From 541cd3ee00a4fe975b22fac6a3bc846bacef37f7 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Wed, 30 Dec 2009 08:23:28 +0000 Subject: phylib: Fix deadlock on resume Sometimes kernel hangs on resume with the following trace: ucc_geth e0102000.ucc: resume INFO: task bash:1764 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. bash D 0fecf43c 0 1764 1763 0x00000000 Call Trace: [cf9a7c10] [c0012868] ret_from_except+0x0/0x14 (unreliable) --- Exception: cf9a7ce0 at __switch_to+0x4c/0x6c LR = 0xcf9a7cc0 [cf9a7cd0] [c0008c14] __switch_to+0x4c/0x6c (unreliable) [cf9a7ce0] [c028bcfc] schedule+0x158/0x260 [cf9a7d10] [c028c720] __mutex_lock_slowpath+0x80/0xd8 [cf9a7d40] [c01cf388] phy_stop+0x20/0x70 [cf9a7d50] [c01d514c] ugeth_resume+0x6c/0x13c [...] Here is why. On suspend: - PM core starts suspending devices, ucc_geth_suspend gets called; - ucc_geth calls phy_stop() on suspend. Note that phy_stop() is mostly asynchronous so it doesn't block ucc_geth's suspend routine, it just sets PHY_HALTED state and disables PHY's interrupts; - Suddenly the state machine gets scheduled, it grabs the phydev->lock mutex and tries to process the PHY_HALTED state, so it calls phydev->adjust_link(phydev->attached_dev). In ucc_geth case adjust_link() calls msleep(), which reschedules the code flow back to PM core, which now finishes suspend and so we end up sleeping with phydev->lock mutex held. On resume: - PM core starts resuming devices (notice that nobody rescheduled the state machine yet, so the mutex is still held), the core calls ucc_geth's resume routine; - ucc_geth_resume restarts the PHY with phy_stop()/phy_start() sequence, and the phy_*() calls are trying to grab the phydev->lock mutex. Here comes the deadlock. This patch fixes the issue by stopping the state machine on suspend and starting it again on resume. Signed-off-by: Anton Vorontsov Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) (limited to 'drivers/net/phy') diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index bd4e8d72dc08..49252d390903 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -303,8 +303,18 @@ static int mdio_bus_suspend(struct device * dev, pm_message_t state) struct phy_driver *phydrv = to_phy_driver(dev->driver); struct phy_device *phydev = to_phy_device(dev); + /* + * We must stop the state machine manually, otherwise it stops out of + * control, possibly with the phydev->lock held. Upon resume, netdev + * may call phy routines that try to grab the same lock, and that may + * lead to a deadlock. + */ + if (phydev->attached_dev) + phy_stop_machine(phydev); + if (!mdio_bus_phy_may_suspend(phydev)) return 0; + return phydrv->suspend(phydev); } @@ -312,10 +322,20 @@ static int mdio_bus_resume(struct device * dev) { struct phy_driver *phydrv = to_phy_driver(dev->driver); struct phy_device *phydev = to_phy_device(dev); + int ret; if (!mdio_bus_phy_may_suspend(phydev)) - return 0; - return phydrv->resume(phydev); + goto no_resume; + + ret = phydrv->resume(phydev); + if (ret < 0) + return ret; + +no_resume: + if (phydev->attached_dev) + phy_start_machine(phydev, NULL); + + return 0; } struct bus_type mdio_bus_type = { -- cgit v1.2.3 From 2f5cb43406d0b29b96248f5328a14a6f6abf8ae6 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Wed, 30 Dec 2009 08:23:30 +0000 Subject: phylib: Properly reinitialize PHYs after hibernation Since hibernation assumes power loss, we should fully reinitialize PHYs (including platform fixups), as if PHYs were just attached. This patch factors phy_init_hw() out of phy_attach_direct(), then converts mdio_bus to dev_pm_ops and adds an appropriate restore() callback. Signed-off-by: Anton Vorontsov Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 50 +++++++++++++++++++++++++++++++++++++------- drivers/net/phy/phy_device.c | 30 +++++++++++++------------- include/linux/phy.h | 1 + 3 files changed, 59 insertions(+), 22 deletions(-) (limited to 'drivers/net/phy') diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 49252d390903..e17b70291bbc 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -264,6 +264,8 @@ static int mdio_bus_match(struct device *dev, struct device_driver *drv) (phydev->phy_id & phydrv->phy_id_mask)); } +#ifdef CONFIG_PM + static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) { struct