* [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad
@ 2015-07-27 11:46 Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 1/4] net/mlx4_core: Preparations for 802.1ad VLAN support Amir Vadai
` (4 more replies)
0 siblings, 5 replies; 8+ messages in thread
From: Amir Vadai @ 2015-07-27 11:46 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Hadar Har-Zion, Or Gerlitz, Tal Alon, Amir Vadai
Hi,
This patchset by Hadar introduces support in Hardware accelerated 802.1ad, for
ConnectX-3pro NIC's. In order to support existing deployment, and due to some
hardware limitations, the feature is disabled by default, and needed to be
enabled using a private flag in ethtool. Ofcourse user can enable the private
flag only if hardware has support.
After being enabled, the standard ethtool -k/-K can be used.
Patchset was applied and tested over commit 71790a2 ("hv_netvsc: Add structs
and handlers for VF messages")
Amir
Hadar Hen Zion (4):
net/mlx4_core: Preparations for 802.1ad VLAN support
net/mlx4_en: Prepare ethtool private flags to support more flags
net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support
net/mlx4_en: Add support for hardware accelerated 802.1ad vlan
drivers/infiniband/hw/mlx4/cq.c | 2 +-
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 51 +++++++++------
drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 46 ++++++++++++++
drivers/net/ethernet/mellanox/mlx4/en_rx.c | 22 +++++--
drivers/net/ethernet/mellanox/mlx4/en_tx.c | 13 ++--
drivers/net/ethernet/mellanox/mlx4/fw.c | 82 +++++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx4/fw.h | 1 +
drivers/net/ethernet/mellanox/mlx4/main.c | 15 +++++
drivers/net/ethernet/mellanox/mlx4/mlx4.h | 3 +
drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 +
include/linux/mlx4/cq.h | 3 +-
include/linux/mlx4/device.h | 5 ++
include/linux/mlx4/qp.h | 3 +-
13 files changed, 218 insertions(+), 29 deletions(-)
--
2.4.3.413.ga5fe668
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH net-next 1/4] net/mlx4_core: Preparations for 802.1ad VLAN support
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
@ 2015-07-27 11:46 ` Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 2/4] net/mlx4_en: Prepare ethtool private flags to support more flags Amir Vadai
` (3 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Amir Vadai @ 2015-07-27 11:46 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Hadar Har-Zion, Or Gerlitz, Tal Alon, Amir Vadai
From: Hadar Hen Zion <hadarh@mellanox.com>
mlx4_core preparation to support hardware accelerated 802.1ad VLAN
device.
To allow 802.1ad accelerated device, "packet has vlan" (phv)
Firmware capability should be available. Firmware without the
phv capability won't behave properly and can't support 802.1ad device
acceleration.
The driver checks the Firmware capability and sets the phv bit
accordingly in SET_PORT command.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/fw.c | 82 +++++++++++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx4/fw.h | 1 +
drivers/net/ethernet/mellanox/mlx4/main.c | 15 ++++++
drivers/net/ethernet/mellanox/mlx4/mlx4.h | 3 ++
include/linux/mlx4/device.h | 5 ++
5 files changed, 106 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index e30bf57..5a1c3d2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -154,6 +154,7 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags)
[26] = "Port ETS Scheduler support",
[27] = "Port beacon support",
[28] = "RX-ALL support",
+ [29] = "802.1ad offload support",
};
int i;
@@ -307,6 +308,7 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
#define QUERY_FUNC_CAP_FLAGS0_FORCE_PHY_WQE_GID 0x80
#define QUERY_FUNC_CAP_SUPPORTS_NON_POWER_OF_2_NUM_EQS (1 << 31)
+#define QUERY_FUNC_CAP_PHV_BIT 0x40
if (vhcr->op_modifier == 1) {
struct mlx4_active_ports actv_ports =
@@ -351,6 +353,12 @@ int mlx4_QUERY_FUNC_CAP_wrapper(struct mlx4_dev *dev, int slave,
MLX4_PUT(outbox->buf, dev->caps.phys_port_id[vhcr->in_modifier],
QUERY_FUNC_CAP_PHYS_PORT_ID);
+ if (dev->caps.phv_bit[port]) {
+ field = QUERY_FUNC_CAP_PHV_BIT;
+ MLX4_PUT(outbox->buf, field,
+ QUERY_FUNC_CAP_FLAGS0_OFFSET);
+ }
+
} else if (vhcr->op_modifier == 0) {
struct mlx4_active_ports actv_ports =
mlx4_get_active_ports(dev, slave);
@@ -600,6 +608,9 @@ int mlx4_QUERY_FUNC_CAP(struct mlx4_dev *dev, u8 gen_or_port,
