Skip to contentSkip to conversation

r8169: mark device as detached in PCI D3 and improve locking

PATCH In review Heiner Kallweit 3 participants 10 messages Jun 20, 2020 – Jun 22, 2020

Subsystems (3)
lore
Share & tools
lore

Heiner Kallweit

author maintainer
#0 lore raw

cover r8169: mark device as detached in PCI D3 and improve locking

Mark the netdevice as detached whenever parent is in PCI D3hot and not accessible. This mainly applies to runtime-suspend state. In addition take RTNL lock in suspend calls, this allows to remove the driver-specific mutex and improve PM callbacks in general.

Heiner Kallweit (7):
net: core: try to runtime-resume detached device in __dev_open
r8169: mark device as not present when in PCI D3
r8169: remove no longer needed checks for device being runtime-active
r8169: add rtl8169_up
r8169: use RTNL to protect critical sections
r8169: remove driver-specific mutex
r8169: improve rtl8169_runtime_resume

 drivers/net/ethernet/realtek/r8169_main.c | 181 +++++-----------------
 net/core/dev.c                            |  10 +-
 2 files changed, 49 insertions(+), 142 deletions(-)
Signature · 1 line
2.27.0
author maintainer
#1 lore raw

PATCH 1/7 net: core: try to runtime-resume detached device in __dev_open

A netdevice may be marked as detached because the parent is runtime-suspended and not accessible whilst interface or link is down. An example are PCI network devices that go into PCI D3hot, see e.g. __igc_shutdown() or rtl8169_net_suspend().
If netdevice is down and marked as detached we can only open it if we runtime-resume it before __dev_open() calls netif_device_present().

Therefore, if netdevice is detached, try to runtime-resume the parent and only return with an error if it's still detached.

---
 net/core/dev.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
net/core/dev.c +8 -2
index 6bc238814..ffa8c371d 100644
@@ -143,6 +143,7 @@
 #include <linux/net_namespace.h>
 #include <linux/indirect_call_wrapper.h>
 #include <net/devlink.h>
+#include <linux/pm_runtime.h>
 
 #include "net-sysfs.h"
 
@@ -1492,8 +1493,13 @@ static int __dev_open(struct net_device *dev, struct netlink_ext_ack *extack)
 
 	ASSERT_RTNL();
 
-	if (!netif_device_present(dev))
-		return -ENODEV;
+	if (!netif_device_present(dev)) {
+		/* may be detached because parent is runtime-suspended */
+		if (dev->dev.parent)
+			pm_runtime_resume(dev->dev.parent);
+		if (!netif_device_present(dev))
+			return -ENODEV;
+	}
 
 	/* Block netpoll from trying to do any rx path servicing.
 	 * If we don't do this there is a chance ndo_poll_controller
Signature · 1 line
2.27.0
author maintainer
#2 lore raw

PATCH 2/7 r8169: mark device as not present when in PCI D3

In reply to Heiner Kallweit

Mark the netdevice as detached whenever we go into PCI D3hot. This allows to remove some checks e.g. from ethtool ops because dev_ethtool() checks for netif_device_present() in the beginning.

In this context move waking up the queue out of rtl_reset_work() because in cases where netif_device_attach() is called afterwards the queue should be woken up by the latter function only.

---
 drivers/net/ethernet/realtek/r8169_main.c | 25 +++++++++++++----------
 1 file changed, 14 insertions(+), 11 deletions(-)
drivers/net/ethernet/realtek/r8169_main.c +14 -11
index 98391797b..91e3ada64 100644
@@ -3980,10 +3980,9 @@ static void rtl8169_cleanup(struct rtl8169_private *tp, bool going_down)
 
 static void rtl_reset_work(struct rtl8169_private *tp)
 {
-	struct net_device *dev = tp->dev;
 	int i;
 
-	netif_stop_queue(dev);
+	netif_stop_queue(tp->dev);
 
 	rtl8169_cleanup(tp, false);
 
@@ -3992,7 +3991,6 @@ static void rtl_reset_work(struct rtl8169_private *tp)
 
 	napi_enable(&tp->napi);
 	rtl_hw_start(tp);
-	netif_wake_queue(dev);
 }
 
 static void rtl8169_tx_timeout(struct net_device *dev, unsigned int txqueue)
@@ -4577,8 +4575,10 @@ static void rtl_task(struct work_struct *work)
 	    !test_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags))
 		goto out_unlock;
 
