/*
* edac_init
* module initialization entry point
*/
static int __init edac_init(void)
{
int err = 0;
edac_printk(KERN_INFO, EDAC_MC, EDAC_VERSION "\n");
/*
* Harvest and clear any boot/initialization PCI parity errors
*
* FIXME: This only clears errors logged by devices present at time of
* module initialization. We should also do an initial clear
* of each newly hotplugged device.
*/
edac_pci_clear_parity_errors();
err = edac_mc_sysfs_init();
if (err)
goto error;
edac_debugfs_init();
/* Setup/Initialize the workq for this core */
err = edac_workqueue_setup();
if (err) {
edac_printk(KERN_ERR, EDAC_MC, "init WorkQueue failure\n");
goto error;
}
return 0;
error:
return err;
}
/**
* zynqmp_ocm_edac_inject_uebitposition1_store - Set UE second bit postion
* @dci: Pointer to the edac device struct
* @data: Pointer to user data
* @count: read the size bytes from buffer
*
* Set the second bit postion for UE Error generation,we need to configure
* any two bitpositions to inject UE Error
* Return: Number of bytes copied.
*/
static ssize_t zynqmp_ocm_edac_inject_uebitpos1_store(
struct edac_device_ctl_info *dci, const char *data,
size_t count)
{
struct zynqmp_ocm_edac_priv *priv = dci->pvt_info;
u32 mask;
if (!data)
return -EFAULT;
if (kstrtou8(data, 0, &priv->ue_bitpos1))
return -EINVAL;
if (priv->ue_bitpos0 == priv->ue_bitpos1) {
edac_printk(KERN_ERR, EDAC_DEVICE,
"Bit positions should not be equal\n");
return -EINVAL;
}
/* If both bit postions are referring to 32 bit data, then configure
* only FID0 register or if it is 64 bit data, then configure only
* FID1 register.
*/
if ((priv->ue_bitpos0 >= 0 && priv->ue_bitpos0 <= 31) &&
(priv->ue_bitpos1 >= 0 && priv->ue_bitpos1 <= 31)) {
mask = (1 << priv->ue_bitpos0);
mask |= (1 << priv->ue_bitpos1);
writel(mask, priv->baseaddr + OCM_FID0_OFST);
writel(0, priv->baseaddr + OCM_FID1_OFST);
} else if ((priv->ue_bitpos0 >= 32 && priv->ue_bitpos0 <= 63) &&
(priv->ue_bitpos1 >= 32 && priv->ue_bitpos1 <= 63)) {
mask = (1 << (priv->ue_bitpos0 - 32));
mask |= (1 << (priv->ue_bitpos1 - 32));
writel(mask, priv->baseaddr + OCM_FID1_OFST);
writel(0, priv->baseaddr + OCM_FID0_OFST);
}
/* If one bit position is referring a bit in 32 bit data and other in
* 64 bit data, just configure FID0/FID1 based on uebitpos1.
*/
if ((priv->ue_bitpos0 >= 0 && priv->ue_bitpos0 <= 31) &&
(priv->ue_bitpos1 >= 32 && priv->ue_bitpos1 <= 63)) {
writel(1 << (priv->ue_bitpos1 - 32),
priv->baseaddr + OCM_FID1_OFST);
} else if ((priv->ue_bitpos0 >= 32 && priv->ue_bitpos0 <= 63) &&
(priv->ue_bitpos1 >= 0 && priv->ue_bitpos1 <= 31)) {
writel(1 << priv->ue_bitpos1,
priv->baseaddr + OCM_FID0_OFST);
} else {
edac_printk(KERN_ERR, EDAC_DEVICE,
"Bit position > 64 is not valid, Valid bits:[63:0]\n");
}
edac_printk(KERN_INFO, EDAC_DEVICE,
"UE at Bit Position0: %d Bit Position1: %d\n",
priv->ue_bitpos0, priv->ue_bitpos1);
return count;
}
/*
* edac_init
* module initialization entry point
*/
static int __init edac_init(void)
{
int err = 0;
edac_printk(KERN_INFO, EDAC_MC, EDAC_VERSION "\n");
/*
* Harvest and clear any boot/initialization PCI parity errors
*
* FIXME: This only clears errors logged by devices present at time of
* module initialization. We should also do an initial clear
* of each newly hotplugged device.
