* [patch net-next v3 10/10] mlxsw: spectrum_acl: Pass mlxsw_sp_port down to ruleset bind/unbind ops
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
No need to convert from mlxsw_sp_port to net_device and back again.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 6 +++--
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c | 4 ++--
.../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 27 +++++++++-------------
3 files changed, 17 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index ab6ada7..525552d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -477,9 +477,11 @@ struct mlxsw_sp_acl_profile_ops {
void *priv, void *ruleset_priv);
void (*ruleset_del)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv);
int (*ruleset_bind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv,
- struct net_device *dev, bool ingress);
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress);
void (*ruleset_unbind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv,
- struct net_device *dev, bool ingress);
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress);
u16 (*ruleset_group_id)(void *ruleset_priv);
size_t rule_priv_size;
int (*rule_add)(struct mlxsw_sp *mlxsw_sp,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
index f98bca9..9439bfa 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
@@ -169,7 +169,7 @@ mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
return ops->ruleset_bind(mlxsw_sp, ruleset->priv,
- binding->mlxsw_sp_port->dev, binding->ingress);
+ binding->mlxsw_sp_port, binding->ingress);
}
static void
@@ -181,7 +181,7 @@ mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
ops->ruleset_unbind(mlxsw_sp, ruleset->priv,
- binding->mlxsw_sp_port->dev, binding->ingress);
+ binding->mlxsw_sp_port, binding->ingress);
}
static bool mlxsw_sp_acl_ruleset_block_bound(struct mlxsw_sp_acl_block *block)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
index 50b2f9a..c6e180c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
@@ -258,15 +258,11 @@ static void mlxsw_sp_acl_tcam_group_del(struct mlxsw_sp *mlxsw_sp,
static int
mlxsw_sp_acl_tcam_group_bind(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_tcam_group *group,
- struct net_device *dev, bool ingress)
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
{
- struct mlxsw_sp_port *mlxsw_sp_port;
char ppbt_pl[MLXSW_REG_PPBT_LEN];
- if (!mlxsw_sp_port_dev_check(dev))
- return -EINVAL;
-
- mlxsw_sp_port = netdev_priv(dev);
mlxsw_reg_ppbt_pack(ppbt_pl, ingress ? MLXSW_REG_PXBT_E_IACL :
MLXSW_REG_PXBT_E_EACL,
MLXSW_REG_PXBT_OP_BIND, mlxsw_sp_port->local_port,
@@ -277,15 +273,11 @@ mlxsw_sp_acl_tcam_group_bind(struct mlxsw_sp *mlxsw_sp,
static void
mlxsw_sp_acl_tcam_group_unbind(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_tcam_group *group,
- struct net_device *dev, bool ingress)
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
{
- struct mlxsw_sp_port *mlxsw_sp_port;
char ppbt_pl[MLXSW_REG_PPBT_LEN];
- if (WARN_ON(!mlxsw_sp_port_dev_check(dev)))
- return;
-
- mlxsw_sp_port = netdev_priv(dev);
mlxsw_reg_ppbt_pack(ppbt_pl, ingress ? MLXSW_REG_PXBT_E_IACL :
MLXSW_REG_PXBT_E_EACL,
MLXSW_REG_PXBT_OP_UNBIND, mlxsw_sp_port->local_port,
@@ -1054,22 +1046,25 @@ mlxsw_sp_acl_tcam_flower_ruleset_del(struct mlxsw_sp *mlxsw_sp,
static int
mlxsw_sp_acl_tcam_flower_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
void *ruleset_priv,
- struct net_device *dev, bool ingress)
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
{
struct mlxsw_sp_acl_tcam_flower_ruleset *ruleset = ruleset_priv;
return mlxsw_sp_acl_tcam_group_bind(mlxsw_sp, &ruleset->group,
- dev, ingress);
+ mlxsw_sp_port, ingress);
}
static void
mlxsw_sp_acl_tcam_flower_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
void *ruleset_priv,
- struct net_device *dev, bool ingress)
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
{
struct mlxsw_sp_acl_tcam_flower_ruleset *ruleset = ruleset_priv;
- mlxsw_sp_acl_tcam_group_unbind(mlxsw_sp, &ruleset->group, dev, ingress);
+ mlxsw_sp_acl_tcam_group_unbind(mlxsw_sp, &ruleset->group,
+ mlxsw_sp_port, ingress);
}
static u16
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 09/10] mlxsw: spectrum_acl: Implement TC block sharing
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Benefit from the prepared TC and in-driver ACL infrastructure and
introduce block sharing offload. For that, a new struct "block" is
introduced in spectrum_acl in order to hold a list of specific
block-port bindings.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v2->v3:
- add tc offload feature handling
v1->v2:
- new patch
---
drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 182 +++++++++++++---
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 37 +++-
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c | 237 ++++++++++++++++++---
.../net/ethernet/mellanox/mlxsw/spectrum_flower.c | 41 ++--
4 files changed, 401 insertions(+), 96 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 3b9c8a0..d5e893a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -1747,72 +1747,186 @@ static int mlxsw_sp_setup_tc_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port,
}
static int
-mlxsw_sp_setup_tc_cls_flower(struct mlxsw_sp_port *mlxsw_sp_port,
- struct tc_cls_flower_offload *f,
- bool ingress)
+mlxsw_sp_setup_tc_cls_flower(struct mlxsw_sp_acl_block *acl_block,
+ struct tc_cls_flower_offload *f)
{
+ struct mlxsw_sp *mlxsw_sp = mlxsw_sp_acl_block_mlxsw_sp(acl_block);
+
switch (f->command) {
case TC_CLSFLOWER_REPLACE:
- return mlxsw_sp_flower_replace(mlxsw_sp_port, ingress, f);
+ return mlxsw_sp_flower_replace(mlxsw_sp, acl_block, f);
case TC_CLSFLOWER_DESTROY:
- mlxsw_sp_flower_destroy(mlxsw_sp_port, ingress, f);
+ mlxsw_sp_flower_destroy(mlxsw_sp, acl_block, f);
return 0;
case TC_CLSFLOWER_STATS:
- return mlxsw_sp_flower_stats(mlxsw_sp_port, ingress, f);
+ return mlxsw_sp_flower_stats(mlxsw_sp, acl_block, f);
default:
return -EOPNOTSUPP;
}
}
-static int mlxsw_sp_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
- void *cb_priv, bool ingress)
+static int mlxsw_sp_setup_tc_block_cb_matchall(enum tc_setup_type type,
+ void *type_data,
+ void *cb_priv, bool ingress)
{
struct mlxsw_sp_port *mlxsw_sp_port = cb_priv;
- if (!tc_can_offload(mlxsw_sp_port->dev))
- return -EOPNOTSUPP;
-
switch (type) {
case TC_SETUP_CLSMATCHALL:
+ if (!tc_can_offload(mlxsw_sp_port->dev))
+ return -EOPNOTSUPP;
+
return mlxsw_sp_setup_tc_cls_matchall(mlxsw_sp_port, type_data,
ingress);
case TC_SETUP_CLSFLOWER:
- return mlxsw_sp_setup_tc_cls_flower(mlxsw_sp_port, type_data,
- ingress);
+ return 0;
default:
return -EOPNOTSUPP;
}
}
-static int mlxsw_sp_setup_tc_block_cb_ig(enum tc_setup_type type,
- void *type_data, void *cb_priv)
+static int mlxsw_sp_setup_tc_block_cb_matchall_ig(enum tc_setup_type type,
+ void *type_data,
+ void *cb_priv)
{
- return mlxsw_sp_setup_tc_block_cb(type, type_data, cb_priv, true);
+ return mlxsw_sp_setup_tc_block_cb_matchall(type, type_data,
+ cb_priv, true);
}
-static int mlxsw_sp_setup_tc_block_cb_eg(enum tc_setup_type type,
- void *type_data, void *cb_priv)
+static int mlxsw_sp_setup_tc_block_cb_matchall_eg(enum tc_setup_type type,
+ void *type_data,
+ void *cb_priv)
{
- return mlxsw_sp_setup_tc_block_cb(type, type_data, cb_priv, false);
+ return mlxsw_sp_setup_tc_block_cb_matchall(type, type_data,
+ cb_priv, false);
+}
+
+static int mlxsw_sp_setup_tc_block_cb_flower(enum tc_setup_type type,
+ void *type_data, void *cb_priv)
+{
+ struct mlxsw_sp_acl_block *acl_block = cb_priv;
+
+ switch (type) {
+ case TC_SETUP_CLSMATCHALL:
+ return 0;
+ case TC_SETUP_CLSFLOWER:
+ if (mlxsw_sp_acl_block_disabled(acl_block))
+ return -EOPNOTSUPP;
+
+ return mlxsw_sp_setup_tc_cls_flower(acl_block, type_data);
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static int
+mlxsw_sp_setup_tc_block_flower_bind(struct mlxsw_sp_port *mlxsw_sp_port,
+ struct tcf_block *block, bool ingress)
+{
+ struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+ struct mlxsw_sp_acl_block *acl_block;
+ struct tcf_block_cb *block_cb;
+ int err;
+
+ block_cb = tcf_block_cb_lookup(block, mlxsw_sp_setup_tc_block_cb_flower,
+ mlxsw_sp);
+ if (!block_cb) {
+ acl_block = mlxsw_sp_acl_block_create(mlxsw_sp, block->net);
+ if (!acl_block)
+ return -ENOMEM;
+ block_cb = __tcf_block_cb_register(block,
+ mlxsw_sp_setup_tc_block_cb_flower,
+ mlxsw_sp, acl_block);
+ if (IS_ERR(block_cb)) {
+ err = PTR_ERR(block_cb);
+ goto err_cb_register;
+ }
+ } else {
+ acl_block = tcf_block_cb_priv(block_cb);
+ }
+ tcf_block_cb_incref(block_cb);
+ err = mlxsw_sp_acl_block_bind(mlxsw_sp, acl_block,
+ mlxsw_sp_port, ingress);
+ if (err)
+ goto err_block_bind;
+
+ if (ingress)
+ mlxsw_sp_port->ing_acl_block = acl_block;
+ else
+ mlxsw_sp_port->eg_acl_block = acl_block;
+
+ return 0;
+
+err_block_bind:
+ if (!tcf_block_cb_decref(block_cb)) {
+ __tcf_block_cb_unregister(block_cb);
+err_cb_register:
+ mlxsw_sp_acl_block_destroy(acl_block);
+ }
+ return err;
+}
+
+static void
+mlxsw_sp_setup_tc_block_flower_unbind(struct mlxsw_sp_port *mlxsw_sp_port,
+ struct tcf_block *block, bool ingress)
+{
+ struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+ struct mlxsw_sp_acl_block *acl_block;
+ struct tcf_block_cb *block_cb;
+ int err;
+
+ block_cb = tcf_block_cb_lookup(block, mlxsw_sp_setup_tc_block_cb_flower,
+ mlxsw_sp);
+ if (!block_cb)
+ return;
+
+ if (ingress)
+ mlxsw_sp_port->ing_acl_block = NULL;
+ else
+ mlxsw_sp_port->eg_acl_block = NULL;
+
+ acl_block = tcf_block_cb_priv(block_cb);
+ err = mlxsw_sp_acl_block_unbind(mlxsw_sp, acl_block,
+ mlxsw_sp_port, ingress);
+ if (!err && !tcf_block_cb_decref(block_cb)) {
+ __tcf_block_cb_unregister(block_cb);
+ mlxsw_sp_acl_block_destroy(acl_block);
+ }
}
static int mlxsw_sp_setup_tc_block(struct mlxsw_sp_port *mlxsw_sp_port,
struct tc_block_offload *f)
{
tc_setup_cb_t *cb;
+ bool ingress;
+ int err;
- if (f->binder_type == TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
- cb = mlxsw_sp_setup_tc_block_cb_ig;
- else if (f->binder_type == TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS)
- cb = mlxsw_sp_setup_tc_block_cb_eg;
- else
+ if (f->binder_type == TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS) {
+ cb = mlxsw_sp_setup_tc_block_cb_matchall_ig;
+ ingress = true;
+ } else if (f->binder_type == TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS) {
+ cb = mlxsw_sp_setup_tc_block_cb_matchall_eg;
+ ingress = false;
+ } else {
return -EOPNOTSUPP;
+ }
switch (f->command) {
case TC_BLOCK_BIND:
- return tcf_block_cb_register(f->block, cb, mlxsw_sp_port,
- mlxsw_sp_port);
+ err = tcf_block_cb_register(f->block, cb, mlxsw_sp_port,
+ mlxsw_sp_port);
+ if (err)
+ return err;
+ err = mlxsw_sp_setup_tc_block_flower_bind(mlxsw_sp_port,
+ f->block, ingress);
+ if (err) {
+ tcf_block_cb_unregister(f->block, cb, mlxsw_sp_port);
+ return err;
+ }
+ return 0;
case TC_BLOCK_UNBIND:
+ mlxsw_sp_setup_tc_block_flower_unbind(mlxsw_sp_port,
+ f->block, ingress);
tcf_block_cb_unregister(f->block, cb, mlxsw_sp_port);
return 0;
default:
@@ -1840,10 +1954,18 @@ static int mlxsw_sp_feature_hw_tc(struct net_device *dev, bool enable)
{
struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
- if (!enable && (mlxsw_sp_port->acl_rule_count ||
- !list_empty(&mlxsw_sp_port->mall_tc_list))) {
- netdev_err(dev, "Active offloaded tc filters, can't turn hw_tc_offload off\n");
- return -EINVAL;
+ if (!enable) {
+ if (mlxsw_sp_acl_block_rule_count(mlxsw_sp_port->ing_acl_block) ||
+ mlxsw_sp_acl_block_rule_count(mlxsw_sp_port->eg_acl_block) ||
+ !list_empty(&mlxsw_sp_port->mall_tc_list)) {
+ netdev_err(dev, "Active offloaded tc filters, can't turn hw_tc_offload off\n");
+ return -EINVAL;
+ }
+ mlxsw_sp_acl_block_disable_inc(mlxsw_sp_port->ing_acl_block);
+ mlxsw_sp_acl_block_disable_inc(mlxsw_sp_port->eg_acl_block);
+ } else {
+ mlxsw_sp_acl_block_disable_dec(mlxsw_sp_port->ing_acl_block);
+ mlxsw_sp_acl_block_disable_dec(mlxsw_sp_port->eg_acl_block);
}
return 0;
}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 523e64e..ab6ada7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -270,7 +270,8 @@ struct mlxsw_sp_port {
struct mlxsw_sp_port_sample *sample;
struct list_head vlans_list;
struct mlxsw_sp_qdisc root_qdisc;
- unsigned acl_rule_count;
+ struct mlxsw_sp_acl_block *ing_acl_block;
+ struct mlxsw_sp_acl_block *eg_acl_block;
};
static inline bool
@@ -498,17 +499,34 @@ struct mlxsw_sp_acl_ops {
enum mlxsw_sp_acl_profile profile);
};
+struct mlxsw_sp_acl_block;
struct mlxsw_sp_acl_ruleset;
/* spectrum_acl.c */
struct mlxsw_afk *mlxsw_sp_acl_afk(struct mlxsw_sp_acl *acl);
+struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block);
+unsigned int mlxsw_sp_acl_block_rule_count(struct mlxsw_sp_acl_block *block);
+void mlxsw_sp_acl_block_disable_inc(struct mlxsw_sp_acl_block *block);
+void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block);
+bool mlxsw_sp_acl_block_disabled(struct mlxsw_sp_acl_block *block);
+struct mlxsw_sp_acl_block *mlxsw_sp_acl_block_create(struct mlxsw_sp *mlxsw_sp,
+ struct net *net);
+void mlxsw_sp_acl_block_destroy(struct mlxsw_sp_acl_block *block);
+int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress);
+int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress);
struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
- bool ingress, u32 chain_index,
+mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
enum mlxsw_sp_acl_profile profile);
struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
- bool ingress, u32 chain_index,
+mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
enum mlxsw_sp_acl_profile profile);
void mlxsw_sp_acl_ruleset_put(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_ruleset *ruleset);
@@ -575,11 +593,14 @@ void mlxsw_sp_acl_fini(struct mlxsw_sp *mlxsw_sp);
extern const struct mlxsw_sp_acl_ops mlxsw_sp_acl_tcam_ops;
/* spectrum_flower.c */
-int mlxsw_sp_flower_replace(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+int mlxsw_sp_flower_replace(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f);
-void mlxsw_sp_flower_destroy(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+void mlxsw_sp_flower_destroy(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f);
-int mlxsw_sp_flower_stats(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f);
/* spectrum_qdisc.c */
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
index 7fb41a4..f98bca9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
@@ -39,6 +39,7 @@
#include <linux/string.h>
#include <linux/rhashtable.h>
#include <linux/netdevice.h>
+#include <net/net_namespace.h>
#include <net/tc_act/tc_vlan.h>
#include "reg.h"
@@ -70,9 +71,23 @@ struct mlxsw_afk *mlxsw_sp_acl_afk(struct mlxsw_sp_acl *acl)
return acl->afk;
}
-struct mlxsw_sp_acl_ruleset_ht_key {
- struct net_device *dev; /* dev this ruleset is bound to */
+struct mlxsw_sp_acl_block_binding {
+ struct list_head list;
+ struct net_device *dev;
+ struct mlxsw_sp_port *mlxsw_sp_port;
bool ingress;
+};
+
+struct mlxsw_sp_acl_block {
+ struct list_head binding_list;
+ struct mlxsw_sp_acl_ruleset *ruleset_zero;
+ struct mlxsw_sp *mlxsw_sp;
+ unsigned int rule_count;
+ unsigned int disable_count;
+};
+
+struct mlxsw_sp_acl_ruleset_ht_key {
+ struct mlxsw_sp_acl_block *block;
u32 chain_index;
const struct mlxsw_sp_acl_profile_ops *ops;
};
@@ -118,27 +133,185 @@ struct mlxsw_sp_fid *mlxsw_sp_acl_dummy_fid(struct mlxsw_sp *mlxsw_sp)
return mlxsw_sp->acl->dummy_fid;
}
-static int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset,
- struct net_device *dev, bool ingress)
+struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block)
+{
+ return block->mlxsw_sp;
+}
+
+unsigned int mlxsw_sp_acl_block_rule_count(struct mlxsw_sp_acl_block *block)
+{
+ return block ? block->rule_count : 0;
+}
+
+void mlxsw_sp_acl_block_disable_inc(struct mlxsw_sp_acl_block *block)
+{
+ if (block)
+ block->disable_count++;
+}
+
+void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block)
+{
+ if (block)
+ block->disable_count--;
+}
+
+bool mlxsw_sp_acl_block_disabled(struct mlxsw_sp_acl_block *block)
{
+ return block->disable_count;
+}
+
+static int
+mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_acl_block_binding *binding)
+{
+ struct mlxsw_sp_acl_ruleset *ruleset = block->ruleset_zero;
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
- return ops->ruleset_bind(mlxsw_sp, ruleset->priv, dev, ingress);
+ return ops->ruleset_bind(mlxsw_sp, ruleset->priv,
+ binding->mlxsw_sp_port->dev, binding->ingress);
}
-static void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset,
- struct net_device *dev, bool ingress)
+static void
+mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_acl_block_binding *binding)
{
+ struct mlxsw_sp_acl_ruleset *ruleset = block->ruleset_zero;
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
- ops->ruleset_unbind(mlxsw_sp, ruleset->priv, dev, ingress);
+ ops->ruleset_unbind(mlxsw_sp, ruleset->priv,
+ binding->mlxsw_sp_port->dev, binding->ingress);
+}
+
+static bool mlxsw_sp_acl_ruleset_block_bound(struct mlxsw_sp_acl_block *block)
+{
+ return block->ruleset_zero;
+}
+
+static int
+mlxsw_sp_acl_ruleset_block_bind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_ruleset *ruleset,
+ struct mlxsw_sp_acl_block *block)
+{
+ struct mlxsw_sp_acl_block_binding *binding;
+ int err;
+
+ block->ruleset_zero = ruleset;
+ list_for_each_entry(binding, &block->binding_list, list) {
+ err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, block, binding);
+ if (err)
+ goto rollback;
+ }
+ return 0;
+
+rollback:
+ list_for_each_entry_continue_reverse(binding, &block->binding_list,
+ list)
+ mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding);
+ block->ruleset_zero = NULL;
+
+ return err;
+}
+
+static void
+mlxsw_sp_acl_ruleset_block_unbind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_ruleset *ruleset,
+ struct mlxsw_sp_acl_block *block)
+{
+ struct mlxsw_sp_acl_block_binding *binding;
+
+ list_for_each_entry(binding, &block->binding_list, list)
+ mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding);
+ block->ruleset_zero = NULL;
+}
+
+struct mlxsw_sp_acl_block *mlxsw_sp_acl_block_create(struct mlxsw_sp *mlxsw_sp,
