eeh.c

linux-2.6.15.6

/**
 * __restore_bars - Restore the Base Address Registers
 * Loads the PCI configuration space base address registers,
 * the expansion ROM base address, the latency timer, and etc.
 * from the saved values in the device node.
 */
static inline void __restore_bars (struct pci_dn *pdn)
{
	int i;

	if (NULL==pdn->phb) return;
	for (i=4; i<10; i++) {
		rtas_write_config(pdn, i*4, 4, pdn->config_space[i]);
	}

	/* 12 == Expansion ROM Address */
	rtas_write_config(pdn, 12*4, 4, pdn->config_space[12]);

#define BYTE_SWAP(OFF) (8*((OFF)/4)+3-(OFF))
#define SAVED_BYTE(OFF) (((u8 *)(pdn->config_space))[BYTE_SWAP(OFF)])

	rtas_write_config (pdn, PCI_CACHE_LINE_SIZE, 1,
	            SAVED_BYTE(PCI_CACHE_LINE_SIZE));

	rtas_write_config (pdn, PCI_LATENCY_TIMER, 1,
	            SAVED_BYTE(PCI_LATENCY_TIMER));

	/* max latency, min grant, interrupt pin and line */
	rtas_write_config(pdn, 15*4, 4, pdn->config_space[15]);
}

/**
 * eeh_restore_bars - restore the PCI config space info
 *
 * This routine performs a recursive walk to the children
 * of this device as well.
 */
void eeh_restore_bars(struct pci_dn *pdn)
{
	struct device_node *dn;
	if (!pdn) 
		return;
	
	if (! pdn->eeh_is_bridge)
		__restore_bars (pdn);

	dn = pdn->node->child;
	while (dn) {
		eeh_restore_bars (PCI_DN(dn));
		dn = dn->sibling;
	}
}

/**
 * eeh_save_bars - save device bars
 *
 * Save the values of the device bars. Unlike the restore
 * routine, this routine is *not* recursive. This is because
 * PCI devices are added individuallly; but, for the restore,
 * an entire slot is reset at a time.
 */
static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn)
{
	int i;

	if (!pdev || !pdn )
		return;
	
	for (i = 0; i < 16; i++)
		pci_read_config_dword(pdev, i * 4, &pdn->config_space[i]);

	if (pdev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
		pdn->eeh_is_bridge = 1;
}

void
rtas_configure_bridge(struct pci_dn *pdn)
{
	int token = rtas_token ("ibm,configure-bridge");
	int rc;

	if (token == RTAS_UNKNOWN_SERVICE)
		return;
	rc = rtas_call(token,3,1, NULL,
	               pdn->eeh_config_addr,
	               BUID_HI(pdn->phb->buid),
	               BUID_LO(pdn->phb->buid));
	if (rc) {
		printk (KERN_WARNING "EEH: Unable to configure device bridge (%d) for %s\n",
		        rc, pdn->node->full_name);
	}
}

/* ------------------------------------------------------------- */
/* 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;
};

/* Enable eeh for the given device node. */
static void *early_enable_eeh(struct device_node *dn, void *data)
{
	struct eeh_early_enable_info *info = data;
	int ret;
	char *status = get_property(dn, "status", NULL);
	u32 *class_code = (u32 *)get_property(dn, "class-code", NULL);
	u32 *vendor_id = (u32 *)get_property(dn, "vendor-id", NULL);
	u32 *device_id = (u32 *)get_property(dn, "device-id", NULL);
	u32 *regs;
	int enable;
	struct pci_dn *pdn = PCI_DN(dn);

	pdn->eeh_mode = 0;
	pdn->eeh_check_count = 0;
	pdn->eeh_freeze_count = 0;

	if (status && strcmp(status, "ok") != 0)
		return NULL;	/* ignore devices with bad status */

	/* Ignore bad nodes. */
	if (!class_code || !vendor_id || !device_id)
		return NULL;

	/* There is nothing to check on PCI to ISA bridges */
	if (dn->type && !strcmp(dn->type, "isa")) {
		pdn->eeh_mode |= EEH_MODE_NOCHECK;
		return NULL;
	}

	/*
	 * Now decide if we are going to "Disable" EEH checking
	 * for this device.  We still run with the EEH hardware active,
	 * but we won't be checking for ff's.  This means a driver
	 * could return bad data (very bad!), an interrupt handler could
	 * hang waiting on status bits that won't change, etc.
	 * But there are a few cases like display devices that make sense.
	 */
	enable = 1;	/* i.e. we will do checking */
	if ((*class_code >> 16) == PCI_BASE_CLASS_DISPLAY)
		enable = 0;

	if (!enable)
		pdn->eeh_mode |= EEH_MODE_NOCHECK;