MLX4_GET(func_cap->phys_port_id, outbox,
QUERY_FUNC_CAP_PHYS_PORT_ID);
+ MLX4_GET(field, outbox, QUERY_FUNC_CAP_FLAGS0_OFFSET);
+ func_cap->flags |= (field & QUERY_FUNC_CAP_PHV_BIT);
+
/* All other resources are allocated by the master, but we still report
* 'num' and 'reserved' capabilities as follows:
* - num remains the maximum resource index
@@ -700,6 +711,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
#define QUERY_DEV_CAP_D_MPT_ENTRY_SZ_OFFSET 0x92
#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET 0x94
#define QUERY_DEV_CAP_CONFIG_DEV_OFFSET 0x94
+#define QUERY_DEV_CAP_PHV_EN_OFFSET 0x96
#define QUERY_DEV_CAP_RSVD_LKEY_OFFSET 0x98
#define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET 0xa0
#define QUERY_DEV_CAP_ETH_BACKPL_OFFSET 0x9c
@@ -898,6 +910,12 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_CONFIG_DEV;
if (field & (1 << 2))
dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_IGNORE_FCS;
+ MLX4_GET(field, outbox, QUERY_DEV_CAP_PHV_EN_OFFSET);
+ if (field & 0x80)
+ dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_PHV_EN;
+ if (field & 0x40)
+ dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN;
+
MLX4_GET(dev_cap->reserved_lkey, outbox,
QUERY_DEV_CAP_RSVD_LKEY_OFFSET);
MLX4_GET(field32, outbox, QUERY_DEV_CAP_ETH_BACKPL_OFFSET);
@@ -1992,6 +2010,10 @@ int mlx4_QUERY_HCA(struct mlx4_dev *dev,
MLX4_GET(param->uar_page_sz, outbox, INIT_HCA_UAR_PAGE_SZ_OFFSET);
MLX4_GET(param->log_uar_sz, outbox, INIT_HCA_LOG_UAR_SZ_OFFSET);
+ /* phv_check enable */
+ MLX4_GET(byte_field, outbox, INIT_HCA_CACHELINE_SZ_OFFSET);
+ if (byte_field & 0x2)
+ param->phv_check_en = 1;
out:
mlx4_free_cmd_mailbox(dev, mailbox);
@@ -2758,3 +2780,63 @@ int mlx4_ACCESS_REG_wrapper(struct mlx4_dev *dev, int slave,
0, MLX4_CMD_ACCESS_REG, MLX4_CMD_TIME_CLASS_C,
MLX4_CMD_NATIVE);
}
+
+static int mlx4_SET_PORT_phv_bit(struct mlx4_dev *dev, u8 port, u8 phv_bit)
+{
+#define SET_PORT_GEN_PHV_VALID 0x10
+#define SET_PORT_GEN_PHV_EN 0x80
+
+ struct mlx4_cmd_mailbox *mailbox;
+ struct mlx4_set_port_general_context *context;
+ u32 in_mod;
+ int err;
+
+ mailbox = mlx4_alloc_cmd_mailbox(dev);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+ context = mailbox->buf;
+
+ context->v_ignore_fcs |= SET_PORT_GEN_PHV_VALID;
+ if (phv_bit)
+ context->phv_en |= SET_PORT_GEN_PHV_EN;
+
+ in_mod = MLX4_SET_PORT_GENERAL << 8 | port;
+ err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE,
+ MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B,
+ MLX4_CMD_NATIVE);
+
+ mlx4_free_cmd_mailbox(dev, mailbox);
+ return err;
+}
+
+int get_phv_bit(struct mlx4_dev *dev, u8 port, int *phv)
+{
+ int err;
+ struct mlx4_func_cap func_cap;
+
+ memset(&func_cap, 0, sizeof(func_cap));
+ err = mlx4_QUERY_FUNC_CAP(dev, 1, &func_cap);
+ if (!err)
+ *phv = func_cap.flags & QUERY_FUNC_CAP_PHV_BIT;
+ return err;
+}
+EXPORT_SYMBOL(get_phv_bit);
+
+int set_phv_bit(struct mlx4_dev *dev, u8 port, int new_val)
+{
+ int ret;
+
+ if (mlx4_is_slave(dev))
+ return -EPERM;
+
+ if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN &&
+ !(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN)) {
+ ret = mlx4_SET_PORT_phv_bit(dev, port, new_val);
+ if (!ret)
+ dev->caps.phv_bit[port] = new_val;
+ return ret;
+ }
+
+ return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL(set_phv_bit);
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h
index 07cb7c2..08de555 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.h
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.h
@@ -204,6 +204,7 @@ struct mlx4_init_hca_param {
u16 cqe_size; /* For use only when CQE stride feature enabled */
u16 eqe_size; /* For use only when EQE stride feature enabled */
u8 rss_ip_frags;
+ u8 phv_check_en; /* for QUERY_HCA */
};
struct mlx4_init_ib_param {
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index d76f425..6f35b6c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -405,6 +405,21 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
dev->caps.max_gso_sz = dev_cap->max_gso_sz;
dev->caps.max_rss_tbl_sz = dev_cap->max_rss_tbl_sz;
+ if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN) {
+ struct mlx4_init_hca_param hca_param;
+
+ memset(&hca_param, 0, sizeof(hca_param));
+ err = mlx4_QUERY_HCA(dev, &hca_param);
+ /* Turn off PHV_EN flag in case phv_check_en is set.