*/
edac_pci_clear_parity_errors();
/*
* perform the registration of the /sys/devices/system/edac class object
*/
if (edac_register_sysfs_edac_name()) {
edac_printk(KERN_ERR, EDAC_MC,
"Error initializing 'edac' kobject\n");
err = -ENODEV;
goto error;
}
/*
* now set up the mc_kset under the edac class object
*/
err = edac_sysfs_setup_mc_kset();
if (err)
goto sysfs_setup_fail;
/* Setup/Initialize the workq for this core */
err = edac_workqueue_setup();
if (err) {
edac_printk(KERN_ERR, EDAC_MC, "init WorkQueue failure\n");
goto workq_fail;
}
return 0;
/* Error teardown stack */
workq_fail:
edac_sysfs_teardown_mc_kset();
sysfs_setup_fail:
edac_unregister_sysfs_edac_name();
error:
return err;
}
static int
dump_syn_reg(struct edac_device_ctl_info *edev_ctl, int err_type, u32 bank)
{
struct llcc_drv_data *drv = edev_ctl->pvt_info;
int ret;
ret = dump_syn_reg_values(drv, bank, err_type);
if (ret)
return ret;
switch (err_type) {
case LLCC_DRAM_CE:
edac_device_handle_ce(edev_ctl, 0, bank,
"LLCC Data RAM correctable Error");
break;
case LLCC_DRAM_UE:
edac_device_handle_ue(edev_ctl, 0, bank,
"LLCC Data RAM uncorrectable Error");
break;
case LLCC_TRAM_CE:
edac_device_handle_ce(edev_ctl, 0, bank,
"LLCC Tag RAM correctable Error");
break;
case LLCC_TRAM_UE:
edac_device_handle_ue(edev_ctl, 0, bank,
"LLCC Tag RAM uncorrectable Error");
break;
default:
ret = -EINVAL;
edac_printk(KERN_CRIT, EDAC_LLCC, "Unexpected error type: %d\n",
err_type);
}
return ret;
}
/**
* zynqmp_ocm_edac_inject_cebitpos_store - Set CE bit postion
* @dci: Pointer to the edac device struct
* @data: Pointer to user data
* @count: read the size bytes from buffer
*
* Set any one bit to inject CE error
* Return: Number of bytes copied.
*/
static ssize_t zynqmp_ocm_edac_inject_cebitpos_store(
struct edac_device_ctl_info *dci, const char *data,
size_t count)
{
struct zynqmp_ocm_edac_priv *priv = dci->pvt_info;
if (!data)
return -EFAULT;
if (kstrtou8(data, 0, &priv->ce_bitpos))
return -EINVAL;
if (priv->ce_bitpos >= 0 && priv->ce_bitpos <= 31) {
writel(1 << priv->ce_bitpos, priv->baseaddr + OCM_FID0_OFST);
writel(0, priv->baseaddr + OCM_FID1_OFST);
} else if (priv->ce_bitpos >= 32 && priv->ce_bitpos <= 63) {
writel(1 << (priv->ce_bitpos - 32),
priv->baseaddr + OCM_FID1_OFST);
writel(0, priv->baseaddr + OCM_FID0_OFST);
} else {
edac_printk(KERN_ERR, EDAC_DEVICE,
"Bit number > 64 is not valid\n");
}
return count;
}
static irqreturn_t
llcc_ecc_irq_handler(int irq, void *edev_ctl)
{
struct edac_device_ctl_info *edac_dev_ctl = edev_ctl;
struct llcc_drv_data *drv = edac_dev_ctl->pvt_info;
irqreturn_t irq_rc = IRQ_NONE;
u32 drp_error, trp_error, i;
int ret;
/* Iterate over the banks and look for Tag RAM or Data RAM errors */
for (i = 0; i < drv->num_banks; i++) {
ret = regmap_read(drv->regmap,
drv->offsets[i] + DRP_INTERRUPT_STATUS,
&drp_error);
if (!ret && (drp_error & SB_ECC_ERROR)) {
edac_printk(KERN_CRIT, EDAC_LLCC,
"Single Bit Error detected in Data RAM\n");
ret = dump_syn_reg(edev_ctl, LLCC_DRAM_CE, i);
} else if (!ret && (drp_error & DB_ECC_ERROR)) {
edac_printk(KERN_CRIT, EDAC_LLCC,
"Double Bit Error detected in Data RAM\n");
ret = dump_syn_reg(edev_ctl, LLCC_DRAM_UE, i);
