[pull request][net 0/9] mlx5 fixes 2021-03-31

[pull request][net 0/9] mlx5 fixes 2021-03-31

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Hi Dave, Jakub,

This series introduces some fixes to mlx5 driver.
Please pull and let me know if there is any problem.

Thanks,
Saeed.

---
The following changes since commit 61431a5907fc36d0738e9a547c7e1556349a03e9:

  net: ensure mac header is set in virtio_net_hdr_to_skb() (2021-03-30 17:40:46 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-fixes-2021-03-31

for you to fetch changes up to 3ff3874fa0b261ef74f2bfb008a82ab1601c11eb:

  net/mlx5e: Guarantee room for XSK wakeup NOP on async ICOSQ (2021-03-31 13:12:24 -0700)

----------------------------------------------------------------
mlx5-fixes-2021-03-31

----------------------------------------------------------------
Ariel Levkovich (1):
      net/mlx5e: Fix mapping of ct_label zero

Aya Levin (1):
      net/mlx5e: Fix ethtool indication of connector type

Daniel Jurgens (1):
      net/mlx5: Don't request more than supported EQs

Dima Chumak (1):
      net/mlx5e: Consider geneve_opts for encap contexts

Maor Dickman (2):
      net/mlx5: Delete auxiliary bus driver eth-rep first
      net/mlx5: E-switch, Create vport miss group only if src rewrite is supported

Tariq Toukan (3):
      net/mlx5e: kTLS, Fix TX counters atomicity
      net/mlx5e: kTLS, Fix RX counters atomicity
      net/mlx5e: Guarantee room for XSK wakeup NOP on async ICOSQ

 drivers/net/ethernet/mellanox/mlx5/core/dev.c      |  4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 36 +++++++++---
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun.h    | 10 ++++
 .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c  | 23 +++-----
 .../ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c | 29 +++++++++
 .../ethernet/mellanox/mlx5/core/en/tc_tun_gre.c    |  1 +
 .../mellanox/mlx5/core/en/tc_tun_mplsoudp.c        |  1 +
 .../ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c  |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  |  6 ++
 .../ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c | 40 ++++++-------
 .../ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c |  5 +-
 .../net/ethernet/mellanox/mlx5/core/en_accel/tls.h |  3 +
 .../mellanox/mlx5/core/en_accel/tls_stats.c        | 49 ++++++++++------
 .../net/ethernet/mellanox/mlx5/core/en_ethtool.c   | 22 +++----
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 21 ++++++-
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 10 ----
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h |  6 --
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       | 13 ++++-
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 68 +++++++++++++---------
 20 files changed, 227 insertions(+), 122 deletions(-)

[net 1/9] net/mlx5e: Fix mapping of ct_label zero

[Saeed Mahameed]

From: Ariel Levkovich <lariel@nvidia.com>

ct_label 0 is a default label each flow has and therefore
there can be rules that match on ct_label=0 without a prior
rule that set the ct_label to this value.

The ct_label value is not used directly in the HW rules and
instead it is mapped to some id within a defined range and this
id is used to set and match the metadata register which carries
the ct_label.

If we have a rule that matches on ct_label=0, the hw rule will
perform matching on a value that is != 0 because of the mapping
from label to id. Since the metadata register default value is
0 and it was never set before to anything else by an action that
sets the ct_label, there will always be a mismatch between that
register and the value in the rule.

To support such rule, a forced mapping of ct_label 0 to id=0
is done so that it will match the metadata register default
value of 0.

Fixes: 54b154ecfb8c ("net/mlx5e: CT: Map 128 bits labels to 32 bit map ID")
Signed-off-by: Ariel Levkovich <lariel@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/tc_ct.c    | 36 +++++++++++++++----
 1 file changed, 29 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index b2cd29847a37..68e54cc1cd16 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -185,6 +185,28 @@ mlx5_tc_ct_entry_has_nat(struct mlx5_ct_entry *entry)
 	return !!(entry->tuple_nat_node.next);
 }
 
+static int
+mlx5_get_label_mapping(struct mlx5_tc_ct_priv *ct_priv,
+		       u32 *labels, u32 *id)
+{
+	if (!memchr_inv(labels, 0, sizeof(u32) * 4)) {
+		*id = 0;
+		return 0;
+	}
+
+	if (mapping_add(ct_priv->labels_mapping, labels, id))
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static void
+mlx5_put_label_mapping(struct mlx5_tc_ct_priv *ct_priv, u32 id)
+{
+	if (id)
+		mapping_remove(ct_priv->labels_mapping, id);
+}
+
 static int
 mlx5_tc_ct_rule_to_tuple(struct mlx5_ct_tuple *tuple, struct flow_rule *rule)
 {
@@ -436,7 +458,7 @@ mlx5_tc_ct_entry_del_rule(struct mlx5_tc_ct_priv *ct_priv,
 	mlx5_tc_rule_delete(netdev_priv(ct_priv->netdev), zone_rule->rule, attr);
 	mlx5e_mod_hdr_detach(ct_priv->dev,
 			     ct_priv->mod_hdr_tbl, zone_rule->mh);
-	mapping_remove(ct_priv->labels_mapping, attr->ct_attr.ct_labels_id);
+	mlx5_put_label_mapping(ct_priv, attr->ct_attr.ct_labels_id);
 	kfree(attr);
 }
 
@@ -639,8 +661,8 @@ mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv,
 	if (!meta)
 		return -EOPNOTSUPP;
 
-	err = mapping_add(ct_priv->labels_mapping, meta->ct_metadata.labels,
-			  &attr->ct_attr.ct_labels_id);
+	err = mlx5_get_label_mapping(ct_priv, meta->ct_metadata.labels,
+				     &attr->ct_attr.ct_labels_id);
 	if (err)
 		return -EOPNOTSUPP;
 	if (nat) {
@@ -677,7 +699,7 @@ mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv,
 
 err_mapping:
 	dealloc_mod_hdr_actions(&mod_acts);
-	mapping_remove(ct_priv->labels_mapping, attr->ct_attr.ct_labels_id);
+	mlx5_put_label_mapping(ct_priv, attr->ct_attr.ct_labels_id);
 	return err;
 }
 
@@ -745,7 +767,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 err_rule:
 	mlx5e_mod_hdr_detach(ct_priv->dev,
 			     ct_priv->mod_hdr_tbl, zone_rule->mh);
-	mapping_remove(ct_priv->labels_mapping, attr->ct_attr.ct_labels_id);
+	mlx5_put_label_mapping(ct_priv, attr->ct_attr.ct_labels_id);
 err_mod_hdr:
 	kfree(attr);
 err_attr:
@@ -1197,7 +1219,7 @@ void mlx5_tc_ct_match_del(struct mlx5_tc_ct_priv *priv, struct mlx5_ct_attr *ct_
 	if (!priv || !ct_attr->ct_labels_id)
 		return;
 
