fix mac not working after system resumed with WoL enabled
Subsystems (6)
v2 cover fix mac not working after system resumed with WoL enabled
Hi,
The issue description is in the commit message.
This patchset is the second version following discussions with Russell. But the name of each patch has changed, so V2 is not marked in each patch.
Thanks.
Clark Wang (2):
net: phylink: add a function to resume phy alone to fix resume issue
with WoL enabled
net: stmmac: resume phy separately before calling stmmac_hw_setup()
.../net/ethernet/stmicro/stmmac/stmmac_main.c | 16 +++++++------- drivers/net/phy/phylink.c | 21 ++++++++++++++++++- include/linux/phylink.h | 1 + 3 files changed, 28 insertions(+), 10 deletions(-)
Signature · 1 line
2.34.1
PATCH v2 1/2 net: phylink: add a function to resume phy alone to fix resume issue with WoL enabled
Issue we met:
On some platforms, mac cannot work after resumed from the suspend with WoL enabled.
The cause of the issue:
1. phylink_resolve() is in a workqueue which will not be executed immediately. This is the call sequence:
phylink_resolve()->phylink_link_up()->pl->mac_ops->mac_link_up()
For stmmac driver, mac_link_up() will set the correct speed/duplex... values which are from link_state.
2. In stmmac_resume(), it will call stmmac_hw_setup() after called the phylink_resume(), because mac need phy rx_clk to do the reset. stmmac_core_init() is called in function stmmac_hw_setup(), which will reset the mac and set the speed/duplex... to default value.
Conclusion: Because phylink_resolve() cannot determine when it is called, it
cannot be guaranteed to be called after stmmac_core_init(). Once stmmac_core_init() is called after phylink_resolve(), the mac will be misconfigured and cannot be used.
In order to avoid this problem, add a function called phylink_phy_resume() to resume phy separately. This eliminates the need to call phylink_resume() before stmmac_hw_setup().
Add another judgement before called phy_start() in phylink_start(). This way phy_start() will not be called multiple times when resumes. At the same time, it may not affect other drivers that do not use phylink_phy_resume().
--- drivers/net/phy/phylink.c | 21 ++++++++++++++++++++- include/linux/phylink.h | 1 + 2 files changed, 21 insertions(+), 1 deletion(-)
drivers/net/phy/phylink.c +20 -1
@@ -1939,7 +1939,7 @@ void phylink_start(struct phylink *pl) }if (poll)mod_timer (&pl->link_poll, jiffies +HZ ); -if (pl->phydev) +if (pl->phydev && pl->phydev->state <PHY_UP )phy_start (pl->phydev);if (pl->sfp_bus)sfp_upstream_start (pl->sfp_bus);
@@ -2020,6 +2020,25 @@ void phylink_suspend(struct phylink *pl, bool mac_wol) }EXPORT_SYMBOL_GPL (phylink_suspend); +/** +* phylink_phy_resume() - resume phy alone +* @pl: a pointer to a &struct phylink returned from phylink_create() +* +* In the MAC driver using phylink, if the MAC needs the clock of the phy +* when it resumes, can call this function to resume the phy separately. +* Then proceed to MAC resume operations. +*/ +void phylink_phy_resume (struct phylink *pl) +{ +ASSERT_RTNL (); + +if (!test_bit (PHYLINK_DISABLE_MAC_WOL , &pl->phylink_disable_state) + && pl->phydev) +phy_start (pl->phydev); + +} +EXPORT_SYMBOL_GPL (phylink_phy_resume); +/** * phylink_resume() - handle a network device resume event * @pl: a pointer to a &struct phylink returned from phylink_create()
include/linux/phylink.h +1 -0
@@ -589,6 +589,7 @@ void phylink_stop(struct phylink *);void phylink_suspend (struct phylink *pl,bool mac_wol);void phylink_resume (struct phylink *pl); +void phylink_phy_resume (struct phylink *pl);void phylink_ethtool_get_wol (struct phylink *,struct ethtool_wolinfo *);int phylink_ethtool_set_wol (struct phylink *,struct ethtool_wolinfo *);
Signature · 1 line
2.34.1
Re: [PATCH 1/2] net: phylink: add a function to resume phy alone to fix resume issue with WoL enabled
Quoted 19 lines · Clark Wang
Issue we met:
On some platforms, mac cannot work after resumed from the suspend with WoL enabled.The cause of the issue:
1.phylink_resolve()is in a workqueue which will not be executed immediately. This is the call sequence:phylink_resolve()->phylink_link_up()->pl->mac_ops->mac_link_up()For stmmac driver,
mac_link_up()will set the correct speed/duplex... values which are fromlink_state.
2. Instmmac_resume(), it will callstmmac_hw_setup()after called thephylink_resume(), because mac need phyrx_clkto do the reset.stmmac_core_init()is called in functionstmmac_hw_setup(), which will reset the mac and set the speed/duplex... to default value.
