summaryrefslogtreecommitdiff
path: root/arch/powerpc/kernel/crash.c
blob: abef75176c079f2f72f71d6af4f10cec4095deca (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
/*
 * Architecture specific (PPC64) functions for kexec based crash dumps.
 *
 * Copyright (C) 2005, IBM Corp.
 *
 * Created by: Haren Myneni
 *
 * This source code is licensed under the GNU General Public License,
 * Version 2.  See the file COPYING for more details.
 *
 */

#include <linux/kernel.h>
#include <linux/smp.h>
#include <linux/reboot.h>
#include <linux/kexec.h>
#include <linux/export.h>
#include <linux/crash_dump.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/irq.h>
#include <linux/types.h>

#include <asm/processor.h>
#include <asm/machdep.h>
#include <asm/kexec.h>
#include <asm/kdump.h>
#include <asm/prom.h>
#include <asm/smp.h>
#include <asm/system.h>
#include <asm/setjmp.h>

/*
 * The primary CPU waits a while for all secondary CPUs to enter. This is to
 * avoid sending an IPI if the secondary CPUs are entering
 * crash_kexec_secondary on their own (eg via a system reset).
 *
 * The secondary timeout has to be longer than the primary. Both timeouts are
 * in milliseconds.
 */
#define PRIMARY_TIMEOUT		500
#define SECONDARY_TIMEOUT	1000

#define IPI_TIMEOUT		10000
#define REAL_MODE_TIMEOUT	10000

/* This keeps a track of which one is the crashing cpu. */
int crashing_cpu = -1;
static int time_to_dump;

#define CRASH_HANDLER_MAX 3
/* NULL terminated list of shutdown handles */
static crash_shutdown_t crash_shutdown_handles[CRASH_HANDLER_MAX+1];
static DEFINE_SPINLOCK(crash_handlers_lock);

static unsigned long crash_shutdown_buf[JMP_BUF_LEN];
static int crash_shutdown_cpu = -1;

static int handle_fault(struct pt_regs *regs)
{
	if (crash_shutdown_cpu == smp_processor_id())
		longjmp(crash_shutdown_buf, 1);
	return 0;
}

#ifdef CONFIG_SMP

static atomic_t cpus_in_crash;
void crash_ipi_callback(struct pt_regs *regs)
{
	static cpumask_t cpus_state_saved = CPU_MASK_NONE;

	int cpu = smp_processor_id();

	if (!cpu_online(cpu))
		return;

	hard_irq_disable();
	if (!cpumask_test_cpu(cpu, &cpus_state_saved)) {
		crash_save_cpu(regs, cpu);
		cpumask_set_cpu(cpu, &cpus_state_saved);
	}

	atomic_inc(&cpus_in_crash);
	smp_mb__after_atomic_inc();

	/*
	 * Starting the kdump boot.
	 * This barrier is needed to make sure that all CPUs are stopped.
	 */
	while (!time_to_dump)
		cpu_relax();

	if (ppc_md.kexec_cpu_down)
		ppc_md.kexec_cpu_down(1, 1);

#ifdef CONFIG_PPC64
	kexec_smp_wait();
#else
	for (;;);	/* FIXME */
#endif

	/* NOTREACHED */
}

static void crash_kexec_prepare_cpus(int cpu)
{
	unsigned int msecs;
	unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */
	int tries = 0;
	int (*old_handler)(struct pt_regs *regs);

	printk(KERN_EMERG "Sending IPI to other CPUs\n");

	crash_send_ipi(crash_ipi_callback);
	smp_wmb();

again:
	/*
	 * FIXME: Until we will have the way to stop other CPUs reliably,
	 * the crash CPU will send an IPI and wait for other CPUs to
	 * respond.
	 */
	msecs = IPI_TIMEOUT;
	while ((atomic_read(&cpus_in_crash) < ncpus) && (--msecs > 0))
		mdelay(1);

	/* Would it be better to replace the trap vector here? */

	if (atomic_read(&cpus_in_crash) >= ncpus) {
		printk(KERN_EMERG "IPI complete\n");
		return;
	}

	printk(KERN_EMERG "ERROR: %d cpu(s) not responding\n",
		ncpus - atomic_read(&cpus_in_crash));

	/*
	 * If we have a panic timeout set then we can't wait indefinitely
	 * for someone to activate system reset. We also give up on the
	 * second time through if system reset fail to work.
	 */
	if ((panic_timeout > 0) || (tries > 0))
		return;

	/*
	 * A system reset will cause all CPUs to take an 0x100 exception.
	 * The primary CPU returns here via setjmp, and the secondary
	 * CPUs reexecute the crash_kexec_secondary path.
	 */
	old_handler = __debugger;
	__debugger = handle_fault;
	crash_shutdown_cpu = smp_processor_id();

	if (setjmp(crash_shutdown_buf) == 0) {
		printk(KERN_EMERG "Activate system reset (dumprestart) "
				  "to stop other cpu(s)\n");

		/*
		 * A system reset will force all CPUs to execute the
		 * crash code again. We need to reset cpus_in_crash so we
		 * wait for everyone to do this.
		 */
		atomic_set(&cpus_in_crash, 0);
		smp_mb();

		while (atomic_read(&cpus_in_crash) < ncpus)
			cpu_relax();
	}

	crash_shutdown_cpu = -1;
	__debugger = old_handler;

	tries++;
	goto again;
}

/*
 * This function will be called by secondary cpus.
 */
void crash_kexec_secondary(struct pt_regs *regs)
{
	unsigned long flags;
	int msecs = SECONDARY_TIMEOUT;

	local_irq_save(flags);