-	mapping_remove(priv->labels_mapping, ct_attr->ct_labels_id);
+	mlx5_put_label_mapping(priv, ct_attr->ct_labels_id);
 }
 
 int
@@ -1280,7 +1302,7 @@ mlx5_tc_ct_match_add(struct mlx5_tc_ct_priv *priv,
 		ct_labels[1] = key->ct_labels[1] & mask->ct_labels[1];
 		ct_labels[2] = key->ct_labels[2] & mask->ct_labels[2];
 		ct_labels[3] = key->ct_labels[3] & mask->ct_labels[3];
-		if (mapping_add(priv->labels_mapping, ct_labels, &ct_attr->ct_labels_id))
+		if (mlx5_get_label_mapping(priv, ct_labels, &ct_attr->ct_labels_id))
 			return -EOPNOTSUPP;
 		mlx5e_tc_match_to_reg_match(spec, LABELS_TO_REG, ct_attr->ct_labels_id,
 					    MLX5_CT_LABELS_MASK);
-- 
2.30.2

[net 2/9] net/mlx5: Delete auxiliary bus driver eth-rep first

[Saeed Mahameed]

From: Maor Dickman <maord@nvidia.com>

Delete auxiliary bus drivers flow deletes the eth driver
first and then the eth-reps driver but eth-reps devices resources
are depend on eth device.

Fixed by changing the delete order of auxiliary bus drivers to delete
the eth-rep driver first and after it the eth driver.

Fixes: 601c10c89cbb ("net/mlx5: Delete custom device management logic")
Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/dev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
index b051417ede67..9153c9bda96f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
@@ -191,12 +191,12 @@ static bool is_ib_supported(struct mlx5_core_dev *dev)
 }
 
 enum {
-	MLX5_INTERFACE_PROTOCOL_ETH_REP,
 	MLX5_INTERFACE_PROTOCOL_ETH,
+	MLX5_INTERFACE_PROTOCOL_ETH_REP,
 
+	MLX5_INTERFACE_PROTOCOL_IB,
 	MLX5_INTERFACE_PROTOCOL_IB_REP,
 	MLX5_INTERFACE_PROTOCOL_MPIB,
-	MLX5_INTERFACE_PROTOCOL_IB,
 
 	MLX5_INTERFACE_PROTOCOL_VNET,
 };
-- 
2.30.2

[net 3/9] net/mlx5e: Fix ethtool indication of connector type

[Saeed Mahameed]

From: Aya Levin <ayal@nvidia.com>

Use connector_type read from PTYS register when it's valid, based on
corresponding capability bit.

Fixes: 5b4793f81745 ("net/mlx5e: Add support for reading connector type from PTYS")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en_ethtool.c  | 22 +++++++++----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index f5f2a8fd0046..53802e18af90 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -758,11 +758,11 @@ static int get_fec_supported_advertised(struct mlx5_core_dev *dev,
 	return 0;
 }
 
-static void ptys2ethtool_supported_advertised_port(struct ethtool_link_ksettings *link_ksettings,
-						   u32 eth_proto_cap,
-						   u8 connector_type, bool ext)
+static void ptys2ethtool_supported_advertised_port(struct mlx5_core_dev *mdev,
+						   struct ethtool_link_ksettings *link_ksettings,
+						   u32 eth_proto_cap, u8 connector_type)
 {
-	if ((!connector_type && !ext) || connector_type >= MLX5E_CONNECTOR_TYPE_NUMBER) {
+	if (!MLX5_CAP_PCAM_FEATURE(mdev, ptys_connector_type)) {
 		if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_10GBASE_CR)
 				   | MLX5E_PROT_MASK(MLX5E_10GBASE_SR)
 				   | MLX5E_PROT_MASK(MLX5E_40GBASE_CR4)
@@ -898,9 +898,9 @@ static int ptys2connector_type[MLX5E_CONNECTOR_TYPE_NUMBER] = {
 		[MLX5E_PORT_OTHER]              = PORT_OTHER,
 	};
 
-static u8 get_connector_port(u32 eth_proto, u8 connector_type, bool ext)
+static u8 get_connector_port(struct mlx5_core_dev *mdev, u32 eth_proto, u8 connector_type)
 {
-	if ((connector_type || ext) && connector_type < MLX5E_CONNECTOR_TYPE_NUMBER)
+	if (MLX5_CAP_PCAM_FEATURE(mdev, ptys_connector_type))
 		return ptys2connector_type[connector_type];
 
 	if (eth_proto &
@@ -1001,11 +1001,11 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv,
 			 data_rate_oper, link_ksettings);
 
 	eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap;
-
-	link_ksettings->base.port = get_connector_port(eth_proto_oper,
-						       connector_type, ext);
-	ptys2ethtool_supported_advertised_port(link_ksettings, eth_proto_admin,
-					       connector_type, ext);
+	connector_type = connector_type < MLX5E_CONNECTOR_TYPE_NUMBER ?
+			 connector_type : MLX5E_PORT_UNKNOWN;
+	link_ksettings->base.port = get_connector_port(mdev, eth_proto_oper, connector_type);
+	ptys2ethtool_supported_advertised_port(mdev, link_ksettings, eth_proto_admin,
+					       connector_type);
 	get_lp_advertising(mdev, eth_proto_lp, link_ksettings);
 
 	if (an_status == MLX5_AN_COMPLETE)
-- 
2.30.2

[net 4/9] net/mlx5: E-switch, Create vport miss group only if src rewrite is supported

[Saeed Mahameed]

From: Maor Dickman <maord@nvidia.com>

Create send to vport miss group was added in order to support traffic
recirculation to root table with metadata source rewrite.
This group is created also in case source rewrite isn't supported.

Fixed by creating send to vport miss group only if source rewrite is
supported by FW.

Fixes: 8e404fefa58b ("net/mlx5e: Match recirculated packet miss in slow table using reg_c1")
Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/eswitch_offloads.c     | 68 +++++++++++--------
 1 file changed, 39 insertions(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 8694b83968b4..d4a2f8d1ee9f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -537,6 +537,14 @@ esw_setup_vport_dests(struct mlx5_flow_destination *dest, struct mlx5_flow_act *
 	return i;
 }
 
+static bool
+esw_src_port_rewrite_supported(struct mlx5_eswitch *esw)
+{
+	return MLX5_CAP_GEN(esw->dev, reg_c_preserve) &&
+	       mlx5_eswitch_vport_match_metadata_enabled(esw) &&
+	       MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level);
+}
+
 static int
 esw_setup_dests(struct mlx5_flow_destination *dest,
 		struct mlx5_flow_act *flow_act,
@@ -550,9 +558,7 @@ esw_setup_dests(struct mlx5_flow_destination *dest,
 	int err = 0;
 
 	if (!mlx5_eswitch_termtbl_required(esw, attr, flow_act, spec) &&
-	    MLX5_CAP_GEN(esw_attr->in_mdev, reg_c_preserve) &&
-	    mlx5_eswitch_vport_match_metadata_enabled(esw) &&
-	    MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level))
+	    esw_src_port_rewrite_supported(esw))
 		attr->flags |= MLX5_ESW_ATTR_FLAG_SRC_REWRITE;
 
 	if (attr->dest_ft) {
@@ -1716,36 +1722,40 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw)
 	}
 	esw->fdb_table.offloads.send_to_vport_grp = g;
 