+ struct net *net)
+{
+ struct mlxsw_sp_acl_block *block;
+
+ block = kzalloc(sizeof(*block), GFP_KERNEL);
+ if (!block)
+ return NULL;
+ INIT_LIST_HEAD(&block->binding_list);
+ block->mlxsw_sp = mlxsw_sp;
+ return block;
+}
+
+void mlxsw_sp_acl_block_destroy(struct mlxsw_sp_acl_block *block)
+{
+ WARN_ON(!list_empty(&block->binding_list));
+ kfree(block);
+}
+
+static struct mlxsw_sp_acl_block_binding *
+mlxsw_sp_acl_block_lookup(struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_port *mlxsw_sp_port, bool ingress)
+{
+ struct mlxsw_sp_acl_block_binding *binding;
+
+ list_for_each_entry(binding, &block->binding_list, list)
+ if (binding->mlxsw_sp_port == mlxsw_sp_port &&
+ binding->ingress == ingress)
+ return binding;
+ return NULL;
+}
+
+int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
+{
+ struct mlxsw_sp_acl_block_binding *binding;
+ int err;
+
+ if (WARN_ON(mlxsw_sp_acl_block_lookup(block, mlxsw_sp_port, ingress)))
+ return -EEXIST;
+
+ binding = kzalloc(sizeof(*binding), GFP_KERNEL);
+ if (!binding)
+ return -ENOMEM;
+ binding->mlxsw_sp_port = mlxsw_sp_port;
+ binding->ingress = ingress;
+
+ if (mlxsw_sp_acl_ruleset_block_bound(block)) {
+ err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, block, binding);
+ if (err)
+ goto err_ruleset_bind;
+ }
+
+ list_add(&binding->list, &block->binding_list);
+ return 0;
+
+err_ruleset_bind:
+ kfree(binding);
+ return err;
+}
+
+int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
+ struct mlxsw_sp_port *mlxsw_sp_port,
+ bool ingress)
+{
+ struct mlxsw_sp_acl_block_binding *binding;
+
+ binding = mlxsw_sp_acl_block_lookup(block, mlxsw_sp_port, ingress);
+ if (!binding)
+ return -ENOENT;
+
+ list_del(&binding->list);
+
+ if (mlxsw_sp_acl_ruleset_block_bound(block))
+ mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding);
+
+ kfree(binding);
+ return 0;
}
static struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
- bool ingress, u32 chain_index,
+mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
const struct mlxsw_sp_acl_profile_ops *ops)
{
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
@@ -151,8 +324,7 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
if (!ruleset)
return ERR_PTR(-ENOMEM);
ruleset->ref_count = 1;
- ruleset->ht_key.dev = dev;
- ruleset->ht_key.ingress = ingress;
+ ruleset->ht_key.block = block;
ruleset->ht_key.chain_index = chain_index;
ruleset->ht_key.ops = ops;
@@ -174,8 +346,7 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
* to be directly bound to device. The rest of the rulesets
* are bound by "Goto action set".
*/
- err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, ruleset,
- dev, ingress);
+ err = mlxsw_sp_acl_ruleset_block_bind(mlxsw_sp, ruleset, block);
if (err)
goto err_ruleset_bind;
}
@@ -198,12 +369,12 @@ static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_ruleset *ruleset)
{
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
+ struct mlxsw_sp_acl_block *block = ruleset->ht_key.block;
+ u32 chain_index = ruleset->ht_key.chain_index;
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
- if (!ruleset->ht_key.chain_index)
- mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, ruleset,
- ruleset->ht_key.dev,
- ruleset->ht_key.ingress);
+ if (!chain_index)
+ mlxsw_sp_acl_ruleset_block_unbind(mlxsw_sp, ruleset, block);
rhashtable_remove_fast(&acl->ruleset_ht, &ruleset->ht_node,
mlxsw_sp_acl_ruleset_ht_params);
ops->ruleset_del(mlxsw_sp, ruleset->priv);
@@ -225,15 +396,14 @@ static void mlxsw_sp_acl_ruleset_ref_dec(struct mlxsw_sp *mlxsw_sp,
}
static struct mlxsw_sp_acl_ruleset *
-__mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp_acl *acl, struct net_device *dev,
- bool ingress, u32 chain_index,
+__mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp_acl *acl,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
const struct mlxsw_sp_acl_profile_ops *ops)
{
struct mlxsw_sp_acl_ruleset_ht_key ht_key;
memset(&ht_key, 0, sizeof(ht_key));
- ht_key.dev = dev;
- ht_key.ingress = ingress;
+ ht_key.block = block;
ht_key.chain_index = chain_index;
ht_key.ops = ops;
return rhashtable_lookup_fast(&acl->ruleset_ht, &ht_key,
@@ -241,8 +411,8 @@ __mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp_acl *acl, struct net_device *dev,
}
struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
- bool ingress, u32 chain_index,
+mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
enum mlxsw_sp_acl_profile profile)
{
const struct mlxsw_sp_acl_profile_ops *ops;
@@ -252,16 +422,15 @@ mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
ops = acl->ops->profile_ops(mlxsw_sp, profile);
if (!ops)
return ERR_PTR(-EINVAL);
- ruleset = __mlxsw_sp_acl_ruleset_lookup(acl, dev, ingress,
- chain_index, ops);
+ ruleset = __mlxsw_sp_acl_ruleset_lookup(acl, block, chain_index, ops);
if (!ruleset)
return ERR_PTR(-ENOENT);
return ruleset;
}
struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
- bool ingress, u32 chain_index,
+mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block, u32 chain_index,
enum mlxsw_sp_acl_profile profile)
{
const struct mlxsw_sp_acl_profile_ops *ops;
@@ -272,14 +441,12 @@ mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
if (!ops)
return ERR_PTR(-EINVAL);
- ruleset = __mlxsw_sp_acl_ruleset_lookup(acl, dev, ingress,
- chain_index, ops);
+ ruleset = __mlxsw_sp_acl_ruleset_lookup(acl, block, chain_index, ops);
if (ruleset) {
mlxsw_sp_acl_ruleset_ref_inc(ruleset);
return ruleset;
}
- return mlxsw_sp_acl_ruleset_create(mlxsw_sp, dev, ingress,
- chain_index, ops);
+ return mlxsw_sp_acl_ruleset_create(mlxsw_sp, block, chain_index, ops);
}
void mlxsw_sp_acl_ruleset_put(struct mlxsw_sp *mlxsw_sp,
@@ -528,6 +695,7 @@ int mlxsw_sp_acl_rule_add(struct mlxsw_sp *mlxsw_sp,
goto err_rhashtable_insert;
list_add_tail(&rule->list, &mlxsw_sp->acl->rules);
+ ruleset->ht_key.block->rule_count++;
return 0;
err_rhashtable_insert:
@@ -541,6 +709,7 @@ void mlxsw_sp_acl_rule_del(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_ruleset *ruleset = rule->ruleset;
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
+ ruleset->ht_key.block->rule_count--;
list_del(&rule->list);
rhashtable_remove_fast(&ruleset->rule_ht, &rule->ht_node,
mlxsw_sp_acl_rule_ht_params);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c
index 42e8a36..cf7b97d4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c
@@ -35,6 +35,7 @@
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/netdevice.h>
+#include <net/net_namespace.h>
#include <net/flow_dissector.h>
#include <net/pkt_cls.h>
#include <net/tc_act/tc_gact.h>
@@ -45,7 +46,7 @@
#include "core_acl_flex_keys.h"
static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp,
- struct net_device *dev, bool ingress,
+ struct mlxsw_sp_acl_block *block,
struct mlxsw_sp_acl_rule_info *rulei,
struct tcf_exts *exts)
{
@@ -80,8 +81,7 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl_ruleset *ruleset;
u16 group_id;
- ruleset = mlxsw_sp_acl_ruleset_lookup(mlxsw_sp, dev,
- ingress,
+ ruleset = mlxsw_sp_acl_ruleset_lookup(mlxsw_sp, block,
chain_index,
MLXSW_SP_ACL_PROFILE_FLOWER);
if (IS_ERR(ruleset))
@@ -104,9 +104,6 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp,
return err;
out_dev = tcf_mirred_dev(a);
- if (out_dev == dev)
- out_dev = NULL;
-
err = mlxsw_sp_acl_rulei_act_fwd(mlxsw_sp, rulei,
out_dev);
if (err)
@@ -265,7 +262,7 @@ static int mlxsw_sp_flower_parse_ip(struct mlxsw_sp *mlxsw_sp,
}
static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp,
- struct net_device *dev, bool ingress,
+ struct mlxsw_sp_acl_block *block,
struct mlxsw_sp_acl_rule_info *rulei,
struct tc_cls_flower_offload *f)
{
@@ -383,21 +380,19 @@ static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp,
if (err)
return err;
- return mlxsw_sp_flower_parse_actions(mlxsw_sp, dev, ingress,
- rulei, f->exts);
+ return mlxsw_sp_flower_parse_actions(mlxsw_sp, block, rulei, f->exts);
}
-int mlxsw_sp_flower_replace(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+int mlxsw_sp_flower_replace(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f)
{
- struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
- struct net_device *dev = mlxsw_sp_port->dev;
struct mlxsw_sp_acl_rule_info *rulei;
struct mlxsw_sp_acl_ruleset *ruleset;
struct mlxsw_sp_acl_rule *rule;
int err;
- ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, dev, ingress,
+ ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, block,
f->common.chain_index,
MLXSW_SP_ACL_PROFILE_FLOWER);
if (IS_ERR(ruleset))
@@ -410,7 +405,7 @@ int mlxsw_sp_flower_replace(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
}
rulei = mlxsw_sp_acl_rule_rulei(rule);
- err = mlxsw_sp_flower_parse(mlxsw_sp, dev, ingress, rulei, f);
+ err = mlxsw_sp_flower_parse(mlxsw_sp, block, rulei, f);
if (err)
goto err_flower_parse;
@@ -423,7 +418,6 @@ int mlxsw_sp_flower_replace(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
goto err_rule_add;
mlxsw_sp_acl_ruleset_put(mlxsw_sp, ruleset);
- mlxsw_sp_port->acl_rule_count++;
return 0;
err_rule_add:
@@ -435,15 +429,15 @@ int mlxsw_sp_flower_replace(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
return err;
}
-void mlxsw_sp_flower_destroy(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+void mlxsw_sp_flower_destroy(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f)
{
- struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
struct mlxsw_sp_acl_ruleset *ruleset;
struct mlxsw_sp_acl_rule *rule;
- ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, mlxsw_sp_port->dev,
- ingress, f->common.chain_index,
+ ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, block,
+ f->common.chain_index,
MLXSW_SP_ACL_PROFILE_FLOWER);
if (IS_ERR(ruleset))
return;
@@ -455,13 +449,12 @@ void mlxsw_sp_flower_destroy(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
}
mlxsw_sp_acl_ruleset_put(mlxsw_sp, ruleset);
- mlxsw_sp_port->acl_rule_count--;
}
-int mlxsw_sp_flower_stats(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
+int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_block *block,
struct tc_cls_flower_offload *f)
{
- struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
struct mlxsw_sp_acl_ruleset *ruleset;
struct mlxsw_sp_acl_rule *rule;
u64 packets;
@@ -469,8 +462,8 @@ int mlxsw_sp_flower_stats(struct mlxsw_sp_port *mlxsw_sp_port, bool ingress,
u64 bytes;
int err;
- ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, mlxsw_sp_port->dev,
- ingress, f->common.chain_index,
+ ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, block,
+ f->common.chain_index,
MLXSW_SP_ACL_PROFILE_FLOWER);
if (WARN_ON(IS_ERR(ruleset)))
return -EINVAL;
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 08/10] mlxsw: spectrum_acl: Don't store netdev and ingress for ruleset unbind
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Instead, pass netdev and ingress flag to ruleset unbind op.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 3 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c | 9 ++++--
.../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 33 +++++++++++-----------
3 files changed, 24 insertions(+), 21 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index a0adcd8..523e64e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -477,7 +477,8 @@ struct mlxsw_sp_acl_profile_ops {
void (*ruleset_del)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv);
int (*ruleset_bind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv,
struct net_device *dev, bool ingress);
- void (*ruleset_unbind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv);
+ void (*ruleset_unbind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv,
+ struct net_device *dev, bool ingress);
u16 (*ruleset_group_id)(void *ruleset_priv);
size_t rule_priv_size;
int (*rule_add)(struct mlxsw_sp *mlxsw_sp,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
index ead4cb8..7fb41a4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
@@ -128,11 +128,12 @@ static int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
}
static void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset)
+ struct mlxsw_sp_acl_ruleset *ruleset,
+ struct net_device *dev, bool ingress)
{
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
- ops->ruleset_unbind(mlxsw_sp, ruleset->priv);
+ ops->ruleset_unbind(mlxsw_sp, ruleset->priv, dev, ingress);
}
static struct mlxsw_sp_acl_ruleset *
@@ -200,7 +201,9 @@ static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
if (!ruleset->ht_key.chain_index)
- mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, ruleset);
+ mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, ruleset,
+ ruleset->ht_key.dev,
+ ruleset->ht_key.ingress);
rhashtable_remove_fast(&acl->ruleset_ht, &ruleset->ht_node,
mlxsw_sp_acl_ruleset_ht_params);
ops->ruleset_del(mlxsw_sp, ruleset->priv);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
index 7e8284b..50b2f9a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c
@@ -154,10 +154,6 @@ struct mlxsw_sp_acl_tcam_group {
struct list_head region_list;
unsigned int region_count;
struct rhashtable chunk_ht;
- struct {
- u16 local_port;
- bool ingress;
- } bound;
struct mlxsw_sp_acl_tcam_group_ops *ops;
const struct mlxsw_sp_acl_tcam_pattern *patterns;
unsigned int patterns_count;
@@ -271,26 +267,28 @@ mlxsw_sp_acl_tcam_group_bind(struct mlxsw_sp *mlxsw_sp,
return -EINVAL;
mlxsw_sp_port = netdev_priv(dev);
- group->bound.local_port = mlxsw_sp_port->local_port;
- group->bound.ingress = ingress;
- mlxsw_reg_ppbt_pack(ppbt_pl,
- group->bound.ingress ? MLXSW_REG_PXBT_E_IACL :
- MLXSW_REG_PXBT_E_EACL,
- MLXSW_REG_PXBT_OP_BIND, group->bound.local_port,
+ mlxsw_reg_ppbt_pack(ppbt_pl, ingress ? MLXSW_REG_PXBT_E_IACL :
+ MLXSW_REG_PXBT_E_EACL,
+ MLXSW_REG_PXBT_OP_BIND, mlxsw_sp_port->local_port,
group->id);
return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ppbt), ppbt_pl);
}
static void
mlxsw_sp_acl_tcam_group_unbind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_tcam_group *group)
+ struct mlxsw_sp_acl_tcam_group *group,
+ struct net_device *dev, bool ingress)
{
+ struct mlxsw_sp_port *mlxsw_sp_port;
char ppbt_pl[MLXSW_REG_PPBT_LEN];
- mlxsw_reg_ppbt_pack(ppbt_pl,
- group->bound.ingress ? MLXSW_REG_PXBT_E_IACL :
- MLXSW_REG_PXBT_E_EACL,
- MLXSW_REG_PXBT_OP_UNBIND, group->bound.local_port,
+ if (WARN_ON(!mlxsw_sp_port_dev_check(dev)))
+ return;
+
+ mlxsw_sp_port = netdev_priv(dev);
+ mlxsw_reg_ppbt_pack(ppbt_pl, ingress ? MLXSW_REG_PXBT_E_IACL :
+ MLXSW_REG_PXBT_E_EACL,
+ MLXSW_REG_PXBT_OP_UNBIND, mlxsw_sp_port->local_port,
group->id);
mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ppbt), ppbt_pl);
}
@@ -1066,11 +1064,12 @@ mlxsw_sp_acl_tcam_flower_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
static void
mlxsw_sp_acl_tcam_flower_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
- void *ruleset_priv)
+ void *ruleset_priv,
+ struct net_device *dev, bool ingress)
{
struct mlxsw_sp_acl_tcam_flower_ruleset *ruleset = ruleset_priv;
- mlxsw_sp_acl_tcam_group_unbind(mlxsw_sp, &ruleset->group);
+ mlxsw_sp_acl_tcam_group_unbind(mlxsw_sp, &ruleset->group, dev, ingress);
}
static u16
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 07/10] mlxsw: spectrum_acl: Reshuffle code around mlxsw_sp_acl_ruleset_create/destroy
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
In order to prepare for follow-up changes, make the bind/unbind helpers
very simple. That required move of ht insertion/removal and bind/unbind
calls into mlxsw_sp_acl_ruleset_create/destroy.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c | 102 ++++++++++-----------
1 file changed, 46 insertions(+), 56 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
index 93dcd31..ead4cb8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c
@@ -118,8 +118,26 @@ struct mlxsw_sp_fid *mlxsw_sp_acl_dummy_fid(struct mlxsw_sp *mlxsw_sp)
return mlxsw_sp->acl->dummy_fid;
}
+static int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_ruleset *ruleset,
+ struct net_device *dev, bool ingress)
+{
+ const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
+
+ return ops->ruleset_bind(mlxsw_sp, ruleset->priv, dev, ingress);
+}
+
+static void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_ruleset *ruleset)
+{
+ const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
+
+ ops->ruleset_unbind(mlxsw_sp, ruleset->priv);
+}
+
static struct mlxsw_sp_acl_ruleset *
-mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
+mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
+ bool ingress, u32 chain_index,
const struct mlxsw_sp_acl_profile_ops *ops)
{
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
@@ -132,6 +150,9 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
if (!ruleset)
return ERR_PTR(-ENOMEM);
ruleset->ref_count = 1;
+ ruleset->ht_key.dev = dev;
+ ruleset->ht_key.ingress = ingress;
+ ruleset->ht_key.chain_index = chain_index;
ruleset->ht_key.ops = ops;
err = rhashtable_init(&ruleset->rule_ht, &mlxsw_sp_acl_rule_ht_params);
@@ -142,68 +163,49 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
if (err)
goto err_ops_ruleset_add;
- return ruleset;
-
-err_ops_ruleset_add:
- rhashtable_destroy(&ruleset->rule_ht);
-err_rhashtable_init:
- kfree(ruleset);
- return ERR_PTR(err);
-}
-
-static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset)
-{
- const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
-
- ops->ruleset_del(mlxsw_sp, ruleset->priv);
- rhashtable_destroy(&ruleset->rule_ht);
- kfree(ruleset);
-}
-
-static int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset,
- struct net_device *dev, bool ingress,
- u32 chain_index)
-{
- const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
- struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
- int err;
-
- ruleset->ht_key.dev = dev;
- ruleset->ht_key.ingress = ingress;
- ruleset->ht_key.chain_index = chain_index;
err = rhashtable_insert_fast(&acl->ruleset_ht, &ruleset->ht_node,
mlxsw_sp_acl_ruleset_ht_params);
if (err)
- return err;
- if (!ruleset->ht_key.chain_index) {
+ goto err_ht_insert;
+
+ if (!chain_index) {
/* We only need ruleset with chain index 0, the implicit one,
* to be directly bound to device. The rest of the rulesets
* are bound by "Goto action set".
