1/* Pseudo NMI support on sparc64 systems.
  2 *
  3 * Copyright (C) 2009 David S. Miller <davem@davemloft.net>
  4 *
  5 * The NMI watchdog support and infrastructure is based almost
  6 * entirely upon the x86 NMI support code.
  7 */
  8#include <linux/kernel.h>
  9#include <linux/param.h>
 10#include <linux/init.h>
 11#include <linux/percpu.h>
 12#include <linux/nmi.h>
 13#include <linux/export.h>
 14#include <linux/kprobes.h>
 15#include <linux/kernel_stat.h>
 16#include <linux/reboot.h>
 17#include <linux/slab.h>
 18#include <linux/kdebug.h>
 19#include <linux/delay.h>
 20#include <linux/smp.h>
 21
 22#include <asm/perf_event.h>
 23#include <asm/ptrace.h>
 24#include <asm/pcr.h>
 25
 26#include "kstack.h"
 27
 28/* We don't have a real NMI on sparc64, but we can fake one
 29 * up using profiling counter overflow interrupts and interrupt
 30 * levels.
 31 *
 32 * The profile overflow interrupts at level 15, so we use
 33 * level 14 as our IRQ off level.
 34 */
 35
 36static int panic_on_timeout;
 37
 38/* nmi_active:
 39 * >0: the NMI watchdog is active, but can be disabled
 40 * <0: the NMI watchdog has not been set up, and cannot be enabled
 41 *  0: the NMI watchdog is disabled, but can be enabled
 42 */
 43atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 44EXPORT_SYMBOL(nmi_active);
 45
 46static unsigned int nmi_hz = HZ;
 47static DEFINE_PER_CPU(short, wd_enabled);
 48static int endflag __initdata;
 49
 50static DEFINE_PER_CPU(unsigned int, last_irq_sum);
 51static DEFINE_PER_CPU(long, alert_counter);
 52static DEFINE_PER_CPU(int, nmi_touch);
 53
 54void touch_nmi_watchdog(void)
 55{
 56	if (atomic_read(&nmi_active)) {
 57		int cpu;
 58
 59		for_each_present_cpu(cpu) {
 60			if (per_cpu(nmi_touch, cpu) != 1)
 61				per_cpu(nmi_touch, cpu) = 1;
 62		}
 63	}
 
 
 64
 65	touch_softlockup_watchdog();
 
 
 66}
 67EXPORT_SYMBOL(touch_nmi_watchdog);
 68
 69static void die_nmi(const char *str, struct pt_regs *regs, int do_panic)
 70{
 71	int this_cpu = smp_processor_id();
 72
 73	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0,
 74		       pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP)
 75		return;
 76
 77	if (do_panic || panic_on_oops)
 78		panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu);
 79	else
 80		WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu);
 81}
 82
 83notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs)
 84{
 85	unsigned int sum, touched = 0;
 86	void *orig_sp;
 87
 88	clear_softint(1 << irq);
 89
 90	local_cpu_data().__nmi_count++;
 91
 92	nmi_enter();
 93
 94	orig_sp = set_hardirq_stack();
 95
 96	if (notify_die(DIE_NMI, "nmi", regs, 0,
 97		       pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP)
 98		touched = 1;
 99	else
100		pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
101
102	sum = local_cpu_data().irq0_irqs;
103	if (__get_cpu_var(nmi_touch)) {
104		__get_cpu_var(nmi_touch) = 0;
105		touched = 1;
106	}
107	if (!touched && __get_cpu_var(last_irq_sum) == sum) {
108		__this_cpu_inc(alert_counter);
109		if (__this_cpu_read(alert_counter) == 30 * nmi_hz)
110			die_nmi("BUG: NMI Watchdog detected LOCKUP",
111				regs, panic_on_timeout);
112	} else {
113		__get_cpu_var(last_irq_sum) = sum;
114		__this_cpu_write(alert_counter, 0);
115	}
116	if (__get_cpu_var(wd_enabled)) {
117		pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
118		pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
119	}
120
121	restore_hardirq_stack(orig_sp);
122
123	nmi_exit();
124}
125
126static inline unsigned int get_nmi_count(int cpu)
127{
128	return cpu_data(cpu).__nmi_count;
129}
130
131static __init void nmi_cpu_busy(void *data)
132{
133	local_irq_enable_in_hardirq();
134	while (endflag == 0)
135		mb();
136}
137
138static void report_broken_nmi(int cpu, int *prev_nmi_count)
139{
140	printk(KERN_CONT "\n");
141
142	printk(KERN_WARNING
143		"WARNING: CPU#%d: NMI appears to be stuck (%d->%d)!\n",
144			cpu, prev_nmi_count[cpu], get_nmi_count(cpu));
145
146	printk(KERN_WARNING
147		"Please report this to bugzilla.kernel.org,\n");
148	printk(KERN_WARNING
149		"and attach the output of the 'dmesg' command.\n");
150
151	per_cpu(wd_enabled, cpu) = 0;
152	atomic_dec(&nmi_active);
153}
154
155void stop_nmi_watchdog(void *unused)
156{
 