-	/* meta send to vport */
-	memset(flow_group_in, 0, inlen);
-	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
-		 MLX5_MATCH_MISC_PARAMETERS_2);
-
-	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
+	if (esw_src_port_rewrite_supported(esw)) {
+		/* meta send to vport */
+		memset(flow_group_in, 0, inlen);
+		MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
+			 MLX5_MATCH_MISC_PARAMETERS_2);
 
-	MLX5_SET(fte_match_param, match_criteria,
-		 misc_parameters_2.metadata_reg_c_0, mlx5_eswitch_get_vport_metadata_mask());
-	MLX5_SET(fte_match_param, match_criteria,
-		 misc_parameters_2.metadata_reg_c_1, ESW_TUN_MASK);
+		match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
 
-	num_vfs = esw->esw_funcs.num_vfs;
-	if (num_vfs) {
-		MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix);
-		MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + num_vfs - 1);
-		ix += num_vfs;
+		MLX5_SET(fte_match_param, match_criteria,
+			 misc_parameters_2.metadata_reg_c_0,
+			 mlx5_eswitch_get_vport_metadata_mask());
+		MLX5_SET(fte_match_param, match_criteria,
+			 misc_parameters_2.metadata_reg_c_1, ESW_TUN_MASK);
 
-		g = mlx5_create_flow_group(fdb, flow_group_in);
-		if (IS_ERR(g)) {
-			err = PTR_ERR(g);
-			esw_warn(dev, "Failed to create send-to-vport meta flow group err(%d)\n",
-				 err);
-			goto send_vport_meta_err;
+		num_vfs = esw->esw_funcs.num_vfs;
+		if (num_vfs) {
+			MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix);
+			MLX5_SET(create_flow_group_in, flow_group_in,
+				 end_flow_index, ix + num_vfs - 1);
+			ix += num_vfs;
+
+			g = mlx5_create_flow_group(fdb, flow_group_in);
+			if (IS_ERR(g)) {
+				err = PTR_ERR(g);
+				esw_warn(dev, "Failed to create send-to-vport meta flow group err(%d)\n",
+					 err);
+				goto send_vport_meta_err;
+			}
+			esw->fdb_table.offloads.send_to_vport_meta_grp = g;
+
+			err = mlx5_eswitch_add_send_to_vport_meta_rules(esw);
+			if (err)
+				goto meta_rule_err;
 		}
-		esw->fdb_table.offloads.send_to_vport_meta_grp = g;
-
-		err = mlx5_eswitch_add_send_to_vport_meta_rules(esw);
-		if (err)
-			goto meta_rule_err;
 	}
 
 	if (MLX5_CAP_ESW(esw->dev, merged_eswitch)) {
-- 
2.30.2

[net 5/9] net/mlx5e: kTLS, Fix TX counters atomicity

[Saeed Mahameed]

From: Tariq Toukan <tariqt@nvidia.com>

Some TLS TX counters increment per socket/connection, and are not
protected against parallel modifications from several cores.
Switch them to atomic counters by taking them out of the SQ stats into
the global atomic TLS stats.

In this patch, we touch a single counter 'tx_tls_ctx' that counts the
number of device-offloaded TX TLS connections added.
Now that this counter can be increased without the for having the SQ
context in hand, move it to the mlx5e_ktls_add_tx() callback where it
really belongs, out of the fast data-path.

This change is not needed for counters that increment only in NAPI
context or under the TX lock, as they are already protected.
Keep them as tls_* counters under 'struct mlx5e_sq_stats'.

Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/en_accel/ktls_tx.c     |  5 +-
 .../mellanox/mlx5/core/en_accel/tls.h         |  1 +
 .../mellanox/mlx5/core/en_accel/tls_stats.c   | 47 +++++++++++--------
 .../ethernet/mellanox/mlx5/core/en_stats.c    |  4 --
 .../ethernet/mellanox/mlx5/core/en_stats.h    |  2 -
 5 files changed, 33 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
index d16def68ecff..51bdf71073f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
 // Copyright (c) 2019 Mellanox Technologies.
 
+#include "en_accel/tls.h"
 #include "en_accel/ktls_txrx.h"
 #include "en_accel/ktls_utils.h"
 
@@ -50,6 +51,7 @@ static int mlx5e_ktls_create_tis(struct mlx5_core_dev *mdev, u32 *tisn)
 struct mlx5e_ktls_offload_context_tx {
 	struct tls_offload_context_tx *tx_ctx;
 	struct tls12_crypto_info_aes_gcm_128 crypto_info;
+	struct mlx5e_tls_sw_stats *sw_stats;
 	u32 expected_seq;
 	u32 tisn;
 	u32 key_id;
@@ -99,6 +101,7 @@ int mlx5e_ktls_add_tx(struct net_device *netdev, struct sock *sk,
 	if (err)
 		goto err_create_key;
 
+	priv_tx->sw_stats = &priv->tls->sw_stats;
 	priv_tx->expected_seq = start_offload_tcp_sn;
 	priv_tx->crypto_info  =
 		*(struct tls12_crypto_info_aes_gcm_128 *)crypto_info;
@@ -111,6 +114,7 @@ int mlx5e_ktls_add_tx(struct net_device *netdev, struct sock *sk,
 		goto err_create_tis;
 
 	priv_tx->ctx_post_pending = true;
+	atomic64_inc(&priv_tx->sw_stats->tx_tls_ctx);
 
 	return 0;
 
@@ -452,7 +456,6 @@ bool mlx5e_ktls_handle_tx_skb(struct tls_context *tls_ctx, struct mlx5e_txqsq *s
 
 	if (unlikely(mlx5e_ktls_tx_offload_test_and_clear_pending(priv_tx))) {
 		mlx5e_ktls_tx_post_param_wqes(sq, priv_tx, false, false);
-		stats->tls_ctx++;
 	}
 
 	seq = ntohl(tcp_hdr(skb)->seq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
index bd270a85c804..5b408904df14 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
@@ -41,6 +41,7 @@
 #include "en.h"
 
 struct mlx5e_tls_sw_stats {
+	atomic64_t tx_tls_ctx;
 	atomic64_t tx_tls_drop_metadata;
 	atomic64_t tx_tls_drop_resync_alloc;
 	atomic64_t tx_tls_drop_no_sync_data;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
index b949b9a7538b..a5aabc5c5236 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
@@ -45,49 +45,58 @@ static const struct counter_desc mlx5e_tls_sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_tls_sw_stats, tx_tls_drop_bypass_required) },
 };
 
+static const struct counter_desc mlx5e_ktls_sw_stats_desc[] = {
+	{ MLX5E_DECLARE_STAT(struct mlx5e_tls_sw_stats, tx_tls_ctx) },
+};
+
 #define MLX5E_READ_CTR_ATOMIC64(ptr, dsc, i) \
 	atomic64_read((atomic64_t *)((char *)(ptr) + (dsc)[i].offset))
 
