Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-04-05
Maciej Fijalkowski says:
We were solving issues around AF_XDP busy poll's not-so-usual scenarios,
such as very big busy poll budgets applied to very small HW rings. This
set carries the things that were found during that work that apply to
net tree.
One thing that was fixed for all in-tree ZC drivers was missing on ice
side all the time - it's about syncing RCU before destroying XDP
resources. Next one fixes the bit that is checked in ice_xsk_wakeup and
third one avoids false setting of DD bits on Tx descriptors.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/Documentation/devicetree/bindings/net/ethernet-controller.yaml b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
index 817794e5..4f15463 100644
--- a/Documentation/devicetree/bindings/net/ethernet-controller.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
@@ -106,6 +106,12 @@
phy-mode:
$ref: "#/properties/phy-connection-type"
+ pcs-handle:
+ $ref: /schemas/types.yaml#/definitions/phandle
+ description:
+ Specifies a reference to a node representing a PCS PHY device on a MDIO
+ bus to link with an external PHY (phy-handle) if exists.
+
phy-handle:
$ref: /schemas/types.yaml#/definitions/phandle
description:
diff --git a/Documentation/devicetree/bindings/net/xilinx_axienet.txt b/Documentation/devicetree/bindings/net/xilinx_axienet.txt
index b8e4894..1aa4c60 100644
--- a/Documentation/devicetree/bindings/net/xilinx_axienet.txt
+++ b/Documentation/devicetree/bindings/net/xilinx_axienet.txt
@@ -26,7 +26,8 @@
specified, the TX/RX DMA interrupts should be on that node
instead, and only the Ethernet core interrupt is optionally
specified here.
-- phy-handle : Should point to the external phy device.
+- phy-handle : Should point to the external phy device if exists. Pointing
+ this to the PCS/PMA PHY is deprecated and should be avoided.
See ethernet.txt file in the same directory.
- xlnx,rxmem : Set to allocated memory buffer for Rx/Tx in the hardware
@@ -68,6 +69,11 @@
required through the core's MDIO interface (i.e. always,
unless the PHY is accessed through a different bus).
+ - pcs-handle: Phandle to the internal PCS/PMA PHY in SGMII or 1000Base-X
+ modes, where "pcs-handle" should be used to point
+ to the PCS/PMA PHY, and "phy-handle" should point to an
+ external PHY if exists.
+
Example:
axi_ethernet_eth: ethernet@40c00000 {
compatible = "xlnx,axi-ethernet-1.00.a";
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index 5f9ab18..c188014 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -2751,7 +2751,7 @@ static int mv643xx_eth_shared_of_add_port(struct platform_device *pdev,
}
ret = of_get_mac_address(pnp, ppd.mac_addr);
- if (ret)
+ if (ret == -EPROBE_DEFER)
return ret;
mv643xx_eth_property(pnp, "tx-queue-size", ppd.tx_queue_size);
diff --git a/drivers/net/ethernet/micrel/Kconfig b/drivers/net/ethernet/micrel/Kconfig
index 1b632cd..830363b 100644
--- a/drivers/net/ethernet/micrel/Kconfig
+++ b/drivers/net/ethernet/micrel/Kconfig
@@ -28,6 +28,7 @@
config KS8851
tristate "Micrel KS8851 SPI"
depends on SPI
+ depends on PTP_1588_CLOCK_OPTIONAL
select MII
select CRC32
select EEPROM_93CX6
diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c
index 83e2723..377df8b 100644
--- a/drivers/net/ethernet/sfc/efx_channels.c
+++ b/drivers/net/ethernet/sfc/efx_channels.c
@@ -1140,7 +1140,7 @@ void efx_start_channels(struct efx_nic *efx)
struct efx_rx_queue *rx_queue;
struct efx_channel *channel;
- efx_for_each_channel(channel, efx) {
+ efx_for_each_channel_rev(channel, efx) {
efx_for_each_channel_tx_queue(tx_queue, channel) {
efx_init_tx_queue(tx_queue);
atomic_inc(&efx->active_queues);
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index d16e031..6983799 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -443,6 +443,9 @@ int efx_xdp_tx_buffers(struct efx_nic *efx, int n, struct xdp_frame **xdpfs,