Conclusion: Becausephylink_resolve()cannot determine when it is called, it
cannot be guaranteed to be called afterstmmac_core_init(). Oncestmmac_core_init()is called afterphylink_resolve(), the mac will be misconfigured and cannot be used.
Quoted 8 lines
In order to avoid this problem, add a function called
phylink_phy_resume()to resume phy separately. This eliminates the need to callphylink_resume()beforestmmac_hw_setup().Add another judgement before called
phy_start()inphylink_start(). This wayphy_start()will not be called multiple times when resumes. At the same time, it may not affect other drivers that do not usephylink_phy_resume().
It'd be nice to see Fixes tag.
Quoted 27 lines
Signed-off-by Clark Wang <[email protected]>--- drivers/net/phy/phylink.c | 21 ++++++++++++++++++++- include/linux/phylink.h | 1 + 2 files changed, 21 insertions(+), 1 deletion(-)drivers/net/phy/phylink.c +5
-1@@ -1939,7 +1939,7 @@ void phylink_start(struct phylink *pl) }if (poll)mod_timer (&pl->link_poll, jiffies +HZ ); -if (pl->phydev) +if (pl->phydev && pl->phydev->state <PHY_UP )phy_start (pl->phydev);if (pl->sfp_bus)sfp_upstream_start (pl->sfp_bus);@@ -2020,6 +2020,25 @@ void phylink_suspend(struct phylink *pl, bool mac_wol) }EXPORT_SYMBOL_GPL (phylink_suspend); +/** +* phylink_phy_resume() - resume phy alone +* @pl: a pointer to a &struct phylink returned from phylink_create() +*
drivers/net/phy/phylink.c void phylink_suspend(struct phylink *pl, bool mac_wol) · line 2036
Quoted 10 lines
drivers/net/phy/phylink.c +10
-0+* In the MAC driver using phylink, if the MAC needs the clock of the phy +* when it resumes, can call this function to resume the phy separately. +* Then proceed to MAC resume operations. +*/ +void phylink_phy_resume (struct phylink *pl) +{ +ASSERT_RTNL (); + +if (!test_bit (PHYLINK_DISABLE_MAC_WOL , &pl->phylink_disable_state) + && pl->phydev)
you can fit && at the end of the previous line.
drivers/net/phy/phylink.c void phylink_suspend(struct phylink *pl, bool mac_wol) · line 2038
Quoted 2 lines
drivers/net/phy/phylink.c +2
-0+phy_start (pl->phydev); +
this blank line is not necessary.
Quoted 21 lines
drivers/net/phy/phylink.c +3
-0+} +EXPORT_SYMBOL_GPL (phylink_phy_resume); +/** * phylink_resume() - handle a network device resume event * @pl: a pointer to a &struct phylink returned from phylink_create() include/linux/phylink.h +1
-0@@ -589,6 +589,7 @@ void phylink_stop(struct phylink *);void phylink_suspend (struct phylink *pl,bool mac_wol);void phylink_resume (struct phylink *pl); +void phylink_phy_resume (struct phylink *pl);void phylink_ethtool_get_wol (struct phylink *,struct ethtool_wolinfo *);int phylink_ethtool_set_wol (struct phylink *,struct ethtool_wolinfo *);Signature · 1 line
2.34.1
Russell King (Oracle)
maintainerRe: [PATCH 1/2] net: phylink: add a function to resume phy alone to fix resume issue with WoL enabled
In reply to Clark WangQuoted 38 lines · Clark Wang
Issue we met:
On some platforms, mac cannot work after resumed from the suspend with WoL enabled.The cause of the issue:
1.phylink_resolve()is in a workqueue which will not be executed immediately. This is the call sequence:phylink_resolve()->phylink_link_up()->pl->mac_ops->mac_link_up()For stmmac driver,
mac_link_up()will set the correct speed/duplex... values which are fromlink_state.
2. Instmmac_resume(), it will callstmmac_hw_setup()after called thephylink_resume(), because mac need phyrx_clkto do the reset.stmmac_core_init()is called in functionstmmac_hw_setup(), which will reset the mac and set the speed/duplex... to default value.