-	if (test_and_clear_bit(RTL_FLAG_TASK_RESET_PENDING, tp->wk.flags))
+	if (test_and_clear_bit(RTL_FLAG_TASK_RESET_PENDING, tp->wk.flags)) {
 		rtl_reset_work(tp);
+		netif_wake_queue(tp->dev);
+	}
 out_unlock:
 	rtl_unlock_work(tp);
 }
@@ -4823,11 +4823,10 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 
 static void rtl8169_net_suspend(struct rtl8169_private *tp)
 {
-	if (!netif_running(tp->dev))
-		return;
-
 	netif_device_detach(tp->dev);
-	rtl8169_down(tp);
+
+	if (netif_running(tp->dev))
+		rtl8169_down(tp);
 }
 
 #ifdef CONFIG_PM
@@ -4843,8 +4842,6 @@ static int __maybe_unused rtl8169_suspend(struct device *device)
 
 static void __rtl8169_resume(struct rtl8169_private *tp)
 {
-	netif_device_attach(tp->dev);
-
 	rtl_pll_power_up(tp);
 	rtl8169_init_phy(tp);
 
@@ -4866,6 +4863,8 @@ static int __maybe_unused rtl8169_resume(struct device *device)
 	if (netif_running(tp->dev))
 		__rtl8169_resume(tp);
 
+	netif_device_attach(tp->dev);
+
 	return 0;
 }
 
@@ -4873,8 +4872,10 @@ static int rtl8169_runtime_suspend(struct device *device)
 {
 	struct rtl8169_private *tp = dev_get_drvdata(device);
 
-	if (!tp->TxDescArray)
+	if (!tp->TxDescArray) {
+		netif_device_detach(tp->dev);
 		return 0;
+	}
 
 	rtl_lock_work(tp);
 	__rtl8169_set_wol(tp, WAKE_PHY);
@@ -4898,6 +4899,8 @@ static int rtl8169_runtime_resume(struct device *device)
 	if (tp->TxDescArray)
 		__rtl8169_resume(tp);
 
+	netif_device_attach(tp->dev);
+
 	return 0;
 }
 
Signature · 1 line
2.27.0
author maintainer
#3 lore raw

PATCH 3/7 r8169: remove no longer needed checks for device being runtime-active

In reply to Heiner Kallweit

Because the netdevice is marked as detached now when parent is not accessible we can remove quite some checks.

---
 drivers/net/ethernet/realtek/r8169_main.c | 54 +++--------------------
 1 file changed, 6 insertions(+), 48 deletions(-)
drivers/net/ethernet/realtek/r8169_main.c +6 -48
index 91e3ada64..c8e0f2bb5 100644
@@ -1422,24 +1422,17 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 static int rtl8169_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
-	struct device *d = tp_to_dev(tp);
 
 	if (wol->wolopts & ~WAKE_ANY)
 		return -EINVAL;
 
-	pm_runtime_get_noresume(d);
-
 	rtl_lock_work(tp);
 
 	tp->saved_wolopts = wol->wolopts;
-
-	if (pm_runtime_active(d))
-		__rtl8169_set_wol(tp, tp->saved_wolopts);
+	__rtl8169_set_wol(tp, tp->saved_wolopts);
 
 	rtl_unlock_work(tp);
 
-	pm_runtime_put_noidle(d);
-
 	return 0;
 }
 
@@ -1657,17 +1650,10 @@ static void rtl8169_get_ethtool_stats(struct net_device *dev,
 				      struct ethtool_stats *stats, u64 *data)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
-	struct device *d = tp_to_dev(tp);
-	struct rtl8169_counters *counters = tp->counters;
-
-	ASSERT_RTNL();
-
-	pm_runtime_get_noresume(d);
-
-	if (pm_runtime_active(d))
-		rtl8169_update_counters(tp);
+	struct rtl8169_counters *counters;
 
-	pm_runtime_put_noidle(d);
+	counters = tp->counters;
+	rtl8169_update_counters(tp);
 
 	data[0] = le64_to_cpu(counters->tx_packets);
 	data[1] = le64_to_cpu(counters->rx_packets);
@@ -1899,48 +1885,26 @@ static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 static int rtl8169_get_eee(struct net_device *dev, struct ethtool_eee *data)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
-	struct device *d = tp_to_dev(tp);
-	int ret;
 
 	if (!rtl_supports_eee(tp))
 		return -EOPNOTSUPP;
 
-	pm_runtime_get_noresume(d);
-
-	if (!pm_runtime_active(d)) {
-		ret = -EOPNOTSUPP;
-	} else {
-		ret = phy_ethtool_get_eee(tp->phydev, data);
-	}
-
-	pm_runtime_put_noidle(d);
-
-	return ret;
+	return phy_ethtool_get_eee(tp->phydev, data);
 }
 
 static int rtl8169_set_eee(struct net_device *dev, struct ethtool_eee *data)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
-	struct device *d = tp_to_dev(tp);
 	int ret;
 
 	if (!rtl_supports_eee(tp))
 		return -EOPNOTSUPP;
 
-	pm_runtime_get_noresume(d);
-
-	if (!pm_runtime_active(d)) {
-		ret = -EOPNOTSUPP;
-		goto out;
-	}
-
 	ret = phy_ethtool_set_eee(tp->phydev, data);
 
 	if (!ret)
 		tp->eee_adv = phy_read_mmd(dev->phydev, MDIO_MMD_AN,
 					   MDIO_AN_EEE_ADV);
-out:
-	pm_runtime_put_noidle(d);
 	return ret;
 }
 
@@ -2222,19 +2186,13 @@ static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr)
 static int rtl_set_mac_address(struct net_device *dev, void *p)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
-	struct device *d = tp_to_dev(tp);
 	int ret;
 
 	ret = eth_mac_addr(dev, p);
 	if (ret)
 		return ret;
 
-	pm_runtime_get_noresume(d);
-
-	if (pm_runtime_active(d))
-		rtl_rar_set(tp, dev->dev_addr);
-
-	pm_runtime_put_noidle(d);
+	rtl_rar_set(tp, dev->dev_addr);
 
 	return 0;
 }
Signature · 1 line
2.27.0
author maintainer
#4 lore raw

PATCH 4/7 r8169: add rtl8169_up

In reply to Heiner Kallweit

Factor out bringing device up to a new function rtl8169_up(), similar to rtl8169_down() for bringing the device down.

---
 drivers/net/ethernet/realtek/r8169_main.c | 48 ++++++++---------------
 1 file changed, 16 insertions(+), 32 deletions(-)
drivers/net/ethernet/realtek/r8169_main.c +16 -32
index c8e0f2bb5..2414df29c 100644
@@ -4616,6 +4616,19 @@ static void rtl8169_down(struct rtl8169_private *tp)
 	rtl_unlock_work(tp);
 }
 
+static void rtl8169_up(struct rtl8169_private *tp)
+{
+	rtl_lock_work(tp);
+	rtl_pll_power_up(tp);
+	rtl8169_init_phy(tp);
+	napi_enable(&tp->napi);
+	set_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags);
+	rtl_reset_work(tp);
+
+	phy_start(tp->phydev);
+	rtl_unlock_work(tp);
+}
+
 static int rtl8169_close(struct net_device *dev)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
@@ -4691,25 +4704,10 @@ static int rtl_open(struct net_device *dev)
 	if (retval)
 		goto err_free_irq;
 
-	rtl_lock_work(tp);
-
-	set_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags);
-
-	napi_enable(&tp->napi);
-
-	rtl8169_init_phy(tp);
-
-	rtl_pll_power_up(tp);
-
-	rtl_hw_start(tp);
-
+	rtl8169_up(tp);
 	rtl8169_init_counter_offsets(tp);
-
-	phy_start(tp->phydev);
 	netif_start_queue(dev);
 
-	rtl_unlock_work(tp);
-
 	pm_runtime_put_sync(&pdev->dev);
 out:
 	return retval;
@@ -4798,20 +4796,6 @@ static int __maybe_unused rtl8169_suspend(struct device *device)
 	return 0;
 }
 
-static void __rtl8169_resume(struct rtl8169_private *tp)
-{
-	rtl_pll_power_up(tp);
-	rtl8169_init_phy(tp);
-
-	phy_start(tp->phydev);
-
-	rtl_lock_work(tp);
-	napi_enable(&tp->napi);
-	set_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags);
-	rtl_reset_work(tp);
-	rtl_unlock_work(tp);
-}
-
 static int __maybe_unused rtl8169_resume(struct device *device)
 {
 	struct rtl8169_private *tp = dev_get_drvdata(device);
@@ -4819,7 +4803,7 @@ static int __maybe_unused rtl8169_resume(struct device *device)
 	rtl_rar_set(tp, tp->dev->dev_addr);
 
 	if (netif_running(tp->dev))
-		__rtl8169_resume(tp);
+		rtl8169_up(tp);
 
 	netif_device_attach(tp->dev);
 
@@ -4855,7 +4839,7 @@ static int rtl8169_runtime_resume(struct device *device)
 	rtl_unlock_work(tp);
 
 	if (tp->TxDescArray)
-		__rtl8169_resume(tp);
+		rtl8169_up(tp);
 
 	netif_device_attach(tp->dev);
 
Signature · 1 line
2.27.0
author maintainer
#5 lore raw

PATCH 5/7 r8169: use RTNL to protect critical sections

In reply to Heiner Kallweit

Most relevant ops (open, close, ethtool ops) are protected with RTNL lock by net core. Make sure that such ops can't be interrupted by e.g. (runtime-)suspending by taking the RTNL lock in suspend ops and the PCI error handler.

---
 drivers/net/ethernet/realtek/r8169_main.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/net/ethernet/realtek/r8169_main.c +8 -0
index 2414df29c..e70797311 100644
@@ -4527,6 +4527,7 @@ static void rtl_task(struct work_struct *work)
 	struct rtl8169_private *tp =
 		container_of(work, struct rtl8169_private, wk.work);
 
+	rtnl_lock();
 	rtl_lock_work(tp);
 
 	if (!netif_running(tp->dev) ||
@@ -4539,6 +4540,7 @@ static void rtl_task(struct work_struct *work)
 	}
 out_unlock:
 	rtl_unlock_work(tp);
+	rtnl_unlock();
 }
 
 static int rtl8169_poll(struct napi_struct *napi, int budget)
@@ -4791,7 +4793,9 @@ static int __maybe_unused rtl8169_suspend(struct device *device)
 {
 	struct rtl8169_private *tp = dev_get_drvdata(device);
 
+	rtnl_lock();
 	rtl8169_net_suspend(tp);
+	rtnl_unlock();
 
 	return 0;
 }
@@ -4819,11 +4823,13 @@ static int rtl8169_runtime_suspend(struct device *device)
 		return 0;
 	}
 
+	rtnl_lock();
 	rtl_lock_work(tp);
 	__rtl8169_set_wol(tp, WAKE_PHY);
 	rtl_unlock_work(tp);
 
 	rtl8169_net_suspend(tp);
+	rtnl_unlock();
 
 	return 0;
 }
@@ -4885,7 +4891,9 @@ static void rtl_shutdown(struct pci_dev *pdev)
 {
 	struct rtl8169_private *tp = pci_get_drvdata(pdev);
 
+	rtnl_lock();
 	rtl8169_net_suspend(tp);
+	rtnl_unlock();
 
 	/* Restore original MAC address */
 	rtl_rar_set(tp, tp->dev->perm_addr);
Signature · 1 line
2.27.0
author maintainer
#6 lore raw

PATCH 6/7 r8169: remove driver-specific mutex

In reply to Heiner Kallweit

Now that the critical sections are protected with RTNL lock, we don't need a separate mutex any longer.

---
 drivers/net/ethernet/realtek/r8169_main.c | 45 -----------------------
 1 file changed, 45 deletions(-)
drivers/net/ethernet/realtek/r8169_main.c +0 -45
index e70797311..e09732c9d 100644
@@ -611,7 +611,6 @@ struct rtl8169_private {
 
 	struct {
 		DECLARE_BITMAP(flags, RTL_FLAG_MAX);
-		struct mutex mutex;
 		struct work_struct work;
 	} wk;
 
@@ -663,16 +662,6 @@ static inline struct device *tp_to_dev(struct rtl8169_private *tp)
 	return &tp->pci_dev->dev;
 }
 
-static void rtl_lock_work(struct rtl8169_private *tp)
-{
-	mutex_lock(&tp->wk.mutex);
-}
-
-static void rtl_unlock_work(struct rtl8169_private *tp)
-{
-	mutex_unlock(&tp->wk.mutex);
-}
-
 static void rtl_lock_config_regs(struct rtl8169_private *tp)
 {
 	RTL_W8(tp, Cfg9346, Cfg9346_Lock);
@@ -1348,10 +1337,8 @@ static void rtl8169_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
 
-	rtl_lock_work(tp);
 	wol->supported = WAKE_ANY;
 	wol->wolopts = tp->saved_wolopts;
-	rtl_unlock_work(tp);
 }
 
 static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
@@ -1426,13 +1413,9 @@ static int rtl8169_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 	if (wol->wolopts & ~WAKE_ANY)
 		return -EINVAL;
 
-	rtl_lock_work(tp);
-
 	tp->saved_wolopts = wol->wolopts;
 	__rtl8169_set_wol(tp, tp->saved_wolopts);
 
-	rtl_unlock_work(tp);
-
 	return 0;
 }
 
@@ -1495,8 +1478,6 @@ static int rtl8169_set_features(struct net_device *dev,
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
 
-	rtl_lock_work(tp);
-
 	rtl_set_rx_config_features(tp, features);
 
 	if (features & NETIF_F_RXCSUM)
@@ -1514,8 +1495,6 @@ static int rtl8169_set_features(struct net_device *dev,
 	RTL_W16(tp, CPlusCmd, tp->cp_cmd);
 	rtl_pci_commit(tp);
 
-	rtl_unlock_work(tp);
-
 	return 0;
 }
 
@@ -1541,10 +1520,8 @@ static void rtl8169_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 	u32 *dw = p;
 	int i;
 
-	rtl_lock_work(tp);
 	for (i = 0; i < R8169_REGS_SIZE; i += 4)
 		memcpy_fromio(dw++, data++, 4);
-	rtl_unlock_work(tp);
 }
 
 static const char rtl8169_gstrings[][ETH_GSTRING_LEN] = {
@@ -1860,8 +1837,6 @@ static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 	units = DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000U, scale);
 	w |= FIELD_PREP(RTL_COALESCE_RX_USECS, units);
 
-	rtl_lock_work(tp);
-
 	RTL_W16(tp, IntrMitigate, w);
 
 	/* Meaning of PktCntrDisable bit changed from RTL8168e-vl */
@@ -1877,8 +1852,6 @@ static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 	RTL_W16(tp, CPlusCmd, tp->cp_cmd);
 	rtl_pci_commit(tp);
 
-	rtl_unlock_work(tp);
-
 	return 0;
 }
 
@@ -2165,8 +2138,6 @@ static void rtl8169_init_phy(struct rtl8169_private *tp)
 
 static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr)
 {
-	rtl_lock_work(tp);
-
 	rtl_unlock_config_regs(tp);
 
 	RTL_W32(tp, MAC4, addr[4] | addr[5] << 8);
@@ -2179,8 +2150,6 @@ static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr)
 		rtl_rar_exgmac_set(tp, addr);
 
 	rtl_lock_config_regs(tp);
-
-	rtl_unlock_work(tp);
 }
 
 static int rtl_set_mac_address(struct net_device *dev, void *p)
@@ -4528,7 +4497,6 @@ static void rtl_task(struct work_struct *work)
 		container_of(work, struct rtl8169_private, wk.work);
 
 	rtnl_lock();
-	rtl_lock_work(tp);
 
 	if (!netif_running(tp->dev) ||
 	    !test_bit(RTL_FLAG_TASK_ENABLED, tp->wk.flags))
@@ -4539,7 +4507,6 @@ static void rtl_task(struct work_struct *work)
 		netif_wake_queue(tp->dev);
 	}
 out_unlock:
-	rtl_unlock_work(tp);
 	rtnl_unlock();
 }
 
@@ -4602,8 +4569,6 @@ static int r8169_phy_connect(struct rtl8169_private *tp)
 
 static void rtl8169_down(struct rtl8169_private *tp)
 {
-	rtl_lock_work(tp);
-
 	/* Clear all task flags */
 	bitmap_zero(tp->wk.flags, RTL_FLAG_MAX);
 