 
157	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
158	__get_cpu_var(wd_enabled) = 0;
159	atomic_dec(&nmi_active);
160}
161
162static int __init check_nmi_watchdog(void)
163{
164	unsigned int *prev_nmi_count;
165	int cpu, err;
166
167	if (!atomic_read(&nmi_active))
168		return 0;
169
170	prev_nmi_count = kmalloc(nr_cpu_ids * sizeof(unsigned int), GFP_KERNEL);
 
171	if (!prev_nmi_count) {
172		err = -ENOMEM;
173		goto error;
174	}
175
176	printk(KERN_INFO "Testing NMI watchdog ... ");
177
178	smp_call_function(nmi_cpu_busy, (void *)&endflag, 0);
179
180	for_each_possible_cpu(cpu)
181		prev_nmi_count[cpu] = get_nmi_count(cpu);
182	local_irq_enable();
183	mdelay((20 * 1000) / nmi_hz); /* wait 20 ticks */
184
185	for_each_online_cpu(cpu) {
186		if (!per_cpu(wd_enabled, cpu))
187			continue;
188		if (get_nmi_count(cpu) - prev_nmi_count[cpu] <= 5)
189			report_broken_nmi(cpu, prev_nmi_count);
190	}
191	endflag = 1;
192	if (!atomic_read(&nmi_active)) {
193		kfree(prev_nmi_count);
194		atomic_set(&nmi_active, -1);
195		err = -ENODEV;
196		goto error;
197	}
198	printk("OK.\n");
199
200	nmi_hz = 1;
201
202	kfree(prev_nmi_count);
203	return 0;
204error:
205	on_each_cpu(stop_nmi_watchdog, NULL, 1);
206	return err;
207}
208
209void start_nmi_watchdog(void *unused)
210{
211	__get_cpu_var(wd_enabled) = 1;
 
 
 
212	atomic_inc(&nmi_active);
213
214	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
215	pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
216
217	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
218}
219
220static void nmi_adjust_hz_one(void *unused)
221{
222	if (!__get_cpu_var(wd_enabled))
223		return;
224
225	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
226	pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
227
228	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
229}
230
231void nmi_adjust_hz(unsigned int new_hz)
232{
233	nmi_hz = new_hz;
234	on_each_cpu(nmi_adjust_hz_one, NULL, 1);
235}
236EXPORT_SYMBOL_GPL(nmi_adjust_hz);
237
238static int nmi_shutdown(struct notifier_block *nb, unsigned long cmd, void *p)
239{
240	on_each_cpu(stop_nmi_watchdog, NULL, 1);
241	return 0;
242}
243
244static struct notifier_block nmi_reboot_notifier = {
245	.notifier_call = nmi_shutdown,
246};
247
248int __init nmi_init(void)
249{
250	int err;
251
252	on_each_cpu(start_nmi_watchdog, NULL, 1);
253
254	err = check_nmi_watchdog();
255	if (!err) {
256		err = register_reboot_notifier(&nmi_reboot_notifier);
257		if (err) {
258			on_each_cpu(stop_nmi_watchdog, NULL, 1);
259			atomic_set(&nmi_active, -1);
260		}
261	}
262
 
 
263	return err;
264}
265
266static int __init setup_nmi_watchdog(char *str)
267{
268	if (!strncmp(str, "panic", 5))
269		panic_on_timeout = 1;
270
271	return 0;
272}
273__setup("nmi_watchdog=", setup_nmi_watchdog);