*/
- err = ops->ruleset_bind(mlxsw_sp, ruleset->priv, dev, ingress);
+ err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, ruleset,
+ dev, ingress);
if (err)
- goto err_ops_ruleset_bind;
+ goto err_ruleset_bind;
}
- return 0;
-err_ops_ruleset_bind:
+ return ruleset;
+
+err_ruleset_bind:
rhashtable_remove_fast(&acl->ruleset_ht, &ruleset->ht_node,
mlxsw_sp_acl_ruleset_ht_params);
- return err;
+err_ht_insert:
+ ops->ruleset_del(mlxsw_sp, ruleset->priv);
+err_ops_ruleset_add:
+ rhashtable_destroy(&ruleset->rule_ht);
+err_rhashtable_init:
+ kfree(ruleset);
+ return ERR_PTR(err);
}
-static void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp,
- struct mlxsw_sp_acl_ruleset *ruleset)
+static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
+ struct mlxsw_sp_acl_ruleset *ruleset)
{
const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops;
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
if (!ruleset->ht_key.chain_index)
- ops->ruleset_unbind(mlxsw_sp, ruleset->priv);
+ mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, ruleset);
rhashtable_remove_fast(&acl->ruleset_ht, &ruleset->ht_node,
mlxsw_sp_acl_ruleset_ht_params);
+ ops->ruleset_del(mlxsw_sp, ruleset->priv);
+ rhashtable_destroy(&ruleset->rule_ht);
+ kfree(ruleset);
}
static void mlxsw_sp_acl_ruleset_ref_inc(struct mlxsw_sp_acl_ruleset *ruleset)
@@ -216,7 +218,6 @@ static void mlxsw_sp_acl_ruleset_ref_dec(struct mlxsw_sp *mlxsw_sp,
{
if (--ruleset->ref_count)
return;
- mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, ruleset);
mlxsw_sp_acl_ruleset_destroy(mlxsw_sp, ruleset);
}
@@ -263,7 +264,6 @@ mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
const struct mlxsw_sp_acl_profile_ops *ops;
struct mlxsw_sp_acl *acl = mlxsw_sp->acl;
struct mlxsw_sp_acl_ruleset *ruleset;
- int err;
ops = acl->ops->profile_ops(mlxsw_sp, profile);
if (!ops)
@@ -275,18 +275,8 @@ mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, struct net_device *dev,
mlxsw_sp_acl_ruleset_ref_inc(ruleset);
return ruleset;
}
- ruleset = mlxsw_sp_acl_ruleset_create(mlxsw_sp, ops);
- if (IS_ERR(ruleset))
- return ruleset;
- err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, ruleset, dev,
- ingress, chain_index);
- if (err)
- goto err_ruleset_bind;
- return ruleset;
-
-err_ruleset_bind:
- mlxsw_sp_acl_ruleset_destroy(mlxsw_sp, ruleset);
- return ERR_PTR(err);
+ return mlxsw_sp_acl_ruleset_create(mlxsw_sp, dev, ingress,
+ chain_index, ops);
}
void mlxsw_sp_acl_ruleset_put(struct mlxsw_sp *mlxsw_sp,
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 06/10] net: sched: allow ingress and clsact qdiscs to share filter blocks
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Benefit from the previously introduced shared filter blocks
infrastructure and allow ingress and clsact qdisc instances to share
filter blocks. The block index is coming from userspace as qdisc option.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v2->v3:
- removed "p_" prefix from block index function args
---
include/uapi/linux/pkt_sched.h | 11 ++++++
net/sched/sch_ingress.c | 89 +++++++++++++++++++++++++++++++++++++++++-
2 files changed, 99 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index af3cc2f..0feee36 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -935,4 +935,15 @@ enum {
#define TCA_CBS_MAX (__TCA_CBS_MAX - 1)
+/* Ingress/clsact */
+
+enum {
+ TCA_CLSACT_UNSPEC,
+ TCA_CLSACT_INGRESS_BLOCK,
+ TCA_CLSACT_EGRESS_BLOCK,
+ __TCA_CLSACT_MAX
+};
+
+#define TCA_CLSACT_MAX (__TCA_CLSACT_MAX - 1)
+
#endif
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index 5ecc38f..3b27b60 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -60,6 +60,29 @@ static void clsact_chain_head_change(struct tcf_proto *tp_head, void *priv)
struct mini_Qdisc_pair *miniqp = priv;
mini_qdisc_pair_swap(miniqp, tp_head);
+};
+
+static const struct nla_policy ingress_policy[TCA_CLSACT_MAX + 1] = {
+ [TCA_CLSACT_INGRESS_BLOCK] = { .type = NLA_U32 },
+};
+
+static int ingress_parse_opt(struct nlattr *opt, u32 *ingress_block_index)
+{
+ struct nlattr *tb[TCA_CLSACT_MAX + 1];
+ int err;
+
+ *ingress_block_index = 0;
+
+ if (!opt)
+ return 0;
+ err = nla_parse_nested(tb, TCA_CLSACT_MAX, opt, ingress_policy, NULL);
+ if (err)
+ return err;
+
+ if (tb[TCA_CLSACT_INGRESS_BLOCK])
+ *ingress_block_index =
+ nla_get_u32(tb[TCA_CLSACT_INGRESS_BLOCK]);
+ return 0;
}
static int ingress_init(struct Qdisc *sch, struct nlattr *opt)
@@ -70,6 +93,11 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt)
mini_qdisc_pair_init(&q->miniqp, sch, &dev->miniq_ingress);
+ err = ingress_parse_opt(opt, &q->block_info.block_index);
+ if (err)
+ return err;
+
+ q->block_info.shareable = true;
q->block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS;
q->block_info.chain_head_change = clsact_chain_head_change;
q->block_info.chain_head_change_priv = &q->miniqp;
@@ -94,11 +122,14 @@ static void ingress_destroy(struct Qdisc *sch)
static int ingress_dump(struct Qdisc *sch, struct sk_buff *skb)
{
+ struct ingress_sched_data *q = qdisc_priv(sch);
struct nlattr *nest;
nest = nla_nest_start(skb, TCA_OPTIONS);
if (nest == NULL)
goto nla_put_failure;
+ if (nla_put_u32(skb, TCA_CLSACT_INGRESS_BLOCK, q->block->index))
+ goto nla_put_failure;
return nla_nest_end(skb, nest);
@@ -166,6 +197,35 @@ static struct tcf_block *clsact_tcf_block(struct Qdisc *sch, unsigned long cl)
}
}
+static const struct nla_policy clsact_policy[TCA_CLSACT_MAX + 1] = {
+ [TCA_CLSACT_INGRESS_BLOCK] = { .type = NLA_U32 },
+ [TCA_CLSACT_EGRESS_BLOCK] = { .type = NLA_U32 },
+};
+
+static int clsact_parse_opt(struct nlattr *opt, u32 *ingress_block_index,
+ u32 *egress_block_index)
+{
+ struct nlattr *tb[TCA_CLSACT_MAX + 1];
+ int err;
+
+ *ingress_block_index = 0;
+ *egress_block_index = 0;
+
+ if (!opt)
+ return 0;
+ err = nla_parse_nested(tb, TCA_CLSACT_MAX, opt, clsact_policy, NULL);
+ if (err)
+ return err;
+
+ if (tb[TCA_CLSACT_INGRESS_BLOCK])
+ *ingress_block_index =
+ nla_get_u32(tb[TCA_CLSACT_INGRESS_BLOCK]);
+ if (tb[TCA_CLSACT_EGRESS_BLOCK])
+ *egress_block_index =
+ nla_get_u32(tb[TCA_CLSACT_EGRESS_BLOCK]);
+ return 0;
+}
+
static int clsact_init(struct Qdisc *sch, struct nlattr *opt)
{
struct clsact_sched_data *q = qdisc_priv(sch);
@@ -174,6 +234,12 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt)
mini_qdisc_pair_init(&q->miniqp_ingress, sch, &dev->miniq_ingress);
+ err = clsact_parse_opt(opt, &q->ingress_block_info.block_index,
+ &q->egress_block_info.block_index);
+ if (err)
+ return err;
+
+ q->ingress_block_info.shareable = true;
q->ingress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS;
q->ingress_block_info.chain_head_change = clsact_chain_head_change;
q->ingress_block_info.chain_head_change_priv = &q->miniqp_ingress;
@@ -184,6 +250,7 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt)
mini_qdisc_pair_init(&q->miniqp_egress, sch, &dev->miniq_egress);
+ q->egress_block_info.shareable = true;
q->egress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS;
q->egress_block_info.chain_head_change = clsact_chain_head_change;
q->egress_block_info.chain_head_change_priv = &q->miniqp_egress;
@@ -215,6 +282,26 @@ static void clsact_destroy(struct Qdisc *sch)
net_dec_egress_queue();
}
+static int clsact_dump(struct Qdisc *sch, struct sk_buff *skb)
+{
+ struct clsact_sched_data *q = qdisc_priv(sch);
+ struct nlattr *nest;
+
+ nest = nla_nest_start(skb, TCA_OPTIONS);
+ if (!nest)
+ goto nla_put_failure;
+ if (nla_put_u32(skb, TCA_CLSACT_INGRESS_BLOCK, q->ingress_block->index))
+ goto nla_put_failure;
+ if (nla_put_u32(skb, TCA_CLSACT_EGRESS_BLOCK, q->egress_block->index))
+ goto nla_put_failure;
+
+ return nla_nest_end(skb, nest);
+
+nla_put_failure:
+ nla_nest_cancel(skb, nest);
+ return -1;
+}
+
static const struct Qdisc_class_ops clsact_class_ops = {
.leaf = ingress_leaf,
.find = clsact_find,
@@ -230,7 +317,7 @@ static struct Qdisc_ops clsact_qdisc_ops __read_mostly = {
.priv_size = sizeof(struct clsact_sched_data),
.init = clsact_init,
.destroy = clsact_destroy,
- .dump = ingress_dump,
+ .dump = clsact_dump,
.owner = THIS_MODULE,
};
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 05/10] net: sched: keep track of offloaded filters and check tc offload feature
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
During block bind, we need to check tc offload feature. If it is
disabled yet still the block contains offloaded filters, forbid the
bind. Also forbid to register callback for a block that already
containes offloaded filters, as the play back is not supported now.
For keeping track of offloaded filters there is a new counter
introduced, alongside with couple of helpers called from cls_* code.
These helpers set and clear TCA_CLS_FLAGS_IN_HW flag.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v2->v3:
- new patch
---
include/net/sch_generic.h | 17 +++++++++++++++
net/sched/cls_api.c | 53 +++++++++++++++++++++++++++++++++++++----------
net/sched/cls_bpf.c | 4 +++-
net/sched/cls_flower.c | 3 ++-
net/sched/cls_matchall.c | 3 ++-
net/sched/cls_u32.c | 13 ++++++------
6 files changed, 73 insertions(+), 20 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index a75bbfa..9c08026 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -283,8 +283,25 @@ struct tcf_block {
struct list_head cb_list;
struct list_head owner_list;
bool keep_dst;
+ unsigned int offloadcnt;
};
+static inline void tcf_block_offload_inc(struct tcf_block *block, u32 *flags)
+{
+ if (*flags & TCA_CLS_FLAGS_IN_HW)
+ return;
+ *flags |= TCA_CLS_FLAGS_IN_HW;
+ block->offloadcnt++;
+}
+
+static inline void tcf_block_offload_dec(struct tcf_block *block, u32 *flags)
+{
+ if (!(*flags & TCA_CLS_FLAGS_IN_HW))
+ return;
+ *flags &= ~TCA_CLS_FLAGS_IN_HW;
+ block->offloadcnt--;
+}
+
static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz)
{
struct qdisc_skb_cb *qcb;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index de9dbcb..ac25142 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -266,31 +266,50 @@ void tcf_chain_put(struct tcf_chain *chain)
}
EXPORT_SYMBOL(tcf_chain_put);
-static void tcf_block_offload_cmd(struct tcf_block *block, struct Qdisc *q,
+static bool tcf_block_offload_in_use(struct tcf_block *block)
+{
+ return block->offloadcnt;
+}
+
+static void tcf_block_offload_cmd(struct tcf_block *block,
+ struct net_device *dev,
struct tcf_block_ext_info *ei,
enum tc_block_command command)
{
- struct net_device *dev = q->dev_queue->dev;
struct tc_block_offload bo = {};
- if (!dev->netdev_ops->ndo_setup_tc)
- return;
bo.command = command;
bo.binder_type = ei->binder_type;
bo.block = block;
dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_BLOCK, &bo);
}
-static void tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q,
- struct tcf_block_ext_info *ei)
+static int tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q,
+ struct tcf_block_ext_info *ei)
{
- tcf_block_offload_cmd(block, q, ei, TC_BLOCK_BIND);
+ struct net_device *dev = q->dev_queue->dev;
+
+ if (!dev->netdev_ops->ndo_setup_tc)
+ return 0;
+
+ /* If tc offload feature is disabled and the block we try to bind
+ * to already has some offloaded filters, forbid to bind.
+ */
+ if (!tc_can_offload(dev) && tcf_block_offload_in_use(block))
+ return -EOPNOTSUPP;
+
+ tcf_block_offload_cmd(block, dev, ei, TC_BLOCK_BIND);
+ return 0;
}
static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q,
struct tcf_block_ext_info *ei)
{
- tcf_block_offload_cmd(block, q, ei, TC_BLOCK_UNBIND);
+ struct net_device *dev = q->dev_queue->dev;
+
+ if (!dev->netdev_ops->ndo_setup_tc)
+ return;
+ tcf_block_offload_cmd(block, dev, ei, TC_BLOCK_UNBIND);
}
static int
@@ -499,10 +518,15 @@ int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
if (err)
goto err_chain_head_change_cb_add;
- tcf_block_offload_bind(block, q, ei);
+ err = tcf_block_offload_bind(block, q, ei);
+ if (err)
+ goto err_block_offload_bind;
+
*p_block = block;
return 0;
+err_block_offload_bind:
+ tcf_chain_head_change_cb_del(tcf_block_chain_zero(block), ei);
err_chain_head_change_cb_add:
tcf_block_owner_del(block, q, ei->binder_type);
err_block_owner_add:
@@ -630,9 +654,16 @@ struct tcf_block_cb *__tcf_block_cb_register(struct tcf_block *block,
{
struct tcf_block_cb *block_cb;
+ /* At this point, playback of previous block cb calls is not supported,
+ * so forbid to register to block which already has some offloaded
+ * filters present.
+ */
+ if (tcf_block_offload_in_use(block))
+ return ERR_PTR(-EOPNOTSUPP);
+
block_cb = kzalloc(sizeof(*block_cb), GFP_KERNEL);
if (!block_cb)
- return NULL;
+ return ERR_PTR(-ENOMEM);
block_cb->cb = cb;
block_cb->cb_ident = cb_ident;
block_cb->cb_priv = cb_priv;
@@ -648,7 +679,7 @@ int tcf_block_cb_register(struct tcf_block *block,
struct tcf_block_cb *block_cb;
block_cb = __tcf_block_cb_register(block, cb, cb_ident, cb_priv);
- return block_cb ? 0 : -ENOMEM;
+ return IS_ERR(block_cb) ? PTR_ERR(block_cb) : 0;
}
EXPORT_SYMBOL(tcf_block_cb_register);
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 69d7e9a..9cf61e7 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -170,8 +170,10 @@ static int cls_bpf_offload_cmd(struct tcf_proto *tp, struct cls_bpf_prog *prog,
cls_bpf_offload_cmd(tp, prog, TC_CLSBPF_DESTROY);
return err;
} else if (err > 0) {
- prog->gen_flags |= TCA_CLS_FLAGS_IN_HW;
+ tcf_block_offload_inc(block, &prog->gen_flags);
}
+ } else {
+ tcf_block_offload_dec(block, &prog->gen_flags);
}
if (addorrep && skip_sw && !(prog->gen_flags & TCA_CLS_FLAGS_IN_HW))
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 6132a73..f61df19 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -229,6 +229,7 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f)
tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER,
&cls_flower, false);
+ tcf_block_offload_dec(block, &f->flags);
}
static int fl_hw_replace_filter(struct tcf_proto *tp,
@@ -256,7 +257,7 @@ static int fl_hw_replace_filter(struct tcf_proto *tp,
fl_hw_destroy_filter(tp, f);
return err;
} else if (err > 0) {
- f->flags |= TCA_CLS_FLAGS_IN_HW;
+ tcf_block_offload_inc(block, &f->flags);
}
if (skip_sw && !(f->flags & TCA_CLS_FLAGS_IN_HW))
diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c
index 66d4e00..d0e57c8 100644
--- a/net/sched/cls_matchall.c
+++ b/net/sched/cls_matchall.c
@@ -81,6 +81,7 @@ static void mall_destroy_hw_filter(struct tcf_proto *tp,
cls_mall.cookie = cookie;
tc_setup_cb_call(block, NULL, TC_SETUP_CLSMATCHALL, &cls_mall, false);
+ tcf_block_offload_dec(block, &head->flags);
}
static int mall_replace_hw_filter(struct tcf_proto *tp,
@@ -103,7 +104,7 @@ static int mall_replace_hw_filter(struct tcf_proto *tp,
mall_destroy_hw_filter(tp, head, cookie);
return err;
} else if (err > 0) {
- head->flags |= TCA_CLS_FLAGS_IN_HW;
+ tcf_block_offload_inc(block, &head->flags);
}
if (skip_sw && !(head->flags & TCA_CLS_FLAGS_IN_HW))
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index ac152b4..0a5a2cb 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -530,16 +530,17 @@ static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h,
return 0;
}
-static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle)
+static void u32_remove_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n)
{
struct tcf_block *block = tp->chain->block;
struct tc_cls_u32_offload cls_u32 = {};
tc_cls_common_offload_init(&cls_u32.common, tp);
cls_u32.command = TC_CLSU32_DELETE_KNODE;
- cls_u32.knode.handle = handle;
+ cls_u32.knode.handle = n->handle;
tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, false);
+ tcf_block_offload_dec(block, &n->flags);
}
static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n,
@@ -568,10 +569,10 @@ static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n,
err = tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, skip_sw);
if (err < 0) {
- u32_remove_hw_knode(tp, n->handle);
+ u32_remove_hw_knode(tp, n);
return err;
} else if (err > 0) {
- n->flags |= TCA_CLS_FLAGS_IN_HW;
+ tcf_block_offload_inc(block, &n->flags);
}
if (skip_sw && !(n->flags & TCA_CLS_FLAGS_IN_HW))
@@ -590,7 +591,7 @@ static void u32_clear_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
RCU_INIT_POINTER(ht->ht[h],
rtnl_dereference(n->next));
tcf_unbind_filter(tp, &n->res);
- u32_remove_hw_knode(tp, n->handle);
+ u32_remove_hw_knode(tp, n);
idr_remove_ext(&ht->handle_idr, n->handle);
if (tcf_exts_get_net(&n->exts))
call_rcu(&n->rcu, u32_delete_key_freepf_rcu);
@@ -683,7 +684,7 @@ static int u32_delete(struct tcf_proto *tp, void *arg, bool *last)
goto out;
if (TC_U32_KEY(ht->handle)) {
- u32_remove_hw_knode(tp, ht->handle);
+ u32_remove_hw_knode(tp, (struct tc_u_knode *)ht);
ret = u32_delete_key(tp, (struct tc_u_knode *)ht);
goto out;
}
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 04/10] net: sched: remove classid and q fields from tcf_proto
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Both are no longer used, so remove them.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
include/net/sch_generic.h | 2 --
net/sched/cls_api.c | 7 ++-----
2 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 70fa3ec..a75bbfa 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -249,8 +249,6 @@ struct tcf_proto {
/* All the rest */
u32 prio;
- u32 classid;
- struct Qdisc *q;
void *data;
const struct tcf_proto_ops *ops;
struct tcf_chain *chain;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 4f51a5d..de9dbcb 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -123,8 +123,7 @@ static inline u32 tcf_auto_prio(struct tcf_proto *tp)
}
static struct tcf_proto *tcf_proto_create(const char *kind, u32 protocol,
- u32 prio, u32 parent, struct Qdisc *q,
- struct tcf_chain *chain)
+ u32 prio, struct tcf_chain *chain)
{
struct tcf_proto *tp;
int err;
@@ -158,8 +157,6 @@ static struct tcf_proto *tcf_proto_create(const char *kind, u32 protocol,
tp->classify = tp->ops->classify;
tp->protocol = protocol;
tp->prio = prio;
- tp->classid = parent;
- tp->q = q;
tp->chain = chain;
err = tp->ops->init(tp);
@@ -1062,7 +1059,7 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n,
prio = tcf_auto_prio(tcf_chain_tp_prev(&chain_info));
tp = tcf_proto_create(nla_data(tca[TCA_KIND]),
- protocol, prio, parent, q, chain);
+ protocol, prio, chain);
if (IS_ERR(tp)) {
err = PTR_ERR(tp);
goto errout;
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 03/10] net: sched: introduce block mechanism to handle netif_keep_dst calls
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Couple of classifiers call netif_keep_dst directly on q->dev. That is
not possible to do directly for shared blocke where multiple qdiscs are
owning the block. So introduce a infrastructure to keep track of the
block owners in list and use this list to implement block variant of
netif_keep_dst.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v1->v2:
- fix binder_type to check "egress" as well as pointed out by Daniel
---
include/net/pkt_cls.h | 1 +
include/net/sch_generic.h | 2 ++
net/sched/cls_api.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++
net/sched/cls_bpf.c | 4 +--
net/sched/cls_flow.c | 2 +-
net/sched/cls_route.c | 2 +-
6 files changed, 76 insertions(+), 4 deletions(-)
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index c04154f..80261e7 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -40,6 +40,7 @@ bool tcf_queue_work(struct work_struct *work);
struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index,
bool create);
void tcf_chain_put(struct tcf_chain *chain);
+void tcf_block_netif_keep_dst(struct tcf_block *block);
int tcf_block_get(struct tcf_block **p_block,
struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q);
int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 9ad1a51..70fa3ec 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -283,6 +283,8 @@ struct tcf_block {
struct net *net;
struct Qdisc *q;
struct list_head cb_list;
+ struct list_head owner_list;
+ bool keep_dst;
};
static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz)
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index de7791c..4f51a5d 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -376,6 +376,7 @@ static struct tcf_block *tcf_block_create(struct net *net, struct Qdisc *q)
return ERR_PTR(-ENOMEM);
INIT_LIST_HEAD(&block->chain_list);
INIT_LIST_HEAD(&block->cb_list);
+ INIT_LIST_HEAD(&block->owner_list);
/* Create chain 0 by default, it has to be always present. */
chain = tcf_chain_create(block, 0);
@@ -406,6 +407,65 @@ static struct tcf_chain *tcf_block_chain_zero(struct tcf_block *block)
return list_first_entry(&block->chain_list, struct tcf_chain, list);
}
+struct tcf_block_owner_item {
+ struct list_head list;
+ struct Qdisc *q;
+ enum tcf_block_binder_type binder_type;
+};
+
+static void
+tcf_block_owner_netif_keep_dst(struct tcf_block *block,
+ struct Qdisc *q,
+ enum tcf_block_binder_type binder_type)
+{
+ if (block->keep_dst &&
+ binder_type != TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS &&
+ binder_type != TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS)
+ netif_keep_dst(qdisc_dev(q));
+}
+
+void tcf_block_netif_keep_dst(struct tcf_block *block)
+{
+ struct tcf_block_owner_item *item;
+
+ block->keep_dst = true;
+ list_for_each_entry(item, &block->owner_list, list)
+ tcf_block_owner_netif_keep_dst(block, item->q,
+ item->binder_type);
+}
+EXPORT_SYMBOL(tcf_block_netif_keep_dst);
+
+static int tcf_block_owner_add(struct tcf_block *block,
+ struct Qdisc *q,
+ enum tcf_block_binder_type binder_type)
+{
+ struct tcf_block_owner_item *item;
+
+ item = kmalloc(sizeof(*item), GFP_KERNEL);
+ if (!item)
+ return -ENOMEM;
+ item->q = q;
+ item->binder_type = binder_type;
+ list_add(&item->list, &block->owner_list);
+ return 0;
+}
+
+static void tcf_block_owner_del(struct tcf_block *block,
+ struct Qdisc *q,
+ enum tcf_block_binder_type binder_type)
+{
+ struct tcf_block_owner_item *item;
+
+ list_for_each_entry(item, &block->owner_list, list) {
+ if (item->q == q && item->binder_type == binder_type) {
+ list_del(&item->list);
+ kfree(item);
+ return;
+ }
+ }
+ WARN_ON(1);
+}
+
int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
struct tcf_block_ext_info *ei)
{
@@ -432,6 +492,12 @@ int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
}
}
+ err = tcf_block_owner_add(block, q, ei->binder_type);
+ if (err)
+ goto err_block_owner_add;
+
+ tcf_block_owner_netif_keep_dst(block, q, ei->binder_type);
+
err = tcf_chain_head_change_cb_add(tcf_block_chain_zero(block), ei);
if (err)
goto err_chain_head_change_cb_add;
@@ -441,6 +507,8 @@ int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
return 0;
err_chain_head_change_cb_add:
+ tcf_block_owner_del(block, q, ei->binder_type);
+err_block_owner_add:
if (created) {
if (ei->shareable)
tcf_block_remove(block, net);
@@ -482,6 +550,7 @@ void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q,
struct tcf_chain *chain, *tmp;
tcf_chain_head_change_cb_del(tcf_block_chain_zero(block), ei);
+ tcf_block_owner_del(block, q, ei->binder_type);
if (--block->refcnt == 0) {
if (ei->shareable)
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 6fe798c..69d7e9a 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -409,8 +409,8 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog,
prog->bpf_name = name;
prog->filter = fp;
- if (fp->dst_needed && !(tp->q->flags & TCQ_F_INGRESS))
- netif_keep_dst(qdisc_dev(tp->q));
+ if (fp->dst_needed)
+ tcf_block_netif_keep_dst(tp->chain->block);
return 0;
}
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index 25c2a88..28cd6fb 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -526,7 +526,7 @@ static int flow_change(struct net *net, struct sk_buff *in_skb,
timer_setup(&fnew->perturb_timer, flow_perturbation, TIMER_DEFERRABLE);
- netif_keep_dst(qdisc_dev(tp->q));
+ tcf_block_netif_keep_dst(tp->chain->block);
if (tb[TCA_FLOW_KEYS]) {
fnew->keymask = keymask;
diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c
index ac9a5b8..a1f2b1b 100644
--- a/net/sched/cls_route.c
+++ b/net/sched/cls_route.c
@@ -527,7 +527,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb,
if (f->handle < f1->handle)
break;
- netif_keep_dst(qdisc_dev(tp->q));
+ tcf_block_netif_keep_dst(tp->chain->block);
rcu_assign_pointer(f->next, f1);
rcu_assign_pointer(*fp, f);
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 02/10] net: sched: avoid usage of tp->q in tcf_classify
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
Use block index in the messages instead.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
net/sched/cls_api.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index dce4c3d..de7791c 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -665,8 +665,9 @@ int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
#ifdef CONFIG_NET_CLS_ACT
reset:
if (unlikely(limit++ >= max_reclassify_loop)) {
- net_notice_ratelimited("%s: reclassify loop, rule prio %u, protocol %02x\n",
- tp->q->ops->id, tp->prio & 0xffff,
+ net_notice_ratelimited("%u: reclassify loop, rule prio %u, protocol %02x\n",
+ tp->chain->block->index,
+ tp->prio & 0xffff,
ntohs(tp->protocol));
return TC_ACT_SHOT;
}
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 01/10] net: sched: introduce support for multiple filter chain pointers registration
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
In-Reply-To: <20171213151038.29665-1-jiri@resnulli.us>
From: Jiri Pirko <jiri@mellanox.com>
So far, there was possible only to register a single filter chain
pointer to block->chain[0]. However, when the blocks will get shareable,
we need to allow multiple filter chain pointers registration.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v2->v3:
- rebased on top of the current net-next
---
include/net/pkt_cls.h | 3 +
include/net/sch_generic.h | 5 +-
net/sched/cls_api.c | 225 ++++++++++++++++++++++++++++++++++++++++------
3 files changed, 206 insertions(+), 27 deletions(-)
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 0105445..c04154f 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -29,6 +29,8 @@ struct tcf_block_ext_info {
enum tcf_block_binder_type binder_type;
tcf_chain_head_change_t *chain_head_change;
void *chain_head_change_priv;
+ bool shareable;
+ u32 block_index;
};
struct tcf_block_cb;
@@ -48,6 +50,7 @@ void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q,
static inline struct Qdisc *tcf_block_q(struct tcf_block *block)
{
+ WARN_ON(block->refcnt != 1);
return block->q;
}
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8f8c0af..9ad1a51 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -269,8 +269,7 @@ typedef void tcf_chain_head_change_t(struct tcf_proto *tp_head, void *priv);
struct tcf_chain {
struct tcf_proto __rcu *filter_chain;
- tcf_chain_head_change_t *chain_head_change;
- void *chain_head_change_priv;
+ struct list_head filter_chain_list;
struct list_head list;
struct tcf_block *block;
u32 index; /* chain index */
@@ -279,6 +278,8 @@ struct tcf_chain {
struct tcf_block {
struct list_head chain_list;
+ u32 index; /* block index for shared blocks */
+ unsigned int refcnt;
struct net *net;
struct Qdisc *q;
struct list_head cb_list;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 5b9b8a6..dce4c3d 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -25,6 +25,7 @@
#include <linux/kmod.h>
#include <linux/err.h>
#include <linux/slab.h>
+#include <linux/idr.h>
#include <net/net_namespace.h>
#include <net/sock.h>
#include <net/netlink.h>
@@ -180,6 +181,12 @@ static void tcf_proto_destroy(struct tcf_proto *tp)
kfree_rcu(tp, rcu);
}
+struct tcf_filter_chain_list_item {
+ struct list_head list;
+ tcf_chain_head_change_t *chain_head_change;
+ void *chain_head_change_priv;
+};
+
static struct tcf_chain *tcf_chain_create(struct tcf_block *block,
u32 chain_index)
{
@@ -188,6 +195,7 @@ static struct tcf_chain *tcf_chain_create(struct tcf_block *block,
chain = kzalloc(sizeof(*chain), GFP_KERNEL);
if (!chain)
return NULL;
+ INIT_LIST_HEAD(&chain->filter_chain_list);
list_add_tail(&chain->list, &block->chain_list);
chain->block = block;
chain->index = chain_index;
@@ -195,12 +203,19 @@ static struct tcf_chain *tcf_chain_create(struct tcf_block *block,
return chain;
}
+static void tcf_chain_head_change_item(struct tcf_filter_chain_list_item *item,
+ struct tcf_proto *tp_head)
+{
+ if (item->chain_head_change)
+ item->chain_head_change(tp_head, item->chain_head_change_priv);
+}
static void tcf_chain_head_change(struct tcf_chain *chain,
struct tcf_proto *tp_head)
{
- if (chain->chain_head_change)
- chain->chain_head_change(tp_head,
- chain->chain_head_change_priv);
+ struct tcf_filter_chain_list_item *item;
+
+ list_for_each_entry(item, &chain->filter_chain_list, list)
+ tcf_chain_head_change_item(item, tp_head);
}
static void tcf_chain_flush(struct tcf_chain *chain)
@@ -281,15 +296,84 @@ static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q,
tcf_block_offload_cmd(block, q, ei, TC_BLOCK_UNBIND);
}
-int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
- struct tcf_block_ext_info *ei)
+static int
+tcf_chain_head_change_cb_add(struct tcf_chain *chain,
+ struct tcf_block_ext_info *ei)
+{
+ struct tcf_filter_chain_list_item *item;
+
+ item = kmalloc(sizeof(*item), GFP_KERNEL);
+ if (!item)
+ return -ENOMEM;
+ item->chain_head_change = ei->chain_head_change;
+ item->chain_head_change_priv = ei->chain_head_change_priv;
+ if (chain->filter_chain)
+ tcf_chain_head_change_item(item, chain->filter_chain);
+ list_add(&item->list, &chain->filter_chain_list);
+ return 0;
+}
+
+static void
+tcf_chain_head_change_cb_del(struct tcf_chain *chain,
+ struct tcf_block_ext_info *ei)
+{
+ struct tcf_filter_chain_list_item *item;
+
+ list_for_each_entry(item, &chain->filter_chain_list, list) {
+ if ((!ei->chain_head_change && !ei->chain_head_change_priv) ||
+ (item->chain_head_change == ei->chain_head_change &&
+ item->chain_head_change_priv == ei->chain_head_change_priv)) {
+ tcf_chain_head_change_item(item, NULL);
+ list_del(&item->list);
+ kfree(item);
+ return;
+ }
+ }
+ WARN_ON(1);
+}
+
+struct tcf_net {
+ struct idr idr;
+};
+
+static unsigned int tcf_net_id;
+
+static int tcf_block_insert(struct tcf_block *block, struct net *net,
+ u32 block_index)
{
- struct tcf_block *block = kzalloc(sizeof(*block), GFP_KERNEL);
+ struct tcf_net *tn = net_generic(net, tcf_net_id);
+ int idr_start;
+ int idr_end;
+ int index;
+
+ if (block_index >= INT_MAX)
+ return -EINVAL;
+ idr_start = block_index ? block_index : 1;
+ idr_end = block_index ? block_index + 1 : INT_MAX;
+
+ index = idr_alloc(&tn->idr, block, idr_start, idr_end, GFP_KERNEL);
+ if (index < 0)
+ return index;
+ block->index = index;
+ return 0;
+}
+
+static void tcf_block_remove(struct tcf_block *block, struct net *net)
+{
+ struct tcf_net *tn = net_generic(net, tcf_net_id);
+
+ idr_remove(&tn->idr, block->index);
+}
+
+static struct tcf_block *tcf_block_create(struct net *net, struct Qdisc *q)
+{
+ struct tcf_block *block;
struct tcf_chain *chain;
int err;
+ block = kzalloc(sizeof(*block), GFP_KERNEL);
if (!block)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
INIT_LIST_HEAD(&block->chain_list);
INIT_LIST_HEAD(&block->cb_list);
@@ -299,17 +383,72 @@ int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
err = -ENOMEM;
goto err_chain_create;
}
- WARN_ON(!ei->chain_head_change);
- chain->chain_head_change = ei->chain_head_change;
- chain->chain_head_change_priv = ei->chain_head_change_priv;
block->net = qdisc_net(q);
+ block->refcnt = 1;
+ block->net = net;
block->q = q;
+ return block;
+
+err_chain_create:
+ kfree(block);
+ return ERR_PTR(err);
+}
+
+static struct tcf_block *tcf_block_lookup(struct net *net, u32 block_index)
+{
+ struct tcf_net *tn = net_generic(net, tcf_net_id);
+
+ return idr_find(&tn->idr, block_index);
+}
+
+static struct tcf_chain *tcf_block_chain_zero(struct tcf_block *block)
+{
+ return list_first_entry(&block->chain_list, struct tcf_chain, list);
+}
+
+int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q,
+ struct tcf_block_ext_info *ei)
+{
+ struct net *net = qdisc_net(q);
+ struct tcf_block *block = NULL;
+ bool created = false;
+ int err;
+
+ if (ei->shareable) {
+ block = tcf_block_lookup(net, ei->block_index);
+ if (block)
+ block->refcnt++;
+ }
+
+ if (!block) {
+ block = tcf_block_create(net, q);
+ if (IS_ERR(block))
+ return PTR_ERR(block);
+ created = true;
+ if (ei->shareable) {
+ err = tcf_block_insert(block, net, ei->block_index);
+ if (err)
+ goto err_block_insert;
+ }
+ }
+
+ err = tcf_chain_head_change_cb_add(tcf_block_chain_zero(block), ei);
+ if (err)
+ goto err_chain_head_change_cb_add;
+
tcf_block_offload_bind(block, q, ei);
*p_block = block;
return 0;
-err_chain_create:
- kfree(block);
+err_chain_head_change_cb_add:
+ if (created) {
+ if (ei->shareable)
+ tcf_block_remove(block, net);
+err_block_insert:
+ kfree(block);
+ } else {
+ block->refcnt--;
+ }
return err;
}
EXPORT_SYMBOL(tcf_block_get_ext);
@@ -342,24 +481,32 @@ void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q,
{
struct tcf_chain *chain, *tmp;
- /* Hold a refcnt for all chains, so that they don't disappear
- * while we are iterating.