	/* Ok... see if this device supports EEH.  Some do, some don't,
	 * and the only way to find out is to check each and every one. */
	regs = (u32 *)get_property(dn, "reg", NULL);
	if (regs) {
		/* 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);

		if (ret == 0) {
			eeh_subsystem_enabled = 1;
			pdn->eeh_mode |= EEH_MODE_SUPPORTED;
			pdn->eeh_config_addr = regs[0];
#ifdef DEBUG
			printk(KERN_DEBUG "EEH: %s: eeh enabled\n", dn->full_name);
#endif
		} else {

			/* This device doesn't support EEH, but it may have an
			 * EEH parent, in which case we mark it as supported. */
			if (dn->parent && PCI_DN(dn->parent)
			    && (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) {
				/* Parent supports EEH. */
				pdn->eeh_mode |= EEH_MODE_SUPPORTED;
				pdn->eeh_config_addr = PCI_DN(dn->parent)->eeh_config_addr;
				return NULL;
			}
		}
	} else {
		printk(KERN_WARNING "EEH: %s: unable to get reg property.\n",
		       dn->full_name);
	}

	return NULL;
}

/*
 * Initialize EEH by trying to enable it for all of the adapters in the system.
 * As a side effect we can determine here if eeh is supported at all.
 * Note that we leave EEH on so failed config cycles won't cause a machine
 * check.  If a user turns off EEH for a particular adapter they are really
 * telling Linux to ignore errors.  Some hardware (e.g. POWER5) won't
 * grant access to a slot if EEH isn't enabled, and so we always enable
 * EEH for all slots/all devices.
 *
 * The eeh-force-off option disables EEH checking globally, for all slots.
 * Even if force-off is set, the EEH hardware is still enabled, so that
 * newer systems can boot.
 */
void __init eeh_init(void)
{
	struct device_node *phb, *np;
	struct eeh_early_enable_info info;

	spin_lock_init(&confirm_error_lock);
	spin_lock_init(&slot_errbuf_lock);

	np = of_find_node_by_path("/rtas");
	if (np == NULL)
		return;

	ibm_set_eeh_option = rtas_token("ibm,set-eeh-option");
	ibm_set_slot_reset = rtas_token("ibm,set-slot-reset");
	ibm_read_slot_reset_state2 = rtas_token("ibm,read-slot-reset-state2");
	ibm_read_slot_reset_state = rtas_token("ibm,read-slot-reset-state");
	ibm_slot_error_detail = rtas_token("ibm,slot-error-detail");

	if (ibm_set_eeh_option == RTAS_UNKNOWN_SERVICE)
		return;

	eeh_error_buf_size = rtas_token("rtas-error-log-max");
	if (eeh_error_buf_size == RTAS_UNKNOWN_SERVICE) {
		eeh_error_buf_size = 1024;
	}
	if (eeh_error_buf_size > RTAS_ERROR_LOG_MAX) {
		printk(KERN_WARNING "EEH: rtas-error-log-max is bigger than allocated "
		      "buffer ! (%d vs %d)", eeh_error_buf_size, RTAS_ERROR_LOG_MAX);
		eeh_error_buf_size = RTAS_ERROR_LOG_MAX;
	}

	/* Enable EEH for all adapters.  Note that eeh requires buid's */
	for (phb = of_find_node_by_name(NULL, "pci"); phb;
	     phb = of_find_node_by_name(phb, "pci")) {
		unsigned long buid;

		buid = get_phb_buid(phb);
		if (buid == 0 || PCI_DN(phb) == NULL)
			continue;

		info.buid_lo = BUID_LO(buid);
		info.buid_hi = BUID_HI(buid);
		traverse_pci_devices(phb, early_enable_eeh, &info);
	}

	if (eeh_subsystem_enabled)
		printk(KERN_INFO "EEH: PCI Enhanced I/O Error Handling Enabled\n");
	else
		printk(KERN_WARNING "EEH: No capable adapters found\n");
}