  1// SPDX-License-Identifier: GPL-2.0-only
  2/* Pseudo NMI support on sparc64 systems.
  3 *
  4 * Copyright (C) 2009 David S. Miller <davem@davemloft.net>
  5 *
  6 * The NMI watchdog support and infrastructure is based almost
  7 * entirely upon the x86 NMI support code.
  8 */
  9#include <linux/kernel.h>
 10#include <linux/param.h>
 11#include <linux/init.h>
 12#include <linux/percpu.h>
 13#include <linux/nmi.h>
 14#include <linux/export.h>
 15#include <linux/kprobes.h>
 16#include <linux/kernel_stat.h>
 17#include <linux/reboot.h>
 18#include <linux/slab.h>
 19#include <linux/kdebug.h>
 20#include <linux/delay.h>
 21#include <linux/smp.h>
 22
 23#include <asm/perf_event.h>
 24#include <asm/ptrace.h>
 25#include <asm/pcr.h>
 26
 27#include "kstack.h"
 28
 29/* We don't have a real NMI on sparc64, but we can fake one
 30 * up using profiling counter overflow interrupts and interrupt
 31 * levels.
 32 *
 33 * The profile overflow interrupts at level 15, so we use
 34 * level 14 as our IRQ off level.
 35 */
 36
 37static int panic_on_timeout;
 38
 39/* nmi_active:
 40 * >0: the NMI watchdog is active, but can be disabled
 41 * <0: the NMI watchdog has not been set up, and cannot be enabled
 42 *  0: the NMI watchdog is disabled, but can be enabled
 43 */
 44atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 45EXPORT_SYMBOL(nmi_active);
 46static int nmi_init_done;
 47static unsigned int nmi_hz = HZ;
 48static DEFINE_PER_CPU(short, wd_enabled);
 49static int endflag __initdata;
 50
 51static DEFINE_PER_CPU(unsigned int, last_irq_sum);
 52static DEFINE_PER_CPU(long, alert_counter);
 53static DEFINE_PER_CPU(int, nmi_touch);
 54
 55void arch_touch_nmi_watchdog(void)
 56{
 57	if (atomic_read(&nmi_active)) {
 58		int cpu;
 59
 60		for_each_present_cpu(cpu) {
 61			if (per_cpu(nmi_touch, cpu) != 1)
 62				per_cpu(nmi_touch, cpu) = 1;
 63		}
 64	}
 65}
 66EXPORT_SYMBOL(arch_touch_nmi_watchdog);
 67
 68int __init watchdog_hardlockup_probe(void)
 69{
 70	return 0;
 71}
 
 72
 73static void die_nmi(const char *str, struct pt_regs *regs, int do_panic)
 74{
 75	int this_cpu = smp_processor_id();
 76
 77	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0,
 78		       pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP)
 79		return;
 80
 81	if (do_panic || panic_on_oops)
 82		panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu);
 83	else
 84		WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu);
 85}
 86
 87notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs)
 88{
 89	unsigned int sum, touched = 0;
 90	void *orig_sp;
 91
 92	clear_softint(1 << irq);
 93
 94	local_cpu_data().__nmi_count++;
 95
 96	nmi_enter();
 97
 98	orig_sp = set_hardirq_stack();
 99
100	if (notify_die(DIE_NMI, "nmi", regs, 0,
101		       pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP)
102		touched = 1;
103	else
104		pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
105
106	sum = local_cpu_data().irq0_irqs;
107	if (__this_cpu_read(nmi_touch)) {
108		__this_cpu_write(nmi_touch, 0);
109		touched = 1;
110	}
111	if (!touched && __this_cpu_read(last_irq_sum) == sum) {
112		__this_cpu_inc(alert_counter);
113		if (__this_cpu_read(alert_counter) == 30 * nmi_hz)
114			die_nmi("BUG: NMI Watchdog detected LOCKUP",
115				regs, panic_on_timeout);
116	} else {
117		__this_cpu_write(last_irq_sum, sum);
118		__this_cpu_write(alert_counter, 0);
119	}
120	if (__this_cpu_read(wd_enabled)) {
121		pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
122		pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
123	}
124
125	restore_hardirq_stack(orig_sp);
126
127	nmi_exit();
128}
129
130static inline unsigned int get_nmi_count(int cpu)
131{
132	return cpu_data(cpu).__nmi_count;
133}
134
135static __init void nmi_cpu_busy(void *data)
136{
 
