Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 17b3a894 authored by Stepan Moskovchenko's avatar Stepan Moskovchenko Committed by Murali Nalajala
Browse files

edac: Allow panic on correctable errors (CE) 



Add an EDAC device flag and associated sysfs entries to
allow an EDAC driver to be configured to panic the kernel
if a correctable error is detected. Though correctable
errors (by definition) have no adverse system effects,
a panic may still be useful, since a correctable error may
be indicative of a marginal system state.

Change-Id: I98921469254aa7b999979c1c7d9186286f982a0c
Signed-off-by: default avatarStepan Moskovchenko <stepanm@codeaurora.org>
Signed-off-by: default avatarVenkata Narendra Kumar Gutta <vnkgutta@codeaurora.org>
[mnalajal@codeaurora.org: Make panic_on_ce as a feature
 to not to break the ABI compatability]
Signed-off-by: default avatarMurali Nalajala <mnalajal@codeaurora.org>
parent 31e5ae30

drivers/edac/Kconfig

+11 −0
Original line number Diff line number Diff line
@@ -481,6 +481,17 @@ config EDAC_TI 
	  Support for error detection and correction on the

          TI SoCs.



config EDAC_PANIC_ON_CE

	depends on EDAC

	bool "Panic on correctable errors"

	help

	  Forcibly cause a kernel panic if an correctable error (CE) is

	  detected, even though the error is (by definition) correctable and

	  would otherwise result in no adverse system effects. This feature reduce

	  debugging times on hardware which may be operating at voltages or

	  frequencies outside normal specification.

	  For production builds, you should say 'N' here.



config EDAC_QCOM

	tristate "QCOM EDAC Controller"

	depends on ARCH_QCOM && QCOM_LLCC

drivers/edac/edac_device.c

+15 −0
Original line number Diff line number Diff line
@@ -549,6 +549,14 @@ static inline int edac_device_get_log_ue(struct edac_device_ctl_info *edac_dev) 
	return edac_dev->log_ue;

}



#ifdef CONFIG_EDAC_PANIC_ON_CE

static inline int edac_device_get_panic_on_ce(struct edac_device_ctl_info

					*edac_dev)

{

	return edac_dev->panic_on_ce;

}

#endif



static inline int edac_device_get_panic_on_ue(struct edac_device_ctl_info

					*edac_dev)

{
@@ -594,6 +602,13 @@ void edac_device_handle_ce(struct edac_device_ctl_info *edac_dev, 
				"CE: %s instance: %s block: %s '%s'\n",

				edac_dev->ctl_name, instance->name,

				block ? block->name : "N/A", msg);



#ifdef CONFIG_EDAC_PANIC_ON_CE

	if (edac_device_get_panic_on_ce(edac_dev))

		panic("EDAC %s: CE instance: %s block %s '%s'\n",

			edac_dev->ctl_name, instance->name,

			block ? block->name : "N/A", msg);

#endif

}

EXPORT_SYMBOL_GPL(edac_device_handle_ce);

drivers/edac/edac_device.h

+3 −0
Original line number Diff line number Diff line
@@ -159,6 +159,9 @@ struct edac_device_ctl_info { 
	/* Per instance controls for this edac_device */

	int log_ue;		/* boolean for logging UEs */

	int log_ce;		/* boolean for logging CEs */

#ifdef CONFIG_EDAC_PANIC_ON_CE

	int panic_on_ce;	/* boolean for panic'ing on an CE */

#endif

	int panic_on_ue;	/* boolean for panic'ing on an UE */

	unsigned poll_msec;	/* number of milliseconds to poll interval */

	unsigned long delay;	/* number of jiffies for poll_msec */

drivers/edac/edac_device_sysfs.c

+34 −0
Original line number Diff line number Diff line
@@ -62,6 +62,15 @@ static ssize_t edac_device_ctl_log_ce_store(struct edac_device_ctl_info 
	return count;

}



#ifdef CONFIG_EDAC_PANIC_ON_CE

/* 'panic_on_ce' */

static ssize_t edac_device_ctl_panic_on_ce_show(struct edac_device_ctl_info

						*ctl_info, char *data)

{

	return snprintf(data, PAGE_SIZE, "%u\n", ctl_info->panic_on_ce);

}

#endif



/* 'panic_on_ue' */

static ssize_t edac_device_ctl_panic_on_ue_show(struct edac_device_ctl_info

						*ctl_info, char *data)
@@ -69,6 +78,23 @@ static ssize_t edac_device_ctl_panic_on_ue_show(struct edac_device_ctl_info 
	return sprintf(data, "%u\n", ctl_info->panic_on_ue);

}



#ifdef CONFIG_EDAC_PANIC_ON_CE

static ssize_t edac_device_ctl_panic_on_ce_store(struct edac_device_ctl_info

						 *ctl_info, const char *data,

						 size_t count)

{

	unsigned long val;



	/* if parameter is zero, turn off flag, if non-zero turn on flag */

	if (kstrtoul(data, 0, &val) < 0)

		return -EINVAL;



	ctl_info->panic_on_ce = !!val;



	return count;

}

#endif



static ssize_t edac_device_ctl_panic_on_ue_store(struct edac_device_ctl_info

						 *ctl_info, const char *data,

						 size_t count)
@@ -156,6 +182,11 @@ CTL_INFO_ATTR(log_ue, S_IRUGO | S_IWUSR, 
	edac_device_ctl_log_ue_show, edac_device_ctl_log_ue_store);

CTL_INFO_ATTR(log_ce, S_IRUGO | S_IWUSR,

	edac_device_ctl_log_ce_show, edac_device_ctl_log_ce_store);

#ifdef CONFIG_EDAC_PANIC_ON_CE

CTL_INFO_ATTR(panic_on_ce, 0644,

	edac_device_ctl_panic_on_ce_show,

	edac_device_ctl_panic_on_ce_store);

#endif

CTL_INFO_ATTR(panic_on_ue, S_IRUGO | S_IWUSR,

	edac_device_ctl_panic_on_ue_show,

	edac_device_ctl_panic_on_ue_store);
@@ -164,6 +195,9 @@ CTL_INFO_ATTR(poll_msec, S_IRUGO | S_IWUSR, 


/* Base Attributes of the EDAC_DEVICE ECC object */

static struct ctl_info_attribute *device_ctrl_attr[] = {

#ifdef CONFIG_EDAC_PANIC_ON_CE

	&attr_ctl_info_panic_on_ce,

#endif

	&attr_ctl_info_panic_on_ue,

	&attr_ctl_info_log_ue,

	&attr_ctl_info_log_ce,