summaryrefslogtreecommitdiffstats
path: root/arch/s390/kernel/machine_kexec.c
blob: cb8b1cc285c9612d695154e1bd3d89fd259ebf98 (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
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright IBM Corp. 2005, 2011
 *
 * Author(s): Rolf Adelsberger,
 *	      Heiko Carstens <heiko.carstens@de.ibm.com>
 *	      Michael Holzheu <holzheu@linux.vnet.ibm.com>
 */

#include <linux/device.h>
#include <linux/mm.h>
#include <linux/kexec.h>
#include <linux/delay.h>
#include <linux/reboot.h>
#include <linux/ftrace.h>
#include <linux/debug_locks.h>
#include <linux/suspend.h>
#include <asm/cio.h>
#include <asm/setup.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/smp.h>
#include <asm/ipl.h>
#include <asm/diag.h>
#include <asm/elf.h>
#include <asm/asm-offsets.h>
#include <asm/cacheflush.h>
#include <asm/os_info.h>
#include <asm/set_memory.h>
#include <asm/stacktrace.h>
#include <asm/switch_to.h>
#include <asm/nmi.h>

typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long);

extern const unsigned char relocate_kernel[];
extern const unsigned long long relocate_kernel_len;

#ifdef CONFIG_CRASH_DUMP

/*
 * PM notifier callback for kdump
 */
static int machine_kdump_pm_cb(struct notifier_block *nb, unsigned long action,
			       void *ptr)
{
	switch (action) {
	case PM_SUSPEND_PREPARE:
	case PM_HIBERNATION_PREPARE:
		if (kexec_crash_image)
			arch_kexec_unprotect_crashkres();
		break;
	case PM_POST_SUSPEND:
	case PM_POST_HIBERNATION:
		if (kexec_crash_image)
			arch_kexec_protect_crashkres();
		break;
	default:
		return NOTIFY_DONE;
	}
	return NOTIFY_OK;
}

static int __init machine_kdump_pm_init(void)
{
	pm_notifier(machine_kdump_pm_cb, 0);
	return 0;
}
arch_initcall(machine_kdump_pm_init);

/*
 * Reset the system, copy boot CPU registers to absolute zero,
 * and jump to the kdump image
 */
static void __do_machine_kdump(void *image)
{
	int (*start_kdump)(int);
	unsigned long prefix;

	/* store_status() saved the prefix register to lowcore */
	prefix = (unsigned long) S390_lowcore.prefixreg_save_area;

	/* Now do the reset  */
	s390_reset_system();

	/*
	 * Copy dump CPU store status info to absolute zero.
	 * This need to be done *after* s390_reset_system set the
	 * prefix register of this CPU to zero
	 */
	memcpy((void *) __LC_FPREGS_SAVE_AREA,
	       (void *)(prefix + __LC_FPREGS_SAVE_AREA), 512);

	__load_psw_mask(PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA);
	start_kdump = (void *)((struct kimage *) image)->start;
	start_kdump(1);

	/* Die if start_kdump returns */
	disabled_wait();
}

/*
 * Start kdump: create a LGR log entry, store status of all CPUs and
 * branch to __do_machine_kdump.
 */
static noinline void __machine_kdump(void *image)
{
	struct mcesa *mcesa;
	union ctlreg2 cr2_old, cr2_new;
	int this_cpu, cpu;

	lgr_info_log();
	/* Get status of the other CPUs */
	this_cpu = smp_find_processor_id(stap());
	for_each_online_cpu(cpu) {
		if (cpu == this_cpu)
			continue;
		if (smp_store_status(cpu))
			continue;
	}
	/* Store status of the boot CPU */
	mcesa = (struct mcesa *)(S390_lowcore.mcesad & MCESA_ORIGIN_MASK);
	if (MACHINE_HAS_VX)
		save_vx_regs((__vector128 *) mcesa->vector_save_area);
	if (MACHINE_HAS_GS) {
		__ctl_store(cr2_old.val, 2, 2);
		cr2_new = cr2_old;
		cr2_new.gse = 1;
		__ctl_load(cr2_new.val, 2, 2);
		save_gs_cb((struct gs_cb *) mcesa->guarded_storage_save_area);
		__ctl_load(cr2_old.val, 2, 2);
	}
	/*
	 * To create a good backchain for this CPU in the dump store_status
	 * is passed the address of a function. The address is saved into
	 * the PSW save area of the boot CPU and the function is invoked as
	 * a tail call of store_status. The backchain in the dump will look
	 * like this:
	 *   restart_int_handler ->  __machine_kexec -> __do_machine_kdump
	 * The call to store_status() will not return.
	 */
	store_status(__do_machine_kdump, image);
}

static unsigned long do_start_kdump(unsigned long addr)
{
	struct kimage *image = (struct kimage *) addr;
	int (*start_kdump)(int) = (void *)image->start;
	int rc;