137	while (endflag == 0)
138		mb();
139}
140
141static void report_broken_nmi(int cpu, int *prev_nmi_count)
142{
143	printk(KERN_CONT "\n");
144
145	printk(KERN_WARNING
146		"WARNING: CPU#%d: NMI appears to be stuck (%d->%d)!\n",
147			cpu, prev_nmi_count[cpu], get_nmi_count(cpu));
148
149	printk(KERN_WARNING
150		"Please report this to bugzilla.kernel.org,\n");
151	printk(KERN_WARNING
152		"and attach the output of the 'dmesg' command.\n");
153
154	per_cpu(wd_enabled, cpu) = 0;
155	atomic_dec(&nmi_active);
156}
157
158void stop_nmi_watchdog(void *unused)
159{
160	if (!__this_cpu_read(wd_enabled))
161		return;
162	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
163	__this_cpu_write(wd_enabled, 0);
164	atomic_dec(&nmi_active);
165}
166
167static int __init check_nmi_watchdog(void)
168{
169	unsigned int *prev_nmi_count;
170	int cpu, err;
171
172	if (!atomic_read(&nmi_active))
173		return 0;
174
175	prev_nmi_count = kmalloc_array(nr_cpu_ids, sizeof(unsigned int),
176				       GFP_KERNEL);
177	if (!prev_nmi_count) {
178		err = -ENOMEM;
179		goto error;
180	}
181
182	printk(KERN_INFO "Testing NMI watchdog ... ");
183
184	smp_call_function(nmi_cpu_busy, (void *)&endflag, 0);
185
186	for_each_possible_cpu(cpu)
187		prev_nmi_count[cpu] = get_nmi_count(cpu);
188	local_irq_enable();
189	mdelay((20 * 1000) / nmi_hz); /* wait 20 ticks */
190
191	for_each_online_cpu(cpu) {
192		if (!per_cpu(wd_enabled, cpu))
193			continue;
194		if (get_nmi_count(cpu) - prev_nmi_count[cpu] <= 5)
195			report_broken_nmi(cpu, prev_nmi_count);
196	}
197	endflag = 1;
198	if (!atomic_read(&nmi_active)) {
199		kfree(prev_nmi_count);
200		atomic_set(&nmi_active, -1);
201		err = -ENODEV;
202		goto error;
203	}
204	printk("OK.\n");
205
206	nmi_hz = 1;
207
208	kfree(prev_nmi_count);
209	return 0;
210error:
211	on_each_cpu(stop_nmi_watchdog, NULL, 1);
212	return err;
213}
214
215void start_nmi_watchdog(void *unused)
216{
217	if (__this_cpu_read(wd_enabled))
218		return;
219
220	__this_cpu_write(wd_enabled, 1);
221	atomic_inc(&nmi_active);
222
223	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
224	pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
225
226	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
227}
228
229static void nmi_adjust_hz_one(void *unused)
230{
231	if (!__this_cpu_read(wd_enabled))
232		return;
233
234	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable);
235	pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz));
236
237	pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable);
238}
239
240void nmi_adjust_hz(unsigned int new_hz)
241{
242	nmi_hz = new_hz;
243	on_each_cpu(nmi_adjust_hz_one, NULL, 1);
244}
245EXPORT_SYMBOL_GPL(nmi_adjust_hz);
246
247static int nmi_shutdown(struct notifier_block *nb, unsigned long cmd, void *p)
248{
249	on_each_cpu(stop_nmi_watchdog, NULL, 1);
250	return 0;
251}
252
253static struct notifier_block nmi_reboot_notifier = {
254	.notifier_call = nmi_shutdown,
255};
256
257int __init nmi_init(void)
258{
259	int err;
260
261	on_each_cpu(start_nmi_watchdog, NULL, 1);
262
263	err = check_nmi_watchdog();
264	if (!err) {
265		err = register_reboot_notifier(&nmi_reboot_notifier);
266		if (err) {
267			on_each_cpu(stop_nmi_watchdog, NULL, 1);
268			atomic_set(&nmi_active, -1);
269		}
270	}
271
272	nmi_init_done = 1;
273
274	return err;
275}
276
277static int __init setup_nmi_watchdog(char *str)
278{
279	if (!strncmp(str, "panic", 5))
280		panic_on_timeout = 1;
281
282	return 1;
283}
284__setup("nmi_watchdog=", setup_nmi_watchdog);
285
286/*
287 * sparc specific NMI watchdog enable function.
288 * Enables watchdog if it is not enabled already.
289 */
290void watchdog_hardlockup_enable(unsigned int cpu)
291{
292	if (atomic_read(&nmi_active) == -1) {
293		pr_warn("NMI watchdog cannot be enabled or disabled\n");
294		return;
295	}
296
297	/*
298	 * watchdog thread could start even before nmi_init is called.
299	 * Just Return in that case. Let nmi_init finish the init
300	 * process first.
301	 */
302	if (!nmi_init_done)
303		return;
304
305	smp_call_function_single(cpu, start_nmi_watchdog, NULL, 1);
306}
307/*
308 * sparc specific NMI watchdog disable function.
309 * Disables watchdog if it is not disabled already.
310 */
311void watchdog_hardlockup_disable(unsigned int cpu)
312{
313	if (atomic_read(&nmi_active) == -1)
314		pr_warn_once("NMI watchdog cannot be enabled or disabled\n");
315	else
316		smp_call_function_single(cpu, stop_nmi_watchdog, NULL, 1);
317}