@@ -4614,13 +4579,10 @@ static void rtl8169_down(struct rtl8169_private *tp)
 	rtl8169_cleanup(tp, true);
 
 	rtl_pll_power_down(tp);
-
-	rtl_unlock_work(tp);
 }
 
 static void rtl8169_up(struct rtl8169_private *tp)
 {
-	rtl_lock_work(tp);
 	rtl_pll_power_up(tp);
 	rtl8169_init_phy(tp);
 	napi_enable(&tp->napi);
@@ -4628,7 +4590,6 @@ static void rtl8169_up(struct rtl8169_private *tp)
 	rtl_reset_work(tp);
 
 	phy_start(tp->phydev);
-	rtl_unlock_work(tp);
 }
 
 static int rtl8169_close(struct net_device *dev)
@@ -4824,10 +4785,7 @@ static int rtl8169_runtime_suspend(struct device *device)
 	}
 
 	rtnl_lock();
-	rtl_lock_work(tp);
 	__rtl8169_set_wol(tp, WAKE_PHY);
-	rtl_unlock_work(tp);
-
 	rtl8169_net_suspend(tp);
 	rtnl_unlock();
 
@@ -4840,9 +4798,7 @@ static int rtl8169_runtime_resume(struct device *device)
 
 	rtl_rar_set(tp, tp->dev->dev_addr);
 
-	rtl_lock_work(tp);
 	__rtl8169_set_wol(tp, tp->saved_wolopts);
-	rtl_unlock_work(tp);
 
 	if (tp->TxDescArray)
 		rtl8169_up(tp);
@@ -5299,7 +5255,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return rc;
 	}
 
-	mutex_init(&tp->wk.mutex);
 	INIT_WORK(&tp->wk.work, rtl_task);
 	u64_stats_init(&tp->rx_stats.syncp);
 	u64_stats_init(&tp->tx_stats.syncp);
Signature · 1 line
2.27.0
author maintainer
#7 lore raw

PATCH 7/7 r8169: improve rtl8169_runtime_resume

In reply to Heiner Kallweit

Simplify rtl8169_runtime_resume() by calling rtl8169_resume().

---
 drivers/net/ethernet/realtek/r8169_main.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)
drivers/net/ethernet/realtek/r8169_main.c +3 -10
index e09732c9d..247fad1c6 100644
@@ -4761,13 +4761,13 @@ static int __maybe_unused rtl8169_suspend(struct device *device)
 	return 0;
 }
 
-static int __maybe_unused rtl8169_resume(struct device *device)
+static int rtl8169_resume(struct device *device)
 {
 	struct rtl8169_private *tp = dev_get_drvdata(device);
 
 	rtl_rar_set(tp, tp->dev->dev_addr);
 
-	if (netif_running(tp->dev))
+	if (tp->TxDescArray)
 		rtl8169_up(tp);
 
 	netif_device_attach(tp->dev);
@@ -4796,16 +4796,9 @@ static int rtl8169_runtime_resume(struct device *device)
 {
 	struct rtl8169_private *tp = dev_get_drvdata(device);
 
-	rtl_rar_set(tp, tp->dev->dev_addr);
-
 	__rtl8169_set_wol(tp, tp->saved_wolopts);
 
-	if (tp->TxDescArray)
-		rtl8169_up(tp);
-
-	netif_device_attach(tp->dev);
-
-	return 0;
+	return rtl8169_resume(device);
 }
 
 static int rtl8169_runtime_idle(struct device *device)
Signature · 1 line
2.27.0
maintainer Reviewed-by on the v1 series
#8 lore raw
In reply to Heiner Kallweit
Quoted 4 lines · On Sat, 20 Jun 2020 22:33:39 +0200 Heiner Kallweit wrote:

Mark the netdevice as detached whenever parent is in PCI D3hot and not accessible. This mainly applies to runtime-suspend state. In addition take RTNL lock in suspend calls, this allows to remove the driver-specific mutex and improve PM callbacks in general.

Not an expert on PM but looks like a nice improvement to me:

maintainer
#9 lore raw

From: Jakub Kicinski <[email protected]>
Date: Mon, 22 Jun 2020 15:27:04 -0700

Quoted 9 lines · On Sat, 20 Jun 2020 22:33:39 +0200 Heiner Kallweit wrote:
Quoted 4 lines

Mark the netdevice as detached whenever parent is in PCI D3hot and not accessible. This mainly applies to runtime-suspend state. In addition take RTNL lock in suspend calls, this allows to remove the driver-specific mutex and improve PM callbacks in general.

Not an expert on PM but looks like a nice improvement to me:

Series applied, thanks everyone.