if (unlikely(!tx_queue))
return -EINVAL;
+ if (!tx_queue->initialised)
+ return -EINVAL;
+
if (efx->xdp_txq_queues_mode != EFX_XDP_TX_QUEUES_DEDICATED)
HARD_TX_LOCK(efx->net_dev, tx_queue->core_txq, cpu);
diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c
index d530cde..9bc8281 100644
--- a/drivers/net/ethernet/sfc/tx_common.c
+++ b/drivers/net/ethernet/sfc/tx_common.c
@@ -101,6 +101,8 @@ void efx_fini_tx_queue(struct efx_tx_queue *tx_queue)
netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
"shutting down TX queue %d\n", tx_queue->queue);
+ tx_queue->initialised = false;
+
if (!tx_queue->buffer)
return;
diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h
index 0f9c88d..d5c1e5c 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet.h
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h
@@ -433,8 +433,6 @@ struct axienet_local {
struct net_device *ndev;
struct device *dev;
- struct device_node *phy_node;
-
struct phylink *phylink;
struct phylink_config phylink_config;
diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index c7eb05e..d6fc3f7 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -2064,25 +2064,33 @@ static int axienet_probe(struct platform_device *pdev)
if (ret)
goto cleanup_clk;
- lp->phy_node = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
- if (lp->phy_node) {
- ret = axienet_mdio_setup(lp);
- if (ret)
- dev_warn(&pdev->dev,
- "error registering MDIO bus: %d\n", ret);
- }
+ ret = axienet_mdio_setup(lp);
+ if (ret)
+ dev_warn(&pdev->dev,
+ "error registering MDIO bus: %d\n", ret);
+
if (lp->phy_mode == PHY_INTERFACE_MODE_SGMII ||
lp->phy_mode == PHY_INTERFACE_MODE_1000BASEX) {
- if (!lp->phy_node) {
- dev_err(&pdev->dev, "phy-handle required for 1000BaseX/SGMII\n");
+ np = of_parse_phandle(pdev->dev.of_node, "pcs-handle", 0);
+ if (!np) {
+ /* Deprecated: Always use "pcs-handle" for pcs_phy.
+ * Falling back to "phy-handle" here is only for
+ * backward compatibility with old device trees.
+ */
+ np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
+ }
+ if (!np) {
+ dev_err(&pdev->dev, "pcs-handle (preferred) or phy-handle required for 1000BaseX/SGMII\n");
ret = -EINVAL;
goto cleanup_mdio;
}
- lp->pcs_phy = of_mdio_find_device(lp->phy_node);
+ lp->pcs_phy = of_mdio_find_device(np);
if (!lp->pcs_phy) {
ret = -EPROBE_DEFER;
+ of_node_put(np);
goto cleanup_mdio;
}
+ of_node_put(np);
lp->pcs.ops = &axienet_pcs_ops;
lp->pcs.poll = true;
}
@@ -2125,8 +2133,6 @@ static int axienet_probe(struct platform_device *pdev)
put_device(&lp->pcs_phy->dev);
if (lp->mii_bus)
axienet_mdio_teardown(lp);
- of_node_put(lp->phy_node);
-
cleanup_clk:
clk_bulk_disable_unprepare(XAE_NUM_MISC_CLOCKS, lp->misc_clks);
clk_disable_unprepare(lp->axi_clk);
@@ -2155,9 +2161,6 @@ static int axienet_remove(struct platform_device *pdev)
clk_bulk_disable_unprepare(XAE_NUM_MISC_CLOCKS, lp->misc_clks);
clk_disable_unprepare(lp->axi_clk);
- of_node_put(lp->phy_node);
- lp->phy_node = NULL;
-
free_netdev(ndev);
return 0;
diff --git a/drivers/net/mdio/mdio-mscc-miim.c b/drivers/net/mdio/mdio-mscc-miim.c
index c483ba6..5829697 100644
--- a/drivers/net/mdio/mdio-mscc-miim.c
+++ b/drivers/net/mdio/mdio-mscc-miim.c
@@ -102,6 +102,9 @@ static int mscc_miim_read(struct mii_bus *bus, int mii_id, int regnum)
u32 val;
int ret;
+ if (regnum & MII_ADDR_C45)
+ return -EOPNOTSUPP;
+
ret = mscc_miim_wait_pending(bus);
if (ret)
goto out;
@@ -145,6 +148,9 @@ static int mscc_miim_write(struct mii_bus *bus, int mii_id,
struct mscc_miim_dev *miim = bus->priv;
int ret;
+ if (regnum & MII_ADDR_C45)
+ return -EOPNOTSUPP;
+
ret = mscc_miim_wait_pending(bus);
if (ret < 0)
goto out;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 5ddfdb2..128ee3b 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -5526,7 +5526,7 @@ int nft_set_elem_expr_clone(const struct nft_ctx *ctx, struct nft_set *set,