- */
- list_for_each_entry(chain, &block->chain_list, list)
- tcf_chain_hold(chain);
+ tcf_chain_head_change_cb_del(tcf_block_chain_zero(block), ei);
- list_for_each_entry(chain, &block->chain_list, list)
- tcf_chain_flush(chain);
+ if (--block->refcnt == 0) {
+ if (ei->shareable)
+ tcf_block_remove(block, block->net);
+
+ /* Hold a refcnt for all chains, except 0, so that they
+ * don't disappear while we are iterating.
+ */
+ list_for_each_entry(chain, &block->chain_list, list)
+ tcf_chain_hold(chain);
+
+ list_for_each_entry(chain, &block->chain_list, list)
+ tcf_chain_flush(chain);
+ }
tcf_block_offload_unbind(block, q, ei);
- /* At this point, all the chains should have refcnt >= 1. */
- list_for_each_entry_safe(chain, tmp, &block->chain_list, list)
- tcf_chain_put(chain);
+ if (block->refcnt == 0) {
+ /* At this point, all the chains should have refcnt >= 1. */
+ list_for_each_entry_safe(chain, tmp, &block->chain_list, list)
+ tcf_chain_put(chain);
- /* Finally, put chain 0 and allow block to be freed. */
- chain = list_first_entry(&block->chain_list, struct tcf_chain, list);
- tcf_chain_put(chain);
+ /* Finally, put chain 0 and allow block to be freed. */
+ tcf_chain_put(tcf_block_chain_zero(block));
+ }
}
EXPORT_SYMBOL(tcf_block_put_ext);
@@ -1246,12 +1393,40 @@ int tc_setup_cb_call(struct tcf_block *block, struct tcf_exts *exts,
}
EXPORT_SYMBOL(tc_setup_cb_call);
+static __net_init int tcf_net_init(struct net *net)
+{
+ struct tcf_net *tn = net_generic(net, tcf_net_id);
+
+ idr_init(&tn->idr);
+ return 0;
+}
+
+static void __net_exit tcf_net_exit(struct net *net)
+{
+ struct tcf_net *tn = net_generic(net, tcf_net_id);
+
+ idr_destroy(&tn->idr);
+}
+
+static struct pernet_operations tcf_net_ops = {
+ .init = tcf_net_init,
+ .exit = tcf_net_exit,
+ .id = &tcf_net_id,
+ .size = sizeof(struct tcf_net),
+};
+
static int __init tc_filter_init(void)
{
+ int err;
+
tc_filter_wq = alloc_ordered_workqueue("tc_filter_workqueue", 0);
if (!tc_filter_wq)
return -ENOMEM;
+ err = register_pernet_subsys(&tcf_net_ops);
+ if (err)
+ return err;
+
rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_ctl_tfilter, NULL, 0);
rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_ctl_tfilter, NULL, 0);
rtnl_register(PF_UNSPEC, RTM_GETTFILTER, tc_ctl_tfilter,
--
2.9.5
^ permalink raw reply related
* [patch net-next v3 00/10] net: sched: allow qdiscs to share filter block instances
From: Jiri Pirko @ 2017-12-13 15:10 UTC (permalink / raw)
To: netdev
Cc: davem, jhs, xiyou.wangcong, mlxsw, andrew, vivien.didelot,
f.fainelli, michael.chan, ganeshgr, saeedm, matanb, leonro,
idosch, jakub.kicinski, simon.horman, pieter.jansenvanvuuren,
john.hurley, alexander.h.duyck, ogerlitz, john.fastabend, daniel
From: Jiri Pirko <jiri@mellanox.com>
Currently the filters added to qdiscs are independent. So for example if you
have 2 netdevices and you create ingress qdisc on both and you want to add
identical filter rules both, you need to add them twice. This patchset
makes this easier and mainly saves resources allowing to share all filters
within a qdisc - I call it a "filter block". Also this helps to save
resources when we do offload to hw for example to expensive TCAM.
So back to the example. First, we create 2 qdiscs. Both will share
block number 22. "22" is just an identification. If we don't pass any
block number, a new one will be generated by kernel:
$ tc qdisc add dev ens7 ingress block 22
^^^^^^^^
$ tc qdisc add dev ens8 ingress block 22
^^^^^^^^
Now if we list the qdiscs, we will see the block index in the output:
$ tc qdisc
qdisc ingress ffff: dev ens7 parent ffff:fff1 block 22
qdisc ingress ffff: dev ens8 parent ffff:fff1 block 22
To make is more visual, the situation looks like this:
ens7 ingress qdisc ens7 ingress qdisc
| |
| |
+----------> block 22 <----------+
Unlimited number of qdiscs may share the same block.
Now we can add filter to any of qdiscs sharing the same block:
$ tc filter add dev ens7 ingress protocol ip pref 25 flower dst_ip 192.168.0.0/16 action drop
We will see the same output if we list filters for ens7 and ens8, including stats:
$ tc -s filter show dev ens7 ingress
filter protocol ip pref 25 flower chain 0
filter protocol ip pref 25 flower chain 0 handle 0x1
eth_type ipv4
dst_ip 192.168.0.0/16
not_in_hw
action order 1: gact action drop
random type none pass val 0
index 1 ref 1 bind 1 installed 39 sec used 2 sec
Action statistics:
Sent 3108 bytes 37 pkt (dropped 37, overlimits 0 requeues 0)
backlog 0b 0p requeues 0
$ tc -s filter show dev ens8 ingress
filter protocol ip pref 25 flower chain 0
filter protocol ip pref 25 flower chain 0 handle 0x1
eth_type ipv4
dst_ip 192.168.0.0/16
not_in_hw
action order 1: gact action drop
random type none pass val 0
index 1 ref 1 bind 1 installed 40 sec used 3 sec
Action statistics:
Sent 3108 bytes 37 pkt (dropped 37, overlimits 0 requeues 0)
backlog 0b 0p requeues 0
---
v2->v3:
- removed original patch 1, removing tp->q cls_bpf dependency. Fixed by
Jakub in the meantime.
- patch 1:
- rebased on top of the current net-next
- patch 5:
- new patch
- patch 6:
- removed "p_" prefix from block index function args
- patch 9:
- add tc offload feature handling
Jiri Pirko (10):
net: sched: introduce support for multiple filter chain pointers
registration
net: sched: avoid usage of tp->q in tcf_classify
net: sched: introduce block mechanism to handle netif_keep_dst calls
net: sched: remove classid and q fields from tcf_proto
net: sched: keep track of offloaded filters and check tc offload
feature
net: sched: allow ingress and clsact qdiscs to share filter blocks
mlxsw: spectrum_acl: Reshuffle code around
mlxsw_sp_acl_ruleset_create/destroy
mlxsw: spectrum_acl: Don't store netdev and ingress for ruleset unbind
mlxsw: spectrum_acl: Implement TC block sharing
mlxsw: spectrum_acl: Pass mlxsw_sp_port down to ruleset bind/unbind
ops
drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 182 +++++++++--
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 44 ++-
drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c | 302 +++++++++++++----
.../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 44 ++-
.../net/ethernet/mellanox/mlxsw/spectrum_flower.c | 41 +--
include/net/pkt_cls.h | 4 +
include/net/sch_generic.h | 26 +-
include/uapi/linux/pkt_sched.h | 11 +
net/sched/cls_api.c | 359 ++++++++++++++++++---
net/sched/cls_bpf.c | 8 +-
net/sched/cls_flow.c | 2 +-
net/sched/cls_flower.c | 3 +-
net/sched/cls_matchall.c | 3 +-
net/sched/cls_route.c | 2 +-
net/sched/cls_u32.c | 13 +-
net/sched/sch_ingress.c | 89 ++++-
16 files changed, 913 insertions(+), 220 deletions(-)
--
2.9.5
^ permalink raw reply
* Re: [PATCH net] tcp: fix potential underestimation on rcv_rtt
From: Neal Cardwell @ 2017-12-13 15:04 UTC (permalink / raw)
To: Wei Wang; +Cc: David Miller, Netdev, Eric Dumazet
In-Reply-To: <20171213002858.74188-1-tracywwnj@gmail.com>
On Tue, Dec 12, 2017 at 7:28 PM, Wei Wang <weiwan@google.com> wrote:
> From: Wei Wang <weiwan@google.com>
>
> When ms timestamp is used, current logic uses 1us in
> tcp_rcv_rtt_update() when the real rcv_rtt is within 1 - 999us.
> This could cause rcv_rtt underestimation.
> Fix it by always using a min value of 1ms if ms timestamp is used.
>
> Fixes: 645f4c6f2ebd ("tcp: switch rcv_rtt_est and rcvq_space to high
> resolution timestamps")
>
> Signed-off-by: Wei Wang <weiwan@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> ---
Acked-by: Neal Cardwell <ncardwell@google.com>
Thanks!
neal
^ permalink raw reply
* Re: AF_VSOCK connection refused errno
From: Stefan Hajnoczi @ 2017-12-13 14:58 UTC (permalink / raw)
To: Jorgen S. Hansen; +Cc: Dexuan Cui, netdev@vger.kernel.org
In-Reply-To: <7CCDA4D3-7683-4E5C-95ED-97B2FC9DFF33@vmware.com>
[-- Attachment #1: Type: text/plain, Size: 1226 bytes --]
On Wed, Dec 13, 2017 at 10:28:30AM +0000, Jorgen S. Hansen wrote:
>
> > On Dec 12, 2017, at 4:53 PM, Stefan Hajnoczi <stefanha@redhat.com> wrote:
> >
> > When connect(2) fails because the peer is not listening the virtio vsock
> > transport returns ECONNRESET. I believe the VMCI transport does the
> > same (based on code inspection).
> >
> > Jorgen: Can you confirm this VMCI transport behavior?
>
> Yes, that is correct.
>
> > I'd like to change to ECONNREFUSED for all transports because developers
> > will be surprised when they get ECONNRESET. It makes porting AF_INET
> > code harder.
> >
> > On the other hand, it may be too late to fix this if there userspace
> > applications that rely on ECONNRESET? I'm not aware of any such
> > applications myself.
>
> In the past, I’ve explained to customers that an ECONNRESET error on connect
> can indicate that the peer isn’t listening on the dest address. Whether they went
> and used that information isn’t clear, but changing this behavior now would
> risk breaking applications. While it is unfortunate that we deviate from INET in
> this case, I would prefer it to stay as is.
That's fine. Thanks for confirming.
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]
^ permalink raw reply
* [PATCH 5/5] VSOCK: add AF_VSOCK test cases
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
In-Reply-To: <20171213144911.6428-1-stefanha@redhat.com>
The vsock_test.c program runs a test suite of AF_VSOCK test cases.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
tools/testing/vsock/Makefile | 5 +-
tools/testing/vsock/vsock_test.c | 321 +++++++++++++++++++++++++++++++++++++++
tools/testing/vsock/.gitignore | 1 +
tools/testing/vsock/README | 1 +
4 files changed, 326 insertions(+), 2 deletions(-)
create mode 100644 tools/testing/vsock/vsock_test.c
diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index cf0650007fa8..9efb9010df2e 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -1,9 +1,10 @@
all: test
-test: vsock_diag_test
+test: vsock_test vsock_diag_test
+vsock_test: vsock_test.o timeout.o control.o util.o
vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o
CFLAGS += -g -O2 -Werror -Wall -I. -I../../include/uapi -I../../include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
.PHONY: all test clean
clean:
- ${RM} *.o *.d vsock_diag_test
+ ${RM} *.o *.d vsock_test vsock_diag_test
-include *.d
diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c
new file mode 100644
index 000000000000..c1fc376eea00
--- /dev/null
+++ b/tools/testing/vsock/vsock_test.c
@@ -0,0 +1,321 @@
+/*
+ * vsock_test - vsock.ko test suite
+ *
+ * Copyright (C) 2017 Red Hat, Inc.
