Netdev List
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeedm@mellanox.com>
To: "David S. Miller" <davem@davemloft.net>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	Vlad Buslov <vladbu@mellanox.com>,
	Jianbo Liu <jianbol@mellanox.com>, Roi Dayan <roid@mellanox.com>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: [net-next 08/13] net/mlx5e: Protect tc flows hashtable with rcu
Date: Mon, 29 Jul 2019 23:50:27 +0000	[thread overview]
Message-ID: <20190729234934.23595-9-saeedm@mellanox.com> (raw)
In-Reply-To: <20190729234934.23595-1-saeedm@mellanox.com>

From: Vlad Buslov <vladbu@mellanox.com>

In order to remove dependency on rtnl lock, access to tc flows hashtable
must be explicitly protected from concurrent flows removal.

Extend tc flow structure with rcu to allow concurrent parallel access. Use
rcu read lock to safely lookup flow in tc flows hash table, and take
reference to it. Use rcu free for flow deletion to accommodate concurrent
stats requests.

Add new DELETED flow flag. Imlement new flow_flag_test_and_set() helper
that is used to set a flag and return its previous value. Use it to
atomically set the flag in mlx5e_delete_flower() to guarantee that flow can
only be deleted once, even when same flow is deleted concurrently by
multiple tasks.

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Reviewed-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 47 ++++++++++++++++---
 1 file changed, 40 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 241157b699df..a39f8a07de0a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -79,6 +79,7 @@ enum {
 	MLX5E_TC_FLOW_FLAG_SLOW		= MLX5E_TC_FLOW_BASE + 3,
 	MLX5E_TC_FLOW_FLAG_DUP		= MLX5E_TC_FLOW_BASE + 4,
 	MLX5E_TC_FLOW_FLAG_NOT_READY	= MLX5E_TC_FLOW_BASE + 5,
+	MLX5E_TC_FLOW_FLAG_DELETED	= MLX5E_TC_FLOW_BASE + 6,
 };
 
 #define MLX5E_TC_MAX_SPLITS 1
@@ -122,6 +123,7 @@ struct mlx5e_tc_flow {
 	struct list_head	peer;    /* flows with peer flow */
 	struct list_head	unready; /* flows not ready to be offloaded (e.g due to missing route) */
 	refcount_t		refcnt;
+	struct rcu_head		rcu_head;
 	union {
 		struct mlx5_esw_flow_attr esw_attr[0];
 		struct mlx5_nic_flow_attr nic_attr[0];
@@ -201,7 +203,7 @@ static void mlx5e_flow_put(struct mlx5e_priv *priv,
 {
 	if (refcount_dec_and_test(&flow->refcnt)) {
 		mlx5e_tc_del_flow(priv, flow);
-		kfree(flow);
+		kfree_rcu(flow, rcu_head);
 	}
 }
 
@@ -214,6 +216,17 @@ static void __flow_flag_set(struct mlx5e_tc_flow *flow, unsigned long flag)
 
 #define flow_flag_set(flow, flag) __flow_flag_set(flow, MLX5E_TC_FLOW_FLAG_##flag)
 
+static bool __flow_flag_test_and_set(struct mlx5e_tc_flow *flow,
+				     unsigned long flag)
+{
+	/* test_and_set_bit() provides all necessary barriers */
+	return test_and_set_bit(flag, &flow->flags);
+}
+
+#define flow_flag_test_and_set(flow, flag)			\
+	__flow_flag_test_and_set(flow,				\
+				 MLX5E_TC_FLOW_FLAG_##flag)
+
 static void __flow_flag_clear(struct mlx5e_tc_flow *flow, unsigned long flag)
 {
 	/* Complete all memory stores before clearing bit. */
@@ -3451,7 +3464,9 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
 	struct mlx5e_tc_flow *flow;
 	int err = 0;
 
-	flow = rhashtable_lookup_fast(tc_ht, &f->cookie, tc_ht_params);
+	rcu_read_lock();
+	flow = rhashtable_lookup(tc_ht, &f->cookie, tc_ht_params);
+	rcu_read_unlock();
 	if (flow) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "flow cookie already exists, ignoring");
@@ -3466,7 +3481,7 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv,
 	if (err)
 		goto out;
 
-	err = rhashtable_insert_fast(tc_ht, &flow->node, tc_ht_params);
+	err = rhashtable_lookup_insert_fast(tc_ht, &flow->node, tc_ht_params);
 	if (err)
 		goto err_free;
 