-#define NUM_TLS_SW_COUNTERS ARRAY_SIZE(mlx5e_tls_sw_stats_desc)
-
-static bool is_tls_atomic_stats(struct mlx5e_priv *priv)
+static const struct counter_desc *get_tls_atomic_stats(struct mlx5e_priv *priv)
 {
-	return priv->tls && !mlx5_accel_is_ktls_device(priv->mdev);
+	if (!priv->tls)
+		return NULL;
+	if (mlx5_accel_is_ktls_device(priv->mdev))
+		return mlx5e_ktls_sw_stats_desc;
+	return mlx5e_tls_sw_stats_desc;
 }
 
 int mlx5e_tls_get_count(struct mlx5e_priv *priv)
 {
-	if (!is_tls_atomic_stats(priv))
+	if (!priv->tls)
 		return 0;
-
-	return NUM_TLS_SW_COUNTERS;
+	if (mlx5_accel_is_ktls_device(priv->mdev))
+		return ARRAY_SIZE(mlx5e_ktls_sw_stats_desc);
+	return ARRAY_SIZE(mlx5e_tls_sw_stats_desc);
 }
 
 int mlx5e_tls_get_strings(struct mlx5e_priv *priv, uint8_t *data)
 {
-	unsigned int i, idx = 0;
+	const struct counter_desc *stats_desc;
+	unsigned int i, n, idx = 0;
 
-	if (!is_tls_atomic_stats(priv))
-		return 0;
+	stats_desc = get_tls_atomic_stats(priv);
+	n = mlx5e_tls_get_count(priv);
 
-	for (i = 0; i < NUM_TLS_SW_COUNTERS; i++)
+	for (i = 0; i < n; i++)
 		strcpy(data + (idx++) * ETH_GSTRING_LEN,
-		       mlx5e_tls_sw_stats_desc[i].format);
+		       stats_desc[i].format);
 
-	return NUM_TLS_SW_COUNTERS;
+	return n;
 }
 
 int mlx5e_tls_get_stats(struct mlx5e_priv *priv, u64 *data)
 {
-	int i, idx = 0;
+	const struct counter_desc *stats_desc;
+	unsigned int i, n, idx = 0;
 
-	if (!is_tls_atomic_stats(priv))
-		return 0;
+	stats_desc = get_tls_atomic_stats(priv);
+	n = mlx5e_tls_get_count(priv);
 
-	for (i = 0; i < NUM_TLS_SW_COUNTERS; i++)
+	for (i = 0; i < n; i++)
 		data[idx++] =
 		    MLX5E_READ_CTR_ATOMIC64(&priv->tls->sw_stats,
-					    mlx5e_tls_sw_stats_desc, i);
+					    stats_desc, i);
 
-	return NUM_TLS_SW_COUNTERS;
+	return n;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 92c5b81427b9..74adaa58189a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -116,7 +116,6 @@ static const struct counter_desc sw_stats_desc[] = {
 #ifdef CONFIG_MLX5_EN_TLS
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_encrypted_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_encrypted_bytes) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_ctx) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_ooo) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_dump_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_dump_bytes) },
@@ -390,7 +389,6 @@ static void mlx5e_stats_grp_sw_update_stats_sq(struct mlx5e_sw_stats *s,
 #ifdef CONFIG_MLX5_EN_TLS
 	s->tx_tls_encrypted_packets += sq_stats->tls_encrypted_packets;
 	s->tx_tls_encrypted_bytes   += sq_stats->tls_encrypted_bytes;
-	s->tx_tls_ctx               += sq_stats->tls_ctx;
 	s->tx_tls_ooo               += sq_stats->tls_ooo;
 	s->tx_tls_dump_bytes        += sq_stats->tls_dump_bytes;
 	s->tx_tls_dump_packets      += sq_stats->tls_dump_packets;
@@ -1650,7 +1648,6 @@ static const struct counter_desc sq_stats_desc[] = {
 #ifdef CONFIG_MLX5_EN_TLS
 	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_packets) },
 	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_bytes) },
-	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_ctx) },
 	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_ooo) },
 	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_dump_packets) },
 	{ MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_dump_bytes) },
@@ -1776,7 +1773,6 @@ static const struct counter_desc qos_sq_stats_desc[] = {
 #ifdef CONFIG_MLX5_EN_TLS
 	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_packets) },
 	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_bytes) },
-	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_ctx) },
 	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_ooo) },
 	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_dump_packets) },
 	{ MLX5E_DECLARE_QOS_TX_STAT(struct mlx5e_sq_stats, tls_dump_bytes) },
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 93c41312fb03..8eb056af79ba 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -191,7 +191,6 @@ struct mlx5e_sw_stats {
 #ifdef CONFIG_MLX5_EN_TLS
 	u64 tx_tls_encrypted_packets;
 	u64 tx_tls_encrypted_bytes;
-	u64 tx_tls_ctx;
 	u64 tx_tls_ooo;
 	u64 tx_tls_dump_packets;
 	u64 tx_tls_dump_bytes;
@@ -364,7 +363,6 @@ struct mlx5e_sq_stats {
 #ifdef CONFIG_MLX5_EN_TLS
 	u64 tls_encrypted_packets;
 	u64 tls_encrypted_bytes;
-	u64 tls_ctx;
 	u64 tls_ooo;
 	u64 tls_dump_packets;
 	u64 tls_dump_bytes;
-- 
2.30.2

[net 6/9] net/mlx5e: kTLS, Fix RX counters atomicity

[Saeed Mahameed]

From: Tariq Toukan <tariqt@nvidia.com>

Some TLS RX counters increment per socket/connection, and are not
protected against parallel modifications from several cores.
Switch them to atomic counters by taking them out of the RQ stats into
the global atomic TLS stats.

In this patch, we touch 'rx_tls_ctx/del' that count the number of
device-offloaded RX TLS connections added/deleted.
These counters are updated in the add/del callbacks, out of the fast
data-path.

This change is not needed for counters that increment only in NAPI
context, as they are protected by the NAPI mechanism.
Keep them as tls_* counters under 'struct mlx5e_rq_stats'.