+ *
+ * Author: Stefan Hajnoczi <stefanha@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <getopt.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+#include <sys/socket.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <linux/list.h>
+#include <linux/net.h>
+#include <linux/netlink.h>
+#include <linux/sock_diag.h>
+#include <netinet/tcp.h>
+
+#include "timeout.h"
+#include "control.h"
+#include "util.h"
+
+static void test_stream_connection_reset(const struct test_opts *opts)
+{
+ union {
+ struct sockaddr sa;
+ struct sockaddr_vm svm;
+ } addr = {
+ .svm = {
+ .svm_family = AF_VSOCK,
+ .svm_port = 1234,
+ .svm_cid = opts->peer_cid,
+ },
+ };
+ int ret;
+ int fd;
+
+ fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+ timeout_begin(TIMEOUT);
+ do {
+ ret = connect(fd, &addr.sa, sizeof(addr.svm));
+ timeout_check("connect");
+ } while (ret < 0 && errno == EINTR);
+ timeout_end();
+
+ if (ret != -1) {
+ fprintf(stderr, "expected connect(2) failure, got %d\n", ret);
+ exit(EXIT_FAILURE);
+ }
+ if (errno != ECONNRESET) {
+ fprintf(stderr, "unexpected connect(2) errno %d\n", errno);
+ exit(EXIT_FAILURE);
+ }
+
+ close(fd);
+}
+
+static void test_stream_client_close_client(const struct test_opts *opts)
+{
+ int fd;
+
+ fd = vsock_stream_connect(opts->peer_cid, 1234);
+ if (fd < 0) {
+ perror("connect");
+ exit(EXIT_FAILURE);
+ }
+
+ send_byte(fd, 1);
+ close(fd);
+ control_writeln("CLOSED");
+}
+
+static void test_stream_client_close_server(const struct test_opts *opts)
+{
+ int fd;
+
+ fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+ if (fd < 0) {
+ perror("accept");
+ exit(EXIT_FAILURE);
+ }
+
+ control_expectln("CLOSED");
+
+ send_byte(fd, -EPIPE);
+ recv_byte(fd, 1);
+ recv_byte(fd, 0);
+ close(fd);
+}
+
+static void test_stream_server_close_client(const struct test_opts *opts)
+{
+ int fd;
+
+ fd = vsock_stream_connect(opts->peer_cid, 1234);
+ if (fd < 0) {
+ perror("connect");
+ exit(EXIT_FAILURE);
+ }
+
+ control_expectln("CLOSED");
+
+ send_byte(fd, -EPIPE);
+ recv_byte(fd, 1);
+ recv_byte(fd, 0);
+ close(fd);
+}
+
+static void test_stream_server_close_server(const struct test_opts *opts)
+{
+ int fd;
+
+ fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+ if (fd < 0) {
+ perror("accept");
+ exit(EXIT_FAILURE);
+ }
+
+ send_byte(fd, 1);
+ close(fd);
+ control_writeln("CLOSED");
+}
+
+#define MULTICONN_NFDS 1000
+
+static void test_stream_multiconn_client(const struct test_opts *opts)
+{
+ int fds[MULTICONN_NFDS];
+ int i;
+
+ for (i = 0; i < MULTICONN_NFDS; i++) {
+ fds[i] = vsock_stream_connect(opts->peer_cid, 1234);
+ if (fds[i] < 0) {
+ perror("connect");
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ for (i = 0; i < MULTICONN_NFDS; i++) {
+ if (i % 1)
+ recv_byte(fds[i], 1);
+ else
+ send_byte(fds[i], 1);
+ }
+
+ for (i = 0; i < MULTICONN_NFDS; i++)
+ close(fds[i]);
+}
+
+static void test_stream_multiconn_server(const struct test_opts *opts)
+{
+ int fds[MULTICONN_NFDS];
+ int i;
+
+ for (i = 0; i < MULTICONN_NFDS; i++) {
+ fds[i] = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
+ if (fds[i] < 0) {
+ perror("accept");
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ for (i = 0; i < MULTICONN_NFDS; i++) {
+ if (i % 1)
+ send_byte(fds[i], 1);
+ else
+ recv_byte(fds[i], 1);
+ }
+
+ for (i = 0; i < MULTICONN_NFDS; i++)
+ close(fds[i]);
+}
+
+static struct test_case test_cases[] = {
+ {
+ .name = "SOCK_STREAM connection reset",
+ .run_client = test_stream_connection_reset,
+ },
+ {
+ .name = "SOCK_STREAM client close",
+ .run_client = test_stream_client_close_client,
+ .run_server = test_stream_client_close_server,
+ },
+ {
+ .name = "SOCK_STREAM server close",
+ .run_client = test_stream_server_close_client,
+ .run_server = test_stream_server_close_server,
+ },
+ {
+ .name = "SOCK_STREAM multiple connections",
+ .run_client = test_stream_multiconn_client,
+ .run_server = test_stream_multiconn_server,
+ },
+ {},
+};
+
+static const char optstring[] = "";
+static const struct option longopts[] = {
+ {
+ .name = "control-host",
+ .has_arg = required_argument,
+ .val = 'H',
+ },
+ {
+ .name = "control-port",
+ .has_arg = required_argument,
+ .val = 'P',
+ },
+ {
+ .name = "mode",
+ .has_arg = required_argument,
+ .val = 'm',
+ },
+ {
+ .name = "peer-cid",
+ .has_arg = required_argument,
+ .val = 'p',
+ },
+ {
+ .name = "help",
+ .has_arg = no_argument,
+ .val = '?',
+ },
+ {},
+};
+
+static void usage(void)
+{
+ fprintf(stderr, "Usage: vsock_test [--help] [--control-host=<host>] --control-port=<port> --mode=client|server --peer-cid=<cid>\n"
+ "\n"
+ " Server: vsock_test --control-port=1234 --mode=server --peer-cid=3\n"
+ " Client: vsock_test --control-host=192.168.0.1 --control-port=1234 --mode=client --peer-cid=2\n"
+ "\n"
+ "Run vsock.ko tests. Must be launched in both guest\n"
+ "and host. One side must use --mode=client and\n"
+ "the other side must use --mode=server.\n"
+ "\n"
+ "A TCP control socket connection is used to coordinate tests\n"
+ "between the client and the server. The server requires a\n"
+ "listen address and the client requires an address to\n"
+ "connect to.\n"
+ "\n"
+ "The CID of the other side must be given with --peer-cid=<cid>.\n");
+ exit(EXIT_FAILURE);
+}
+
+int main(int argc, char **argv)
+{
+ const char *control_host = NULL;
+ const char *control_port = NULL;
+ struct test_opts opts = {
+ .mode = TEST_MODE_UNSET,
+ .peer_cid = VMADDR_CID_ANY,
+ };
+
+ init_signals();
+
+ for (;;) {
+ int opt = getopt_long(argc, argv, optstring, longopts, NULL);
+
+ if (opt == -1)
+ break;
+
+ switch (opt) {
+ case 'H':
+ control_host = optarg;
+ break;
+ case 'm':
+ if (strcmp(optarg, "client") == 0)
+ opts.mode = TEST_MODE_CLIENT;
+ else if (strcmp(optarg, "server") == 0)
+ opts.mode = TEST_MODE_SERVER;
+ else {
+ fprintf(stderr, "--mode must be \"client\" or \"server\"\n");
+ return EXIT_FAILURE;
+ }
+ break;
+ case 'p':
+ opts.peer_cid = parse_cid(optarg);
+ break;
+ case 'P':
+ control_port = optarg;
+ break;
+ case '?':
+ default:
+ usage();
+ }
+ }
+
+ if (!control_port)
+ usage();
+ if (opts.mode == TEST_MODE_UNSET)
+ usage();
+ if (opts.peer_cid == VMADDR_CID_ANY)
+ usage();
+
+ if (!control_host) {
+ if (opts.mode != TEST_MODE_SERVER)
+ usage();
+ control_host = "0.0.0.0";
+ }
+
+ control_init(control_host, control_port,
+ opts.mode == TEST_MODE_SERVER);
+
+ run_tests(test_cases, &opts);
+
+ control_cleanup();
+ return EXIT_SUCCESS;
+}
diff --git a/tools/testing/vsock/.gitignore b/tools/testing/vsock/.gitignore
index dc5f11faf530..7f7a2ccc30c4 100644
--- a/tools/testing/vsock/.gitignore
+++ b/tools/testing/vsock/.gitignore
@@ -1,2 +1,3 @@
*.d
+vsock_test
vsock_diag_test
diff --git a/tools/testing/vsock/README b/tools/testing/vsock/README
index 2cc6d7302db6..3014d9e4959a 100644
--- a/tools/testing/vsock/README
+++ b/tools/testing/vsock/README
@@ -5,6 +5,7 @@ Hyper-V.
The following tests are available:
+ * vsock_test - core AF_VSOCK socket functionality
* vsock_diag_test - vsock_diag.ko module for listing open sockets
The following prerequisite steps are not automated and must be performed prior
--
2.14.3
^ permalink raw reply related
* [PATCH 4/5] VSOCK: add send_byte()/recv_byte() test utilities
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
In-Reply-To: <20171213144911.6428-1-stefanha@redhat.com>
Test cases will want to transfer data. This patch adds utility
functions to do this.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
tools/testing/vsock/util.h | 2 +
tools/testing/vsock/util.c | 99 ++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 101 insertions(+)
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index 3e6971f15caa..cf43d42591bc 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -35,6 +35,8 @@ unsigned int parse_cid(const char *str);
int vsock_stream_connect(unsigned int cid, unsigned int port);
int vsock_stream_accept(unsigned int cid, unsigned int port,
struct sockaddr_vm *clientaddrp);
+void send_byte(int fd, int expected_ret);
+void recv_byte(int fd, int expected_ret);
void run_tests(const struct test_case *test_cases,
const struct test_opts *opts);
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index 2be923fe9922..e28085a2a429 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -13,6 +13,7 @@
#include <errno.h>
#include <stdio.h>
+#include <stdint.h>
#include <stdlib.h>
#include <signal.h>
#include <unistd.h>
@@ -153,6 +154,104 @@ int vsock_stream_accept(unsigned int cid, unsigned int port,
return client_fd;
}
+/* Transmit one byte and check the return value.
+ *
+ * expected_ret:
+ * <0 Negative errno (for testing errors)
+ * 0 End-of-file
+ * 1 Success
+ */
+void send_byte(int fd, int expected_ret)
+{
+ const uint8_t byte = 'A';
+ ssize_t nwritten;
+
+ timeout_begin(TIMEOUT);
+ do {
+ nwritten = write(fd, &byte, sizeof(byte));
+ timeout_check("write");
+ } while (nwritten < 0 && errno == EINTR);
+ timeout_end();
+
+ if (expected_ret < 0) {
+ if (nwritten != -1) {
+ fprintf(stderr, "bogus write(2) return value %zd\n",
+ nwritten);
+ exit(EXIT_FAILURE);
+ }
+ if (errno != -expected_ret) {
+ perror("write");
+ exit(EXIT_FAILURE);
+ }
+ return;
+ }
+
+ if (nwritten < 0) {
+ perror("write");
+ exit(EXIT_FAILURE);
+ }
+ if (nwritten == 0) {
+ if (expected_ret == 0)
+ return;
+
+ fprintf(stderr, "unexpected EOF while sending byte\n");
+ exit(EXIT_FAILURE);
+ }
+ if (nwritten != sizeof(byte)) {
+ fprintf(stderr, "bogus write(2) return value %zd\n", nwritten);
+ exit(EXIT_FAILURE);
+ }
+}
+
+/* Receive one byte and check the return value.
+ *
+ * expected_ret:
+ * <0 Negative errno (for testing errors)
+ * 0 End-of-file
+ * 1 Success
+ */
+void recv_byte(int fd, int expected_ret)
+{
+ uint8_t byte;
+ ssize_t nread;
+
+ timeout_begin(TIMEOUT);
+ do {
+ nread = read(fd, &byte, sizeof(byte));
+ timeout_check("read");
+ } while (nread < 0 && errno == EINTR);
+ timeout_end();
+
+ if (expected_ret < 0) {
+ if (nread != -1) {
+ fprintf(stderr, "bogus read(2) return value %zd\n",
+ nread);
+ exit(EXIT_FAILURE);
+ }
+ if (errno != -expected_ret) {
+ perror("read");
+ exit(EXIT_FAILURE);
+ }
+ return;
+ }
+
+ if (nread < 0) {
+ perror("read");
+ exit(EXIT_FAILURE);
+ }
+ if (nread == 0) {
+ if (expected_ret == 0)
+ return;
+
+ fprintf(stderr, "unexpected EOF while receiving byte\n");
+ exit(EXIT_FAILURE);
+ }
+ if (nread != sizeof(byte)) {
+ fprintf(stderr, "bogus read(2) return value %zd\n", nread);
+ exit(EXIT_FAILURE);
+ }
+}
+
/* Run test cases. The program terminates if a failure occurs. */
void run_tests(const struct test_case *test_cases,
const struct test_opts *opts)
--
2.14.3
^ permalink raw reply related
* [PATCH 3/5] VSOCK: add full barrier between test cases
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
In-Reply-To: <20171213144911.6428-1-stefanha@redhat.com>
See code comment for details.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
tools/testing/vsock/util.c | 18 ++++++++++++++++--
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index 75a78f295b37..2be923fe9922 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -165,10 +165,24 @@ void run_tests(const struct test_case *test_cases,
printf("%s...", test_cases[i].name);
fflush(stdout);
- if (opts->mode == TEST_MODE_CLIENT)
+ if (opts->mode == TEST_MODE_CLIENT) {
+ /* Full barrier before executing the next test. This
+ * ensures that client and server are executing the
+ * same test case. In particular, it means whoever is
+ * faster will not see the peer still executing the
+ * last test. This is important because port numbers
+ * can be used by multiple test cases.
+ */
+ control_expectln("NEXT");
+ control_writeln("NEXT");
+
run = test_cases[i].run_client;
- else
+ } else {
+ control_writeln("NEXT");
+ control_expectln("NEXT");
+
run = test_cases[i].run_server;
+ }
if (run)
run(opts);
--
2.14.3
^ permalink raw reply related
* [PATCH 2/5] VSOCK: extract connect/accept functions from vsock_diag_test.c
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
In-Reply-To: <20171213144911.6428-1-stefanha@redhat.com>
Many test cases will need to connect to the server or accept incoming
connections. This patch extracts these operations into utility
functions that can be reused.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
tools/testing/vsock/util.h | 6 ++
tools/testing/vsock/util.c | 108 ++++++++++++++++++++++++++++++++++
tools/testing/vsock/vsock_diag_test.c | 81 ++-----------------------
3 files changed, 119 insertions(+), 76 deletions(-)
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
index a51e17578ccf..3e6971f15caa 100644
--- a/tools/testing/vsock/util.h
+++ b/tools/testing/vsock/util.h
@@ -1,6 +1,9 @@
#ifndef UTIL_H
#define UTIL_H
+#include <sys/socket.h>
+#include "../../../include/uapi/linux/vm_sockets.h"
+
/* Tests can either run as the client or the server */
enum test_mode {
TEST_MODE_UNSET,
@@ -29,6 +32,9 @@ struct test_case {
void init_signals(void);
unsigned int parse_cid(const char *str);
+int vsock_stream_connect(unsigned int cid, unsigned int port);
+int vsock_stream_accept(unsigned int cid, unsigned int port,
+ struct sockaddr_vm *clientaddrp);
void run_tests(const struct test_case *test_cases,
const struct test_opts *opts);
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
index 2567abd90153..75a78f295b37 100644
--- a/tools/testing/vsock/util.c
+++ b/tools/testing/vsock/util.c
@@ -15,8 +15,10 @@
#include <stdio.h>
#include <stdlib.h>
#include <signal.h>
+#include <unistd.h>
#include "timeout.h"
+#include "control.h"
#include "util.h"
/* Install signal handlers */
@@ -45,6 +47,112 @@ unsigned int parse_cid(const char *str)
return n;
}
+/* Connect to <cid, port> and return the file descriptor. */
+int vsock_stream_connect(unsigned int cid, unsigned int port)
+{
+ union {
+ struct sockaddr sa;
+ struct sockaddr_vm svm;
+ } addr = {
+ .svm = {
+ .svm_family = AF_VSOCK,
+ .svm_port = port,
+ .svm_cid = cid,
+ },
+ };
+ int ret;
+ int fd;
+
+ control_expectln("LISTENING");
+
+ fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+ timeout_begin(TIMEOUT);
+ do {
+ ret = connect(fd, &addr.sa, sizeof(addr.svm));
+ timeout_check("connect");
+ } while (ret < 0 && errno == EINTR);
+ timeout_end();
+
+ if (ret < 0) {
+ int old_errno = errno;
+
+ close(fd);
+ fd = -1;
+ errno = old_errno;
+ }
+ return fd;
+}
+
+/* Listen on <cid, port> and return the first incoming connection. The remote
+ * address is stored to clientaddrp. clientaddrp may be NULL.
