Merge branch 'several-fixes-for-indirect-flow_blocks-offload'

wenxu says:

====================
several fixes for indirect flow_blocks offload

v2:
patch2: store the cb_priv of representor to the flow_block_cb->indr.cb_priv
in the driver. And make the correct check with the statments
this->indr.cb_priv == cb_priv

patch4: del the driver list only in the indriect cleanup callbacks

v3:
add the cover letter and changlogs.

v4:
collapsed 1/4, 2/4, 4/4 in v3 to one fix
Add the prepare patch 1 and 2

v5:
patch1: place flow_indr_block_cb_alloc() right before
flow_indr_dev_setup_offload() to avoid moving flow_block_indr_init()

This series fixes commit 1fac52da ("net: flow_offload: consolidate
indirect flow_block infrastructure") that revists the flow_block
infrastructure.

patch #1 #2: prepare for fix patch #3
add and use flow_indr_block_cb_alloc/remove function

patch #3: fix flow_indr_dev_unregister path
If the representor is removed, then identify the indirect flow_blocks
that need to be removed by the release callback and the port representor
structure. To identify the port representor structure, a new
indr.cb_priv field needs to be introduced. The flow_block also needs to
be removed from the driver list from the cleanup path

patch#4 fix block->nooffloaddevcnt warning dmesg log.
When a indr device add in offload success. The block->nooffloaddevcnt
should be 0. After the representor go away. When the dir device go away
the flow_block UNBIND operation with -EOPNOTSUPP which lead the warning
demesg log.
The block->nooffloaddevcnt should always count for indr block.
even the indr block offload successful. The representor maybe
gone away and the ingress qdisc can work in software mode.
====================
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c

@@ -1889,7 +1889,8 @@ static void bnxt_tc_setup_indr_rel(void *cb_priv)
 }
 
 static int bnxt_tc_setup_indr_block(struct net_device *netdev, struct bnxt *bp,
-				    struct flow_block_offload *f)
+				    struct flow_block_offload *f, void *data,
+				    void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	struct bnxt_flower_indr_block_cb_priv *cb_priv;
 	struct flow_block_cb *block_cb;
@@ -1907,9 +1908,10 @@ static int bnxt_tc_setup_indr_block(struct net_device *netdev, struct bnxt *bp,
 		cb_priv->bp = bp;
 		list_add(&cb_priv->list, &bp->tc_indr_block_list);
 
-		block_cb = flow_block_cb_alloc(bnxt_tc_setup_indr_block_cb,
-					       cb_priv, cb_priv,
-					       bnxt_tc_setup_indr_rel);
+		block_cb = flow_indr_block_cb_alloc(bnxt_tc_setup_indr_block_cb,
+						    cb_priv, cb_priv,
+						    bnxt_tc_setup_indr_rel, f,
+						    netdev, data, bp, cleanup);
 		if (IS_ERR(block_cb)) {
 			list_del(&cb_priv->list);
 			kfree(cb_priv);
@@ -1930,7 +1932,7 @@ static int bnxt_tc_setup_indr_block(struct net_device *netdev, struct bnxt *bp,
 		if (!block_cb)
 			return -ENOENT;
 
-		flow_block_cb_remove(block_cb, f);
+		flow_indr_block_cb_remove(block_cb, f);
 		list_del(&block_cb->driver_list);
 		break;
 	default:
@@ -1945,14 +1947,17 @@ static bool bnxt_is_netdev_indr_offload(struct net_device *netdev)
 }
 
 static int bnxt_tc_setup_indr_cb(struct net_device *netdev, void *cb_priv,
-				 enum tc_setup_type type, void *type_data)
+				 enum tc_setup_type type, void *type_data,
+				 void *data,
+				 void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	if (!bnxt_is_netdev_indr_offload(netdev))
 		return -EOPNOTSUPP;
 
 	switch (type) {
 	case TC_SETUP_BLOCK:
-		return bnxt_tc_setup_indr_block(netdev, cb_priv, type_data);
+		return bnxt_tc_setup_indr_block(netdev, cb_priv, type_data, data,
+						cleanup);
 	default:
 		break;
 	}
@@ -2074,7 +2079,7 @@ void bnxt_shutdown_tc(struct bnxt *bp)
 		return;
 
 	flow_indr_dev_unregister(bnxt_tc_setup_indr_cb, bp,
-				 bnxt_tc_setup_indr_block_cb);
+				 bnxt_tc_setup_indr_rel);
 	rhashtable_destroy(&tc_info->flow_table);
 	rhashtable_destroy(&tc_info->l2_table);
 	rhashtable_destroy(&tc_info->decap_l2_table);

drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c

@@ -407,7 +407,9 @@ static int
 mlx5e_rep_indr_setup_block(struct net_device *netdev,
 			   struct mlx5e_rep_priv *rpriv,
 			   struct flow_block_offload *f,
-			   flow_setup_cb_t *setup_cb)
+			   flow_setup_cb_t *setup_cb,
+			   void *data,
+			   void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	struct mlx5e_priv *priv = netdev_priv(rpriv->netdev);
 	struct mlx5e_rep_indr_block_priv *indr_priv;
@@ -438,8 +440,10 @@ mlx5e_rep_indr_setup_block(struct net_device *netdev,
 		list_add(&indr_priv->list,
 			 &rpriv->uplink_priv.tc_indr_block_priv_list);
 
-		block_cb = flow_block_cb_alloc(setup_cb, indr_priv, indr_priv,
-					       mlx5e_rep_indr_block_unbind);
+		block_cb = flow_indr_block_cb_alloc(setup_cb, indr_priv, indr_priv,
+						    mlx5e_rep_indr_block_unbind,
+						    f, netdev, data, rpriv,
+						    cleanup);
 		if (IS_ERR(block_cb)) {
 			list_del(&indr_priv->list);
 			kfree(indr_priv);
@@ -458,7 +462,7 @@ mlx5e_rep_indr_setup_block(struct net_device *netdev,
 		if (!block_cb)
 			return -ENOENT;
 
-		flow_block_cb_remove(block_cb, f);
+		flow_indr_block_cb_remove(block_cb, f);
 		list_del(&block_cb->driver_list);
 		return 0;
 	default:
@@ -469,15 +473,19 @@ mlx5e_rep_indr_setup_block(struct net_device *netdev,
 
 static
 int mlx5e_rep_indr_setup_cb(struct net_device *netdev, void *cb_priv,
-			    enum tc_setup_type type, void *type_data)
+			    enum tc_setup_type type, void *type_data,
+			    void *data,
+			    void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	switch (type) {
 	case TC_SETUP_BLOCK:
 		return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data,
-						  mlx5e_rep_indr_setup_tc_cb);
+						  mlx5e_rep_indr_setup_tc_cb,
+						  data, cleanup);
 	case TC_SETUP_FT:
 		return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data,
-						  mlx5e_rep_indr_setup_ft_cb);
+						  mlx5e_rep_indr_setup_ft_cb,
+						  data, cleanup);
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -496,7 +504,7 @@ int mlx5e_rep_tc_netdevice_event_register(struct mlx5e_rep_priv *rpriv)
 void mlx5e_rep_tc_netdevice_event_unregister(struct mlx5e_rep_priv *rpriv)
 {
 	flow_indr_dev_unregister(mlx5e_rep_indr_setup_cb, rpriv,
-				 mlx5e_rep_indr_setup_tc_cb);
+				 mlx5e_rep_indr_block_unbind);
 }
 
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)

drivers/net/ethernet/netronome/nfp/flower/main.c

@@ -861,7 +861,7 @@ static void nfp_flower_clean(struct nfp_app *app)
 	flush_work(&app_priv->cmsg_work);
 
 	flow_indr_dev_unregister(nfp_flower_indr_setup_tc_cb, app,
-				 nfp_flower_setup_indr_block_cb);
+				 nfp_flower_setup_indr_tc_release);
 
 	if (app_priv->flower_ext_feats & NFP_FL_FEATS_VF_RLIM)
 		nfp_flower_qos_cleanup(app);

drivers/net/ethernet/netronome/nfp/flower/main.h

@@ -459,9 +459,10 @@ int nfp_flower_setup_qos_offload(struct nfp_app *app, struct net_device *netdev,
 				 struct tc_cls_matchall_offload *flow);
 void nfp_flower_stats_rlim_reply(struct nfp_app *app, struct sk_buff *skb);
 int nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv,
-				enum tc_setup_type type, void *type_data);
-int nfp_flower_setup_indr_block_cb(enum tc_setup_type type, void *type_data,
-				   void *cb_priv);
+				enum tc_setup_type type, void *type_data,
+				void *data,
+				void (*cleanup)(struct flow_block_cb *block_cb));
+void nfp_flower_setup_indr_tc_release(void *cb_priv);
 
 void
 __nfp_flower_non_repr_priv_get(struct nfp_flower_non_repr_priv *non_repr_priv);

drivers/net/ethernet/netronome/nfp/flower/offload.c

@@ -1619,8 +1619,8 @@ nfp_flower_indr_block_cb_priv_lookup(struct nfp_app *app,
 	return NULL;
 }
 
-int nfp_flower_setup_indr_block_cb(enum tc_setup_type type,
-				   void *type_data, void *cb_priv)
+static int nfp_flower_setup_indr_block_cb(enum tc_setup_type type,
+					  void *type_data, void *cb_priv)
 {
 	struct nfp_flower_indr_block_cb_priv *priv = cb_priv;
 	struct flow_cls_offload *flower = type_data;
@@ -1637,7 +1637,7 @@ int nfp_flower_setup_indr_block_cb(enum tc_setup_type type,
 	}
 }
 
-static void nfp_flower_setup_indr_tc_release(void *cb_priv)
+void nfp_flower_setup_indr_tc_release(void *cb_priv)
 {
 	struct nfp_flower_indr_block_cb_priv *priv = cb_priv;
 
@@ -1647,7 +1647,8 @@ static void nfp_flower_setup_indr_tc_release(void *cb_priv)
 
 static int
 nfp_flower_setup_indr_tc_block(struct net_device *netdev, struct nfp_app *app,
-			       struct flow_block_offload *f)
+			       struct flow_block_offload *f, void *data,
+			       void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	struct nfp_flower_indr_block_cb_priv *cb_priv;
 	struct nfp_flower_priv *priv = app->priv;
@@ -1676,9 +1677,10 @@ nfp_flower_setup_indr_tc_block(struct net_device *netdev, struct nfp_app *app,
 		cb_priv->app = app;
 		list_add(&cb_priv->list, &priv->indr_block_cb_priv);
 
-		block_cb = flow_block_cb_alloc(nfp_flower_setup_indr_block_cb,
-					       cb_priv, cb_priv,
-					       nfp_flower_setup_indr_tc_release);
+		block_cb = flow_indr_block_cb_alloc(nfp_flower_setup_indr_block_cb,
+						    cb_priv, cb_priv,
+						    nfp_flower_setup_indr_tc_release,
+						    f, netdev, data, app, cleanup);
 		if (IS_ERR(block_cb)) {
 			list_del(&cb_priv->list);
 			kfree(cb_priv);
@@ -1699,7 +1701,7 @@ nfp_flower_setup_indr_tc_block(struct net_device *netdev, struct nfp_app *app,
 		if (!block_cb)
 			return -ENOENT;
 
-		flow_block_cb_remove(block_cb, f);
+		flow_indr_block_cb_remove(block_cb, f);
 		list_del(&block_cb->driver_list);
 		return 0;
 	default:
@@ -1710,7 +1712,9 @@ nfp_flower_setup_indr_tc_block(struct net_device *netdev, struct nfp_app *app,
 
 int
 nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv,
-			    enum tc_setup_type type, void *type_data)
+			    enum tc_setup_type type, void *type_data,
+			    void *data,
+			    void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	if (!nfp_fl_is_netdev_to_offload(netdev))
 		return -EOPNOTSUPP;
@@ -1718,7 +1722,7 @@ nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv,
 	switch (type) {
 	case TC_SETUP_BLOCK:
 		return nfp_flower_setup_indr_tc_block(netdev, cb_priv,
-						      type_data);
+						      type_data, data, cleanup);
 	default:
 		return -EOPNOTSUPP;
 	}

include/net/flow_offload.h

@@ -450,6 +450,7 @@ struct flow_block_indr {
 	struct net_device		*dev;
 	enum flow_block_binder_type	binder_type;
 	void				*data;
+	void				*cb_priv;
 	void				(*cleanup)(struct flow_block_cb *block_cb);
 };
 
@@ -467,6 +468,13 @@ struct flow_block_cb {
 struct flow_block_cb *flow_block_cb_alloc(flow_setup_cb_t *cb,
 					  void *cb_ident, void *cb_priv,
 					  void (*release)(void *cb_priv));
+struct flow_block_cb *flow_indr_block_cb_alloc(flow_setup_cb_t *cb,
+					       void *cb_ident, void *cb_priv,
+					       void (*release)(void *cb_priv),
+					       struct flow_block_offload *bo,
+					       struct net_device *dev, void *data,
+					       void *indr_cb_priv,
+					       void (*cleanup)(struct flow_block_cb *block_cb));
 void flow_block_cb_free(struct flow_block_cb *block_cb);
 
 struct flow_block_cb *flow_block_cb_lookup(struct flow_block *block,
@@ -488,6 +496,13 @@ static inline void flow_block_cb_remove(struct flow_block_cb *block_cb,
 	list_move(&block_cb->list, &offload->cb_list);
 }
 
+static inline void flow_indr_block_cb_remove(struct flow_block_cb *block_cb,
+					     struct flow_block_offload *offload)
+{
+	list_del(&block_cb->indr.list);
+	list_move(&block_cb->list, &offload->cb_list);
+}
+
 bool flow_block_cb_is_busy(flow_setup_cb_t *cb, void *cb_ident,
 			   struct list_head *driver_block_list);
 
@@ -532,11 +547,13 @@ static inline void flow_block_init(struct flow_block *flow_block)
 }
 
 typedef int flow_indr_block_bind_cb_t(struct net_device *dev, void *cb_priv,
-				      enum tc_setup_type type, void *type_data);
+				      enum tc_setup_type type, void *type_data,
+				      void *data,
+				      void (*cleanup)(struct flow_block_cb *block_cb));
 
 int flow_indr_dev_register(flow_indr_block_bind_cb_t *cb, void *cb_priv);
 void flow_indr_dev_unregister(flow_indr_block_bind_cb_t *cb, void *cb_priv,
-			      flow_setup_cb_t *setup_cb);
+			      void (*release)(void *cb_priv));
 int flow_indr_dev_setup_offload(struct net_device *dev,
 				enum tc_setup_type type, void *data,
 				struct flow_block_offload *bo,

net/core/flow_offload.c

@@ -372,14 +372,15 @@ int flow_indr_dev_register(flow_indr_block_bind_cb_t *cb, void *cb_priv)
 }
 EXPORT_SYMBOL(flow_indr_dev_register);
 
-static void __flow_block_indr_cleanup(flow_setup_cb_t *setup_cb, void *cb_priv,
+static void __flow_block_indr_cleanup(void (*release)(void *cb_priv),
+				      void *cb_priv,
 				      struct list_head *cleanup_list)
 {
 	struct flow_block_cb *this, *next;
 
 	list_for_each_entry_safe(this, next, &flow_block_indr_list, indr.list) {
-		if (this->cb == setup_cb &&
-		    this->cb_priv == cb_priv) {
+		if (this->release == release &&
+		    this->indr.cb_priv == cb_priv) {
 			list_move(&this->indr.list, cleanup_list);
 			return;
 		}
@@ -397,7 +398,7 @@ static void flow_block_indr_notify(struct list_head *cleanup_list)
 }
 
 void flow_indr_dev_unregister(flow_indr_block_bind_cb_t *cb, void *cb_priv,
-			      flow_setup_cb_t *setup_cb)
+			      void (*release)(void *cb_priv))
 {
 	struct flow_indr_dev *this, *next, *indr_dev = NULL;
 	LIST_HEAD(cleanup_list);
@@ -418,7 +419,7 @@ void flow_indr_dev_unregister(flow_indr_block_bind_cb_t *cb, void *cb_priv,
 		return;
 	}
 
-	__flow_block_indr_cleanup(setup_cb, cb_priv, &cleanup_list);
+	__flow_block_indr_cleanup(release, cb_priv, &cleanup_list);
 	mutex_unlock(&flow_indr_block_lock);
 
 	flow_block_indr_notify(&cleanup_list);
@@ -429,32 +430,37 @@ EXPORT_SYMBOL(flow_indr_dev_unregister);
 static void flow_block_indr_init(struct flow_block_cb *flow_block,
 				 struct flow_block_offload *bo,
 				 struct net_device *dev, void *data,
+				 void *cb_priv,
 				 void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	flow_block->indr.binder_type = bo->binder_type;
 	flow_block->indr.data = data;
+	flow_block->indr.cb_priv = cb_priv;
 	flow_block->indr.dev = dev;
 	flow_block->indr.cleanup = cleanup;
 }
 
-static void __flow_block_indr_binding(struct flow_block_offload *bo,
-				      struct net_device *dev, void *data,
-				      void (*cleanup)(struct flow_block_cb *block_cb))
+struct flow_block_cb *flow_indr_block_cb_alloc(flow_setup_cb_t *cb,
+					       void *cb_ident, void *cb_priv,
+					       void (*release)(void *cb_priv),
+					       struct flow_block_offload *bo,
+					       struct net_device *dev, void *data,
+					       void *indr_cb_priv,
+					       void (*cleanup)(struct flow_block_cb *block_cb))
 {
 	struct flow_block_cb *block_cb;
 