Fixes: 76c1e1ac2aae ("net/mlx5e: kTLS, Add kTLS RX stats")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/en_accel/ktls_rx.c     | 22 ++++++++++---------
 .../mellanox/mlx5/core/en_accel/tls.h         |  2 ++
 .../mellanox/mlx5/core/en_accel/tls_stats.c   |  2 ++
 .../ethernet/mellanox/mlx5/core/en_stats.c    |  6 -----
 .../ethernet/mellanox/mlx5/core/en_stats.h    |  4 ----
 5 files changed, 16 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
index d06532d0baa4..57c5ebd597a7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
@@ -46,7 +46,8 @@ struct mlx5e_ktls_offload_context_rx {
 	struct tls12_crypto_info_aes_gcm_128 crypto_info;
 	struct accel_rule rule;
 	struct sock *sk;
-	struct mlx5e_rq_stats *stats;
+	struct mlx5e_rq_stats *rq_stats;
+	struct mlx5e_tls_sw_stats *sw_stats;
 	struct completion add_ctx;
 	u32 tirn;
 	u32 key_id;
@@ -218,7 +219,7 @@ static int post_rx_param_wqes(struct mlx5e_channel *c,
 	return err;
 
 err_out:
-	priv_rx->stats->tls_resync_req_skip++;
+	priv_rx->rq_stats->tls_resync_req_skip++;
 	err = PTR_ERR(cseg);
 	complete(&priv_rx->add_ctx);
 	goto unlock;
@@ -322,7 +323,7 @@ resync_post_get_progress_params(struct mlx5e_icosq *sq,
 err_free:
 	kfree(buf);
 err_out:
-	priv_rx->stats->tls_resync_req_skip++;
+	priv_rx->rq_stats->tls_resync_req_skip++;
 	return err;
 }
 
@@ -378,13 +379,13 @@ static int resync_handle_seq_match(struct mlx5e_ktls_offload_context_rx *priv_rx
 
 	cseg = post_static_params(sq, priv_rx);
 	if (IS_ERR(cseg)) {
-		priv_rx->stats->tls_resync_res_skip++;
+		priv_rx->rq_stats->tls_resync_res_skip++;
 		err = PTR_ERR(cseg);
 		goto unlock;
 	}
 	/* Do not increment priv_rx refcnt, CQE handling is empty */
 	mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, cseg);
-	priv_rx->stats->tls_resync_res_ok++;
+	priv_rx->rq_stats->tls_resync_res_ok++;
 unlock:
 	spin_unlock_bh(&c->async_icosq_lock);
 
@@ -420,13 +421,13 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi,
 	auth_state = MLX5_GET(tls_progress_params, ctx, auth_state);
 	if (tracker_state != MLX5E_TLS_PROGRESS_PARAMS_RECORD_TRACKER_STATE_TRACKING ||
 	    auth_state != MLX5E_TLS_PROGRESS_PARAMS_AUTH_STATE_NO_OFFLOAD) {
-		priv_rx->stats->tls_resync_req_skip++;
+		priv_rx->rq_stats->tls_resync_req_skip++;
 		goto out;
 	}
 
 	hw_seq = MLX5_GET(tls_progress_params, ctx, hw_resync_tcp_sn);
 	tls_offload_rx_resync_async_request_end(priv_rx->sk, cpu_to_be32(hw_seq));
-	priv_rx->stats->tls_resync_req_end++;
+	priv_rx->rq_stats->tls_resync_req_end++;
 out:
 	mlx5e_ktls_priv_rx_put(priv_rx);
 	dma_unmap_single(dev, buf->dma_addr, PROGRESS_PARAMS_PADDED_SIZE, DMA_FROM_DEVICE);
@@ -609,7 +610,8 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk,
 	priv_rx->rxq = rxq;
 	priv_rx->sk = sk;
 
-	priv_rx->stats = &priv->channel_stats[rxq].rq;
+	priv_rx->rq_stats = &priv->channel_stats[rxq].rq;
+	priv_rx->sw_stats = &priv->tls->sw_stats;
 	mlx5e_set_ktls_rx_priv_ctx(tls_ctx, priv_rx);
 
 	rqtn = priv->direct_tir[rxq].rqt.rqtn;
@@ -630,7 +632,7 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk,
 	if (err)
 		goto err_post_wqes;
 
-	priv_rx->stats->tls_ctx++;
+	atomic64_inc(&priv_rx->sw_stats->rx_tls_ctx);
 
 	return 0;
 
@@ -666,7 +668,7 @@ void mlx5e_ktls_del_rx(struct net_device *netdev, struct tls_context *tls_ctx)
 	if (cancel_work_sync(&resync->work))
 		mlx5e_ktls_priv_rx_put(priv_rx);
 
-	priv_rx->stats->tls_del++;
+	atomic64_inc(&priv_rx->sw_stats->rx_tls_del);
 	if (priv_rx->rule.rule)
 		mlx5e_accel_fs_del_sk(priv_rx->rule.rule);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
index 5b408904df14..4c9274d390da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h
@@ -46,6 +46,8 @@ struct mlx5e_tls_sw_stats {
 	atomic64_t tx_tls_drop_resync_alloc;
 	atomic64_t tx_tls_drop_no_sync_data;
 	atomic64_t tx_tls_drop_bypass_required;
+	atomic64_t rx_tls_ctx;
+	atomic64_t rx_tls_del;
 	atomic64_t rx_tls_drop_resync_request;
 	atomic64_t rx_tls_resync_request;
 	atomic64_t rx_tls_resync_reply;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
index a5aabc5c5236..29463bdb7715 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_stats.c
@@ -47,6 +47,8 @@ static const struct counter_desc mlx5e_tls_sw_stats_desc[] = {
 
 static const struct counter_desc mlx5e_ktls_sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_tls_sw_stats, tx_tls_ctx) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_tls_sw_stats, rx_tls_ctx) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_tls_sw_stats, rx_tls_del) },
 };
 
 #define MLX5E_READ_CTR_ATOMIC64(ptr, dsc, i) \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 74adaa58189a..88a01c59ce61 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -179,8 +179,6 @@ static const struct counter_desc sw_stats_desc[] = {
 #ifdef CONFIG_MLX5_EN_TLS
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_decrypted_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_decrypted_bytes) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_ctx) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_del) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_resync_req_pkt) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_resync_req_start) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_tls_resync_req_end) },
@@ -341,8 +339,6 @@ static void mlx5e_stats_grp_sw_update_stats_rq_stats(struct mlx5e_sw_stats *s,
 #ifdef CONFIG_MLX5_EN_TLS
 	s->rx_tls_decrypted_packets   += rq_stats->tls_decrypted_packets;
 	s->rx_tls_decrypted_bytes     += rq_stats->tls_decrypted_bytes;
-	s->rx_tls_ctx                 += rq_stats->tls_ctx;
-	s->rx_tls_del                 += rq_stats->tls_del;
 	s->rx_tls_resync_req_pkt      += rq_stats->tls_resync_req_pkt;
 	s->rx_tls_resync_req_start    += rq_stats->tls_resync_req_start;
 	s->rx_tls_resync_req_end      += rq_stats->tls_resync_req_end;
@@ -1620,8 +1616,6 @@ static const struct counter_desc rq_stats_desc[] = {
 #ifdef CONFIG_MLX5_EN_TLS
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_decrypted_packets) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_decrypted_bytes) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_ctx) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_del) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_resync_req_pkt) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_resync_req_start) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, tls_resync_req_end) },
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 8eb056af79ba..adf9b7b8b712 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -201,8 +201,6 @@ struct mlx5e_sw_stats {
 
 	u64 rx_tls_decrypted_packets;
 	u64 rx_tls_decrypted_bytes;
-	u64 rx_tls_ctx;
-	u64 rx_tls_del;
 	u64 rx_tls_resync_req_pkt;
 	u64 rx_tls_resync_req_start;
 	u64 rx_tls_resync_req_end;
@@ -333,8 +331,6 @@ struct mlx5e_rq_stats {
 #ifdef CONFIG_MLX5_EN_TLS
 	u64 tls_decrypted_packets;
 	u64 tls_decrypted_bytes;
-	u64 tls_ctx;
-	u64 tls_del;
 	u64 tls_resync_req_pkt;
 	u64 tls_resync_req_start;
 	u64 tls_resync_req_end;
-- 
2.30.2