device_driver *drv = phydev->dev.driver; @@ -295,10 +297,7 @@ static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) return true; } -/* Suspend and resume. Copied from platform_suspend and - * platform_resume - */ -static int mdio_bus_suspend(struct device * dev, pm_message_t state) +static int mdio_bus_suspend(struct device *dev) { struct phy_driver *phydrv = to_phy_driver(dev->driver); struct phy_device *phydev = to_phy_device(dev); @@ -318,7 +317,7 @@ static int mdio_bus_suspend(struct device * dev, pm_message_t state) return phydrv->suspend(phydev); } -static int mdio_bus_resume(struct device * dev) +static int mdio_bus_resume(struct device *dev) { struct phy_driver *phydrv = to_phy_driver(dev->driver); struct phy_device *phydev = to_phy_device(dev); @@ -338,11 +337,48 @@ no_resume: return 0; } +static int mdio_bus_restore(struct device *dev) +{ + struct phy_device *phydev = to_phy_device(dev); + struct net_device *netdev = phydev->attached_dev; + int ret; + + if (!netdev) + return 0; + + ret = phy_init_hw(phydev); + if (ret < 0) + return ret; + + /* The PHY needs to renegotiate. */ + phydev->link = 0; + phydev->state = PHY_UP; + + phy_start_machine(phydev, NULL); + + return 0; +} + +static struct dev_pm_ops mdio_bus_pm_ops = { + .suspend = mdio_bus_suspend, + .resume = mdio_bus_resume, + .freeze = mdio_bus_suspend, + .thaw = mdio_bus_resume, + .restore = mdio_bus_restore, +}; + +#define MDIO_BUS_PM_OPS (&mdio_bus_pm_ops) + +#else + +#define MDIO_BUS_PM_OPS NULL + +#endif /* CONFIG_PM */ + struct bus_type mdio_bus_type = { .name = "mdio_bus", .match = mdio_bus_match, - .suspend = mdio_bus_suspend, - .resume = mdio_bus_resume, + .pm = MDIO_BUS_PM_OPS, }; EXPORT_SYMBOL(mdio_bus_type); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index b10fedd82143..8212b2b93422 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -378,6 +378,20 @@ void phy_disconnect(struct phy_device *phydev) } EXPORT_SYMBOL(phy_disconnect); +int phy_init_hw(struct phy_device *phydev) +{ + int ret; + + if (!phydev->drv || !phydev->drv->config_init) + return 0; + + ret = phy_scan_fixups(phydev); + if (ret < 0) + return ret; + + return phydev->drv->config_init(phydev); +} + /** * phy_attach_direct - attach a network device to a given PHY device pointer * @dev: network device to attach @@ -425,21 +439,7 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, /* Do initial configuration here, now that * we have certain key parameters * (dev_flags and interface) */ - if (phydev->drv->config_init) { - int err; - - err = phy_scan_fixups(phydev); - - if (err < 0) - return err; - - err = phydev->drv->config_init(phydev); - - if (err < 0) - return err; - } - - return 0; + return phy_init_hw(phydev); } EXPORT_SYMBOL(phy_attach_direct); diff --git a/include/linux/phy.h b/include/linux/phy.h index b1368b8f6572..7968defd2fa7 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -447,6 +447,7 @@ struct phy_device* get_phy_device(struct mii_bus *bus, int addr); int phy_device_register(struct phy_device *phy); int phy_clear_interrupt(struct phy_device *phydev); int phy_config_interrupt(struct phy_device *phydev, u32 interrupts); +int phy_init_hw(struct phy_device *phydev); int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface); struct phy_device * phy_attach(struct net_device *dev, -- cgit v1.2.3 From 7ec4e7d3cfee9d7846dbd02ad442c40cb58512e8 Mon Sep 17 00:00:00 2001 From: roel kluin Date: Wed, 30 Dec 2009 06:43:06 +0000 Subject: broadcom: Fix &&/|| confusion in bcm54xx_adjust_rxrefclk() This always evaluates to true. Signed-off-by: Roel Kluin Signed-off-by: David S. Miller --- drivers/net/phy/broadcom.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/phy') diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index c13cf64095b6..33c4b12a63ba 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -331,8 +331,8 @@ static void bcm54xx_adjust_rxrefclk(struct phy_device *phydev) bool clk125en = true; /* Abort if we are using an untested phy. */ - if (BRCM_PHY_MODEL(phydev) != PHY_ID_BCM57780 || - BRCM_PHY_MODEL(phydev) != PHY_ID_BCM50610 || + if (BRCM_PHY_MODEL(phydev) != PHY_ID_BCM57780 && + BRCM_PHY_MODEL(phydev) != PHY_ID_BCM50610 && BRCM_PHY_MODEL(phydev) != PHY_ID_BCM50610M) return; -- cgit v1.2.3