}
if (!ret)
irq_rc = IRQ_HANDLED;
ret = regmap_read(drv->regmap,
drv->offsets[i] + TRP_INTERRUPT_0_STATUS,
&trp_error);
if (!ret && (trp_error & SB_ECC_ERROR)) {
edac_printk(KERN_CRIT, EDAC_LLCC,
"Single Bit Error detected in Tag RAM\n");
ret = dump_syn_reg(edev_ctl, LLCC_TRAM_CE, i);
} else if (!ret && (trp_error & DB_ECC_ERROR)) {
edac_printk(KERN_CRIT, EDAC_LLCC,
"Double Bit Error detected in Tag RAM\n");
ret = dump_syn_reg(edev_ctl, LLCC_TRAM_UE, i);
}
if (!ret)
irq_rc = IRQ_HANDLED;
}
return irq_rc;
}
/* Dump Syndrome registers data for Tag RAM, Data RAM bit errors*/
static int
dump_syn_reg_values(struct llcc_drv_data *drv, u32 bank, int err_type)
{
struct llcc_edac_reg_data reg_data = edac_reg_data[err_type];
int err_cnt, err_ways, ret, i;
u32 synd_reg, synd_val;
for (i = 0; i < reg_data.reg_cnt; i++) {
synd_reg = reg_data.synd_reg + (i * 4);
ret = regmap_read(drv->regmap, drv->offsets[bank] + synd_reg,
&synd_val);
if (ret)
goto clear;
edac_printk(KERN_CRIT, EDAC_LLCC, "%s: ECC_SYN%d: 0x%8x\n",
reg_data.name, i, synd_val);
}
ret = regmap_read(drv->regmap,
drv->offsets[bank] + reg_data.count_status_reg,
&err_cnt);
if (ret)
goto clear;
err_cnt &= reg_data.count_mask;
err_cnt >>= reg_data.count_shift;
edac_printk(KERN_CRIT, EDAC_LLCC, "%s: Error count: 0x%4x\n",
reg_data.name, err_cnt);
ret = regmap_read(drv->regmap,
drv->offsets[bank] + reg_data.ways_status_reg,
&err_ways);
if (ret)
goto clear;
err_ways &= reg_data.ways_mask;
err_ways >>= reg_data.ways_shift;
edac_printk(KERN_CRIT, EDAC_LLCC, "%s: Error ways: 0x%4x\n",
reg_data.name, err_ways);
clear:
return qcom_llcc_clear_error_status(err_type, drv);
}
/*
* ->get_sdram_scrub_rate() return value semantics same as above.
*/
static ssize_t mci_sdram_scrub_rate_show(struct device *dev,
struct device_attribute *mattr,
char *data)
{
struct mem_ctl_info *mci = to_mci(dev);
int bandwidth = 0;
bandwidth = mci->get_sdram_scrub_rate(mci);
if (bandwidth < 0) {
edac_printk(KERN_DEBUG, EDAC_MC, "Error reading scrub rate\n");
return bandwidth;
}
return sprintf(data, "%d\n", bandwidth);
}
/**
* zynqmp_ocm_edac_inject_uebitpos0_store - set UE bit position0
* @dci: Pointer to the edac device struct
* @data: Pointer to user data
* @count: read the size bytes from buffer
*
* Set the first bit postion for UE Error generation,we need to configure
* any two bitpositions to inject UE Error
* Return: Number of bytes copied.
*/
static ssize_t zynqmp_ocm_edac_inject_uebitpos0_store(
struct edac_device_ctl_info *dci,
const char *data, size_t count)
{
struct zynqmp_ocm_edac_priv *priv = dci->pvt_info;
if (!data)
return -EFAULT;
if (kstrtou8(data, 0, &priv->ue_bitpos0))
return -EINVAL;
if (priv->ue_bitpos0 >= 0 && priv->ue_bitpos0 <= 31)
writel(1 << priv->ue_bitpos0, priv->baseaddr + OCM_FID0_OFST);
else if (priv->ue_bitpos0 >= 32 && priv->ue_bitpos0 <= 63)
writel(1 << (priv->ue_bitpos0 - 32),
priv->baseaddr + OCM_FID1_OFST);
else
edac_printk(KERN_ERR, EDAC_DEVICE,
"Bit position > 64 is not valid\n");
edac_printk(KERN_INFO, EDAC_DEVICE,
"Set another bit position for UE\n");
return count;
}
/* Memory scrubbing interface:
*
* A MC driver can limit the scrubbing bandwidth based on the CPU type.
* Therefore, ->set_sdram_scrub_rate should be made to return the actual
* bandwidth that is accepted or 0 when scrubbing is to be disabled.
*
* Negative value still means that an error has occurred while setting
* the scrub rate.
*/
static ssize_t mci_sdram_scrub_rate_store(struct device *dev,
struct device_attribute *mattr,
const char *data, size_t count)
{
struct mem_ctl_info *mci = to_mci(dev);
unsigned long bandwidth = 0;
int new_bw = 0;
if (kstrtoul(data, 10, &bandwidth) < 0)
return -EINVAL;
new_bw = mci->set_sdram_scrub_rate(mci, bandwidth);
if (new_bw < 0) {
edac_printk(KERN_WARNING, EDAC_MC,
"Error setting scrub rate to: %lu\n", bandwidth);
return -EINVAL;
}
return count;
}
/* Clear the error interrupt and counter registers */
static int
qcom_llcc_clear_error_status(int err_type, struct llcc_drv_data *drv)
{
int ret = 0;
switch (err_type) {
case LLCC_DRAM_CE:
case LLCC_DRAM_UE:
ret = regmap_write(drv->bcast_regmap, DRP_INTERRUPT_CLEAR,
DRP_TRP_INT_CLEAR);
if (ret)
return ret;
ret = regmap_write(drv->bcast_regmap, DRP_ECC_ERROR_CNTR_CLEAR,
DRP_TRP_CNT_CLEAR);
if (ret)
return ret;
break;
case LLCC_TRAM_CE:
case LLCC_TRAM_UE:
ret = regmap_write(drv->bcast_regmap, TRP_INTERRUPT_0_CLEAR,
DRP_TRP_INT_CLEAR);
if (ret)
return ret;
ret = regmap_write(drv->bcast_regmap, TRP_ECC_ERROR_CNTR_CLEAR,
DRP_TRP_CNT_CLEAR);
if (ret)
return ret;
break;
default:
ret = -EINVAL;
edac_printk(KERN_CRIT, EDAC_LLCC, "Unexpected error type: %d\n",
err_type);
}
return ret;
}
/*
* PCI Parity polling
*
* Fucntion to retrieve the current parity status
* and decode it
*
*/
static void edac_pci_dev_parity_test(struct pci_dev *dev)
{
unsigned long flags;
u16 status;
u8 header_type;
/* stop any interrupts until we can acquire the status */
local_irq_save(flags);
/* read the STATUS register on this device */
status = get_pci_parity_status(dev, 0);
/* read the device TYPE, looking for bridges */
pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type);
local_irq_restore(flags);
debugf4("PCI STATUS= 0x%04x %s\n", status, dev_name(&dev->dev));
/* check the status reg for errors on boards NOT marked as broken
* if broken, we cannot trust any of the status bits
*/
if (status && !dev->broken_parity_status) {
if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) {
edac_printk(KERN_CRIT, EDAC_PCI,
"Signaled System Error on %s\n",
pci_name(dev));
atomic_inc(&pci_nonparity_count);
}
if (status & (PCI_STATUS_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI,
"Master Data Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
if (status & (PCI_STATUS_DETECTED_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI,
"Detected Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
}
debugf4("PCI HEADER TYPE= 0x%02x %s\n", header_type, dev_name(&dev->dev));
if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
/* On bridges, need to examine secondary status register */
status = get_pci_parity_status(dev, 1);
debugf4("PCI SEC_STATUS= 0x%04x %s\n", status, dev_name(&dev->dev));
/* check the secondary status reg for errors,
* on NOT broken boards
*/
if (status && !dev->broken_parity_status) {
if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) {
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Signaled System Error on %s\n",
pci_name(dev));
atomic_inc(&pci_nonparity_count);
}
if (status & (PCI_STATUS_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Master Data Parity Error on "
"%s\n", pci_name(dev));
atomic_inc(&pci_parity_count);
}
if (status & (PCI_STATUS_DETECTED_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Detected Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
}
}
}
/*
* PCI Parity polling
*
*/
static void edac_pci_dev_parity_test(struct pci_dev *dev)
{
u16 status;
u8 header_type;
/* read the STATUS register on this device
*/
status = get_pci_parity_status(dev, 0);
debugf2("PCI STATUS= 0x%04x %s\n", status, dev->dev.bus_id );
/* check the status reg for errors */
if (status) {
if (status & (PCI_STATUS_SIG_SYSTEM_ERROR))
edac_printk(KERN_CRIT, EDAC_PCI,
"Signaled System Error on %s\n",
pci_name(dev));
if (status & (PCI_STATUS_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI,
"Master Data Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
if (status & (PCI_STATUS_DETECTED_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI,
"Detected Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
}
/* read the device TYPE, looking for bridges */
pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type);
debugf2("PCI HEADER TYPE= 0x%02x %s\n", header_type, dev->dev.bus_id );
if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
/* On bridges, need to examine secondary status register */
status = get_pci_parity_status(dev, 1);
debugf2("PCI SEC_STATUS= 0x%04x %s\n",
status, dev->dev.bus_id );
/* check the secondary status reg for errors */
if (status) {
if (status & (PCI_STATUS_SIG_SYSTEM_ERROR))
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Signaled System Error on %s\n",
pci_name(dev));
if (status & (PCI_STATUS_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Master Data Parity Error on "
"%s\n", pci_name(dev));
atomic_inc(&pci_parity_count);
}
if (status & (PCI_STATUS_DETECTED_PARITY)) {
edac_printk(KERN_CRIT, EDAC_PCI, "Bridge "
"Detected Parity Error on %s\n",
pci_name(dev));
atomic_inc(&pci_parity_count);
}
}
}
}
/**
* zynqmp_ocm_edac_probe - Check controller and bind driver
* @pdev: Pointer to the platform_device struct
*
* Probes a specific controller instance for binding with the driver.
*
* Return: 0 if the controller instance was successfully bound to the
* driver; otherwise, < 0 on error.
*/
static int zynqmp_ocm_edac_probe(struct platform_device *pdev)
{
struct edac_device_ctl_info *dci;
struct zynqmp_ocm_edac_priv *priv;
int irq, status;
struct resource *res;
void __iomem *baseaddr;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
baseaddr = devm_ioremap_resource(&pdev->dev, res);
if (IS_ERR(baseaddr))
return PTR_ERR(baseaddr);
if (!zynqmp_ocm_edac_get_eccstate(baseaddr)) {
edac_printk(KERN_INFO, EDAC_DEVICE,
"ECC not enabled - Disabling EDAC driver\n");
return -ENXIO;
}
dci = edac_device_alloc_ctl_info(sizeof(*priv), ZYNQMP_OCM_EDAC_STRING,
1, ZYNQMP_OCM_EDAC_STRING, 1, 0, NULL, 0, 0);
if (!dci) {
edac_printk(KERN_ERR, EDAC_DEVICE,
"Unable to allocate EDAC device\n");
return -ENOMEM;
}
priv = dci->pvt_info;
platform_set_drvdata(pdev, dci);
dci->dev = &pdev->dev;
priv->baseaddr = baseaddr;
dci->mod_name = pdev->dev.driver->name;
dci->ctl_name = ZYNQMP_OCM_EDAC_STRING;
dci->dev_name = dev_name(&pdev->dev);
zynqmp_set_ocm_edac_sysfs_attributes(dci);
if (edac_device_add_device(dci))
goto free_dev_ctl;
irq = platform_get_irq(pdev, 0);
if (irq < 0) {
edac_printk(KERN_ERR, EDAC_DEVICE,
"No irq %d in DT\n", irq);
return irq;
}
status = devm_request_irq(&pdev->dev, irq,
zynqmp_ocm_edac_intr_handler,
0, dev_name(&pdev->dev), dci);
if (status < 0) {
edac_printk(KERN_ERR, EDAC_DEVICE, "Failed to request Irq\n");
goto free_edac_dev;
}
writel(OCM_CEUE_MASK, priv->baseaddr + OCM_IEN_OFST);
return 0;
free_edac_dev:
edac_device_del_device(&pdev->dev);
free_dev_ctl:
edac_device_free_ctl_info(dci);
return -1;
}