int err, i, k;
for (i = 0; i < set->num_exprs; i++) {
- expr = kzalloc(set->exprs[i]->ops->size, GFP_KERNEL);
+ expr = kzalloc(set->exprs[i]->ops->size, GFP_KERNEL_ACCOUNT);
if (!expr)
goto err_expr;
diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c
index 38caa66..f590ee1 100644
--- a/net/netfilter/nft_bitwise.c
+++ b/net/netfilter/nft_bitwise.c
@@ -290,7 +290,7 @@ static bool nft_bitwise_reduce(struct nft_regs_track *track,
if (!track->regs[priv->sreg].selector)
return false;
- bitwise = nft_expr_priv(expr);
+ bitwise = nft_expr_priv(track->regs[priv->dreg].selector);
if (track->regs[priv->sreg].selector == track->regs[priv->dreg].selector &&
track->regs[priv->sreg].num_reg == 0 &&
track->regs[priv->dreg].bitwise &&
@@ -442,7 +442,7 @@ static bool nft_bitwise_fast_reduce(struct nft_regs_track *track,
if (!track->regs[priv->sreg].selector)
return false;
- bitwise = nft_expr_priv(expr);
+ bitwise = nft_expr_priv(track->regs[priv->dreg].selector);
if (track->regs[priv->sreg].selector == track->regs[priv->dreg].selector &&
track->regs[priv->dreg].bitwise &&
track->regs[priv->dreg].bitwise->ops == expr->ops &&
diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c
index 9de1462..d657f99 100644
--- a/net/netfilter/nft_connlimit.c
+++ b/net/netfilter/nft_connlimit.c
@@ -77,7 +77,7 @@ static int nft_connlimit_do_init(const struct nft_ctx *ctx,
invert = true;
}
- priv->list = kmalloc(sizeof(*priv->list), GFP_KERNEL);
+ priv->list = kmalloc(sizeof(*priv->list), GFP_KERNEL_ACCOUNT);
if (!priv->list)
return -ENOMEM;
diff --git a/net/netfilter/nft_counter.c b/net/netfilter/nft_counter.c
index da90836..f4d3573e 100644
--- a/net/netfilter/nft_counter.c
+++ b/net/netfilter/nft_counter.c
@@ -62,7 +62,7 @@ static int nft_counter_do_init(const struct nlattr * const tb[],
struct nft_counter __percpu *cpu_stats;
struct nft_counter *this_cpu;
- cpu_stats = alloc_percpu(struct nft_counter);
+ cpu_stats = alloc_percpu_gfp(struct nft_counter, GFP_KERNEL_ACCOUNT);
if (cpu_stats == NULL)
return -ENOMEM;
diff --git a/net/netfilter/nft_last.c b/net/netfilter/nft_last.c
index 43d0d4a..bb15a55 100644
--- a/net/netfilter/nft_last.c
+++ b/net/netfilter/nft_last.c
@@ -30,7 +30,7 @@ static int nft_last_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
u64 last_jiffies;
int err;
- last = kzalloc(sizeof(*last), GFP_KERNEL);
+ last = kzalloc(sizeof(*last), GFP_KERNEL_ACCOUNT);
if (!last)
return -ENOMEM;
diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c
index d4a6cf3c..04ea8b9 100644
--- a/net/netfilter/nft_limit.c
+++ b/net/netfilter/nft_limit.c
@@ -90,7 +90,7 @@ static int nft_limit_init(struct nft_limit_priv *priv,
priv->rate);
}
- priv->limit = kmalloc(sizeof(*priv->limit), GFP_KERNEL);
+ priv->limit = kmalloc(sizeof(*priv->limit), GFP_KERNEL_ACCOUNT);
if (!priv->limit)
return -ENOMEM;
diff --git a/net/netfilter/nft_quota.c b/net/netfilter/nft_quota.c
index d7db57e..e6b0df6 100644
--- a/net/netfilter/nft_quota.c
+++ b/net/netfilter/nft_quota.c
@@ -90,7 +90,7 @@ static int nft_quota_do_init(const struct nlattr * const tb[],
return -EOPNOTSUPP;
}
- priv->consumed = kmalloc(sizeof(*priv->consumed), GFP_KERNEL);
+ priv->consumed = kmalloc(sizeof(*priv->consumed), GFP_KERNEL_ACCOUNT);
if (!priv->consumed)
return -ENOMEM;
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 7056cb1b..1b5d730 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -1051,7 +1051,7 @@ static int clone(struct datapath *dp, struct sk_buff *skb,
int rem = nla_len(attr);
bool dont_clone_flow_key;
- /* The first action is always 'OVS_CLONE_ATTR_ARG'. */
+ /* The first action is always 'OVS_CLONE_ATTR_EXEC'. */
clone_arg = nla_data(attr);
dont_clone_flow_key = nla_get_u32(clone_arg);
actions = nla_next(clone_arg, &rem);
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index cc282a5..7176156 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -2317,6 +2317,62 @@ static struct sw_flow_actions *nla_alloc_flow_actions(int size)
return sfa;
}
+static void ovs_nla_free_nested_actions(const struct nlattr *actions, int len);
+
+static void ovs_nla_free_check_pkt_len_action(const struct nlattr *action)
+{
+ const struct nlattr *a;
+ int rem;
+
+ nla_for_each_nested(a, action, rem) {
+ switch (nla_type(a)) {
+ case OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_LESS_EQUAL:
+ case OVS_CHECK_PKT_LEN_ATTR_ACTIONS_IF_GREATER:
+ ovs_nla_free_nested_actions(nla_data(a), nla_len(a));
+ break;
+ }
+ }
+}
+
+static void ovs_nla_free_clone_action(const struct nlattr *action)
+{
+ const struct nlattr *a = nla_data(action);
+ int rem = nla_len(action);
+
+ switch (nla_type(a)) {
+ case OVS_CLONE_ATTR_EXEC:
+ /* The real list of actions follows this attribute. */
+ a = nla_next(a, &rem);
+ ovs_nla_free_nested_actions(a, rem);
+ break;
+ }
+}
+
+static void ovs_nla_free_dec_ttl_action(const struct nlattr *action)
+{
+ const struct nlattr *a = nla_data(action);
+
+ switch (nla_type(a)) {
+ case OVS_DEC_TTL_ATTR_ACTION:
+ ovs_nla_free_nested_actions(nla_data(a), nla_len(a));
+ break;
+ }
+}
+
+static void ovs_nla_free_sample_action(const struct nlattr *action)
+{
+ const struct nlattr *a = nla_data(action);
+ int rem = nla_len(action);
+
+ switch (nla_type(a)) {
+ case OVS_SAMPLE_ATTR_ARG:
+ /* The real list of actions follows this attribute. */
+ a = nla_next(a, &rem);
+ ovs_nla_free_nested_actions(a, rem);
+ break;
+ }
+}
+
static void ovs_nla_free_set_action(const struct nlattr *a)
{
const struct nlattr *ovs_key = nla_data(a);
@@ -2330,25 +2386,54 @@ static void ovs_nla_free_set_action(const struct nlattr *a)
}
}
-void ovs_nla_free_flow_actions(struct sw_flow_actions *sf_acts)
+static void ovs_nla_free_nested_actions(const struct nlattr *actions, int len)
{
const struct nlattr *a;
int rem;
- if (!sf_acts)
+ /* Whenever new actions are added, the need to update this
+ * function should be considered.
+ */
+ BUILD_BUG_ON(OVS_ACTION_ATTR_MAX != 23);
+
+ if (!actions)
return;
- nla_for_each_attr(a, sf_acts->actions, sf_acts->actions_len, rem) {
+ nla_for_each_attr(a, actions, len, rem) {
switch (nla_type(a)) {
- case OVS_ACTION_ATTR_SET:
- ovs_nla_free_set_action(a);
+ case OVS_ACTION_ATTR_CHECK_PKT_LEN:
+ ovs_nla_free_check_pkt_len_action(a);
break;
+
+ case OVS_ACTION_ATTR_CLONE:
+ ovs_nla_free_clone_action(a);
+ break;
+
case OVS_ACTION_ATTR_CT:
ovs_ct_free_action(a);
break;
+
+ case OVS_ACTION_ATTR_DEC_TTL:
+ ovs_nla_free_dec_ttl_action(a);
+ break;
+
+ case OVS_ACTION_ATTR_SAMPLE:
+ ovs_nla_free_sample_action(a);
+ break;
+
+ case OVS_ACTION_ATTR_SET:
+ ovs_nla_free_set_action(a);
+ break;
}
}
+}
+void ovs_nla_free_flow_actions(struct sw_flow_actions *sf_acts)
+{
+ if (!sf_acts)
+ return;
+
+ ovs_nla_free_nested_actions(sf_acts->actions, sf_acts->actions_len);
kfree(sf_acts);
}
@@ -3458,7 +3543,9 @@ static int clone_action_to_attr(const struct nlattr *attr,
if (!start)
return -EMSGSIZE;
- err = ovs_nla_put_actions(nla_data(attr), rem, skb);
+ /* Skipping the OVS_CLONE_ATTR_EXEC that is always the first attribute. */
+ attr = nla_next(nla_data(attr), &rem);
+ err = ovs_nla_put_actions(attr, rem, skb);
if (err)
nla_nest_cancel(skb, start);
diff --git a/net/rxrpc/net_ns.c b/net/rxrpc/net_ns.c
index 25bbc4c..f15d694 100644
--- a/net/rxrpc/net_ns.c
+++ b/net/rxrpc/net_ns.c
@@ -113,8 +113,8 @@ static __net_exit void rxrpc_exit_net(struct net *net)
struct rxrpc_net *rxnet = rxrpc_net(net);
rxnet->live = false;
- del_timer_sync(&rxnet->peer_keepalive_timer);
cancel_work_sync(&rxnet->peer_keepalive_work);
+ del_timer_sync(&rxnet->peer_keepalive_timer);
rxrpc_destroy_all_calls(rxnet);
rxrpc_destroy_all_connections(rxnet);
rxrpc_destroy_all_peers(rxnet);