@@ -3492,16 +3507,32 @@ int mlx5e_delete_flower(struct net_device *dev, struct mlx5e_priv *priv,
 {
 	struct rhashtable *tc_ht = get_tc_ht(priv, flags);
 	struct mlx5e_tc_flow *flow;
+	int err;
 
+	rcu_read_lock();
 	flow = rhashtable_lookup_fast(tc_ht, &f->cookie, tc_ht_params);
-	if (!flow || !same_flow_direction(flow, flags))
-		return -EINVAL;
+	if (!flow || !same_flow_direction(flow, flags)) {
+		err = -EINVAL;
+		goto errout;
+	}
 
+	/* Only delete the flow if it doesn't have MLX5E_TC_FLOW_DELETED flag
+	 * set.
+	 */
+	if (flow_flag_test_and_set(flow, DELETED)) {
+		err = -EINVAL;
+		goto errout;
+	}
 	rhashtable_remove_fast(tc_ht, &flow->node, tc_ht_params);
+	rcu_read_unlock();
 
 	mlx5e_flow_put(priv, flow);
 
 	return 0;
+
+errout:
+	rcu_read_unlock();
+	return err;
 }
 
 int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv,
@@ -3517,8 +3548,10 @@ int mlx5e_stats_flower(struct net_device *dev, struct mlx5e_priv *priv,
 	u64 bytes = 0;
 	int err = 0;
 
-	flow = mlx5e_flow_get(rhashtable_lookup_fast(tc_ht, &f->cookie,
-						     tc_ht_params));
+	rcu_read_lock();
+	flow = mlx5e_flow_get(rhashtable_lookup(tc_ht, &f->cookie,
+						tc_ht_params));
+	rcu_read_unlock();
 	if (IS_ERR(flow))
 		return PTR_ERR(flow);
 
-- 
2.21.0


  parent reply	other threads:[~2019-07-29 23:50 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-29 23:50 [pull request][net-next 00/13] Mellanox, mlx5 tc flow handling for concurrent execution (Part 1) Saeed Mahameed
2019-07-29 23:50 ` [net-next 01/13] net/mlx5e: Print a warning when LRO feature is dropped or not allowed Saeed Mahameed
2019-07-30 15:52   ` Willem de Bruijn
2019-07-30 20:07     ` Saeed Mahameed
2019-07-30 21:34       ` Willem de Bruijn
2019-07-29 23:50 ` [net-next 02/13] net/mlx5e: Avoid warning print when not required Saeed Mahameed
2019-07-29 23:50 ` [net-next 03/13] net/mlx5e: Improve ethtool rxnfc callback structure Saeed Mahameed
2019-07-29 23:50 ` [net-next 04/13] net/mlx5e: Fix unnecessary flow_block_cb_is_busy call Saeed Mahameed
2019-07-29 23:50 ` [net-next 05/13] net/mlx5e: Simplify get_route_and_out_devs helper function Saeed Mahameed
2019-07-29 23:50 ` [net-next 06/13] net/mlx5e: Extend tc flow struct with reference counter Saeed Mahameed
2019-07-29 23:50 ` [net-next 07/13] net/mlx5e: Change flow flags type to unsigned long Saeed Mahameed
2019-07-29 23:50 ` Saeed Mahameed [this message]
2019-07-30 16:15   ` [net-next 08/13] net/mlx5e: Protect tc flows hashtable with rcu Willem de Bruijn
2019-07-30 16:37     ` Willem de Bruijn
2019-07-30 20:09       ` Saeed Mahameed
2019-07-29 23:50 ` [net-next 09/13] net/mlx5e: Protect unready flows with dedicated lock Saeed Mahameed
2019-07-29 23:50 ` [net-next 10/13] net/mlx5e: Eswitch, change offloads num_flows type to atomic64 Saeed Mahameed
2019-07-29 23:50 ` [net-next 11/13] net/mlx5e: Eswitch, use state_lock to synchronize vlan change Saeed Mahameed
2019-07-29 23:50 ` [net-next 12/13] net/mlx5e: Rely on rcu instead of rtnl lock when getting upper dev Saeed Mahameed
2019-07-29 23:50 ` [net-next 13/13] net/mlx5e: Protect tc flow table with mutex Saeed Mahameed
2019-07-31 22:48 ` [pull request][net-next 00/13] Mellanox, mlx5 tc flow handling for concurrent execution (Part 1) David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190729234934.23595-9-saeedm@mellanox.com \
    --to=saeedm@mellanox.com \
    --cc=davem@davemloft.net \
    --cc=jianbol@mellanox.com \
    --cc=netdev@vger.kernel.org \
    --cc=roid@mellanox.com \
    --cc=vladbu@mellanox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox