From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Wed, 17 Jan 2018 18:50:59 -0800
Subject: nfp: protect each repr pointer individually with RCU
Patch-mainline: v4.16-rc1
Git-commit: 3eb47dfca0b245d88a4c30b8e41204036e0882e4
References: bsc#1109837

Representors are grouped in sets by type.  Currently the whole
sets are under RCU protection, but individual representor pointers
are not.  This causes some inconveniences when representors have
to be destroyed, because we have to allocate new sets to remove
any representors.  Protect the individual pointers with RCU.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
---
 drivers/net/ethernet/netronome/nfp/flower/main.c  |   45 ++++++++-----
 drivers/net/ethernet/netronome/nfp/nfp_net_repr.c |   71 ++++++++++------------
 drivers/net/ethernet/netronome/nfp/nfp_net_repr.h |   15 ++--
 3 files changed, 68 insertions(+), 63 deletions(-)

--- a/drivers/net/ethernet/netronome/nfp/flower/main.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.c
@@ -99,7 +99,7 @@ nfp_flower_repr_get(struct nfp_app *app,
 	if (port >= reprs->num_reprs)
 		return NULL;
 
-	return reprs->reprs[port];
+	return rcu_dereference(reprs->reprs[port]);
 }
 
 static int
@@ -114,15 +114,19 @@ nfp_flower_reprs_reify(struct nfp_app *a
 	if (!reprs)
 		return 0;
 
-	for (i = 0; i < reprs->num_reprs; i++)
-		if (reprs->reprs[i]) {
-			struct nfp_repr *repr = netdev_priv(reprs->reprs[i]);
+	for (i = 0; i < reprs->num_reprs; i++) {
+		struct net_device *netdev;
+
+		netdev = nfp_repr_get_locked(app, reprs, i);
+		if (netdev) {
+			struct nfp_repr *repr = netdev_priv(netdev);
 
 			err = nfp_flower_cmsg_portreify(repr, exists);
 			if (err)
 				return err;
 			count++;
 		}
+	}
 
 	return count;
 }
@@ -234,19 +238,21 @@ nfp_flower_spawn_vnic_reprs(struct nfp_a
 		return -ENOMEM;
 
 	for (i = 0; i < cnt; i++) {
+		struct net_device *repr;
 		struct nfp_port *port;
 		u32 port_id;
 
-		reprs->reprs[i] = nfp_repr_alloc(app);
-		if (!reprs->reprs[i]) {
+		repr = nfp_repr_alloc(app);
+		if (!repr) {
 			err = -ENOMEM;
 			goto err_reprs_clean;
 		}
+		RCU_INIT_POINTER(reprs->reprs[i], repr);
 
 		/* For now we only support 1 PF */
 		WARN_ON(repr_type == NFP_REPR_TYPE_PF && i);
 
-		port = nfp_port_alloc(app, port_type, reprs->reprs[i]);
+		port = nfp_port_alloc(app, port_type, repr);
 		if (repr_type == NFP_REPR_TYPE_PF) {
 			port->pf_id = i;
 			port->vnic = priv->nn->dp.ctrl_bar;
@@ -257,11 +263,11 @@ nfp_flower_spawn_vnic_reprs(struct nfp_a
 				app->pf->vf_cfg_mem + i * NFP_NET_CFG_BAR_SZ;
 		}
 
-		eth_hw_addr_random(reprs->reprs[i]);
+		eth_hw_addr_random(repr);
 
 		port_id = nfp_flower_cmsg_pcie_port(nfp_pcie, vnic_type,
 						    i, queue);
-		err = nfp_repr_init(app, reprs->reprs[i],
+		err = nfp_repr_init(app, repr,
 				    port_id, port, priv->nn->dp.netdev);
 		if (err) {
 			nfp_port_free(port);
@@ -270,7 +276,7 @@ nfp_flower_spawn_vnic_reprs(struct nfp_a
 
 		nfp_info(app->cpp, "%s%d Representor(%s) created\n",
 			 repr_type == NFP_REPR_TYPE_PF ? "PF" : "VF", i,
-			 reprs->reprs[i]->name);
+			 repr->name);
 	}
 
 	nfp_app_reprs_set(app, repr_type, reprs);
@@ -291,7 +297,7 @@ nfp_flower_spawn_vnic_reprs(struct nfp_a
 err_reprs_remove:
 	reprs = nfp_app_reprs_set(app, repr_type, NULL);
 err_reprs_clean:
-	nfp_reprs_clean_and_free(reprs);
+	nfp_reprs_clean_and_free(app, reprs);
 	return err;
 }
 
@@ -329,17 +335,18 @@ nfp_flower_spawn_phy_reprs(struct nfp_ap
 
 	for (i = 0; i < eth_tbl->count; i++) {
 		unsigned int phys_port = eth_tbl->ports[i].index;
+		struct net_device *repr;
 		struct nfp_port *port;
 		u32 cmsg_port_id;
 
-		reprs->reprs[phys_port] = nfp_repr_alloc(app);
-		if (!reprs->reprs[phys_port]) {
+		repr = nfp_repr_alloc(app);
+		if (!repr) {
 			err = -ENOMEM;
 			goto err_reprs_clean;
 		}
+		RCU_INIT_POINTER(reprs->reprs[phys_port], repr);
 
-		port = nfp_port_alloc(app, NFP_PORT_PHYS_PORT,
-				      reprs->reprs[phys_port]);
+		port = nfp_port_alloc(app, NFP_PORT_PHYS_PORT, repr);
 		if (IS_ERR(port)) {
 			err = PTR_ERR(port);
 			goto err_reprs_clean;
@@ -350,11 +357,11 @@ nfp_flower_spawn_phy_reprs(struct nfp_ap
 			goto err_reprs_clean;
 		}
 
-		SET_NETDEV_DEV(reprs->reprs[phys_port], &priv->nn->pdev->dev);
+		SET_NETDEV_DEV(repr, &priv->nn->pdev->dev);
 		nfp_net_get_mac_addr(app->pf, port);
 
 		cmsg_port_id = nfp_flower_cmsg_phys_port(phys_port);
-		err = nfp_repr_init(app, reprs->reprs[phys_port],
+		err = nfp_repr_init(app, repr,
 				    cmsg_port_id, port, priv->nn->dp.netdev);
 		if (err) {
 			nfp_port_free(port);
@@ -367,7 +374,7 @@ nfp_flower_spawn_phy_reprs(struct nfp_ap
 					     phys_port);
 
 		nfp_info(app->cpp, "Phys Port %d Representor(%s) created\n",
-			 phys_port, reprs->reprs[phys_port]->name);
+			 phys_port, repr->name);
 	}
 
 	nfp_app_reprs_set(app, NFP_REPR_TYPE_PHYS_PORT, reprs);
@@ -397,7 +404,7 @@ nfp_flower_spawn_phy_reprs(struct nfp_ap
 err_reprs_remove:
 	reprs = nfp_app_reprs_set(app, NFP_REPR_TYPE_PHYS_PORT, NULL);
 err_reprs_clean:
-	nfp_reprs_clean_and_free(reprs);
+	nfp_reprs_clean_and_free(app, reprs);
 err_free_ctrl_skb:
 	kfree_skb(ctrl_skb);
 	return err;
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
@@ -46,6 +46,13 @@
 #include "nfp_net_sriov.h"
 #include "nfp_port.h"
 
+struct net_device *
+nfp_repr_get_locked(struct nfp_app *app, struct nfp_reprs *set, unsigned int id)
+{
+	return rcu_dereference_protected(set->reprs[id],
+					 lockdep_is_held(&app->pf->lock));
+}
+
 static void
 nfp_repr_inc_tx_stats(struct net_device *netdev, unsigned int len,
 		      int tx_status)
@@ -369,21 +376,24 @@ static void nfp_repr_clean_and_free(stru
 	nfp_repr_free(repr);
 }
 
-void nfp_reprs_clean_and_free(struct nfp_reprs *reprs)
+void nfp_reprs_clean_and_free(struct nfp_app *app, struct nfp_reprs *reprs)
 {
+	struct net_device *netdev;
 	unsigned int i;
 
-	for (i = 0; i < reprs->num_reprs; i++)
-		if (reprs->reprs[i])
-			nfp_repr_clean_and_free(netdev_priv(reprs->reprs[i]));
+	for (i = 0; i < reprs->num_reprs; i++) {
+		netdev = nfp_repr_get_locked(app, reprs, i);
+		if (netdev)
+			nfp_repr_clean_and_free(netdev_priv(netdev));
+	}
 
 	kfree(reprs);
 }
 
 void
-nfp_reprs_clean_and_free_by_type(struct nfp_app *app,
-				 enum nfp_repr_type type)
+nfp_reprs_clean_and_free_by_type(struct nfp_app *app, enum nfp_repr_type type)
 {
+	struct net_device *netdev;
 	struct nfp_reprs *reprs;
 	int i;
 
@@ -395,14 +405,16 @@ nfp_reprs_clean_and_free_by_type(struct
 	/* Preclean must happen before we remove the reprs reference from the
 	 * app below.
 	 */
-	for (i = 0; i < reprs->num_reprs; i++)
-		if (reprs->reprs[i])
-			nfp_app_repr_preclean(app, reprs->reprs[i]);
+	for (i = 0; i < reprs->num_reprs; i++) {
+		netdev = nfp_repr_get_locked(app, reprs, i);
+		if (netdev)
+			nfp_app_repr_preclean(app, netdev);
+	}
 
 	reprs = nfp_app_reprs_set(app, type, NULL);
 
 	synchronize_rcu();
-	nfp_reprs_clean_and_free(reprs);
+	nfp_reprs_clean_and_free(app, reprs);
 }
 
 struct nfp_reprs *nfp_reprs_alloc(unsigned int num_reprs)
@@ -420,46 +432,29 @@ struct nfp_reprs *nfp_reprs_alloc(unsign
 
 int nfp_reprs_resync_phys_ports(struct nfp_app *app)
 {
-	struct nfp_reprs *reprs, *old_reprs;
+	struct net_device *netdev;
+	struct nfp_reprs *reprs;
 	struct nfp_repr *repr;
 	int i;
 
-	old_reprs = nfp_reprs_get_locked(app, NFP_REPR_TYPE_PHYS_PORT);
-	if (!old_reprs)
-		return 0;
-
-	reprs = nfp_reprs_alloc(old_reprs->num_reprs);
+	reprs = nfp_reprs_get_locked(app, NFP_REPR_TYPE_PHYS_PORT);
 	if (!reprs)
-		return -ENOMEM;
-
-	for (i = 0; i < old_reprs->num_reprs; i++) {
-		if (!old_reprs->reprs[i])
-			continue;
-
-		repr = netdev_priv(old_reprs->reprs[i]);
-		if (repr->port->type == NFP_PORT_INVALID) {
-			nfp_app_repr_preclean(app, old_reprs->reprs[i]);
-			continue;
-		}
-
-		reprs->reprs[i] = old_reprs->reprs[i];
-	}
-
-	old_reprs = nfp_app_reprs_set(app, NFP_REPR_TYPE_PHYS_PORT, reprs);
-	synchronize_rcu();
+		return 0;
 
-	/* Now we free up removed representors */
-	for (i = 0; i < old_reprs->num_reprs; i++) {
-		if (!old_reprs->reprs[i])
+	for (i = 0; i < reprs->num_reprs; i++) {
+		netdev = nfp_repr_get_locked(app, reprs, i);
+		if (!netdev)
 			continue;
 
-		repr = netdev_priv(old_reprs->reprs[i]);
+		repr = netdev_priv(netdev);
 		if (repr->port->type != NFP_PORT_INVALID)
 			continue;
 
+		nfp_app_repr_preclean(app, netdev);
+		rcu_assign_pointer(reprs->reprs[i], NULL);
+		synchronize_rcu();
 		nfp_repr_clean(repr);
 	}
 
-	kfree(old_reprs);
 	return 0;
 }
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.h
@@ -35,6 +35,7 @@
 #define NFP_NET_REPR_H
 
 struct metadata_dst;
+struct nfp_app;
 struct nfp_net;
 struct nfp_port;
 
@@ -47,7 +48,7 @@ struct nfp_port;
  */
 struct nfp_reprs {
 	unsigned int num_reprs;
-	struct net_device *reprs[0];
+	struct net_device __rcu *reprs[0];
 };
 
 /**
@@ -114,16 +115,18 @@ static inline int nfp_repr_get_port_id(s
 	return priv->dst->u.port_info.port_id;
 }
 
+struct net_device *
+nfp_repr_get_locked(struct nfp_app *app, struct nfp_reprs *set,
+		    unsigned int id);
+
 void nfp_repr_inc_rx_stats(struct net_device *netdev, unsigned int len);
 int nfp_repr_init(struct nfp_app *app, struct net_device *netdev,
 		  u32 cmsg_port_id, struct nfp_port *port,
 		  struct net_device *pf_netdev);
 struct net_device *nfp_repr_alloc(struct nfp_app *app);
-void
-nfp_reprs_clean_and_free(struct nfp_reprs *reprs);
-void
-nfp_reprs_clean_and_free_by_type(struct nfp_app *app,
-				 enum nfp_repr_type type);
+void nfp_reprs_clean_and_free(struct nfp_app *app, struct nfp_reprs *reprs);
+void nfp_reprs_clean_and_free_by_type(struct nfp_app *app,
+				      enum nfp_repr_type type);
 struct nfp_reprs *nfp_reprs_alloc(unsigned int num_reprs);
 int nfp_reprs_resync_phys_ports(struct nfp_app *app);