diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index dbdffb2fe42999500929c799d9650512bad6b80b..27515476ad6c2cd2c0d0469188a61b02c356d837 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -3,4 +3,4 @@ obj-y := pci.o lpar.o hvCall.o nvram.o reconfig.o \ obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_XICS) += xics.o -obj-$(CONFIG_EEH) += eeh.o +obj-$(CONFIG_EEH) += eeh.o eeh_event.o diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index 9df1d5018363e9d32494cd8b02407638d2ab576c..1fec99d53311bb6f99875e1875172428547dabda 100644 --- a/arch/powerpc/platforms/pseries/eeh.c +++ b/arch/powerpc/platforms/pseries/eeh.c @@ -19,7 +19,6 @@ #include #include -#include #include #include #include @@ -27,12 +26,12 @@ #include #include #include +#include #include #include +#include #include -#include #include -#include #undef DEBUG @@ -70,14 +69,6 @@ * and sent out for processing. */ -/* EEH event workqueue setup. */ -static DEFINE_SPINLOCK(eeh_eventlist_lock); -LIST_HEAD(eeh_eventlist); -static void eeh_event_handler(void *); -DECLARE_WORK(eeh_event_wq, eeh_event_handler, NULL); - -static struct notifier_block *eeh_notifier_chain; - /* If a device driver keeps reading an MMIO register in an interrupt * handler after a slot isolation event has occurred, we assume it * is broken and panic. This sets the threshold for how many read @@ -420,24 +411,6 @@ void eeh_slot_error_detail (struct pci_dn *pdn, int severity) spin_unlock_irqrestore(&slot_errbuf_lock, flags); } -/** - * eeh_register_notifier - Register to find out about EEH events. - * @nb: notifier block to callback on events - */ -int eeh_register_notifier(struct notifier_block *nb) -{ - return notifier_chain_register(&eeh_notifier_chain, nb); -} - -/** - * eeh_unregister_notifier - Unregister to an EEH event notifier. - * @nb: notifier block to callback on events - */ -int eeh_unregister_notifier(struct notifier_block *nb) -{ - return notifier_chain_unregister(&eeh_notifier_chain, nb); -} - /** * read_slot_reset_state - Read the reset state of a device node's slot * @dn: device node to read @@ -460,73 +433,6 @@ static int read_slot_reset_state(struct pci_dn *pdn, int rets[]) BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid)); } -/** - * eeh_panic - call panic() for an eeh event that cannot be handled. - * The philosophy of this routine is that it is better to panic and - * halt the OS than it is to risk possible data corruption by - * oblivious device drivers that don't know better. - * - * @dev pci device that had an eeh event - * @reset_state current reset state of the device slot - */ -static void eeh_panic(struct pci_dev *dev, int reset_state) -{ - /* - * XXX We should create a separate sysctl for this. - * - * Since the panic_on_oops sysctl is used to halt the system - * in light of potential corruption, we can use it here. - */ - if (panic_on_oops) { - struct device_node *dn = pci_device_to_OF_node(dev); - eeh_slot_error_detail (PCI_DN(dn), 2 /* Permanent Error */); - panic("EEH: MMIO failure (%d) on device:%s\n", reset_state, - pci_name(dev)); - } - else { - __get_cpu_var(ignored_failures)++; - printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n", - reset_state, pci_name(dev)); - } -} - -/** - * eeh_event_handler - dispatch EEH events. The detection of a frozen - * slot can occur inside an interrupt, where it can be hard to do - * anything about it. The goal of this routine is to pull these - * detection events out of the context of the interrupt handler, and - * re-dispatch them for processing at a later time in a normal context. - * - * @dummy - unused - */ -static void eeh_event_handler(void *dummy) -{ - unsigned long flags; - struct eeh_event *event; - - while (1) { - spin_lock_irqsave(&eeh_eventlist_lock, flags); - event = NULL; - if (!list_empty(&eeh_eventlist)) { - event = list_entry(eeh_eventlist.next, struct eeh_event, list); - list_del(&event->list); - } - spin_unlock_irqrestore(&eeh_eventlist_lock, flags); - if (event == NULL) - break; - - printk(KERN_INFO "EEH: MMIO failure (%d), notifiying device " - "%s\n", event->reset_state, - pci_name(event->dev)); - - notifier_call_chain (&eeh_notifier_chain, - EEH_NOTIFY_FREEZE, event); - - pci_dev_put(event->dev); - kfree(event); - } -} - /** * eeh_token_to_phys - convert EEH address token to phys address * @token i/o token, should be address in the form 0xA.... @@ -613,8 +519,6 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) int ret; int rets[3]; unsigned long flags; - int reset_state; - struct eeh_event *event; struct pci_dn *pdn; struct device_node *pe_dn; int rc = 0; @@ -722,33 +626,12 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) __eeh_mark_slot (pe_dn); spin_unlock_irqrestore(&confirm_error_lock, flags); - reset_state = rets[0]; - - eeh_slot_error_detail (pdn, 1 /* Temporary Error */); - - printk(KERN_INFO "EEH: MMIO failure (%d) on device: %s %s\n", - rets[0], dn->name, dn->full_name); - event = kmalloc(sizeof(*event), GFP_ATOMIC); - if (event == NULL) { - eeh_panic(dev, reset_state); - return 1; - } - - event->dev = dev; - event->dn = dn; - event->reset_state = reset_state; - - /* We may or may not be called in an interrupt context */ - spin_lock_irqsave(&eeh_eventlist_lock, flags); - list_add(&event->list, &eeh_eventlist); - spin_unlock_irqrestore(&eeh_eventlist_lock, flags); - + eeh_send_failure_event (dn, dev, rets[0], rets[2]); + /* Most EEH events are due to device driver bugs. Having * a stack trace will help the device-driver authors figure * out what happened. So print that out. */ if (rets[0] != 5) dump_stack(); - schedule_work(&eeh_event_wq); - return 1; dn_unlock: @@ -793,6 +676,14 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon EXPORT_SYMBOL(eeh_check_failure); +/* ------------------------------------------------------------- */ +/* The code below deals with enabling EEH for devices during the + * early boot sequence. EEH must be enabled before any PCI probing + * can be done. + */ + +#define EEH_ENABLE 1 + struct eeh_early_enable_info { unsigned int buid_hi; unsigned int buid_lo; @@ -850,8 +741,9 @@ static void *early_enable_eeh(struct device_node *dn, void *data) /* First register entry is addr (00BBSS00) */ /* Try to enable eeh */ ret = rtas_call(ibm_set_eeh_option, 4, 1, NULL, - regs[0], info->buid_hi, info->buid_lo, - EEH_ENABLE); + regs[0], info->buid_hi, info->buid_lo, + EEH_ENABLE); + if (ret == 0) { eeh_subsystem_enabled = 1; pdn->eeh_mode |= EEH_MODE_SUPPORTED; diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c new file mode 100644 index 0000000000000000000000000000000000000000..92497333c2b65af53a4f007e2b1a09c5954f0f4a --- /dev/null +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -0,0 +1,155 @@ +/* + * eeh_event.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2005 Linas Vepstas + */ + +#include +#include +#include + +/** Overview: + * EEH error states may be detected within exception handlers; + * however, the recovery processing needs to occur asynchronously + * in a normal kernel context and not an interrupt context. + * This pair of routines creates an event and queues it onto a + * work-queue, where a worker thread can drive recovery. + */ + +/* EEH event workqueue setup. */ +static spinlock_t eeh_eventlist_lock = SPIN_LOCK_UNLOCKED; +LIST_HEAD(eeh_eventlist); +static void eeh_thread_launcher(void *); +DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); + +/** + * eeh_panic - call panic() for an eeh event that cannot be handled. + * The philosophy of this routine is that it is better to panic and + * halt the OS than it is to risk possible data corruption by + * oblivious device drivers that don't know better. + * + * @dev pci device that had an eeh event + * @reset_state current reset state of the device slot + */ +static void eeh_panic(struct pci_dev *dev, int reset_state) +{ + /* + * Since the panic_on_oops sysctl is used to halt the system + * in light of potential corruption, we can use it here. + */ + if (panic_on_oops) { + panic("EEH: MMIO failure (%d) on device:%s\n", reset_state, + pci_name(dev)); + } + else { + printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n", + reset_state, pci_name(dev)); + } +} + +/** + * eeh_event_handler - dispatch EEH events. The detection of a frozen + * slot can occur inside an interrupt, where it can be hard to do + * anything about it. The goal of this routine is to pull these + * detection events out of the context of the interrupt handler, and + * re-dispatch them for processing at a later time in a normal context. + * + * @dummy - unused + */ +static int eeh_event_handler(void * dummy) +{ + unsigned long flags; + struct eeh_event *event; + + daemonize ("eehd"); + + while (1) { + set_current_state(TASK_INTERRUPTIBLE); + + spin_lock_irqsave(&eeh_eventlist_lock, flags); + event = NULL; + if (!list_empty(&eeh_eventlist)) { + event = list_entry(eeh_eventlist.next, struct eeh_event, list); + list_del(&event->list); + } + spin_unlock_irqrestore(&eeh_eventlist_lock, flags); + if (event == NULL) + break; + + printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n", + pci_name(event->dev)); + + eeh_panic (event->dev, event->state); + + kfree(event); + } + + return 0; +} + +/** + * eeh_thread_launcher + * + * @dummy - unused + */ +static void eeh_thread_launcher(void *dummy) +{ + if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0) + printk(KERN_ERR "Failed to start EEH daemon\n"); +} + +/** + * eeh_send_failure_event - generate a PCI error event + * @dev pci device + * + * This routine