+ */
+int vsock_stream_accept(unsigned int cid, unsigned int port,
+ struct sockaddr_vm *clientaddrp)
+{
+ union {
+ struct sockaddr sa;
+ struct sockaddr_vm svm;
+ } addr = {
+ .svm = {
+ .svm_family = AF_VSOCK,
+ .svm_port = port,
+ .svm_cid = cid,
+ },
+ };
+ union {
+ struct sockaddr sa;
+ struct sockaddr_vm svm;
+ } clientaddr;
+ socklen_t clientaddr_len = sizeof(clientaddr.svm);
+ int fd;
+ int client_fd;
+ int old_errno;
+
+ fd = socket(AF_VSOCK, SOCK_STREAM, 0);
+
+ if (bind(fd, &addr.sa, sizeof(addr.svm)) < 0) {
+ perror("bind");
+ exit(EXIT_FAILURE);
+ }
+
+ if (listen(fd, 1) < 0) {
+ perror("listen");
+ exit(EXIT_FAILURE);
+ }
+
+ control_writeln("LISTENING");
+
+ timeout_begin(TIMEOUT);
+ do {
+ client_fd = accept(fd, &clientaddr.sa, &clientaddr_len);
+ timeout_check("accept");
+ } while (client_fd < 0 && errno == EINTR);
+ timeout_end();
+
+ old_errno = errno;
+ close(fd);
+ errno = old_errno;
+
+ if (client_fd < 0)
+ return client_fd;
+
+ if (clientaddr_len != sizeof(clientaddr.svm)) {
+ fprintf(stderr, "unexpected addrlen from accept(2), %zu\n",
+ (size_t)clientaddr_len);
+ exit(EXIT_FAILURE);
+ }
+ if (clientaddr.sa.sa_family != AF_VSOCK) {
+ fprintf(stderr, "expected AF_VSOCK from accept(2), got %d\n",
+ clientaddr.sa.sa_family);
+ exit(EXIT_FAILURE);
+ }
+
+ if (clientaddrp)
+ *clientaddrp = clientaddr.svm;
+ return client_fd;
+}
+
/* Run test cases. The program terminates if a failure occurs. */
void run_tests(const struct test_case *test_cases,
const struct test_opts *opts)
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index f1ace5d96e00..654bbe2fc211 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -17,7 +17,6 @@
#include <string.h>
#include <errno.h>
#include <unistd.h>
-#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <linux/list.h>
@@ -26,7 +25,6 @@
#include <linux/sock_diag.h>
#include <netinet/tcp.h>
-#include "../../../include/uapi/linux/vm_sockets.h"
#include "../../../include/uapi/linux/vm_sockets_diag.h"
#include "timeout.h"
@@ -383,33 +381,12 @@ static void test_listen_socket_server(const struct test_opts *opts)
static void test_connect_client(const struct test_opts *opts)
{
- union {
- struct sockaddr sa;
- struct sockaddr_vm svm;
- } addr = {
- .svm = {
- .svm_family = AF_VSOCK,
- .svm_port = 1234,
- .svm_cid = opts->peer_cid,
- },
- };
int fd;
- int ret;
LIST_HEAD(sockets);
struct vsock_stat *st;
- control_expectln("LISTENING");
-
- fd = socket(AF_VSOCK, SOCK_STREAM, 0);
-
- timeout_begin(TIMEOUT);
- do {
- ret = connect(fd, &addr.sa, sizeof(addr.svm));
- timeout_check("connect");
- } while (ret < 0 && errno == EINTR);
- timeout_end();
-
- if (ret < 0) {
+ fd = vsock_stream_connect(opts->peer_cid, 1234);
+ if (fd < 0) {
perror("connect");
exit(EXIT_FAILURE);
}
@@ -429,66 +406,19 @@ static void test_connect_client(const struct test_opts *opts)
static void test_connect_server(const struct test_opts *opts)
{
- union {
- struct sockaddr sa;
- struct sockaddr_vm svm;
- } addr = {
- .svm = {
- .svm_family = AF_VSOCK,
- .svm_port = 1234,
- .svm_cid = VMADDR_CID_ANY,
- },
- };
- union {
- struct sockaddr sa;
- struct sockaddr_vm svm;
- } clientaddr;
- socklen_t clientaddr_len = sizeof(clientaddr.svm);
- LIST_HEAD(sockets);
struct vsock_stat *st;
- int fd;
+ LIST_HEAD(sockets);
int client_fd;
- fd = socket(AF_VSOCK, SOCK_STREAM, 0);
-
- if (bind(fd, &addr.sa, sizeof(addr.svm)) < 0) {
- perror("bind");
- exit(EXIT_FAILURE);
- }
-
- if (listen(fd, 1) < 0) {
- perror("listen");
- exit(EXIT_FAILURE);
- }
-
- control_writeln("LISTENING");
-
- timeout_begin(TIMEOUT);
- do {
- client_fd = accept(fd, &clientaddr.sa, &clientaddr_len);
- timeout_check("accept");
- } while (client_fd < 0 && errno == EINTR);
- timeout_end();
-
+ client_fd = vsock_stream_accept(VMADDR_CID_ANY, 1234, NULL);
if (client_fd < 0) {
perror("accept");
exit(EXIT_FAILURE);
}
- if (clientaddr.sa.sa_family != AF_VSOCK) {
- fprintf(stderr, "expected AF_VSOCK from accept(2), got %d\n",
- clientaddr.sa.sa_family);
- exit(EXIT_FAILURE);
- }
- if (clientaddr.svm.svm_cid != opts->peer_cid) {
- fprintf(stderr, "expected peer CID %u from accept(2), got %u\n",
- opts->peer_cid, clientaddr.svm.svm_cid);
- exit(EXIT_FAILURE);
- }
read_vsock_stat(&sockets);
- check_num_sockets(&sockets, 2);
- find_vsock_stat(&sockets, fd);
+ check_num_sockets(&sockets, 1);
st = find_vsock_stat(&sockets, client_fd);
check_socket_state(st, TCP_ESTABLISHED);
@@ -496,7 +426,6 @@ static void test_connect_server(const struct test_opts *opts)
control_expectln("DONE");
close(client_fd);
- close(fd);
free_sock_stat(&sockets);
}
--
2.14.3
^ permalink raw reply related
* [PATCH 1/5] VSOCK: extract utility functions from vsock_diag_test.c
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
In-Reply-To: <20171213144911.6428-1-stefanha@redhat.com>
Move useful functions into a separate file in preparation for more vsock
test programs.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
tools/testing/vsock/Makefile | 2 +-
tools/testing/vsock/util.h | 35 +++++++++++++
tools/testing/vsock/util.c | 70 ++++++++++++++++++++++++++
tools/testing/vsock/vsock_diag_test.c | 92 +++++++++--------------------------
4 files changed, 128 insertions(+), 71 deletions(-)
create mode 100644 tools/testing/vsock/util.h
create mode 100644 tools/testing/vsock/util.c
diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index 66ba0924194d..cf0650007fa8 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -1,6 +1,6 @@
all: test
test: vsock_diag_test
-vsock_diag_test: vsock_diag_test.o timeout.o control.o
+vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o
CFLAGS += -g -O2 -Werror -Wall -I. -I../../include/uapi -I../../include -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -D_GNU_SOURCE
.PHONY: all test clean
diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h
new file mode 100644
index 000000000000..a51e17578ccf
--- /dev/null
+++ b/tools/testing/vsock/util.h
@@ -0,0 +1,35 @@
+#ifndef UTIL_H
+#define UTIL_H
+
+/* Tests can either run as the client or the server */
+enum test_mode {
+ TEST_MODE_UNSET,
+ TEST_MODE_CLIENT,
+ TEST_MODE_SERVER
+};
+
+/* Test runner options */
+struct test_opts {
+ enum test_mode mode;
+ unsigned int peer_cid;
+};
+
+/* A test case definition. Test functions must print failures to stderr and
+ * terminate with exit(EXIT_FAILURE).
+ */
+struct test_case {
+ const char *name; /* human-readable name */
+
+ /* Called when test mode is TEST_MODE_CLIENT */
+ void (*run_client)(const struct test_opts *opts);
+
+ /* Called when test mode is TEST_MODE_SERVER */
+ void (*run_server)(const struct test_opts *opts);
+};
+
+void init_signals(void);
+unsigned int parse_cid(const char *str);
+void run_tests(const struct test_case *test_cases,
+ const struct test_opts *opts);
+
+#endif /* UTIL_H */
diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c
new file mode 100644
index 000000000000..2567abd90153
--- /dev/null
+++ b/tools/testing/vsock/util.c
@@ -0,0 +1,70 @@
+/*
+ * vsock test utilities
+ *
+ * Copyright (C) 2017 Red Hat, Inc.
+ *
+ * Author: Stefan Hajnoczi <stefanha@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <signal.h>
+
+#include "timeout.h"
+#include "util.h"
+
+/* Install signal handlers */
+void init_signals(void)
+{
+ struct sigaction act = {
+ .sa_handler = sigalrm,
+ };
+
+ sigaction(SIGALRM, &act, NULL);
+ signal(SIGPIPE, SIG_IGN);
+}
+
+/* Parse a CID in string representation */
+unsigned int parse_cid(const char *str)
+{
+ char *endptr = NULL;
+ unsigned long int n;
+
+ errno = 0;
+ n = strtoul(str, &endptr, 10);
+ if (errno || *endptr != '\0') {
+ fprintf(stderr, "malformed CID \"%s\"\n", str);
+ exit(EXIT_FAILURE);
+ }
+ return n;
+}
+
+/* Run test cases. The program terminates if a failure occurs. */
+void run_tests(const struct test_case *test_cases,
+ const struct test_opts *opts)
+{
+ int i;
+
+ for (i = 0; test_cases[i].name; i++) {
+ void (*run)(const struct test_opts *opts);
+
+ printf("%s...", test_cases[i].name);
+ fflush(stdout);
+
+ if (opts->mode == TEST_MODE_CLIENT)
+ run = test_cases[i].run_client;
+ else
+ run = test_cases[i].run_server;
+
+ if (run)
+ run(opts);
+
+ printf("ok\n");
+ }
+}
diff --git a/tools/testing/vsock/vsock_diag_test.c b/tools/testing/vsock/vsock_diag_test.c
index e896a4af52f4..f1ace5d96e00 100644
--- a/tools/testing/vsock/vsock_diag_test.c
+++ b/tools/testing/vsock/vsock_diag_test.c
@@ -13,12 +13,10 @@
#include <getopt.h>
#include <stdio.h>
-#include <stdbool.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
-#include <signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/types.h>
@@ -33,12 +31,7 @@
#include "timeout.h"
#include "control.h"
-
-enum test_mode {
- TEST_MODE_UNSET,
- TEST_MODE_CLIENT,
- TEST_MODE_SERVER
-};
+#include "util.h"
/* Per-socket status */
struct vsock_stat {
@@ -339,7 +332,7 @@ static void free_sock_stat(struct list_head *sockets)
free(st);
}
-static void test_no_sockets(unsigned int peer_cid)
+static void test_no_sockets(const struct test_opts *opts)
{
LIST_HEAD(sockets);
@@ -350,7 +343,7 @@ static void test_no_sockets(unsigned int peer_cid)
free_sock_stat(&sockets);
}
-static void test_listen_socket_server(unsigned int peer_cid)
+static void test_listen_socket_server(const struct test_opts *opts)
{
union {
struct sockaddr sa;
@@ -388,7 +381,7 @@ static void test_listen_socket_server(unsigned int peer_cid)
free_sock_stat(&sockets);
}
-static void test_connect_client(unsigned int peer_cid)
+static void test_connect_client(const struct test_opts *opts)
{
union {
struct sockaddr sa;
@@ -397,7 +390,7 @@ static void test_connect_client(unsigned int peer_cid)
.svm = {
.svm_family = AF_VSOCK,
.svm_port = 1234,
- .svm_cid = peer_cid,
+ .svm_cid = opts->peer_cid,
},
};
int fd;
@@ -434,7 +427,7 @@ static void test_connect_client(unsigned int peer_cid)
free_sock_stat(&sockets);
}
-static void test_connect_server(unsigned int peer_cid)
+static void test_connect_server(const struct test_opts *opts)
{
union {
struct sockaddr sa;
@@ -486,9 +479,9 @@ static void test_connect_server(unsigned int peer_cid)
clientaddr.sa.sa_family);
exit(EXIT_FAILURE);
}
- if (clientaddr.svm.svm_cid != peer_cid) {
+ if (clientaddr.svm.svm_cid != opts->peer_cid) {
fprintf(stderr, "expected peer CID %u from accept(2), got %u\n",
- peer_cid, clientaddr.svm.svm_cid);
+ opts->peer_cid, clientaddr.svm.svm_cid);
exit(EXIT_FAILURE);
}
@@ -507,11 +500,7 @@ static void test_connect_server(unsigned int peer_cid)
free_sock_stat(&sockets);
}
-static struct {
- const char *name;
- void (*run_client)(unsigned int peer_cid);
- void (*run_server)(unsigned int peer_cid);
-} test_cases[] = {
+static struct test_case test_cases[] = {
{
.name = "No sockets",
.run_server = test_no_sockets,
@@ -528,30 +517,6 @@ static struct {
{},
};
-static void init_signals(void)
-{
- struct sigaction act = {
- .sa_handler = sigalrm,
- };
-
- sigaction(SIGALRM, &act, NULL);
- signal(SIGPIPE, SIG_IGN);
-}
-
-static unsigned int parse_cid(const char *str)
-{
- char *endptr = NULL;
- unsigned long int n;
-
- errno = 0;
- n = strtoul(str, &endptr, 10);
- if (errno || *endptr != '\0') {
- fprintf(stderr, "malformed CID \"%s\"\n", str);
- exit(EXIT_FAILURE);
- }
- return n;
-}
-
static const char optstring[] = "";
static const struct option longopts[] = {
{
@@ -606,9 +571,10 @@ int main(int argc, char **argv)
{
const char *control_host = NULL;
const char *control_port = NULL;
- int mode = TEST_MODE_UNSET;
- unsigned int peer_cid = VMADDR_CID_ANY;
- int i;
+ struct test_opts opts = {
+ .mode = TEST_MODE_UNSET,
+ .peer_cid = VMADDR_CID_ANY,
+ };
init_signals();
@@ -624,16 +590,16 @@ int main(int argc, char **argv)
break;
case 'm':
if (strcmp(optarg, "client") == 0)
- mode = TEST_MODE_CLIENT;
+ opts.mode = TEST_MODE_CLIENT;
else if (strcmp(optarg, "server") == 0)
- mode = TEST_MODE_SERVER;
+ opts.mode = TEST_MODE_SERVER;
else {
fprintf(stderr, "--mode must be \"client\" or \"server\"\n");
return EXIT_FAILURE;
}
break;
case 'p':
- peer_cid = parse_cid(optarg);
+ opts.peer_cid = parse_cid(optarg);
break;
case 'P':
control_port = optarg;
@@ -646,35 +612,21 @@ int main(int argc, char **argv)
if (!control_port)
usage();
- if (mode == TEST_MODE_UNSET)
+ if (opts.mode == TEST_MODE_UNSET)
usage();
- if (peer_cid == VMADDR_CID_ANY)
+ if (opts.peer_cid == VMADDR_CID_ANY)
usage();
if (!control_host) {
- if (mode != TEST_MODE_SERVER)
+ if (opts.mode != TEST_MODE_SERVER)
usage();
control_host = "0.0.0.0";
}
- control_init(control_host, control_port, mode == TEST_MODE_SERVER);
+ control_init(control_host, control_port,
+ opts.mode == TEST_MODE_SERVER);
- for (i = 0; test_cases[i].name; i++) {
- void (*run)(unsigned int peer_cid);
-
- printf("%s...", test_cases[i].name);
- fflush(stdout);
-
- if (mode == TEST_MODE_CLIENT)
- run = test_cases[i].run_client;
- else
- run = test_cases[i].run_server;
-
- if (run)
- run(peer_cid);
-
- printf("ok\n");
- }
+ run_tests(test_cases, &opts);
control_cleanup();
return EXIT_SUCCESS;
--
2.14.3
^ permalink raw reply related
* [PATCH 0/5] VSOCK: add vsock_test test suite
From: Stefan Hajnoczi @ 2017-12-13 14:49 UTC (permalink / raw)
To: netdev; +Cc: Jorgen Hansen, Dexuan Cui, Stefan Hajnoczi
The vsock_diag.ko module already has a test suite but the core AF_VSOCK
functionality has no tests. This patch series adds several test cases that
exercise AF_VSOCK SOCK_STREAM socket semantics (send/recv, connect/accept,
half-closed connections, simultaneous connections).
The test suite is modest but I hope to cover additional cases in the future.
My goal is to have a shared test suite so VMCI, Hyper-V, and KVM can ensure
that our transports behave the same.
I have tested virtio-vsock.
Jorgen: Please test the VMCI transport and let me know if anything needs to be
adjusted. See tools/testing/vsock/README for information on how to run the
test suite.
Dexuan: I'm not sure if this test suite is useful for the Hyper-V transport
since the host is Windows and uses a different API for AF_HYPERV?
Stefan Hajnoczi (5):
VSOCK: extract utility functions from vsock_diag_test.c
VSOCK: extract connect/accept functions from vsock_diag_test.c
VSOCK: add full barrier between test cases
VSOCK: add send_byte()/recv_byte() test utilities
VSOCK: add AF_VSOCK test cases
tools/testing/vsock/Makefile | 7 +-
tools/testing/vsock/util.h | 43 +++++
tools/testing/vsock/util.c | 291 ++++++++++++++++++++++++++++++
tools/testing/vsock/vsock_diag_test.c | 167 +++---------------
tools/testing/vsock/vsock_test.c | 321 ++++++++++++++++++++++++++++++++++
tools/testing/vsock/.gitignore | 1 +
tools/testing/vsock/README | 1 +
7 files changed, 685 insertions(+), 146 deletions(-)
create mode 100644 tools/testing/vsock/util.h
create mode 100644 tools/testing/vsock/util.c
create mode 100644 tools/testing/vsock/vsock_test.c
--
2.14.3
^ permalink raw reply
* [PATCH v2 net-next] net: dsa: lan9303: Introduce lan9303_read_wait
From: Egil Hjelmeland @ 2017-12-13 14:42 UTC (permalink / raw)
To: andrew, vivien.didelot, f.fainelli, netdev, linux-kernel; +Cc: Egil Hjelmeland
Simplify lan9303_indirect_phy_wait_for_completion()
and lan9303_switch_wait_for_completion() by using a new function
lan9303_read_wait()
Changes v1 -> v2:
- param 'mask' type u32
- removed param 'value' (will probably never be used)
- add newline before return
Signed-off-by: Egil Hjelmeland <privat@egil-hjelmeland.no>
---
drivers/net/dsa/lan9303-core.c | 59 +++++++++++++++++++-----------------------
1 file changed, 27 insertions(+), 32 deletions(-)
diff --git a/drivers/net/dsa/lan9303-core.c b/drivers/net/dsa/lan9303-core.c
index c1b004fa64d9..f412aad58253 100644
--- a/drivers/net/dsa/lan9303-core.c
+++ b/drivers/net/dsa/lan9303-core.c
@@ -249,6 +249,29 @@ static int lan9303_read(struct regmap *regmap, unsigned int offset, u32 *reg)
return -EIO;
}
+/* Wait a while until mask & reg == value. Otherwise return timeout. */
+static int lan9303_read_wait(struct lan9303 *chip, int offset, u32 mask)
+{
+ int i;
+
+ for (i = 0; i < 25; i++) {
+ u32 reg;
+ int ret;
+
+ ret = lan9303_read(chip->regmap, offset, ®);
+ if (ret) {
+ dev_err(chip->dev, "%s failed to read offset %d: %d\n",
+ __func__, offset, ret);
+ return ret;
+ }
+ if (!(reg & mask))
+ return 0;
+ usleep_range(1000, 2000);
+ }
+
+ return -ETIMEDOUT;
+}
+
static int lan9303_virt_phy_reg_read(struct lan9303 *chip, int regnum)
{
int ret;
@@ -274,22 +297,8 @@ static int lan9303_virt_phy_reg_write(struct lan9303 *chip, int regnum, u16 val)
static int lan9303_indirect_phy_wait_for_completion(struct lan9303 *chip)
{
- int ret, i;
- u32 reg;
-
- for (i = 0; i < 25; i++) {
- ret = lan9303_read(chip->regmap, LAN9303_PMI_ACCESS, ®);
- if (ret) {
- dev_err(chip->dev,
- "Failed to read pmi access status: %d\n", ret);
- return ret;
- }
- if (!(reg & LAN9303_PMI_ACCESS_MII_BUSY))
- return 0;
- usleep_range(1000, 2000);
- }
-
- return -EIO;
+ return lan9303_read_wait(chip, LAN9303_PMI_ACCESS,
+ LAN9303_PMI_ACCESS_MII_BUSY);
}
static int lan9303_indirect_phy_read(struct lan9303 *chip, int addr, int regnum)
@@ -366,22 +375,8 @@ EXPORT_SYMBOL_GPL(lan9303_indirect_phy_ops);
static int lan9303_switch_wait_for_completion(struct lan9303 *chip)
{
- int ret, i;
- u32 reg;
-
- for (i = 0; i < 25; i++) {
- ret = lan9303_read(chip->regmap, LAN9303_SWITCH_CSR_CMD, ®);
- if (ret) {
- dev_err(chip->dev,
- "Failed to read csr command status: %d\n", ret);
- return ret;
- }
- if (!(reg & LAN9303_SWITCH_CSR_CMD_BUSY))
- return 0;
- usleep_range(1000, 2000);
- }
-
- return -EIO;
+ return lan9303_read_wait(chip, LAN9303_SWITCH_CSR_CMD,
+ LAN9303_SWITCH_CSR_CMD_BUSY);
}
static int lan9303_write_switch_reg(struct lan9303 *chip, u16 regnum, u32 val)
--
2.14.1
^ permalink raw reply related
* Re: Setting large MTU size on slave interfaces may stall the whole system
From: Or Gerlitz @ 2017-12-13 14:28 UTC (permalink / raw)
To: Qing Huang
Cc: Linux Netdev List, Linux Kernel, Jay Vosburgh, Veaceslav Falico,
Andy Gospodarek, Aviv Heller, Moni Shoua
In-Reply-To: <ea15b7f9-4c02-3d4e-1c01-4e2a90d2a23c@oracle.com>
On Tue, Dec 12, 2017 at 5:21 AM, Qing Huang <qing.huang@oracle.com> wrote:
> Hi,
>
> We found an issue with the bonding driver when testing Mellanox devices.