[net 7/9] net/mlx5: Don't request more than supported EQs

[Saeed Mahameed]

From: Daniel Jurgens <danielj@mellanox.com>

Calculating the number of compeltion EQs based on the number of
available IRQ vectors doesn't work now that all async EQs share one IRQ.
Thus the max number of EQs can be exceeded on systems with more than
approximately 256 CPUs. Take this into account when calculating the
number of available completion EQs.

Fixes: 81bfa206032a ("net/mlx5: Use a single IRQ for all async EQs")
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 174dfbc996c6..1fa9c18563da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -931,13 +931,24 @@ void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev)
 	mutex_unlock(&table->lock);
 }
 
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+#define MLX5_MAX_ASYNC_EQS 4
+#else
+#define MLX5_MAX_ASYNC_EQS 3
+#endif
+
 int mlx5_eq_table_create(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *eq_table = dev->priv.eq_table;
+	int num_eqs = MLX5_CAP_GEN(dev, max_num_eqs) ?
+		      MLX5_CAP_GEN(dev, max_num_eqs) :
+		      1 << MLX5_CAP_GEN(dev, log_max_eq);
 	int err;
 
 	eq_table->num_comp_eqs =
-		mlx5_irq_get_num_comp(eq_table->irq_table);
+		min_t(int,
+		      mlx5_irq_get_num_comp(eq_table->irq_table),
+		      num_eqs - MLX5_MAX_ASYNC_EQS);
 
 	err = create_async_eqs(dev);
 	if (err) {
-- 
2.30.2

[net 8/9] net/mlx5e: Consider geneve_opts for encap contexts

[Saeed Mahameed]

From: Dima Chumak <dchumak@nvidia.com>

Current algorithm for encap keys is legacy from initial vxlan
implementation and doesn't take into account all possible fields of a
tunnel. For example, for a Geneve tunnel, which may have additional TLV
options, they are ignored when comparing encap keys and a rule can be
attached to an incorrect encap entry.

Fix that by introducing encap_info_equal() operation in
struct mlx5e_tc_tunnel. Geneve tunnel type uses custom implementation,
which extends generic algorithm and considers options if they are set.

Fixes: 7f1a546e3222 ("net/mlx5e: Consider tunnel type for encap contexts")
Signed-off-by: Dima Chumak <dchumak@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/en/tc_tun.h   | 10 +++++++
 .../mellanox/mlx5/core/en/tc_tun_encap.c      | 23 ++++++---------
 .../mellanox/mlx5/core/en/tc_tun_geneve.c     | 29 +++++++++++++++++++
 .../mellanox/mlx5/core/en/tc_tun_gre.c        |  1 +
 .../mellanox/mlx5/core/en/tc_tun_mplsoudp.c   |  1 +
 .../mellanox/mlx5/core/en/tc_tun_vxlan.c      |  1 +
 6 files changed, 51 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
index 67de2bf36861..e1271998b937 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h
@@ -21,6 +21,11 @@ enum {
 	MLX5E_TC_TUNNEL_TYPE_MPLSOUDP,
 };
 
+struct mlx5e_encap_key {
+	const struct ip_tunnel_key *ip_tun_key;
+	struct mlx5e_tc_tunnel     *tc_tunnel;
+};
+
 struct mlx5e_tc_tunnel {
 	int tunnel_type;
 	enum mlx5_flow_match_level match_level;
@@ -44,6 +49,8 @@ struct mlx5e_tc_tunnel {
 			    struct flow_cls_offload *f,
 			    void *headers_c,
 			    void *headers_v);
+	bool (*encap_info_equal)(struct mlx5e_encap_key *a,
+				 struct mlx5e_encap_key *b);
 };
 
 extern struct mlx5e_tc_tunnel vxlan_tunnel;
@@ -101,6 +108,9 @@ int mlx5e_tc_tun_parse_udp_ports(struct mlx5e_priv *priv,
 				 void *headers_c,
 				 void *headers_v);
 
+bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
+					   struct mlx5e_encap_key *b);
+
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif //__MLX5_EN_TC_TUNNEL_H__
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index 7f7b0f6dcdf9..9f16ad2c0710 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -476,16 +476,11 @@ void mlx5e_detach_decap(struct mlx5e_priv *priv,
 	mlx5e_decap_dealloc(priv, d);
 }
 
-struct encap_key {
-	const struct ip_tunnel_key *ip_tun_key;
-	struct mlx5e_tc_tunnel *tc_tunnel;
-};
-
-static int cmp_encap_info(struct encap_key *a,
-			  struct encap_key *b)
+bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
+					   struct mlx5e_encap_key *b)
 {
-	return memcmp(a->ip_tun_key, b->ip_tun_key, sizeof(*a->ip_tun_key)) ||
-		a->tc_tunnel->tunnel_type != b->tc_tunnel->tunnel_type;
+	return memcmp(a->ip_tun_key, b->ip_tun_key, sizeof(*a->ip_tun_key)) == 0 &&
+		a->tc_tunnel->tunnel_type == b->tc_tunnel->tunnel_type;
 }
 
 static int cmp_decap_info(struct mlx5e_decap_key *a,
@@ -494,7 +489,7 @@ static int cmp_decap_info(struct mlx5e_decap_key *a,
 	return memcmp(&a->key, &b->key, sizeof(b->key));
 }
 
-static int hash_encap_info(struct encap_key *key)
+static int hash_encap_info(struct mlx5e_encap_key *key)
 {
 	return jhash(key->ip_tun_key, sizeof(*key->ip_tun_key),
 		     key->tc_tunnel->tunnel_type);
@@ -516,18 +511,18 @@ static bool mlx5e_decap_take(struct mlx5e_decap_entry *e)
 }
 
 static struct mlx5e_encap_entry *
-mlx5e_encap_get(struct mlx5e_priv *priv, struct encap_key *key,
+mlx5e_encap_get(struct mlx5e_priv *priv, struct mlx5e_encap_key *key,
 		uintptr_t hash_key)
 {
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
+	struct mlx5e_encap_key e_key;
 	struct mlx5e_encap_entry *e;
-	struct encap_key e_key;
 
 	hash_for_each_possible_rcu(esw->offloads.encap_tbl, e,
 				   encap_hlist, hash_key) {
 		e_key.ip_tun_key = &e->tun_info->key;
 		e_key.tc_tunnel = e->tunnel;
-		if (!cmp_encap_info(&e_key, key) &&
+		if (e->tunnel->encap_info_equal(&e_key, key) &&
 		    mlx5e_encap_take(e))
 			return e;
 	}
@@ -694,8 +689,8 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv,
 	struct mlx5_flow_attr *attr = flow->attr;
 	const struct ip_tunnel_info *tun_info;
 	unsigned long tbl_time_before = 0;
-	struct encap_key key;
 	struct mlx5e_encap_entry *e;
+	struct mlx5e_encap_key key;
 	bool entry_created = false;
 	unsigned short family;
 	uintptr_t hash_key;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
index 7ed3f9f79f11..f5b26f5a7de4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c
@@ -329,6 +329,34 @@ static int mlx5e_tc_tun_parse_geneve(struct mlx5e_priv *priv,
 	return mlx5e_tc_tun_parse_geneve_options(priv, spec, f);
 }
 