	/* Wait for the primary crash CPU to signal its progress */
	while (crashing_cpu < 0) {
		if (--msecs < 0) {
			/* No response, kdump image may not have been loaded */
			local_irq_restore(flags);
			return;
		}

		mdelay(1);
	}

	crash_ipi_callback(regs);
}

#else	/* ! CONFIG_SMP */

static void crash_kexec_prepare_cpus(int cpu)
{
	/*
	 * move the secondaries to us so that we can copy
	 * the new kernel 0-0x100 safely
	 *
	 * do this if kexec in setup.c ?
	 */
#ifdef CONFIG_PPC64
	smp_release_cpus();
#else
	/* FIXME */
#endif
}

void crash_kexec_secondary(struct pt_regs *regs)
{
}
#endif	/* CONFIG_SMP */

/* wait for all the CPUs to hit real mode but timeout if they don't come in */
#if defined(CONFIG_SMP) && defined(CONFIG_PPC_STD_MMU_64)
static void crash_kexec_wait_realmode(int cpu)
{
	unsigned int msecs;
	int i;

	msecs = REAL_MODE_TIMEOUT;
	for (i=0; i < nr_cpu_ids && msecs > 0; i++) {
		if (i == cpu)
			continue;

		while (paca[i].kexec_state < KEXEC_STATE_REAL_MODE) {
			barrier();
			if (!cpu_possible(i) || !cpu_online(i) || (msecs <= 0))
				break;
			msecs--;
			mdelay(1);
		}
	}
	mb();
}
#else
static inline void crash_kexec_wait_realmode(int cpu) {}
#endif	/* CONFIG_SMP && CONFIG_PPC_STD_MMU_64 */

/*
 * Register a function to be called on shutdown.  Only use this if you
 * can't reset your device in the second kernel.
 */
int crash_shutdown_register(crash_shutdown_t handler)
{
	unsigned int i, rc;

	spin_lock(&crash_handlers_lock);
	for (i = 0 ; i < CRASH_HANDLER_MAX; i++)
		if (!crash_shutdown_handles[i]) {
			/* Insert handle at first empty entry */
			crash_shutdown_handles[i] = handler;
			rc = 0;
			break;
		}

	if (i == CRASH_HANDLER_MAX) {
		printk(KERN_ERR "Crash shutdown handles full, "
		       "not registered.\n");
		rc = 1;
	}

	spin_unlock(&crash_handlers_lock);
	return rc;
}
EXPORT_SYMBOL(crash_shutdown_register);

int crash_shutdown_unregister(crash_shutdown_t handler)
{
	unsigned int i, rc;

	spin_lock(&crash_handlers_lock);
	for (i = 0 ; i < CRASH_HANDLER_MAX; i++)
		if (crash_shutdown_handles[i] == handler)
			break;

	if (i == CRASH_HANDLER_MAX) {
		printk(KERN_ERR "Crash shutdown handle not found\n");
		rc = 1;
	} else {
		/* Shift handles down */
		for (; crash_shutdown_handles[i]; i++)
			crash_shutdown_handles[i] =
				crash_shutdown_handles[i+1];
		rc = 0;
	}

	spin_unlock(&crash_handlers_lock);
	return rc;
}
EXPORT_SYMBOL(crash_shutdown_unregister);

void default_machine_crash_shutdown(struct pt_regs *regs)
{
	unsigned int i;
	int (*old_handler)(struct pt_regs *regs);

	/*
	 * This function is only called after the system
	 * has panicked or is otherwise in a critical state.
	 * The minimum amount of code to allow a kexec'd kernel
	 * to run successfully needs to happen here.
	 *
	 * In practice this means stopping other cpus in
	 * an SMP system.
	 * The kernel is broken so disable interrupts.
	 */
	hard_irq_disable();

	/*
	 * Make a note of crashing cpu. Will be used in machine_kexec
	 * such that another IPI will not be sent.
	 */
	crashing_cpu = smp_processor_id();

	/*
	 * If we came in via system reset, wait a while for the secondary
	 * CPUs to enter.
	 */
	if (TRAP(regs) == 0x100)
		mdelay(PRIMARY_TIMEOUT);

	crash_kexec_prepare_cpus(crashing_cpu);

	crash_save_cpu(regs, crashing_cpu);

	time_to_dump = 1;

	crash_kexec_wait_realmode(crashing_cpu);

	machine_kexec_mask_interrupts();

	/*
	 * Call registered shutdown routines safely.  Swap out
	 * __debugger_fault_handler, and replace on exit.
	 */
	old_handler = __debugger_fault_handler;
	__debugger_fault_handler = handle_fault;
	crash_shutdown_cpu = smp_processor_id();
	for (i = 0; crash_shutdown_handles[i]; i++) {
		if (setjmp(crash_shutdown_buf) == 0) {
			/*
			 * Insert syncs and delay to ensure
			 * instructions in the dangerous region don't
			 * leak away from this protected region.
			 */
			asm volatile("sync; isync");
			/* dangerous region */
			crash_shutdown_handles[i]();
			asm volatile("sync; isync");
		}
	}
	crash_shutdown_cpu = -1;
	__debugger_fault_handler = old_handler;

	if (ppc_md.kexec_cpu_down)
		ppc_md.kexec_cpu_down(1, 0);
}