/**
 * eeh_add_device_early - enable EEH for the indicated device_node
 * @dn: device node for which to set up EEH
 *
 * This routine must be used to perform EEH initialization for PCI
 * devices that were added after system boot (e.g. hotplug, dlpar).
 * This routine must be called before any i/o is performed to the
 * adapter (inluding any config-space i/o).
 * Whether this actually enables EEH or not for this device depends
 * on the CEC architecture, type of the device, on earlier boot
 * command-line arguments & etc.
 */
void eeh_add_device_early(struct device_node *dn)
{
	struct pci_controller *phb;
	struct eeh_early_enable_info info;

	if (!dn || !PCI_DN(dn))
		return;
	phb = PCI_DN(dn)->phb;
	if (NULL == phb || 0 == phb->buid) {
		printk(KERN_WARNING "EEH: Expected buid but found none for %s\n",
		       dn->full_name);
		dump_stack();
		return;
	}

	info.buid_hi = BUID_HI(phb->buid);
	info.buid_lo = BUID_LO(phb->buid);
	early_enable_eeh(dn, &info);
}
EXPORT_SYMBOL_GPL(eeh_add_device_early);

/**
 * eeh_add_device_late - perform EEH initialization for the indicated pci device
 * @dev: pci device for which to set up EEH
 *
 * This routine must be used to complete EEH initialization for PCI
 * devices that were added after system boot (e.g. hotplug, dlpar).
 */
void eeh_add_device_late(struct pci_dev *dev)
{
	struct device_node *dn;
	struct pci_dn *pdn;

	if (!dev || !eeh_subsystem_enabled)
		return;

#ifdef DEBUG
	printk(KERN_DEBUG "EEH: adding device %s\n", pci_name(dev));
#endif

	pci_dev_get (dev);
	dn = pci_device_to_OF_node(dev);
	pdn = PCI_DN(dn);
	pdn->pcidev = dev;

	pci_addr_cache_insert_device (dev);
	eeh_save_bars(dev, pdn);
}
EXPORT_SYMBOL_GPL(eeh_add_device_late);

/**
 * 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).
 */
void eeh_remove_device(struct pci_dev *dev)
{
	struct device_node *dn;
	if (!dev || !eeh_subsystem_enabled)
		return;

	/* Unregister the device with the EEH/PCI address search system */
#ifdef DEBUG
	printk(KERN_DEBUG "EEH: remove device %s\n", pci_name(dev));
#endif
	pci_addr_cache_remove_device(dev);

	dn = pci_device_to_OF_node(dev);
	PCI_DN(dn)->pcidev = NULL;
	pci_dev_put (dev);
}
EXPORT_SYMBOL_GPL(eeh_remove_device);

static int proc_eeh_show(struct seq_file *m, void *v)
{
	unsigned int cpu;
	unsigned long ffs = 0, positives = 0, failures = 0;
	unsigned long resets = 0;
	unsigned long no_dev = 0, no_dn = 0, no_cfg = 0, no_check = 0;

	for_each_cpu(cpu) {
		ffs += per_cpu(total_mmio_ffs, cpu);
		positives += per_cpu(false_positives, cpu);
		failures += per_cpu(ignored_failures, cpu);
		resets += per_cpu(slot_resets, cpu);
		no_dev += per_cpu(no_device, cpu);
		no_dn += per_cpu(no_dn, cpu);
		no_cfg += per_cpu(no_cfg_addr, cpu);
		no_check += per_cpu(ignored_check, cpu);
	}

	if (0 == eeh_subsystem_enabled) {
		seq_printf(m, "EEH Subsystem is globally disabled\n");
		seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs);
	} else {
		seq_printf(m, "EEH Subsystem is enabled\n");
		seq_printf(m,
				"no device=%ld\n"
				"no device node=%ld\n"
				"no config address=%ld\n"
				"check not wanted=%ld\n"
				"eeh_total_mmio_ffs=%ld\n"
				"eeh_false_positives=%ld\n"
				"eeh_ignored_failures=%ld\n"
				"eeh_slot_resets=%ld\n",
				no_dev, no_dn, no_cfg, no_check,
				ffs, positives, failures, resets);
	}

	return 0;
}

static int proc_eeh_open(struct inode *inode, struct file *file)
{
	return single_open(file, proc_eeh_show, NULL);
}

static struct file_operations proc_eeh_operations = {
	.open      = proc_eeh_open,
	.read      = seq_read,
	.llseek    = seq_lseek,
	.release   = single_release,
};

static int __init eeh_init_proc(void)
{
	struct proc_dir_entry *e;

	if (platform_is_pseries()) {
		e = create_proc_entry("ppc64/eeh", 0, NULL);
		if (e)
			e->proc_fops = &proc_eeh_operations;
	}

	return 0;
}
__initcall(eeh_init_proc);