+ * phv_check_en is a HW check that parse the packet and verify
+ * phv bit was reported correctly in the wqe. To allow QinQ
+ * PHV_EN flag should be set and phv_check_en must be cleared
+ * otherwise QinQ packets will be drop by the HW.
+ */
+ if (err || hca_param.phv_check_en)
+ dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_PHV_EN;
+ }
+
/* Sense port always allowed on supported devices for ConnectX-1 and -2 */
if (mlx4_priv(dev)->pci_dev_data & MLX4_PCI_DEV_FORCE_SENSE_PORT)
dev->caps.flags |= MLX4_DEV_CAP_FLAG_SENSE_SUPPORT;
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index a092c5c..232b2b5 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -787,6 +787,9 @@ struct mlx4_set_port_general_context {
u8 pprx;
u8 pfcrx;
u16 reserved4;
+ u32 reserved5;
+ u8 phv_en;
+ u8 reserved6[3];
};
struct mlx4_set_port_rqp_calc_context {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index fd13c1c..bcbf8c7 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -211,6 +211,8 @@ enum {
MLX4_DEV_CAP_FLAG2_ETS_CFG = 1LL << 26,
MLX4_DEV_CAP_FLAG2_PORT_BEACON = 1LL << 27,
MLX4_DEV_CAP_FLAG2_IGNORE_FCS = 1LL << 28,
+ MLX4_DEV_CAP_FLAG2_PHV_EN = 1LL << 29,
+ MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN = 1LL << 30,
};
enum {
@@ -581,6 +583,7 @@ struct mlx4_caps {
u64 phys_port_id[MLX4_MAX_PORTS + 1];
int tunnel_offload_mode;
u8 rx_checksum_flags_port[MLX4_MAX_PORTS + 1];
+ u8 phv_bit[MLX4_MAX_PORTS + 1];
u8 alloc_res_qp_mask;
u32 dmfs_high_rate_qpn_base;
u32 dmfs_high_rate_qpn_range;
@@ -1332,6 +1335,8 @@ int mlx4_SET_PORT_BEACON(struct mlx4_dev *dev, u8 port, u16 time);
int mlx4_SET_PORT_fcs_check(struct mlx4_dev *dev, u8 port,
u8 ignore_fcs_value);
int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable);
+int set_phv_bit(struct mlx4_dev *dev, u8 port, int new_val);
+int get_phv_bit(struct mlx4_dev *dev, u8 port, int *phv);
int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx);
int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx);
int mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index);
--
2.4.3.413.ga5fe668
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 2/4] net/mlx4_en: Prepare ethtool private flags to support more flags
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 1/4] net/mlx4_core: Preparations for 802.1ad VLAN support Amir Vadai
@ 2015-07-27 11:46 ` Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 3/4] net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support Amir Vadai
` (2 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Amir Vadai @ 2015-07-27 11:46 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Hadar Har-Zion, Or Gerlitz, Tal Alon, Amir Vadai
From: Hadar Hen Zion <hadarh@mellanox.com>
Currently we support only one ethtool private flag. Prepare
mlx4_en_set_priv_flags function to support more than one private flag.
Will be used in the next patch to support hardware accelerated 802.1ad
vlan.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 35 ++++++++++++-------------
1 file changed, 17 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 99ba1c5..70f6553 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -1801,30 +1801,29 @@ static int mlx4_en_set_priv_flags(struct net_device *dev, u32 flags)
bool bf_enabled_old = !!(priv->pflags & MLX4_EN_PRIV_FLAGS_BLUEFLAME);
int i;
- if (bf_enabled_new == bf_enabled_old)
- return 0; /* Nothing to do */
+ if (bf_enabled_new != bf_enabled_old) {
+ if (bf_enabled_new) {
+ bool bf_supported = true;
- if (bf_enabled_new) {
- bool bf_supported = true;
+ for (i = 0; i < priv->tx_ring_num; i++)
+ bf_supported &= priv->tx_ring[i]->bf_alloced;
- for (i = 0; i < priv->tx_ring_num; i++)
- bf_supported &= priv->tx_ring[i]->bf_alloced;
+ if (!bf_supported) {
+ en_err(priv, "BlueFlame is not supported\n");
+ return -EINVAL;
+ }
- if (!bf_supported) {
- en_err(priv, "BlueFlame is not supported\n");
- return -EINVAL;
+ priv->pflags |= MLX4_EN_PRIV_FLAGS_BLUEFLAME;
+ } else {
+ priv->pflags &= ~MLX4_EN_PRIV_FLAGS_BLUEFLAME;
}
- priv->pflags |= MLX4_EN_PRIV_FLAGS_BLUEFLAME;
- } else {
- priv->pflags &= ~MLX4_EN_PRIV_FLAGS_BLUEFLAME;
- }
-
- for (i = 0; i < priv->tx_ring_num; i++)
- priv->tx_ring[i]->bf_enabled = bf_enabled_new;
+ for (i = 0; i < priv->tx_ring_num; i++)
+ priv->tx_ring[i]->bf_enabled = bf_enabled_new;
- en_info(priv, "BlueFlame %s\n",
- bf_enabled_new ? "Enabled" : "Disabled");
+ en_info(priv, "BlueFlame %s\n",
+ bf_enabled_new ? "Enabled" : "Disabled");
+ }
return 0;
}
--
2.4.3.413.ga5fe668
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 3/4] net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 1/4] net/mlx4_core: Preparations for 802.1ad VLAN support Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 2/4] net/mlx4_en: Prepare ethtool private flags to support more flags Amir Vadai
@ 2015-07-27 11:46 ` Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 4/4] net/mlx4_en: Add support for hardware accelerated 802.1ad vlan Amir Vadai
2015-07-27 22:00 ` [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad David Miller
4 siblings, 0 replies; 8+ messages in thread
From: Amir Vadai @ 2015-07-27 11:46 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Hadar Har-Zion, Or Gerlitz, Tal Alon, Amir Vadai
From: Hadar Hen Zion <hadarh@mellanox.com>
To add Hardware accelerated support in 802.1ad vlan, replace
Current VLAN macros to CVLAN.
Replace:
MLX4_WQE_CTRL_INS_VLAN
MLX4_CQE_VLAN_PRESENT_MASK
With:
MLX4_WQE_CTRL_INS_CVLAN
MLX4_CQE_CVLAN_PRESENT_MASK
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
drivers/infiniband/hw/mlx4/cq.c | 2 +-
drivers/net/ethernet/mellanox/mlx4/en_rx.c | 6 +++---
drivers/net/ethernet/mellanox/mlx4/en_tx.c | 2 +-
include/linux/mlx4/cq.h | 2 +-
include/linux/mlx4/qp.h | 2 +-
5 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 36eb3d0..180a8f7 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -871,7 +871,7 @@ repoll:
if (is_eth) {
wc->sl = be16_to_cpu(cqe->sl_vid) >> 13;
if (be32_to_cpu(cqe->vlan_my_qpn) &
- MLX4_CQE_VLAN_PRESENT_MASK) {
+ MLX4_CQE_CVLAN_PRESENT_MASK) {
wc->vlan_id = be16_to_cpu(cqe->sl_vid) &
MLX4_CQE_VID_MASK;
} else {
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 12c65e1..10f6c2f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -726,7 +726,7 @@ static int check_csum(struct mlx4_cqe *cqe, struct sk_buff *skb, void *va,