+static bool mlx5e_tc_tun_encap_info_equal_geneve(struct mlx5e_encap_key *a,
+						 struct mlx5e_encap_key *b)
+{
+	struct ip_tunnel_info *a_info;
+	struct ip_tunnel_info *b_info;
+	bool a_has_opts, b_has_opts;
+
+	if (!mlx5e_tc_tun_encap_info_equal_generic(a, b))
+		return false;
+
+	a_has_opts = !!(a->ip_tun_key->tun_flags & TUNNEL_GENEVE_OPT);
+	b_has_opts = !!(b->ip_tun_key->tun_flags & TUNNEL_GENEVE_OPT);
+
+	/* keys are equal when both don't have any options attached */
+	if (!a_has_opts && !b_has_opts)
+		return true;
+
+	if (a_has_opts != b_has_opts)
+		return false;
+
+	/* geneve options stored in memory next to ip_tunnel_info struct */
+	a_info = container_of(a->ip_tun_key, struct ip_tunnel_info, key);
+	b_info = container_of(b->ip_tun_key, struct ip_tunnel_info, key);
+
+	return a_info->options_len == b_info->options_len &&
+		memcmp(a_info + 1, b_info + 1, a_info->options_len) == 0;
+}
+
 struct mlx5e_tc_tunnel geneve_tunnel = {
 	.tunnel_type          = MLX5E_TC_TUNNEL_TYPE_GENEVE,
 	.match_level          = MLX5_MATCH_L4,
@@ -338,4 +366,5 @@ struct mlx5e_tc_tunnel geneve_tunnel = {
 	.generate_ip_tun_hdr  = mlx5e_gen_ip_tunnel_header_geneve,
 	.parse_udp_ports      = mlx5e_tc_tun_parse_udp_ports_geneve,
 	.parse_tunnel         = mlx5e_tc_tun_parse_geneve,
+	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_geneve,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c
index 2805416c32a3..ada14f0574dc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c
@@ -94,4 +94,5 @@ struct mlx5e_tc_tunnel gre_tunnel = {
 	.generate_ip_tun_hdr  = mlx5e_gen_ip_tunnel_header_gretap,
 	.parse_udp_ports      = NULL,
 	.parse_tunnel         = mlx5e_tc_tun_parse_gretap,
+	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_generic,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c
index 3479672e84cf..60952b33b568 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c
@@ -131,4 +131,5 @@ struct mlx5e_tc_tunnel mplsoudp_tunnel = {
 	.generate_ip_tun_hdr  = generate_ip_tun_hdr,
 	.parse_udp_ports      = parse_udp_ports,
 	.parse_tunnel         = parse_tunnel,
+	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_generic,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
index 038a0f1cecec..4267f3a1059e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
@@ -150,4 +150,5 @@ struct mlx5e_tc_tunnel vxlan_tunnel = {
 	.generate_ip_tun_hdr  = mlx5e_gen_ip_tunnel_header_vxlan,
 	.parse_udp_ports      = mlx5e_tc_tun_parse_udp_ports_vxlan,
 	.parse_tunnel         = mlx5e_tc_tun_parse_vxlan,
+	.encap_info_equal     = mlx5e_tc_tun_encap_info_equal_generic,
 };
-- 
2.30.2

[net 9/9] net/mlx5e: Guarantee room for XSK wakeup NOP on async ICOSQ

[Saeed Mahameed]

From: Tariq Toukan <tariqt@nvidia.com>

XSK wakeup flow triggers an IRQ by posting a NOP WQE and hitting
the doorbell on the async ICOSQ.
It maintains its state so that it doesn't issue another NOP WQE
if it has an outstanding one already.

For this flow to work properly, the NOP post must not fail.
Make sure to reserve room for the NOP WQE in all WQE posts to the
async ICOSQ.

Fixes: 8d94b590f1e4 ("net/mlx5e: Turn XSK ICOSQ into a general asynchronous one")
Fixes: 1182f3659357 ("net/mlx5e: kTLS, Add kTLS RX HW offload support")
Fixes: 0419d8c9d8f8 ("net/mlx5e: kTLS, Add kTLS RX resync support")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  1 +
 .../net/ethernet/mellanox/mlx5/core/en/txrx.h |  6 ++++++
 .../mellanox/mlx5/core/en_accel/ktls_rx.c     | 18 +++++++---------
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 21 ++++++++++++++++++-
 4 files changed, 34 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 304b296fe8b9..bc6f77ea0a31 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -516,6 +516,7 @@ struct mlx5e_icosq {
 	struct mlx5_wq_cyc         wq;
 	void __iomem              *uar_map;
 	u32                        sqn;
+	u16                        reserved_room;
 	unsigned long              state;
 
 	/* control path */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 2371b83dad9c..055c3bc23733 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -441,4 +441,10 @@ static inline u16 mlx5e_stop_room_for_wqe(u16 wqe_size)
 	return wqe_size * 2 - 1;
 }
 
+static inline bool mlx5e_icosq_can_post_wqe(struct mlx5e_icosq *sq, u16 wqe_size)
+{
+	u16 room = sq->reserved_room + mlx5e_stop_room_for_wqe(wqe_size);
+
+	return mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, room);
+}
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
index 57c5ebd597a7..19d22a63313f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c
@@ -138,11 +138,10 @@ post_static_params(struct mlx5e_icosq *sq,
 {
 	struct mlx5e_set_tls_static_params_wqe *wqe;
 	struct mlx5e_icosq_wqe_info wi;
-	u16 pi, num_wqebbs, room;
+	u16 pi, num_wqebbs;
 
 	num_wqebbs = MLX5E_TLS_SET_STATIC_PARAMS_WQEBBS;
-	room = mlx5e_stop_room_for_wqe(num_wqebbs);
-	if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, room)))
+	if (unlikely(!mlx5e_icosq_can_post_wqe(sq, num_wqebbs)))
 		return ERR_PTR(-ENOSPC);
 
 	pi = mlx5e_icosq_get_next_pi(sq, num_wqebbs);
@@ -169,11 +168,10 @@ post_progress_params(struct mlx5e_icosq *sq,
 {
 	struct mlx5e_set_tls_progress_params_wqe *wqe;
 	struct mlx5e_icosq_wqe_info wi;
-	u16 pi, num_wqebbs, room;
+	u16 pi, num_wqebbs;
 
 	num_wqebbs = MLX5E_TLS_SET_PROGRESS_PARAMS_WQEBBS;
-	room = mlx5e_stop_room_for_wqe(num_wqebbs);
-	if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, room)))
+	if (unlikely(!mlx5e_icosq_can_post_wqe(sq, num_wqebbs)))
 		return ERR_PTR(-ENOSPC);
 
 	pi = mlx5e_icosq_get_next_pi(sq, num_wqebbs);
@@ -278,17 +276,15 @@ resync_post_get_progress_params(struct mlx5e_icosq *sq,
 
 	buf->priv_rx = priv_rx;
 