Conclusion: Becausephylink_resolve()cannot determine when it is called, it
cannot be guaranteed to be called afterstmmac_core_init(). Oncestmmac_core_init()is called afterphylink_resolve(), the mac will be misconfigured and cannot be used.In order to avoid this problem, add a function called
phylink_phy_resume()to resume phy separately. This eliminates the need to callphylink_resume()beforestmmac_hw_setup().Add another judgement before called
phy_start()inphylink_start(). This wayphy_start()will not be called multiple times when resumes. At the same time, it may not affect other drivers that do not usephylink_phy_resume().Signed-off-by Clark Wang <[email protected]>--- drivers/net/phy/phylink.c | 21 ++++++++++++++++++++- include/linux/phylink.h | 1 + 2 files changed, 21 insertions(+), 1 deletion(-)drivers/net/phy/phylink.c +0
-0
drivers/net/phy/phylink.c void phylink_start(struct phylink *pl) · line 1942
Quoted 5 lines
drivers/net/phy/phylink.c +1
-1@@ -1939,7 +1939,7 @@ void phylink_start(struct phylink *pl) }if (poll)mod_timer (&pl->link_poll, jiffies +HZ ); -if (pl->phydev) +if (pl->phydev && pl->phydev->state <PHY_UP )
I'm really not happy with this - not only does this subvert the checks in phy_start(), it's a layering violation, and it delves into internals of phylib in an unprotected way.
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
PATCH v2 2/2 net: stmmac: resume phy separately before calling stmmac_hw_setup()
In reply to Clark WangOn some platforms, mac cannot work after resumed from the suspend with WoL enabled.
We found the stmmac_hw_setup() when system resumes will called after the stmmac_mac_link_up(). So the correct values set in stmmac_mac_link_up() are overwritten by stmmac_core_init() in phylink_resume().
So call the new added function in phylink to resume phy fristly. Then can call the stmmac_hw_setup() before calling phy_resume().
--- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +7 -9
@@ -7532,16 +7532,9 @@ int stmmac_resume(struct device *dev) }rtnl_lock (); -if (device_may_wakeup (priv->device) && priv->plat->pmt) { -phylink_resume (priv->phylink); - }else { -phylink_resume (priv->phylink); -if (device_may_wakeup (priv->device)) -phylink_speed_up (priv->phylink); - } -rtnl_unlock (); -rtnl_lock (); +phylink_phy_resume (priv->phylink); +mutex_lock (&priv->lock);stmmac_reset_queues_param (priv);
@@ -7559,6 +7552,11 @@ int stmmac_resume(struct device *dev)stmmac_enable_all_dma_irq (priv);mutex_unlock (&priv->lock); + +phylink_resume (priv->phylink); +if (device_may_wakeup (priv->device) && !priv->plat->pmt) +phylink_speed_up (priv->phylink); +rtnl_unlock ();netif_device_attach (ndev);
Signature · 1 line
2.34.1
Re: [PATCH 2/2] net: stmmac: resume phy separately before calling stmmac_hw_setup()
Quoted 8 lines · Clark Wang
On some platforms, mac cannot work after resumed from the suspend with WoL enabled.
We found the
stmmac_hw_setup()when system resumes will called after thestmmac_mac_link_up(). So the correct values set instmmac_mac_link_up()are overwritten bystmmac_core_init()inphylink_resume().So call the new added function in phylink to resume phy fristly.
first.
Quoted 1 line
Then can call the
stmmac_hw_setup()before callingphy_resume().
It'd be nice to add Fixes tag with appropriate commit id. Other than that looks fine.
Quoted 44 lines
Signed-off-by Clark Wang <[email protected]>--- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-)drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +7
-9@@ -7532,16 +7532,9 @@ int stmmac_resume(struct device *dev) }rtnl_lock (); -if (device_may_wakeup (priv->device) && priv->plat->pmt) { -phylink_resume (priv->phylink); - }else { -phylink_resume (priv->phylink); -if (device_may_wakeup (priv->device)) -phylink_speed_up (priv->phylink); - } -rtnl_unlock (); -rtnl_lock (); +phylink_phy_resume (priv->phylink); +mutex_lock (&priv->lock);stmmac_reset_queues_param (priv);@@ -7559,6 +7552,11 @@ int stmmac_resume(struct device *dev)stmmac_enable_all_dma_irq (priv);mutex_unlock (&priv->lock); + +phylink_resume (priv->phylink); +if (device_may_wakeup (priv->device) && !priv->plat->pmt) +phylink_speed_up (priv->phylink); +rtnl_unlock ();netif_device_attach (ndev);Signature · 1 line
2.34.1
Quoted 8 lines · Clark Wang
Hi,
The issue description is in the commit message.This patchset is the second version following discussions with Russell. But the name of each patch has changed, so V2 is not marked in each patch.
Thanks.
Please include target tree name in the subject eg. [net 1/2]
Quoted 13 lines
Clark Wang (2):
net: phylink: add a function to resume phy alone to fix resume issue
with WoL enabled
net: stmmac: resume phy separately before callingstmmac_hw_setup().../net/ethernet/stmicro/stmmac/stmmac_main.c | 16 +++++++------- drivers/net/phy/phylink.c | 21 ++++++++++++++++++- include/linux/phylink.h | 1 + 3 files changed, 28 insertions(+), 10 deletions(-)Signature · 1 line
2.34.1