hw_checksum = csum_unfold((__force __sum16)cqe->checksum);
- if (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_VLAN_PRESENT_MASK) &&
+ if (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_CVLAN_PRESENT_MASK) &&
!(dev_features & NETIF_F_HW_VLAN_CTAG_RX)) {
hw_checksum = get_fixed_vlan_csum(hw_checksum, hdr);
hdr += sizeof(struct vlan_hdr);
@@ -907,7 +907,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
gro_skb->csum_level = 1;
if ((cqe->vlan_my_qpn &
- cpu_to_be32(MLX4_CQE_VLAN_PRESENT_MASK)) &&
+ cpu_to_be32(MLX4_CQE_CVLAN_PRESENT_MASK)) &&
(dev->features & NETIF_F_HW_VLAN_CTAG_RX)) {
u16 vid = be16_to_cpu(cqe->sl_vid);
@@ -970,7 +970,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
PKT_HASH_TYPE_L3);
if ((be32_to_cpu(cqe->vlan_my_qpn) &
- MLX4_CQE_VLAN_PRESENT_MASK) &&
+ MLX4_CQE_CVLAN_PRESENT_MASK) &&
(dev->features & NETIF_F_HW_VLAN_CTAG_RX))
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), be16_to_cpu(cqe->sl_vid));
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index c10d98f..7c858f6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -958,7 +958,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
ring->bf.offset ^= ring->bf.buf_size;
} else {
tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag);
- tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_VLAN *
+ tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN *
!!skb_vlan_tag_present(skb);
tx_desc->ctrl.fence_size = real_size;
diff --git a/include/linux/mlx4/cq.h b/include/linux/mlx4/cq.h
index e7ecc12..899a97b 100644
--- a/include/linux/mlx4/cq.h
+++ b/include/linux/mlx4/cq.h
@@ -88,7 +88,7 @@ struct mlx4_ts_cqe {
enum {
MLX4_CQE_L2_TUNNEL_IPOK = 1 << 31,
- MLX4_CQE_VLAN_PRESENT_MASK = 1 << 29,
+ MLX4_CQE_CVLAN_PRESENT_MASK = 1 << 29,
MLX4_CQE_L2_TUNNEL = 1 << 27,
MLX4_CQE_L2_TUNNEL_CSUM = 1 << 26,
MLX4_CQE_L2_TUNNEL_IPV4 = 1 << 25,
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 6fed539..6c61900 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -272,7 +272,7 @@ enum {
MLX4_WQE_CTRL_SOLICITED = 1 << 1,
MLX4_WQE_CTRL_IP_CSUM = 1 << 4,
MLX4_WQE_CTRL_TCP_UDP_CSUM = 1 << 5,
- MLX4_WQE_CTRL_INS_VLAN = 1 << 6,
+ MLX4_WQE_CTRL_INS_CVLAN = 1 << 6,
MLX4_WQE_CTRL_STRONG_ORDER = 1 << 7,
MLX4_WQE_CTRL_FORCE_LOOPBACK = 1 << 0,
};
--
2.4.3.413.ga5fe668
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 4/4] net/mlx4_en: Add support for hardware accelerated 802.1ad vlan
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
` (2 preceding siblings ...)
2015-07-27 11:46 ` [PATCH net-next 3/4] net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support Amir Vadai
@ 2015-07-27 11:46 ` Amir Vadai
2015-07-27 22:00 ` [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad David Miller
4 siblings, 0 replies; 8+ messages in thread
From: Amir Vadai @ 2015-07-27 11:46 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Hadar Har-Zion, Or Gerlitz, Tal Alon, Amir Vadai
From: Hadar Hen Zion <hadarh@mellanox.com>
To enable device support in accelerated 802.1ad vlan, the port
capability "packet has vlan enable" (phv_en) should be set.
Firmware won't work properly, in case phv_en is not set.
The user can enable "phv_en" port capability with the new ethtool
private flag phv-bit. The phv-bit private flag default value is OFF,
users who are interested in 802.1ad hardware acceleration should turn ON
the phv-bit private flag:
$ ethtool --set-priv-flags eth1 phv-bit on
Once the private flag is set, the device is ready for 802.1ad vlan
acceleration.
The user should also change the interface device features and turn on
"tx-vlan-stag-hw-insert" which is off by default:
$ ethtool -K eth1 tx-vlan-stag-hw-insert on
"phv-bit" private flag setting is available only for Physical
Functions(PF), the Virtual Function (VF) will be able to use the feature
by setting "tx-vlan-stag-hw-insert" ethtool device feature only if the
feature was enabled by the Hypervisor.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 16 +++++++++
drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 46 +++++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx4/en_rx.c | 16 ++++++++-
drivers/net/ethernet/mellanox/mlx4/en_tx.c | 13 ++++---
drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 +
include/linux/mlx4/cq.h | 1 +
include/linux/mlx4/qp.h | 1 +
7 files changed, 89 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 70f6553..f79d812 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -102,6 +102,7 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo)
static const char mlx4_en_priv_flags[][ETH_GSTRING_LEN] = {
"blueflame",
+ "phv-bit"
};
static const char main_strings[][ETH_GSTRING_LEN] = {
@@ -1797,9 +1798,13 @@ static int mlx4_en_get_ts_info(struct net_device *dev,
static int mlx4_en_set_priv_flags(struct net_device *dev, u32 flags)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
+ struct mlx4_en_dev *mdev = priv->mdev;
bool bf_enabled_new = !!(flags & MLX4_EN_PRIV_FLAGS_BLUEFLAME);
bool bf_enabled_old = !!(priv->pflags & MLX4_EN_PRIV_FLAGS_BLUEFLAME);
+ bool phv_enabled_new = !!(flags & MLX4_EN_PRIV_FLAGS_PHV);
+ bool phv_enabled_old = !!(priv->pflags & MLX4_EN_PRIV_FLAGS_PHV);
int i;
+ int ret = 0;
if (bf_enabled_new != bf_enabled_old) {
if (bf_enabled_new) {
@@ -1825,6 +1830,17 @@ static int mlx4_en_set_priv_flags(struct net_device *dev, u32 flags)
bf_enabled_new ? "Enabled" : "Disabled");
}
+ if (phv_enabled_new != phv_enabled_old) {
+ ret = set_phv_bit(mdev->dev, priv->port, (int)phv_enabled_new);
+ if (ret)
+ return ret;
+ else if (phv_enabled_new)
+ priv->pflags |= MLX4_EN_PRIV_FLAGS_PHV;
+ else
+ priv->pflags &= ~MLX4_EN_PRIV_FLAGS_PHV;
+ en_info(priv, "PHV bit %s\n",
+ phv_enabled_new ? "Enabled" : "Disabled");
+ }
return 0;
}
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index e0de2fd..4726122 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2184,6 +2184,25 @@ static int mlx4_en_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
}
}
+static netdev_features_t mlx4_en_fix_features(struct net_device *netdev,
+ netdev_features_t features)
+{
+ struct mlx4_en_priv *en_priv = netdev_priv(netdev);
+ struct mlx4_en_dev *mdev = en_priv->mdev;
+
+ /* Since there is no support for separate RX C-TAG/S-TAG vlan accel
+ * enable/disable make sure S-TAG flag is always in same state as
+ * C-TAG.
+ */
+ if (features & NETIF_F_HW_VLAN_CTAG_RX &&
+ !(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN))
+ features |= NETIF_F_HW_VLAN_STAG_RX;
+ else
+ features &= ~NETIF_F_HW_VLAN_STAG_RX;
+
+ return features;
+}
+
static int mlx4_en_set_features(struct net_device *netdev,
netdev_features_t features)
{
@@ -2218,6 +2237,10 @@ static int mlx4_en_set_features(struct net_device *netdev,
en_info(priv, "Turn %s TX vlan strip offload\n",
(features & NETIF_F_HW_VLAN_CTAG_TX) ? "ON" : "OFF");
+ if (DEV_FEATURE_CHANGED(netdev, features, NETIF_F_HW_VLAN_STAG_TX))
+ en_info(priv, "Turn %s TX S-VLAN strip offload\n",
+ (features & NETIF_F_HW_VLAN_STAG_TX) ? "ON" : "OFF");
+
if (DEV_FEATURE_CHANGED(netdev, features, NETIF_F_LOOPBACK)) {
en_info(priv, "Turn %s loopback\n",
(features & NETIF_F_LOOPBACK) ? "ON" : "OFF");
@@ -2460,6 +2483,7 @@ static const struct net_device_ops mlx4_netdev_ops = {
.ndo_poll_controller = mlx4_en_netpoll,
#endif
.ndo_set_features = mlx4_en_set_features,
+ .ndo_fix_features = mlx4_en_fix_features,
.ndo_setup_tc = mlx4_en_setup_tc,
#ifdef CONFIG_RFS_ACCEL
.ndo_rx_flow_steer = mlx4_en_filter_rfs,
@@ -2500,6 +2524,7 @@ static const struct net_device_ops mlx4_netdev_ops_master = {
.ndo_poll_controller = mlx4_en_netpoll,
#endif
.ndo_set_features = mlx4_en_set_features,
+ .ndo_fix_features = mlx4_en_fix_features,
.ndo_setup_tc = mlx4_en_setup_tc,
#ifdef CONFIG_RFS_ACCEL
.ndo_rx_flow_steer = mlx4_en_filter_rfs,
@@ -2931,6 +2956,27 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
dev->hw_features |= NETIF_F_LOOPBACK |
NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
+ if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN)) {
+ dev->features |= NETIF_F_HW_VLAN_STAG_RX |
+ NETIF_F_HW_VLAN_STAG_FILTER;
+ dev->hw_features |= NETIF_F_HW_VLAN_STAG_RX;
+ }
+
+ if (mlx4_is_slave(mdev->dev)) {
+ int phv;
+
+ err = get_phv_bit(mdev->dev, port, &phv);
+ if (!err && phv) {
+ dev->hw_features |= NETIF_F_HW_VLAN_STAG_TX;
+ priv->pflags |= MLX4_EN_PRIV_FLAGS_PHV;
+ }
+ } else {
+ if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN &&
+ !(mdev->dev->caps.flags2 &
+ MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN))
+ dev->hw_features |= NETIF_F_HW_VLAN_STAG_TX;
+ }
+
if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP)
dev->hw_features |= NETIF_F_RXFCS;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 10f6c2f..a67fbb9 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -912,6 +912,12 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
u16 vid = be16_to_cpu(cqe->sl_vid);
__vlan_hwaccel_put_tag(gro_skb, htons(ETH_P_8021Q), vid);
+ } else if ((be32_to_cpu(cqe->vlan_my_qpn) &
+ MLX4_CQE_SVLAN_PRESENT_MASK) &&
+ (dev->features & NETIF_F_HW_VLAN_STAG_RX)) {
+ __vlan_hwaccel_put_tag(gro_skb,
+ htons(ETH_P_8021AD),
+ be16_to_cpu(cqe->sl_vid));
}
if (dev->features & NETIF_F_RXHASH)
@@ -973,6 +979,11 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
MLX4_CQE_CVLAN_PRESENT_MASK) &&
(dev->features & NETIF_F_HW_VLAN_CTAG_RX))
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), be16_to_cpu(cqe->sl_vid));
+ else if ((be32_to_cpu(cqe->vlan_my_qpn) &
+ MLX4_CQE_SVLAN_PRESENT_MASK) &&
+ (dev->features & NETIF_F_HW_VLAN_STAG_RX))
+ __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021AD),
+ be16_to_cpu(cqe->sl_vid));
if (ring->hwtstamp_rx_filter == HWTSTAMP_FILTER_ALL) {
timestamp = mlx4_en_get_cqe_ts(cqe);
@@ -1070,7 +1081,10 @@ static const int frag_sizes[] = {
void mlx4_en_calc_rx_buf(struct net_device *dev)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
- int eff_mtu = dev->mtu + ETH_HLEN + VLAN_HLEN;
+ /* VLAN_HLEN is added twice,to support skb vlan tagged with multiple
+ * headers. (For example: ETH_P_8021Q and ETH_P_8021AD).
+ */
+ int eff_mtu = dev->mtu + ETH_HLEN + (2 * VLAN_HLEN);
int buf_size = 0;
int i = 0;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 7c858f6..494e776 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -718,6 +718,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
u32 index, bf_index;
__be32 op_own;
u16 vlan_tag = 0;
+ u16 vlan_proto = 0;
int i_frag;
int lso_header_size;
void *fragptr = NULL;
@@ -750,9 +751,10 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
goto tx_drop;
}
- if (skb_vlan_tag_present(skb))
+ if (skb_vlan_tag_present(skb)) {
vlan_tag = skb_vlan_tag_get(skb);
-
+ vlan_proto = be16_to_cpu(skb->vlan_proto);
+ }
netdev_txq_bql_enqueue_prefetchw(ring->tx_queue);
@@ -958,8 +960,11 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
ring->bf.offset ^= ring->bf.buf_size;
} else {
tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag);
- tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN *
- !!skb_vlan_tag_present(skb);
+ if (vlan_proto == ETH_P_8021AD)
+ tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN;
+ else if (vlan_proto == ETH_P_8021Q)
+ tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN;
+
tx_desc->ctrl.fence_size = real_size;
/* Ensure new descriptor hits memory
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 666d166..defcf8c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -95,6 +95,7 @@
*/
#define MLX4_EN_PRIV_FLAGS_BLUEFLAME 1
+#define MLX4_EN_PRIV_FLAGS_PHV 2
#define MLX4_EN_WATCHDOG_TIMEOUT (15 * HZ)
diff --git a/include/linux/mlx4/cq.h b/include/linux/mlx4/cq.h
index 899a97b..09cebe5 100644
--- a/include/linux/mlx4/cq.h
+++ b/include/linux/mlx4/cq.h
@@ -89,6 +89,7 @@ struct mlx4_ts_cqe {
enum {
MLX4_CQE_L2_TUNNEL_IPOK = 1 << 31,
MLX4_CQE_CVLAN_PRESENT_MASK = 1 << 29,
+ MLX4_CQE_SVLAN_PRESENT_MASK = 1 << 30,
MLX4_CQE_L2_TUNNEL = 1 << 27,
MLX4_CQE_L2_TUNNEL_CSUM = 1 << 26,
MLX4_CQE_L2_TUNNEL_IPV4 = 1 << 25,
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 6c61900..de45a51 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -273,6 +273,7 @@ enum {
MLX4_WQE_CTRL_IP_CSUM = 1 << 4,
MLX4_WQE_CTRL_TCP_UDP_CSUM = 1 << 5,
MLX4_WQE_CTRL_INS_CVLAN = 1 << 6,
+ MLX4_WQE_CTRL_INS_SVLAN = 1 << 7,
MLX4_WQE_CTRL_STRONG_ORDER = 1 << 7,
MLX4_WQE_CTRL_FORCE_LOOPBACK = 1 << 0,
};
--
2.4.3.413.ga5fe668
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
` (3 preceding siblings ...)
2015-07-27 11:46 ` [PATCH net-next 4/4] net/mlx4_en: Add support for hardware accelerated 802.1ad vlan Amir Vadai
@ 2015-07-27 22:00 ` David Miller
2015-07-28 7:51 ` Or Gerlitz
4 siblings, 1 reply; 8+ messages in thread
From: David Miller @ 2015-07-27 22:00 UTC (permalink / raw)
To: amirv; +Cc: netdev, hadarh, ogerlitz, talal
From: Amir Vadai <amirv@mellanox.com>
Date: Mon, 27 Jul 2015 14:46:30 +0300
> This patchset by Hadar introduces support in Hardware accelerated 802.1ad, for
> ConnectX-3pro NIC's. In order to support existing deployment, and due to some
> hardware limitations, the feature is disabled by default, and needed to be
> enabled using a private flag in ethtool. Ofcourse user can enable the private
> flag only if hardware has support.
> After being enabled, the standard ethtool -k/-K can be used.
>
> Patchset was applied and tested over commit 71790a2 ("hv_netvsc: Add structs
> and handlers for VF messages")
Series applied, thanks.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad
2015-07-27 22:00 ` [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad David Miller
@ 2015-07-28 7:51 ` Or Gerlitz
2015-07-28 8:05 ` David Miller
0 siblings, 1 reply; 8+ messages in thread
From: Or Gerlitz @ 2015-07-28 7:51 UTC (permalink / raw)
To: David Miller; +Cc: amirv, netdev, hadarh, talal
On 7/28/2015 1:00 AM, David Miller wrote:
> Series applied, thanks.
Hi Dave,
I don't see this on your kernel.org clone.. maybe forgot to press on the
push button?
Or.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad
2015-07-28 7:51 ` Or Gerlitz
@ 2015-07-28 8:05 ` David Miller
0 siblings, 0 replies; 8+ messages in thread
From: David Miller @ 2015-07-28 8:05 UTC (permalink / raw)
To: ogerlitz; +Cc: amirv, netdev, hadarh, talal
From: Or Gerlitz <ogerlitz@mellanox.com>
Date: Tue, 28 Jul 2015 10:51:08 +0300
> On 7/28/2015 1:00 AM, David Miller wrote:
>> Series applied, thanks.
>
> Hi Dave,
>
> I don't see this on your kernel.org clone.. maybe forgot to press on
> the push button?
It should be there now.
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2015-07-28 8:05 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-07-27 11:46 [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 1/4] net/mlx4_core: Preparations for 802.1ad VLAN support Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 2/4] net/mlx4_en: Prepare ethtool private flags to support more flags Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 3/4] net/mlx4: Prepare VLAN macros for 802.1ad Hardware accelerated support Amir Vadai
2015-07-27 11:46 ` [PATCH net-next 4/4] net/mlx4_en: Add support for hardware accelerated 802.1ad vlan Amir Vadai
2015-07-27 22:00 ` [PATCH net-next 0/4] net/mlx4_en: Hardware accelerated 802.1ad David Miller
2015-07-28 7:51 ` Or Gerlitz
2015-07-28 8:05 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).