-	list_for_each_entry(block_cb, &bo->cb_list, list) {
-		switch (bo->command) {
-		case FLOW_BLOCK_BIND:
-			flow_block_indr_init(block_cb, bo, dev, data, cleanup);
-			list_add(&block_cb->indr.list, &flow_block_indr_list);
-			break;
-		case FLOW_BLOCK_UNBIND:
-			list_del(&block_cb->indr.list);
-			break;
-		}
-	}
+	block_cb = flow_block_cb_alloc(cb, cb_ident, cb_priv, release);
+	if (IS_ERR(block_cb))
+		goto out;
+
+	flow_block_indr_init(block_cb, bo, dev, data, indr_cb_priv, cleanup);
+	list_add(&block_cb->indr.list, &flow_block_indr_list);
+
+out:
+	return block_cb;
 }
+EXPORT_SYMBOL(flow_indr_block_cb_alloc);
 
 int flow_indr_dev_setup_offload(struct net_device *dev,
 				enum tc_setup_type type, void *data,
@@ -465,9 +471,8 @@ int flow_indr_dev_setup_offload(struct net_device *dev,
 
 	mutex_lock(&flow_indr_block_lock);
 	list_for_each_entry(this, &flow_block_indr_dev_list, list)
-		this->cb(dev, this->cb_priv, type, bo);
+		this->cb(dev, this->cb_priv, type, bo, data, cleanup);
 
-	__flow_block_indr_binding(bo, dev, data, cleanup);
 	mutex_unlock(&flow_indr_block_lock);
 
 	return list_empty(&bo->cb_list) ? -EOPNOTSUPP : 0;

net/netfilter/nf_flow_table_offload.c

@@ -950,6 +950,7 @@ static void nf_flow_table_indr_cleanup(struct flow_block_cb *block_cb)
 	nf_flow_table_gc_cleanup(flowtable, dev);
 	down_write(&flowtable->flow_block_lock);
 	list_del(&block_cb->list);
+	list_del(&block_cb->driver_list);
 	flow_block_cb_free(block_cb);
 	up_write(&flowtable->flow_block_lock);
 }

net/netfilter/nf_tables_offload.c

@@ -296,6 +296,7 @@ static void nft_indr_block_cleanup(struct flow_block_cb *block_cb)
 	nft_flow_block_offload_init(&bo, dev_net(dev), FLOW_BLOCK_UNBIND,
 				    basechain, &extack);
 	mutex_lock(&net->nft.commit_mutex);
+	list_del(&block_cb->driver_list);
 	list_move(&block_cb->list, &bo.cb_list);
 	nft_flow_offload_unbind(&bo, basechain);
 	mutex_unlock(&net->nft.commit_mutex);

net/sched/cls_api.c

@@ -652,6 +652,7 @@ static void tc_block_indr_cleanup(struct flow_block_cb *block_cb)
 			       &block->flow_block, tcf_block_shared(block),
 			       &extack);
 	down_write(&block->cb_lock);
+	list_del(&block_cb->driver_list);
 	list_move(&block_cb->list, &bo.cb_list);
 	up_write(&block->cb_lock);
 	rtnl_lock();
@@ -671,25 +672,29 @@ static int tcf_block_offload_cmd(struct tcf_block *block,
 				 struct netlink_ext_ack *extack)
 {
 	struct flow_block_offload bo = {};
-	int err;
 
 	tcf_block_offload_init(&bo, dev, command, ei->binder_type,
 			       &block->flow_block, tcf_block_shared(block),
 			       extack);
 
-	if (dev->netdev_ops->ndo_setup_tc)
+	if (dev->netdev_ops->ndo_setup_tc) {
+		int err;
+
 		err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_BLOCK, &bo);
-	else
-		err = flow_indr_dev_setup_offload(dev, TC_SETUP_BLOCK, block,
-						  &bo, tc_block_indr_cleanup);
+		if (err < 0) {
+			if (err != -EOPNOTSUPP)
+				NL_SET_ERR_MSG(extack, "Driver ndo_setup_tc failed");
+			return err;
+		}
 
-	if (err < 0) {
-		if (err != -EOPNOTSUPP)
-			NL_SET_ERR_MSG(extack, "Driver ndo_setup_tc failed");
-		return err;
+		return tcf_block_setup(block, &bo);
 	}
 
-	return tcf_block_setup(block, &bo);
+	flow_indr_dev_setup_offload(dev, TC_SETUP_BLOCK, block, &bo,
+				    tc_block_indr_cleanup);
+	tcf_block_setup(block, &bo);
+
+	return -EOPNOTSUPP;
 }
 
 static int tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q,