can be called within an interrupt context; + * the actual event will be delivered in a normal context + * (from a workqueue). + */ +int eeh_send_failure_event (struct device_node *dn, + struct pci_dev *dev, + int state, + int time_unavail) +{ + unsigned long flags; + struct eeh_event *event; + + event = kmalloc(sizeof(*event), GFP_ATOMIC); + if (event == NULL) { + printk (KERN_ERR "EEH: out of memory, event not handled\n"); + return 1; + } + + if (dev) + pci_dev_get(dev); + + event->dn = dn; + event->dev = dev; + event->state = state; + event->time_unavail = time_unavail; + + /* We may or may not be called in an interrupt context */ + spin_lock_irqsave(&eeh_eventlist_lock, flags); + list_add(&event->list, &eeh_eventlist); + spin_unlock_irqrestore(&eeh_eventlist_lock, flags); + + schedule_work(&eeh_event_wq); + + return 0; +} + +/********************** END OF FILE ******************************/ diff --git a/include/asm-powerpc/eeh_event.h b/include/asm-powerpc/eeh_event.h new file mode 100644 index 0000000000000000000000000000000000000000..d168a30b3866ea11283ac10155befa7c2234c01b --- /dev/null +++ b/include/asm-powerpc/eeh_event.h @@ -0,0 +1,52 @@ +/* + * eeh_event.h + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2005 Linas Vepstas + */ + +#ifndef ASM_PPC64_EEH_EVENT_H +#define ASM_PPC64_EEH_EVENT_H + +/** EEH event -- structure holding pci controller data that describes + * a change in the isolation status of a PCI slot. A pointer + * to this struct is passed as the data pointer in a notify callback. + */ +struct eeh_event { + struct list_head list; + struct device_node *dn; /* struct device node */ + struct pci_dev *dev; /* affected device */ + int state; + int time_unavail; /* milliseconds until device might be available */ +}; + +/** + * eeh_send_failure_event - generate a PCI error event + * @dev pci device + * + * This routine builds a PCI error event which will be delivered + * to all listeners on the peh_notifier_chain. + * + * This routine can be called within an interrupt context; + * the actual event will be delivered in a normal context + * (from a workqueue). + */ +int eeh_send_failure_event (struct device_node *dn, + struct pci_dev *dev, + int reset_state, + int time_unavail); + +#endif /* ASM_PPC64_EEH_EVENT_H */ diff --git a/include/asm-ppc64/eeh.h b/include/asm-ppc64/eeh.h index 40c8eb57493eb124170ad721cbdf378813f23a05..89f26ab31908df6e72932291568340243ba7b4b3 100644 --- a/include/asm-ppc64/eeh.h +++ b/include/asm-ppc64/eeh.h @@ -1,4 +1,4 @@ -/* +/* * eeh.h * Copyright (C) 2001 Dave Engebretsen & Todd Inglett IBM Corporation. * @@ -6,12 +6,12 @@ * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA @@ -27,8 +27,6 @@ struct pci_dev; struct device_node; -struct device_node; -struct notifier_block; #ifdef CONFIG_EEH @@ -37,6 +35,10 @@ struct notifier_block; #define EEH_MODE_NOCHECK (1<<1) #define EEH_MODE_ISOLATED (1<<2) +/* Max number of EEH freezes allowed before we consider the device + * to be permanently disabled. */ +#define EEH_MAX_ALLOWED_FREEZES 5 + void __init eeh_init(void); unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned long val); @@ -59,36 +61,14 @@ void eeh_add_device_late(struct pci_dev *); * eeh_remove_device - undo EEH setup for the indicated pci device * @dev: pci device to be removed * - * This routine should be when a device is removed from a running - * system (e.g. by hotplug or dlpar). + * This routine should be called when a device is removed from + * a running system (e.g. by hotplug or dlpar). It unregisters + * the PCI device from the EEH subsystem. I/O errors affecting + * this device will no longer be detected after this call; thus, + * i/o errors affecting this slot may leave this device unusable. */ void eeh_remove_device(struct pci_dev *); -#define EEH_DISABLE 0 -#define EEH_ENABLE 1 -#define EEH_RELEASE_LOADSTORE 2 -#define EEH_RELEASE_DMA 3 - -/** - * Notifier event flags. - */ -#define EEH_NOTIFY_FREEZE 1 - -/** EEH event -- structure holding pci slot data that describes - * a change in the isolation status of a PCI slot. A pointer - * to this struct is passed as the data pointer in a notify callback. - */ -struct eeh_event { - struct list_head list; - struct pci_dev *dev; - struct device_node *dn; - int reset_state; -}; - -/** Register to find out about EEH events. */ -int eeh_register_notifier(struct notifier_block *nb); -int eeh_unregister_notifier(struct notifier_block *nb); - /** * EEH_POSSIBLE_ERROR() -- test for possible MMIO failure. * @@ -129,7 +109,7 @@ static inline void eeh_remove_device(struct pci_dev *dev) { } #define EEH_IO_ERROR_VALUE(size) (-1UL) #endif /* CONFIG_EEH */ -/* +/* * MMIO read/write operations with EEH support. */ static inline u8 eeh_readb(const volatile void __iomem *addr)