the link event
* @pf: PF that the link event is associated with
@@ -1003,7 +1040,7 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
pi->lport, ice_stat_str(status),
ice_aq_str(pi->hw->adminq.sq_last_status));
- ice_check_module_power(pf, pi->phy.link_info.link_cfg_err);
+ ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err);
/* Check if the link state is up after updating link info, and treat
* this event as an UP event since the link is actually UP now.
@@ -1081,7 +1118,8 @@ static int ice_init_link_events(struct ice_port_info *pi)
u16 mask;
mask = ~((u16)(ICE_AQ_LINK_EVENT_UPDOWN | ICE_AQ_LINK_EVENT_MEDIA_NA |
- ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL));
+ ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL |
+ ICE_AQ_LINK_EVENT_PHY_FW_LOAD_FAIL));
if (ice_aq_set_event_mask(pi->hw, pi->lport, mask, NULL)) {
dev_dbg(ice_hw_to_dev(pi->hw), "Failed to set link event mask for port %d\n",
@@ -2152,7 +2190,7 @@ static void ice_check_media_subtask(struct ice_pf *pf)
if (err)
return;
- ice_check_module_power(pf, pi->phy.link_info.link_cfg_err);
+ ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err);
if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) {
if (!test_bit(ICE_PHY_INIT_COMPLETE, pf->state))
@@ -4600,7 +4638,8 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent)
ice_init_link_dflt_override(pf->hw.port_info);
- ice_check_module_power(pf, pf->hw.port_info->phy.link_info.link_cfg_err);
+ ice_check_link_cfg_err(pf,
+ pf->hw.port_info->phy.link_info.link_cfg_err);
/* if media available, initialize PHY settings */
if (pf->hw.port_info->phy.link_info.link_info &
@@ -8402,7 +8441,7 @@ int ice_open_internal(struct net_device *netdev)
return -EIO;
}
- ice_check_module_power(pf, pi->phy.link_info.link_cfg_err);
+ ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err);
/* Set PHY if there is media, otherwise, turn off PHY */
if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) {
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 8/9] ice: Fix clang -Wimplicit-fallthrough in ice_pull_qvec_from_rc()
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (6 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 7/9] ice: Add support to print error on PHY FW load failure Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-28 18:06 ` [PATCH net-next 9/9] ice: fix error return code in ice_get_recp_frm_fw() Tony Nguyen
2021-10-29 12:10 ` [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 patchwork-bot+netdevbpf
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba
Cc: Nathan Chancellor, netdev, anthony.l.nguyen, Jesse Brandeburg,
Gurucharan G
From: Nathan Chancellor <nathan@kernel.org>
Clang warns:
drivers/net/ethernet/intel/ice/ice_lib.c:1906:2: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
default:
^
drivers/net/ethernet/intel/ice/ice_lib.c:1906:2: note: insert 'break;' to avoid fall-through
default:
^
break;
1 error generated.
Clang is a little more pedantic than GCC, which does not warn when
falling through to a case that is just break or return. Clang's version
is more in line with the kernel's own stance in deprecated.rst, which
states that all switch/case blocks must end in either break,
fallthrough, continue, goto, or return. Add the missing break to silence
the warning.