> The following test commands will stall the whole system sometimes, with
> serial console
> flooded with log messages from the bond_miimon_inspect() function. Setting
> mtu size
> to be 1500 seems okay but very rarely it may hit the same problem too.
>
> ip address flush dev ens3f0
> ip link set dev ens3f0 down
> ip address flush dev ens3f1
> ip link set dev ens3f1 down
> [root@ca-hcl629 etc]# modprobe bonding mode=0 miimon=250 use_carrier=1
> updelay=500 downdelay=500
> [root@ca-hcl629 etc]# ifconfig bond0 up
> [root@ca-hcl629 etc]# ifenslave bond0 ens3f0 ens3f1
> [root@ca-hcl629 etc]# ip link set bond0 mtu 4500 up
> Seiral console output:
>
> ** 4 printk messages dropped ** [ 3717.743761] bond0: link status down for
> interface ens3f0, disabling it in 500 ms
[..]
> It seems that when setting a large mtu size on an RoCE interface, the RTNL
> mutex may be held too long by the slave
> interface, causing bond_mii_monitor() to be called repeatedly at an interval
> of 1 tick (1K HZ kernel configuration) and kernel to become unresponsive.
Did you try/managed to reproduce that also with other NIC drivers?
Or.
^ permalink raw reply
* Re: [RFC][PATCH] new byteorder primitives - ..._{replace,get}_bits()
From: Al Viro @ 2017-12-13 14:22 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: Linus Torvalds, netdev, linux-kernel
In-Reply-To: <20171212184400.13b27cf8@cakuba.netronome.com>
On Tue, Dec 12, 2017 at 06:44:00PM -0800, Jakub Kicinski wrote:
> On Wed, 13 Dec 2017 01:51:25 +0000, Al Viro wrote:
> > On Tue, Dec 12, 2017 at 05:35:28PM -0800, Jakub Kicinski wrote:
> >
> > > It used to be __always_inline, but apparently LLVM/clang doesn't
> > > propagate constants :(
> > >
> > > 4e59532541c8 ("nfp: don't depend on compiler constant propagation")
> >
> > Doesn't propagate constants or doesn't have exact same set of
> > rules for __builtin_constant_p()? IOW, if you dropped that
> > BUILD_BUG_ON(), what would be left after optimizations?
>
> Hm. You're right. It just doesn't recognize the parameter as constant
> in __builtin_constant_p().
FWIW, clang does propagate them well enough to detect and optimize
multiplication by constant power of two. With the variant I've posted.
The check on field overflow (which works on gcc builds) does nothing on
clang - __builtin_constant_p() gives a flat-out false for arguments of
inline function there. I can understand their reasoning, even though
it's inconvenient in cases like this one - semantics of __builtin_constant_p()
is a fucking mess and the less you rely upon its details, the safer you
are...
Hell knows - a part of that can be recovered by taking the check into a
wrapper; as in
static __always_inline __le32_replace_bits(__le32 old, u32 v, u32 mask)
{
__le32 m = cpu_to_le32(mask);
return (old & ~m) | cpu_to_le32(v * mask/mask_to_multiplier(mask)));
}
#define le32_replace_bits(old, v, mask) ({ \
typeof(v) ____v = (v); \
typeof(mask) ____m = (mask); \
if (__builtin_constant_p(____v)) \
if (v & ~(____m / mask_to_multiplier(____m))) \
__field_overflow(); \
__l32_replace_bits(old, ____v, ____m); \
})
That would give that sanity check a better coverage on clang builds, but...
does it really buy us enough to bother, especially since all those macros
would have to be spelled out; you can have a macro expanding to definition
of static inline, but you can't have a macro expanding to anything that
would contain a preprocessor directive, including #define. And with that
kind of sanity checks, the first build with gcc will catch everything
missed by clang builds anyway.
IMO it's not worth the trouble; let's go with the check inside of
static inline and accept that on clang builds it'll do nothing.
Next question: where do we put that bunch? I've put it into
linux/byteorder/generic.h, so that anything picking fixed-endian primitives
would pick those as well; I hadn't thought of linux/bitfield.h at the time.
We certainly could put it there instead - it's never pulled by other headers,
so adding #include <asm/byteorder.h> into linux/bitfield.h is not going to
cause header order problems. Not sure...
Linus, do you have any preferences in that area?
^ permalink raw reply
* Re: [bpf-next V1-RFC PATCH 06/14] mlx4: setup xdp_rxq_info
From: Jesper Dangaard Brouer @ 2017-12-13 14:00 UTC (permalink / raw)
To: Tariq Toukan
Cc: Daniel Borkmann, Alexei Starovoitov, netdev, dsahern, gospo,
bjorn.topel, michael.chan, brouer
In-Reply-To: <5b7d2132-40c7-71f5-b939-64539c6c53bb@mellanox.com>
On Wed, 13 Dec 2017 14:42:25 +0200
Tariq Toukan <tariqt@mellanox.com> wrote:
> > @@ -650,6 +658,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
> > int cq_ring = cq->ring;
> > bool doorbell_pending;
> > struct mlx4_cqe *cqe;
> > + struct xdp_buff xdp;
> > int polled = 0;
> > int index;
> >
> > @@ -664,6 +673,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
> > /* Protect accesses to: ring->xdp_prog, priv->mac_hash list */
> > rcu_read_lock();
> > xdp_prog = rcu_dereference(ring->xdp_prog);
> > + xdp.rxq = &ring->xdp_rxq;
>
> You moved this to update it only once, and not per packet. Right?
> This is because all fields of struct xdp_buff used to be specific
> per-packet and filled in every iteration. Now you introduce a new field
> which holds a context.
>
> Well, I still need to go over the infrastructure in your other patches,
> but from first glance it seems that we can use two separated structs:
> one for context, and one for per-packet info.
This are two separate structs. I guess, what you are suggesting is
passing them as separate structs to bpf_prog_run_xdp() ?
The reason I like/want to have xdp_buff point to xdp_rxq_info, is that
I want this information transferred through to the XDP_REDIRECT calls.
The plan (after this patchset) is to implement a kfree_xdp_buff() that
can free/return xdp frames directly to the driver (needed in err cases
in redirect code) by checking if the rxq->dev have defined an
ndo_xdp_return() function.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply
* Re: [bpf-next V1-RFC PATCH 02/14] xdp/mlx5: setup xdp_rxq_info and extend with qtype
From: Jesper Dangaard Brouer @ 2017-12-13 13:44 UTC (permalink / raw)
To: Tariq Toukan
Cc: Daniel Borkmann, Alexei Starovoitov, netdev, dsahern, Matan Barak,
Saeed Mahameed, gospo, bjorn.topel, michael.chan, brouer
In-Reply-To: <e54da3e8-9944-6a7a-37a9-be438c88884b@mellanox.com>
On Wed, 13 Dec 2017 14:27:08 +0200
Tariq Toukan <tariqt@mellanox.com> wrote:
> Hi Jesper,
> Thanks for taking care of the drop RQ.
>
> In general, mlx5 part looks ok to me.
> Find a few comments below. Mostly pointing out some typos.
>
> On 13/12/2017 1:19 PM, Jesper Dangaard Brouer wrote:
> > The mlx5 driver have a special drop-RQ queue (one per interface) that
> > simply drops all incoming traffic. It helps driver keep other HW
> > objects (flow steering) alive upon down/up operations. It is
> > temporarily pointed by flow steering objects during the interface
> > setup, and when interface is down. It lacks many fields that are set
> > in a regular RQ (for example its state is never switched to
> > MLX5_RQC_STATE_RDY). (Thanks to Tariq Toukan for explaination).
> typo: explanation
Fixed
> >
> > The XDP RX-queue info API is extended with a queue-type, and mlx5 uses
> > this kind of drop/sink-type (RXQ_TYPE_SINK) for this kind of sink queue.
> >
> > Driver hook points for xdp_rxq_info:
> > * init+reg: mlx5e_alloc_rq()
> > * init+reg: mlx5e_alloc_drop_rq()
> > * unreg : mlx5e_free_rq()
> >
> > Tested on actual hardware with samples/bpf program
> >
> > Cc: Saeed Mahameed <saeedm@mellanox.com>
> > Cc: Matan Barak <matanb@mellanox.com>
> > Cc: Tariq Toukan <tariqt@mellanox.com>
> > Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
> > ---
> > drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++++
> > drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 14 +++++++++++++
> > drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 1 +
> > include/net/xdp.h | 23 +++++++++++++++++++++
> > net/core/xdp.c | 6 +++++
> > 5 files changed, 48 insertions(+)
> >
> > diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
> > index c0872b3284cb..fe10a042783b 100644
> > --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
> > +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
> > @@ -46,6 +46,7 @@
> > #include <linux/mlx5/transobj.h>
> > #include <linux/rhashtable.h>
> > #include <net/switchdev.h>
> > +#include <net/xdp.h>
> > #include "wq.h"
> > #include "mlx5_core.h"
> > #include "en_stats.h"
> > @@ -568,6 +569,9 @@ struct mlx5e_rq {
> > u32 rqn;
> > struct mlx5_core_dev *mdev;
> > struct mlx5_core_mkey umr_mkey;
> > +
> > + /* XDP read-mostly */
> > + struct xdp_rxq_info xdp_rxq;
> > } ____cacheline_aligned_in_smp;
> >
> > struct mlx5e_channel {
> > diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > index 0f5c012de52e..ea44b5f25e11 100644
> > --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > @@ -582,6 +582,12 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c,
> > rq->ix = c->ix;
> > rq->mdev = mdev;
> >
> > + /* XDP RX-queue info */
> > + xdp_rxq_info_init(&rq->xdp_rxq);
> > + rq->xdp_rxq.dev = rq->netdev;
> > + rq->xdp_rxq.queue_index = rq->ix;
> > + xdp_rxq_info_reg(&rq->xdp_rxq);
> > +
> You don't set type here. This is ok as long as the following hold:
> 1) RXQ_TYPE_DEFAULT is zero
True
> 2) xdp_rxq is zalloc'ed.
xdp_rxq memory area is part of rq allocation, but in
xdp_rxq_info_init() I memset/zero the area explicit.
> > rq->xdp_prog = params->xdp_prog ?
> > bpf_prog_inc(params->xdp_prog) : NULL; if (IS_ERR(rq->xdp_prog)) {
> > err = PTR_ERR(rq->xdp_prog);
> > @@ -695,6 +701,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel
> > *c, err_rq_wq_destroy:
> > if (rq->xdp_prog)
> > bpf_prog_put(rq->xdp_prog);
> > + xdp_rxq_info_unreg(&rq->xdp_rxq);
> > mlx5_wq_destroy(&rq->wq_ctrl);
> >
> > return err;
> > @@ -707,6 +714,8 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq)
> > if (rq->xdp_prog)
> > bpf_prog_put(rq->xdp_prog);
> >
> > + xdp_rxq_info_unreg(&rq->xdp_rxq);
> > +
> > switch (rq->wq_type) {
> > case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
> > mlx5e_rq_free_mpwqe_info(rq);
> > @@ -2768,6 +2777,11 @@ static int mlx5e_alloc_drop_rq(struct
> > mlx5_core_dev *mdev, if (err)
> > return err;
> >
> > + /* XDP RX-queue info for "Drop-RQ", packets never reach
> > XDP */
> > + xdp_rxq_info_init(&rq->xdp_rxq);
> > + xdp_rxq_info_type(&rq->xdp_rxq, RXQ_TYPE_SINK);
> > + xdp_rxq_info_reg(&rq->xdp_rxq);
> > +
> > rq->mdev = mdev;
> >
> > return 0;
> > diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
> > b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index
> > 5b499c7a698f..7b38480811d4 100644 ---
> > a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++
> > b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -812,6 +812,7
> > @@ static inline int mlx5e_xdp_handle(struct mlx5e_rq *rq,
> > xdp_set_data_meta_invalid(&xdp); xdp.data_end = xdp.data + *len;
> > xdp.data_hard_start = va;
> > + xdp.rxq = &rq->xdp_rxq;
> >
> > act = bpf_prog_run_xdp(prog, &xdp);
> > switch (act) {
> > diff --git a/include/net/xdp.h b/include/net/xdp.h
> > index e4acd198fd60..5be560d943e1 100644
> > --- a/include/net/xdp.h
> > +++ b/include/net/xdp.h
> > @@ -36,10 +36,33 @@ struct xdp_rxq_info {
> > struct net_device *dev;
> > u32 queue_index;
> > u32 reg_state;
> > + u32 qtype;
> > } ____cacheline_aligned; /* perf critical, avoid false-sharing */
> >
> > void xdp_rxq_info_init(struct xdp_rxq_info *xdp_rxq);
> > void xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq);
> > void xdp_rxq_info_unreg(struct xdp_rxq_info *xdp_rxq);
> >
> > +/**
> > + * DOC: XDP RX-queue type
> > + *
> > + * The XDP RX-queue info can have associated a type.
> > + *
> > + * @RXQ_TYPE_DEFAULT: default no specifik queue type need to be
> > specified
>
> typo: specific
Thanks, this is a Danish typo (it's spelled that way in Danish).
> > + *
> > + * @RXQ_TYPE_SINK: indicate a fake queue that never reach XDP RX
> > + * code. Some drivers have a need to maintain a lower layer
> > + * RX-queue as a sink queue, while reconfiguring other
> > RX-queues.
> > + */
> > +#define RXQ_TYPE_DEFAULT 0
> > +#define RXQ_TYPE_SINK 1
> > +#define RXQ_TYPE_MAX RXQ_TYPE_SINK
>
> Definitions of incremental numbers, enum might be best here, you can
> give them some enum type and use it in xdp_rxq_info->qtype.
I use defines to make the below BUILD_BUG_ON work, as enums does not
get expanded to their values in the C-preprocessor stage.
> > +
> > +static inline
> > +void xdp_rxq_info_type(struct xdp_rxq_info *xdp_rxq, u32 qtype)
> > +{
> > + BUILD_BUG_ON(qtype > RXQ_TYPE_MAX);
> > + xdp_rxq->qtype = qtype;
> > +}
> > +
> > #endif /* __LINUX_NET_XDP_H__ */
> > diff --git a/net/core/xdp.c b/net/core/xdp.c
> > index a9d2dd7b1ede..2a111f5987f6 100644
> > --- a/net/core/xdp.c
> > +++ b/net/core/xdp.c
> > @@ -32,8 +32,14 @@ EXPORT_SYMBOL_GPL(xdp_rxq_info_init);
> >
> > void xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq)
> > {
> > + if (xdp_rxq->qtype == RXQ_TYPE_SINK)
> > + goto skip_content_check;
> > +
> > + /* Check information setup by driver code */
> > WARN(!xdp_rxq->dev, "Missing net_device from driver");
> > WARN(xdp_rxq->queue_index == U32_MAX, "Miss queue_index from driver"); +
> > +skip_content_check:
> > WARN(!(xdp_rxq->reg_state == REG_STATE_NEW),"API violation, miss init");
> > xdp_rxq->reg_state = REG_STATE_REGISTRED;
> typo: REGISTERED (introduced in a previous patch)
Thanks for catching that! :-)
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply
* RE: [PATCH V3 net-next 3/8] net: hns3: Add HNS3 VF HCL(Hardware Compatibility Layer) Support
From: Salil Mehta @ 2017-12-13 13:02 UTC (permalink / raw)
To: Philippe Ombredanne
Cc: David S. Miller, Zhuangyuzeng (Yisen), lipeng (Y), Salil Mehta,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, LKML,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Linuxarm
In-Reply-To: <CAOFm3uFH_X5Bfagpx=ai_EdEKAmBHrcPb-xhdO=E-gDhtq0rLw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Hi Philippe,
> -----Original Message-----
> From: Philippe Ombredanne [mailto:pombredanne@nexb.com]
> Sent: Wednesday, December 13, 2017 12:54 PM
> To: Salil Mehta <salil.mehta@huawei.com>
> Cc: David S. Miller <davem@davemloft.net>; Zhuangyuzeng (Yisen)
> <yisen.zhuang@huawei.com>; lipeng (Y) <lipeng321@huawei.com>; Salil
> Mehta <mehta.salil.lnk@gmail.com>; netdev@vger.kernel.org; LKML <linux-
> kernel@vger.kernel.org>; linux-rdma@vger.kernel.org; Linuxarm
> <linuxarm@huawei.com>
> Subject: Re: [PATCH V3 net-next 3/8] net: hns3: Add HNS3 VF
> HCL(Hardware Compatibility Layer) Support
>
> Dear Salil,
>
> On Wed, Dec 13, 2017 at 11:35 AM, Salil Mehta <salil.mehta@huawei.com>
> wrote:
>
> > I could also make out from different articles, including from the
> below,
> > Linus suggesting moving to "//" type instead of starred ones for
> headers.
> >
> > It looks SPDX change is still a suggestion?
>
> Well, I think this has been discussed and agreed among maintainers at
> the plumbers summit and here on this list. And when I see Thomas doc
> patches, Greg patches, other key maintainers as well as Linus comments
> and reviews and guidelines, I can can hardly see this as just being a
> "suggestion", but that's your call to consider it this way.
Ok. I understand. Thanks for confirming it. I will do the changes as
suggested by you and thanks for bringing this into notice.
Best regards
Salil
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox