diff options
Diffstat (limited to 'arch/powerpc/platforms/pseries/ras.c')
| -rw-r--r-- | arch/powerpc/platforms/pseries/ras.c | 309 |
1 files changed, 206 insertions, 103 deletions
diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c index a4fc6da87c2..9c5778e6ed4 100644 --- a/arch/powerpc/platforms/pseries/ras.c +++ b/arch/powerpc/platforms/pseries/ras.c @@ -16,37 +16,15 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -/* Change Activity: - * 2001/09/21 : engebret : Created with minimal EPOW and HW exception support. - * End Change Activity - */ - -#include <linux/errno.h> -#include <linux/threads.h> -#include <linux/kernel_stat.h> -#include <linux/signal.h> #include <linux/sched.h> -#include <linux/ioport.h> #include <linux/interrupt.h> -#include <linux/timex.h> -#include <linux/init.h> -#include <linux/delay.h> #include <linux/irq.h> -#include <linux/random.h> -#include <linux/sysrq.h> -#include <linux/bitops.h> - -#include <asm/uaccess.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/irq.h> -#include <asm/cache.h> -#include <asm/prom.h> -#include <asm/ptrace.h> +#include <linux/of.h> +#include <linux/fs.h> +#include <linux/reboot.h> + #include <asm/machdep.h> #include <asm/rtas.h> -#include <asm/udbg.h> #include <asm/firmware.h> #include "pseries.h" @@ -54,9 +32,9 @@ static unsigned char ras_log_buf[RTAS_ERROR_LOG_MAX]; static DEFINE_SPINLOCK(ras_log_buf_lock); -static char mce_data_buf[RTAS_ERROR_LOG_MAX]; +static char global_mce_data_buf[RTAS_ERROR_LOG_MAX]; +static DEFINE_PER_CPU(__u64, mce_data_buf); -static int ras_get_sensor_state_token; static int ras_check_exception_token; #define EPOW_SENSOR_TOKEN 9 @@ -74,7 +52,6 @@ static int __init init_ras_IRQ(void) { struct device_node *np; - ras_get_sensor_state_token = rtas_token("get-sensor-state"); ras_check_exception_token = rtas_token("check-exception"); /* Internal Errors */ @@ -94,26 +71,126 @@ static int __init init_ras_IRQ(void) return 0; } -__initcall(init_ras_IRQ); +subsys_initcall(init_ras_IRQ); -/* - * Handle power subsystem events (EPOW). - * - * Presently we just log the event has occurred. This should be fixed - * to examine the type of power failure and take appropriate action where - * the time horizon permits something useful to be done. - */ +#define EPOW_SHUTDOWN_NORMAL 1 +#define EPOW_SHUTDOWN_ON_UPS 2 +#define EPOW_SHUTDOWN_LOSS_OF_CRITICAL_FUNCTIONS 3 +#define EPOW_SHUTDOWN_AMBIENT_TEMPERATURE_TOO_HIGH 4 + +static void handle_system_shutdown(char event_modifier) +{ + switch (event_modifier) { + case EPOW_SHUTDOWN_NORMAL: + pr_emerg("Firmware initiated power off"); + orderly_poweroff(true); + break; + + case EPOW_SHUTDOWN_ON_UPS: + pr_emerg("Loss of power reported by firmware, system is " + "running on UPS/battery"); + break; + + case EPOW_SHUTDOWN_LOSS_OF_CRITICAL_FUNCTIONS: + pr_emerg("Loss of system critical functions reported by " + "firmware"); + pr_emerg("Check RTAS error log for details"); + orderly_poweroff(true); + break; + + case EPOW_SHUTDOWN_AMBIENT_TEMPERATURE_TOO_HIGH: + pr_emerg("Ambient temperature too high reported by firmware"); + pr_emerg("Check RTAS error log for details"); + orderly_poweroff(true); + break; + + default: + pr_err("Unknown power/cooling shutdown event (modifier %d)", + event_modifier); + } +} + +struct epow_errorlog { + unsigned char sensor_value; + unsigned char event_modifier; + unsigned char extended_modifier; + unsigned char reserved; + unsigned char platform_reason; +}; + +#define EPOW_RESET 0 +#define EPOW_WARN_COOLING 1 +#define EPOW_WARN_POWER 2 +#define EPOW_SYSTEM_SHUTDOWN 3 +#define EPOW_SYSTEM_HALT 4 +#define EPOW_MAIN_ENCLOSURE 5 +#define EPOW_POWER_OFF 7 + +void rtas_parse_epow_errlog(struct rtas_error_log *log) +{ + struct pseries_errorlog *pseries_log; + struct epow_errorlog *epow_log; + char action_code; + char modifier; + + pseries_log = get_pseries_errorlog(log, PSERIES_ELOG_SECT_ID_EPOW); + if (pseries_log == NULL) + return; + + epow_log = (struct epow_errorlog *)pseries_log->data; + action_code = epow_log->sensor_value & 0xF; /* bottom 4 bits */ + modifier = epow_log->event_modifier & 0xF; /* bottom 4 bits */ + + switch (action_code) { + case EPOW_RESET: + pr_err("Non critical power or cooling issue cleared"); + break; + + case EPOW_WARN_COOLING: + pr_err("Non critical cooling issue reported by firmware"); + pr_err("Check RTAS error log for details"); + break; + + case EPOW_WARN_POWER: + pr_err("Non critical power issue reported by firmware"); + pr_err("Check RTAS error log for details"); + break; + + case EPOW_SYSTEM_SHUTDOWN: + handle_system_shutdown(epow_log->event_modifier); + break; + + case EPOW_SYSTEM_HALT: + pr_emerg("Firmware initiated power off"); + orderly_poweroff(true); + break; + + case EPOW_MAIN_ENCLOSURE: + case EPOW_POWER_OFF: + pr_emerg("Critical power/cooling issue reported by firmware"); + pr_emerg("Check RTAS error log for details"); + pr_emerg("Immediate power off"); + emergency_sync(); + kernel_power_off(); + break; + + default: + pr_err("Unknown power/cooling event (action code %d)", + action_code); + } +} + +/* Handle environmental and power warning (EPOW) interrupts. */ static irqreturn_t ras_epow_interrupt(int irq, void *dev_id) { - int status = 0xdeadbeef; - int state = 0; + int status; + int state; int critical; - status = rtas_call(ras_get_sensor_state_token, 2, 2, &state, - EPOW_SENSOR_TOKEN, EPOW_SENSOR_INDEX); + status = rtas_get_sensor(EPOW_SENSOR_TOKEN, EPOW_SENSOR_INDEX, &state); if (state > 3) - critical = 1; /* Time Critical */ + critical = 1; /* Time Critical */ else critical = 0; @@ -121,19 +198,15 @@ static irqreturn_t ras_epow_interrupt(int irq, void *dev_id) status = rtas_call(ras_check_exception_token, 6, 1, NULL, RTAS_VECTOR_EXTERNAL_INTERRUPT, - irq_map[irq].hwirq, - RTAS_EPOW_WARNING | RTAS_POWERMGM_EVENTS, + virq_to_hw(irq), + RTAS_EPOW_WARNING, critical, __pa(&ras_log_buf), rtas_get_error_log_max()); - udbg_printf("EPOW <0x%lx 0x%x 0x%x>\n", - *((unsigned long *)&ras_log_buf), status, state); - printk(KERN_WARNING "EPOW <0x%lx 0x%x 0x%x>\n", - *((unsigned long *)&ras_log_buf), status, state); - - /* format and print the extended information */ log_error(ras_log_buf, ERR_TYPE_RTAS_LOG, 0); + rtas_parse_epow_errlog((struct rtas_error_log *)ras_log_buf); + spin_unlock(&ras_log_buf_lock); return IRQ_HANDLED; } @@ -149,21 +222,22 @@ static irqreturn_t ras_epow_interrupt(int irq, void *dev_id) static irqreturn_t ras_error_interrupt(int irq, void *dev_id) { struct rtas_error_log *rtas_elog; - int status = 0xdeadbeef; + int status; int fatal; spin_lock(&ras_log_buf_lock); status = rtas_call(ras_check_exception_token, 6, 1, NULL, RTAS_VECTOR_EXTERNAL_INTERRUPT, - irq_map[irq].hwirq, - RTAS_INTERNAL_ERROR, 1 /*Time Critical */, + virq_to_hw(irq), + RTAS_INTERNAL_ERROR, 1 /* Time Critical */, __pa(&ras_log_buf), rtas_get_error_log_max()); rtas_elog = (struct rtas_error_log *)ras_log_buf; - if ((status == 0) && (rtas_elog->severity >= RTAS_SEVERITY_ERROR_SYNC)) + if (status == 0 && + rtas_error_severity(rtas_elog) >= RTAS_SEVERITY_ERROR_SYNC) fatal = 1; else fatal = 0; @@ -172,36 +246,37 @@ static irqreturn_t ras_error_interrupt(int irq, void *dev_id) log_error(ras_log_buf, ERR_TYPE_RTAS_LOG, fatal); if (fatal) { - udbg_printf("Fatal HW Error <0x%lx 0x%x>\n", - *((unsigned long *)&ras_log_buf), status); - printk(KERN_EMERG "Error: Fatal hardware error <0x%lx 0x%x>\n", - *((unsigned long *)&ras_log_buf), status); - -#ifndef DEBUG_RTAS_POWER_OFF - /* Don't actually power off when debugging so we can test - * without actually failing while injecting errors. - * Error data will not be logged to syslog. - */ - ppc_md.power_off(); -#endif + pr_emerg("Fatal hardware error reported by firmware"); + pr_emerg("Check RTAS error log for details"); + pr_emerg("Immediate power off"); + emergency_sync(); + kernel_power_off(); } else { - udbg_printf("Recoverable HW Error <0x%lx 0x%x>\n", - *((unsigned long *)&ras_log_buf), status); - printk(KERN_WARNING - "Warning: Recoverable hardware error <0x%lx 0x%x>\n", - *((unsigned long *)&ras_log_buf), status); + pr_err("Recoverable hardware error reported by firmware"); } spin_unlock(&ras_log_buf_lock); return IRQ_HANDLED; } -/* Get the error information for errors coming through the +/* + * Some versions of FWNMI place the buffer inside the 4kB page starting at + * 0x7000. Other versions place it inside the rtas buffer. We check both. + */ +#define VALID_FWNMI_BUFFER(A) \ + ((((A) >= 0x7000) && ((A) < 0x7ff0)) || \ + (((A) >= rtas.base) && ((A) < (rtas.base + rtas.size - 16)))) + +/* + * Get the error information for errors coming through the * FWNMI vectors. The pt_regs' r3 will be updated to reflect * the actual r3 if possible, and a ptr to the error log entry * will be returned if found. * - * The mce_data_buf does not have any locks or protection around it, + * If the RTAS error is not of the extended type, then we put it in a per + * cpu 64bit buffer. If it is the extended type we use global_mce_data_buf. + * + * The global_mce_data_buf does not have any locks or protection around it, * if a second machine check comes in, or a system reset is done * before we have logged the error, then we will get corruption in the * error log. This is preferable over holding off on calling @@ -210,20 +285,35 @@ static irqreturn_t ras_error_interrupt(int irq, void *dev_id) */ static struct rtas_error_log *fwnmi_get_errinfo(struct pt_regs *regs) { - unsigned long errdata = regs->gpr[3]; - struct rtas_error_log *errhdr = NULL; unsigned long *savep; + struct rtas_error_log *h, *errhdr = NULL; + + /* Mask top two bits */ + regs->gpr[3] &= ~(0x3UL << 62); - if ((errdata >= 0x7000 && errdata < 0x7fff0) || - (errdata >= rtas.base && errdata < rtas.base + rtas.size - 16)) { - savep = __va(errdata); - regs->gpr[3] = savep[0]; /* restore original r3 */ - memset(mce_data_buf, 0, RTAS_ERROR_LOG_MAX); - memcpy(mce_data_buf, (char *)(savep + 1), RTAS_ERROR_LOG_MAX); - errhdr = (struct rtas_error_log *)mce_data_buf; + if (!VALID_FWNMI_BUFFER(regs->gpr[3])) { + printk(KERN_ERR "FWNMI: corrupt r3 0x%016lx\n", regs->gpr[3]); + return NULL; + } + + savep = __va(regs->gpr[3]); + regs->gpr[3] = savep[0]; /* restore original r3 */ + + /* If it isn't an extended log we can use the per cpu 64bit buffer */ + h = (struct rtas_error_log *)&savep[1]; + if (!rtas_error_extended(h)) { + memcpy(&__get_cpu_var(mce_data_buf), h, sizeof(__u64)); + errhdr = (struct rtas_error_log *)&__get_cpu_var(mce_data_buf); } else { - printk("FWNMI: corrupt r3\n"); + int len, error_log_length; + + error_log_length = 8 + rtas_error_extended_log_length(h); + len = max_t(int, error_log_length, RTAS_ERROR_LOG_MAX); + memset(global_mce_data_buf, 0, RTAS_ERROR_LOG_MAX); + memcpy(global_mce_data_buf, h, len); + errhdr = (struct rtas_error_log *)global_mce_data_buf; } + return errhdr; } @@ -235,7 +325,7 @@ static void fwnmi_release_errinfo(void) { int ret = rtas_call(rtas_token("ibm,nmi-interlock"), 0, 1, NULL); if (ret != 0) - printk("FWNMI: nmi-interlock failed: %d\n", ret); + printk(KERN_ERR "FWNMI: nmi-interlock failed: %d\n", ret); } int pSeries_system_reset_exception(struct pt_regs *regs) @@ -259,31 +349,44 @@ int pSeries_system_reset_exception(struct pt_regs *regs) * Return 1 if corrected (or delivered a signal). * Return 0 if there is nothing we can do. */ -static int recover_mce(struct pt_regs *regs, struct rtas_error_log * err) +static int recover_mce(struct pt_regs *regs, struct rtas_error_log *err) { - int nonfatal = 0; + int recovered = 0; + int disposition = rtas_error_disposition(err); + + if (!(regs->msr & MSR_RI)) { + /* If MSR_RI isn't set, we cannot recover */ + recovered = 0; - if (err->disposition == RTAS_DISP_FULLY_RECOVERED) { + } else if (disposition == RTAS_DISP_FULLY_RECOVERED) { /* Platform corrected itself */ - nonfatal = 1; - } else if ((regs->msr & MSR_RI) && - user_mode(regs) && - err->severity == RTAS_SEVERITY_ERROR_SYNC && - err->disposition == RTAS_DISP_NOT_RECOVERED && - err->target == RTAS_TARGET_MEMORY && - err->type == RTAS_TYPE_ECC_UNCORR && - !(current->pid == 0 || is_global_init(current))) { - /* Kill off a user process with an ECC error */ - printk(KERN_ERR "MCE: uncorrectable ecc error for pid %d\n", - current->pid); - /* XXX something better for ECC error? */ - _exception(SIGBUS, regs, BUS_ADRERR, regs->nip); - nonfatal = 1; + recovered = 1; + + } else if (disposition == RTAS_DISP_LIMITED_RECOVERY) { + /* Platform corrected itself but could be degraded */ + printk(KERN_ERR "MCE: limited recovery, system may " + "be degraded\n"); + recovered = 1; + + } else if (user_mode(regs) && !is_global_init(current) && + rtas_error_severity(err) == RTAS_SEVERITY_ERROR_SYNC) { + + /* + * If we received a synchronous error when in userspace + * kill the task. Firmware may report details of the fail + * asynchronously, so we can't rely on the target and type + * fields being valid here. + */ + printk(KERN_ERR "MCE: uncorrectable error, killing task " + "%s:%d\n", current->comm, current->pid); + + _exception(SIGBUS, regs, BUS_MCEERR_AR, regs->nip); + recovered = 1; } - log_error((char *)err, ERR_TYPE_RTAS_LOG, !nonfatal); + log_error((char *)err, ERR_TYPE_RTAS_LOG, 0); - return nonfatal; + return recovered; } /* |