Link: https://github.com/ClangBuiltLinux/linux/issues/1482
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice_lib.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 77dceab9fbbe..159c52b9b9d4 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -1983,6 +1983,7 @@ static struct ice_q_vector *ice_pull_qvec_from_rc(struct ice_ring_container *rc)
case ICE_TX_CONTAINER:
if (rc->tx_ring)
return rc->tx_ring->q_vector;
+ break;
default:
break;
}
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* [PATCH net-next 9/9] ice: fix error return code in ice_get_recp_frm_fw()
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (7 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 8/9] ice: Fix clang -Wimplicit-fallthrough in ice_pull_qvec_from_rc() Tony Nguyen
@ 2021-10-28 18:06 ` Tony Nguyen
2021-10-29 12:10 ` [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 patchwork-bot+netdevbpf
9 siblings, 0 replies; 11+ messages in thread
From: Tony Nguyen @ 2021-10-28 18:06 UTC (permalink / raw)
To: davem, kuba; +Cc: Wang Hai, netdev, anthony.l.nguyen, Hulk Robot
From: Wang Hai <wanghai38@huawei.com>
Return error code if devm_kmemdup() fails in ice_get_recp_frm_fw()
Fixes: fd2a6b71e300 ("ice: create advanced switch recipe")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
drivers/net/ethernet/intel/ice/ice_switch.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 2af03b9845eb..793f4a9fc2cd 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -1363,8 +1363,10 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
recps[rid].root_buf = devm_kmemdup(ice_hw_to_dev(hw), tmp,
recps[rid].n_grp_count * sizeof(*recps[rid].root_buf),
GFP_KERNEL);
- if (!recps[rid].root_buf)
+ if (!recps[rid].root_buf) {
+ status = ICE_ERR_NO_MEMORY;
goto err_unroll;
+ }
/* Copy result indexes */
bitmap_copy(recps[rid].res_idxs, result_bm, ICE_MAX_FV_WORDS);
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread* Re: [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28
2021-10-28 18:06 [PATCH net-next 0/9][pull request] 100GbE Intel Wired LAN Driver Updates 2021-10-28 Tony Nguyen
` (8 preceding siblings ...)
2021-10-28 18:06 ` [PATCH net-next 9/9] ice: fix error return code in ice_get_recp_frm_fw() Tony Nguyen
@ 2021-10-29 12:10 ` patchwork-bot+netdevbpf
9 siblings, 0 replies; 11+ messages in thread
From: patchwork-bot+netdevbpf @ 2021-10-29 12:10 UTC (permalink / raw)
To: Tony Nguyen; +Cc: davem, kuba, netdev
Hello:
This series was applied to netdev/net-next.git (master)
by Tony Nguyen <anthony.l.nguyen@intel.com>:
On Thu, 28 Oct 2021 11:06:50 -0700 you wrote:
> This series contains updates to ice driver only.
>
> Michal adds support for eswitch drop and redirect filters from and to
> tunnel devices. From meaning from uplink to VF and to means from VF to
> uplink. This is accomplished by adding support for indirect TC tunnel
> notifications and adding appropriate training packets and match fields
> for UDP tunnel headers. He also adds returning virtchannel responses for
> blocked operations as returning a response is still needed.
>
> [...]
Here is the summary with links:
- [net-next,1/9] ice: support for indirect notification
https://git.kernel.org/netdev/net-next/c/195bb48fccde
- [net-next,2/9] ice: VXLAN and Geneve TC support
https://git.kernel.org/netdev/net-next/c/9e300987d4a8
- [net-next,3/9] ice: low level support for tunnels
https://git.kernel.org/netdev/net-next/c/8b032a55c1bd
- [net-next,4/9] ice: support for GRE in eswitch
https://git.kernel.org/netdev/net-next/c/f0a35040adbe
- [net-next,5/9] ice: send correct vc status in switchdev
https://git.kernel.org/netdev/net-next/c/e492c2e12d7b
- [net-next,6/9] ice: Add support for changing MTU on PR in switchdev mode
https://git.kernel.org/netdev/net-next/c/e984c4408fc9
- [net-next,7/9] ice: Add support to print error on PHY FW load failure
https://git.kernel.org/netdev/net-next/c/99d407524cdf
- [net-next,8/9] ice: Fix clang -Wimplicit-fallthrough in ice_pull_qvec_from_rc()
https://git.kernel.org/netdev/net-next/c/370764e60b18
- [net-next,9/9] ice: fix error return code in ice_get_recp_frm_fw()
https://git.kernel.org/netdev/net-next/c/c8e51a012214
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 11+ messages in thread