	__arch_local_irq_stnsm(0xfb); /* disable DAT */
	rc = start_kdump(0);
	__arch_local_irq_stosm(0x04); /* enable DAT */
	return rc;
}

#endif /* CONFIG_CRASH_DUMP */

/*
 * Check if kdump checksums are valid: We call purgatory with parameter "0"
 */
static bool kdump_csum_valid(struct kimage *image)
{
#ifdef CONFIG_CRASH_DUMP
	int rc;

	preempt_disable();
	rc = CALL_ON_STACK(do_start_kdump, S390_lowcore.nodat_stack, 1, image);
	preempt_enable();
	return rc == 0;
#else
	return false;
#endif
}

#ifdef CONFIG_CRASH_DUMP

void crash_free_reserved_phys_range(unsigned long begin, unsigned long end)
{
	unsigned long addr, size;

	for (addr = begin; addr < end; addr += PAGE_SIZE)
		free_reserved_page(pfn_to_page(addr >> PAGE_SHIFT));
	size = begin - crashk_res.start;
	if (size)
		os_info_crashkernel_add(crashk_res.start, size);
	else
		os_info_crashkernel_add(0, 0);
}

static void crash_protect_pages(int protect)
{
	unsigned long size;

	if (!crashk_res.end)
		return;
	size = resource_size(&crashk_res);
	if (protect)
		set_memory_ro(crashk_res.start, size >> PAGE_SHIFT);
	else
		set_memory_rw(crashk_res.start, size >> PAGE_SHIFT);
}

void arch_kexec_protect_crashkres(void)
{
	crash_protect_pages(1);
}

void arch_kexec_unprotect_crashkres(void)
{
	crash_protect_pages(0);
}

#endif

/*
 * Give back memory to hypervisor before new kdump is loaded
 */
static int machine_kexec_prepare_kdump(void)
{
#ifdef CONFIG_CRASH_DUMP
	if (MACHINE_IS_VM)
		diag10_range(PFN_DOWN(crashk_res.start),
			     PFN_DOWN(crashk_res.end - crashk_res.start + 1));
	return 0;
#else
	return -EINVAL;
#endif
}

int machine_kexec_prepare(struct kimage *image)
{
	void *reboot_code_buffer;

	if (image->type == KEXEC_TYPE_CRASH)
		return machine_kexec_prepare_kdump();

	/* We don't support anything but the default image type for now. */
	if (image->type != KEXEC_TYPE_DEFAULT)
		return -EINVAL;

	/* Get the destination where the assembler code should be copied to.*/
	reboot_code_buffer = (void *) page_to_phys(image->control_code_page);

	/* Then copy it */
	memcpy(reboot_code_buffer, relocate_kernel, relocate_kernel_len);
	return 0;
}

void machine_kexec_cleanup(struct kimage *image)
{
}

void arch_crash_save_vmcoreinfo(void)
{
	VMCOREINFO_SYMBOL(lowcore_ptr);
	VMCOREINFO_SYMBOL(high_memory);
	VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS);
	vmcoreinfo_append_str("SDMA=%lx\n", __sdma);
	vmcoreinfo_append_str("EDMA=%lx\n", __edma);
	vmcoreinfo_append_str("KERNELOFFSET=%lx\n", kaslr_offset());
	mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note());
}

void machine_shutdown(void)
{
}

void machine_crash_shutdown(struct pt_regs *regs)
{
	set_os_info_reipl_block();
}

/*
 * Do normal kexec
 */
static void __do_machine_kexec(void *data)
{
	relocate_kernel_t data_mover;
	struct kimage *image = data;

	s390_reset_system();
	data_mover = (relocate_kernel_t) page_to_phys(image->control_code_page);

	__arch_local_irq_stnsm(0xfb); /* disable DAT - avoid no-execute */
	/* Call the moving routine */
	(*data_mover)(&image->head, image->start);

	/* Die if kexec returns */
	disabled_wait();
}

/*
 * Reset system and call either kdump or normal kexec
 */
static void __machine_kexec(void *data)
{
	__arch_local_irq_stosm(0x04); /* enable DAT */
	pfault_fini();
	tracing_off();
	debug_locks_off();
#ifdef CONFIG_CRASH_DUMP
	if (((struct kimage *) data)->type == KEXEC_TYPE_CRASH)
		__machine_kdump(data);
#endif
	__do_machine_kexec(data);
}

/*
 * Do either kdump or normal kexec. In case of kdump we first ask
 * purgatory, if kdump checksums are valid.
 */
void machine_kexec(struct kimage *image)
{
	if (image->type == KEXEC_TYPE_CRASH && !kdump_csum_valid(image))
		return;
	tracer_disable();
	smp_send_stop();
	smp_call_ipl_cpu(__machine_kexec, image);
}