-	BUILD_BUG_ON(MLX5E_KTLS_GET_PROGRESS_WQEBBS != 1);
-
 	spin_lock_bh(&sq->channel->async_icosq_lock);
 
-	if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, 1))) {
+	if (unlikely(!mlx5e_icosq_can_post_wqe(sq, MLX5E_KTLS_GET_PROGRESS_WQEBBS))) {
 		spin_unlock_bh(&sq->channel->async_icosq_lock);
 		err = -ENOSPC;
 		goto err_dma_unmap;
 	}
 
-	pi = mlx5e_icosq_get_next_pi(sq, 1);
+	pi = mlx5e_icosq_get_next_pi(sq, MLX5E_KTLS_GET_PROGRESS_WQEBBS);
 	wqe = MLX5E_TLS_FETCH_GET_PROGRESS_PARAMS_WQE(sq, pi);
 
 #define GET_PSV_DS_CNT (DIV_ROUND_UP(sizeof(*wqe), MLX5_SEND_WQE_DS))
@@ -308,7 +304,7 @@ resync_post_get_progress_params(struct mlx5e_icosq *sq,
 
 	wi = (struct mlx5e_icosq_wqe_info) {
 		.wqe_type = MLX5E_ICOSQ_WQE_GET_PSV_TLS,
-		.num_wqebbs = 1,
+		.num_wqebbs = MLX5E_KTLS_GET_PROGRESS_WQEBBS,
 		.tls_get_params.buf = buf,
 	};
 	icosq_fill_wi(sq, pi, &wi);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 158f947a8503..5db63b9f3b70 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1091,6 +1091,7 @@ static int mlx5e_alloc_icosq(struct mlx5e_channel *c,
 
 	sq->channel   = c;
 	sq->uar_map   = mdev->mlx5e_res.bfreg.map;
+	sq->reserved_room = param->stop_room;
 
 	param->wq.db_numa_node = cpu_to_node(c->cpu);
 	err = mlx5_wq_cyc_create(mdev, &param->wq, sqc_wq, wq, &sq->wq_ctrl);
@@ -2350,6 +2351,24 @@ void mlx5e_build_icosq_param(struct mlx5e_priv *priv,
 	mlx5e_build_ico_cq_param(priv, log_wq_size, &param->cqp);
 }
 
+static void mlx5e_build_async_icosq_param(struct mlx5e_priv *priv,
+					  struct mlx5e_params *params,
+					  u8 log_wq_size,
+					  struct mlx5e_sq_param *param)
+{
+	void *sqc = param->sqc;
+	void *wq = MLX5_ADDR_OF(sqc, sqc, wq);
+
+	mlx5e_build_sq_param_common(priv, param);
+
+	/* async_icosq is used by XSK only if xdp_prog is active */
+	if (params->xdp_prog)
+		param->stop_room = mlx5e_stop_room_for_wqe(1); /* for XSK NOP */
+	MLX5_SET(sqc, sqc, reg_umr, MLX5_CAP_ETH(priv->mdev, reg_umr_sq));
+	MLX5_SET(wq, wq, log_wq_sz, log_wq_size);
+	mlx5e_build_ico_cq_param(priv, log_wq_size, &param->cqp);
+}
+
 void mlx5e_build_xdpsq_param(struct mlx5e_priv *priv,
 			     struct mlx5e_params *params,
 			     struct mlx5e_sq_param *param)
@@ -2398,7 +2417,7 @@ static void mlx5e_build_channel_param(struct mlx5e_priv *priv,
 	mlx5e_build_sq_param(priv, params, &cparam->txq_sq);
 	mlx5e_build_xdpsq_param(priv, params, &cparam->xdp_sq);
 	mlx5e_build_icosq_param(priv, icosq_log_wq_sz, &cparam->icosq);
-	mlx5e_build_icosq_param(priv, async_icosq_log_wq_sz, &cparam->async_icosq);
+	mlx5e_build_async_icosq_param(priv, params, async_icosq_log_wq_sz, &cparam->async_icosq);
 }
 
 int mlx5e_open_channels(struct mlx5e_priv *priv,
-- 
2.30.2

Re: [net 1/9] net/mlx5e: Fix mapping of ct_label zero

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (refs/heads/master):

On Wed, 31 Mar 2021 13:14:16 -0700 you wrote:
> From: Ariel Levkovich <lariel@nvidia.com>
> 
> ct_label 0 is a default label each flow has and therefore
> there can be rules that match on ct_label=0 without a prior
> rule that set the ct_label to this value.
> 
> The ct_label value is not used directly in the HW rules and
> instead it is mapped to some id within a defined range and this
> id is used to set and match the metadata register which carries
> the ct_label.
> 
> [...]

Here is the summary with links:
  - [net,1/9] net/mlx5e: Fix mapping of ct_label zero
    https://git.kernel.org/netdev/net/c/d24f847e5421
  - [net,2/9] net/mlx5: Delete auxiliary bus driver eth-rep first
    https://git.kernel.org/netdev/net/c/1f90aedfb496
  - [net,3/9] net/mlx5e: Fix ethtool indication of connector type
    https://git.kernel.org/netdev/net/c/3211434dfe7a
  - [net,4/9] net/mlx5: E-switch, Create vport miss group only if src rewrite is supported
    https://git.kernel.org/netdev/net/c/e929e3da537e
  - [net,5/9] net/mlx5e: kTLS, Fix TX counters atomicity
    https://git.kernel.org/netdev/net/c/a51bce9698e9
  - [net,6/9] net/mlx5e: kTLS, Fix RX counters atomicity
    https://git.kernel.org/netdev/net/c/6f4fdd530a09
  - [net,7/9] net/mlx5: Don't request more than supported EQs
    https://git.kernel.org/netdev/net/c/a7b76002ae78
  - [net,8/9] net/mlx5e: Consider geneve_opts for encap contexts
    https://git.kernel.org/netdev/net/c/929a2faddd55
  - [net,9/9] net/mlx5e: Guarantee room for XSK wakeup NOP on async ICOSQ
    https://git.kernel.org/netdev/net/c/3ff3874fa0b2

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html