[mirror_ubuntu-bionic-kernel.git] / arch / powerpc / kernel / machine_kexec_64.c

/*
 * machine_kexec.c - handle transition of Linux booting another kernel
 *
 * Copyright (C) 2004-2005, IBM Corp.
 *
 * Created by: Milton D Miller II
 *
 * This source code is licensed under the GNU General Public License,
 * Version 2.  See the file COPYING for more details.
 */


#include <linux/cpumask.h>
#include <linux/kexec.h>
#include <linux/smp.h>
#include <linux/thread_info.h>
#include <linux/errno.h>

#include <asm/page.h>
#include <asm/current.h>
#include <asm/machdep.h>
#include <asm/cacheflush.h>
#include <asm/paca.h>
#include <asm/mmu.h>
#include <asm/sections.h>	/* _end */
#include <asm/prom.h>
#include <asm/smp.h>

#define HASH_GROUP_SIZE 0x80	/* size of each hash group, asm/mmu.h */

/* Have this around till we move it into crash specific file */
note_buf_t crash_notes[NR_CPUS];

/* Dummy for now. Not sure if we need to have a crash shutdown in here
 * and if what it will achieve. Letting it be now to compile the code
 * in generic kexec environment
 */
void machine_crash_shutdown(struct pt_regs *regs)
{
	/* do nothing right now */
	/* smp_relase_cpus() if we want smp on panic kernel */
	/* cpu_irq_down to isolate us until we are ready */
}

int machine_kexec_prepare(struct kimage *image)
{
	int i;
	unsigned long begin, end;	/* limits of segment */
	unsigned long low, high;	/* limits of blocked memory range */
	struct device_node *node;
	unsigned long *basep;
	unsigned int *sizep;

	if (!ppc_md.hpte_clear_all)
		return -ENOENT;

	/*
	 * Since we use the kernel fault handlers and paging code to
	 * handle the virtual mode, we must make sure no destination
	 * overlaps kernel static data or bss.
	 */
	for (i = 0; i < image->nr_segments; i++)
		if (image->segment[i].mem < __pa(_end))
			return -ETXTBSY;

	/*
	 * For non-LPAR, we absolutely can not overwrite the mmu hash
	 * table, since we are still using the bolted entries in it to
	 * do the copy.  Check that here.
	 *
	 * It is safe if the end is below the start of the blocked
	 * region (end <= low), or if the beginning is after the
	 * end of the blocked region (begin >= high).  Use the
	 * boolean identity !(a || b)  === (!a && !b).
	 */
	if (htab_address) {
		low = __pa(htab_address);
		high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE;

		for (i = 0; i < image->nr_segments; i++) {
			begin = image->segment[i].mem;
			end = begin + image->segment[i].memsz;

			if ((begin < high) && (end > low))
				return -ETXTBSY;
		}
	}

	/* We also should not overwrite the tce tables */
	for (node = of_find_node_by_type(NULL, "pci"); node != NULL;
			node = of_find_node_by_type(node, "pci")) {
		basep = (unsigned long *)get_property(node, "linux,tce-base",
							NULL);
		sizep = (unsigned int *)get_property(node, "linux,tce-size",
							NULL);
		if (basep == NULL || sizep == NULL)
			continue;

		low = *basep;
		high = low + (*sizep);

		for (i = 0; i < image->nr_segments; i++) {
			begin = image->segment[i].mem;
			end = begin + image->segment[i].memsz;

			if ((begin < high) && (end > low))
				return -ETXTBSY;
		}
	}

	return 0;
}

void machine_kexec_cleanup(struct kimage *image)
{
	/* we do nothing in prepare that needs to be undone */
}

#define IND_FLAGS (IND_DESTINATION | IND_INDIRECTION | IND_DONE | IND_SOURCE)

static void copy_segments(unsigned long ind)
{
	unsigned long entry;
	unsigned long *ptr;
	void *dest;
	void *addr;

	/*
	 * We rely on kexec_load to create a lists that properly
	 * initializes these pointers before they are used.
	 * We will still crash if the list is wrong, but at least
	 * the compiler will be quiet.
	 */
	ptr = NULL;
	dest = NULL;

	for (entry = ind; !(entry & IND_DONE); entry = *ptr++) {
		addr = __va(entry & PAGE_MASK);

		switch (entry & IND_FLAGS) {
		case IND_DESTINATION:
			dest = addr;
			break;
		case IND_INDIRECTION:
			ptr = addr;
			break;
		case IND_SOURCE:
			copy_page(dest, addr);
			dest += PAGE_SIZE;
		}
	}
}

void kexec_copy_flush(struct kimage *image)
{
	long i, nr_segments = image->nr_segments;
	struct  kexec_segment ranges[KEXEC_SEGMENT_MAX];

	/* save the ranges on the stack to efficiently flush the icache */
	memcpy(ranges, image->segment, sizeof(ranges));

	/*
	 * After this call we may not use anything allocated in dynamic
	 * memory, including *image.
	 *
	 * Only globals and the stack are allowed.
	 */
	copy_segments(image->head);

	/*
	 * we need to clear the icache for all dest pages sometime,
	 * including ones that were in place on the original copy
	 */
	for (i = 0; i < nr_segments; i++)
		flush_icache_range(ranges[i].mem + KERNELBASE,
				ranges[i].mem + KERNELBASE +
				ranges[i].memsz);
}

#ifdef CONFIG_SMP

/* FIXME: we should schedule this function to be called on all cpus based
 * on calling the interrupts, but we would like to call it off irq level
 * so that the interrupt controller is clean.
 */
void kexec_smp_down(void *arg)
{
	if (ppc_md.kexec_cpu_down)
		ppc_md.kexec_cpu_down(0, 1);

	local_irq_disable();
	kexec_smp_wait();
	/* NOTREACHED */
}

static void kexec_prepare_cpus(void)
{
	int my_cpu, i, notified=-1;

	smp_call_function(kexec_smp_down, NULL, 0, /* wait */0);
	my_cpu = get_cpu();

	/* check the others cpus are now down (via paca hw cpu id == -1) */
	for (i=0; i < NR_CPUS; i++) {
		if (i == my_cpu)
			continue;

		while (paca[i].hw_cpu_id != -1) {
			barrier();
			if (!cpu_possible(i)) {
				printk("kexec: cpu %d hw_cpu_id %d is not"
						" possible, ignoring\n",
						i, paca[i].hw_cpu_id);
				break;
			}
			if (!cpu_online(i)) {
				/* Fixme: this can be spinning in
				 * pSeries_secondary_wait with a paca
				 * waiting for it to go online.
				 */
				printk("kexec: cpu %d hw_cpu_id %d is not"
						" online, ignoring\n",
						i, paca[i].hw_cpu_id);
				break;
			}
			if (i != notified) {
				printk( "kexec: waiting for cpu %d (physical"
						" %d) to go down\n",
						i, paca[i].hw_cpu_id);
				notified = i;
			}
		}
	}

	/* after we tell the others to go down */
	if (ppc_md.kexec_cpu_down)
		ppc_md.kexec_cpu_down(0, 0);

	put_cpu();

	local_irq_disable();
}

#else /* ! SMP */

static void kexec_prepare_cpus(void)
{
	/*
	 * move the secondarys to us so that we can copy
	 * the new kernel 0-0x100 safely
	 *
	 * do this if kexec in setup.c ?
	 *
	 * We need to release the cpus if we are ever going from an
	 * UP to an SMP kernel.
	 */
	smp_release_cpus();
	if (ppc_md.kexec_cpu_down)
		ppc_md.kexec_cpu_down(0, 0);
	local_irq_disable();
}

#endif /* SMP */

/*
 * kexec thread structure and stack.
 *
 * We need to make sure that this is 16384-byte aligned due to the
 * way process stacks are handled.  It also must be statically allocated
 * or allocated as part of the kimage, because everything else may be
 * overwritten when we copy the kexec image.  We piggyback on the
 * "init_task" linker section here to statically allocate a stack.
 *
 * We could use a smaller stack if we don't care about anything using
 * current, but that audit has not been performed.
 */
union thread_union kexec_stack
	__attribute__((__section__(".data.init_task"))) = { };

/* Our assembly helper, in kexec_stub.S */
extern NORET_TYPE void kexec_sequence(void *newstack, unsigned long start,
					void *image, void *control,
					void (*clear_all)(void)) ATTRIB_NORET;

/* too late to fail here */
void machine_kexec(struct kimage *image)
{

	/* prepare control code if any */

	/* shutdown other cpus into our wait loop and quiesce interrupts */
	kexec_prepare_cpus();

	/* switch to a staticly allocated stack.  Based on irq stack code.
	 * XXX: the task struct will likely be invalid once we do the copy!
	 */
	kexec_stack.thread_info.task = current_thread_info()->task;
	kexec_stack.thread_info.flags = 0;

	/* Some things are best done in assembly.  Finding globals with
	 * a toc is easier in C, so pass in what we can.
	 */
	kexec_sequence(&kexec_stack, image->start, image,
			page_address(image->control_code_page),
			ppc_md.hpte_clear_all);
	/* NOTREACHED */
}

/* Values we need to export to the second kernel via the device tree. */
static unsigned long htab_base, htab_size, kernel_end;

static struct property htab_base_prop = {
	.name = "linux,htab-base",
	.length = sizeof(unsigned long),
	.value = (unsigned char *)&htab_base,
};

static struct property htab_size_prop = {
	.name = "linux,htab-size",
	.length = sizeof(unsigned long),
	.value = (unsigned char *)&htab_size,
};

static struct property kernel_end_prop = {
	.name = "linux,kernel-end",
	.length = sizeof(unsigned long),
	.value = (unsigned char *)&kernel_end,
};

static void __init export_htab_values(void)
{
	struct device_node *node;

	node = of_find_node_by_path("/chosen");
	if (!node)
		return;

	kernel_end = __pa(_end);
	prom_add_property(node, &kernel_end_prop);

	/* On machines with no htab htab_address is NULL */
	if (NULL == htab_address)
		goto out;

	htab_base = __pa(htab_address);
	prom_add_property(node, &htab_base_prop);

	htab_size = 1UL << ppc64_pft_size;
	prom_add_property(node, &htab_size_prop);

 out:
	of_node_put(node);
}

void __init kexec_setup(void)
{
	export_htab_values();
}
Commit	Line	Data
fce0d574 S	1	/*
	2	* machine_kexec.c - handle transition of Linux booting another kernel
	3	*
	4	* Copyright (C) 2004-2005, IBM Corp.
	5	*
	6	* Created by: Milton D Miller II
	7	*
	8	* This source code is licensed under the GNU General Public License,
	9	* Version 2. See the file COPYING for more details.
	10	*/
	11
	12
	13	#include <linux/cpumask.h>
	14	#include <linux/kexec.h>
	15	#include <linux/smp.h>
	16	#include <linux/thread_info.h>
	17	#include <linux/errno.h>
	18
	19	#include <asm/page.h>
	20	#include <asm/current.h>
	21	#include <asm/machdep.h>
	22	#include <asm/cacheflush.h>
	23	#include <asm/paca.h>
	24	#include <asm/mmu.h>
	25	#include <asm/sections.h> /* _end */
	26	#include <asm/prom.h>
2249ca9d	27	#include <asm/smp.h>
fce0d574 S	28
	29	#define HASH_GROUP_SIZE 0x80 /* size of each hash group, asm/mmu.h */
	30
	31	/* Have this around till we move it into crash specific file */
	32	note_buf_t crash_notes[NR_CPUS];
	33
	34	/* Dummy for now. Not sure if we need to have a crash shutdown in here
	35	* and if what it will achieve. Letting it be now to compile the code
	36	* in generic kexec environment
	37	*/
6e274d14	38	void machine_crash_shutdown(struct pt_regs *regs)
fce0d574 S	39	{
	40	/* do nothing right now */
	41	/* smp_relase_cpus() if we want smp on panic kernel */
	42	/* cpu_irq_down to isolate us until we are ready */
	43	}
	44
	45	int machine_kexec_prepare(struct kimage *image)
	46	{
	47	int i;
	48	unsigned long begin, end; /* limits of segment */
	49	unsigned long low, high; /* limits of blocked memory range */
	50	struct device_node *node;
	51	unsigned long *basep;
	52	unsigned int *sizep;
	53
	54	if (!ppc_md.hpte_clear_all)
	55	return -ENOENT;
	56
	57	/*
	58	* Since we use the kernel fault handlers and paging code to
	59	* handle the virtual mode, we must make sure no destination
	60	* overlaps kernel static data or bss.
	61	*/
72414d3f	62	for (i = 0; i < image->nr_segments; i++)
fce0d574 S	63	if (image->segment[i].mem < __pa(_end))
	64	return -ETXTBSY;
	65
	66	/*
	67	* For non-LPAR, we absolutely can not overwrite the mmu hash
	68	* table, since we are still using the bolted entries in it to
	69	* do the copy. Check that here.
	70	*
	71	* It is safe if the end is below the start of the blocked
	72	* region (end <= low), or if the beginning is after the
	73	* end of the blocked region (begin >= high). Use the
	74	* boolean identity !(a \|\| b) === (!a && !b).
	75	*/
	76	if (htab_address) {
	77	low = __pa(htab_address);
	78	high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE;
	79
72414d3f	80	for (i = 0; i < image->nr_segments; i++) {
fce0d574 S	81	begin = image->segment[i].mem;
	82	end = begin + image->segment[i].memsz;
	83
	84	if ((begin < high) && (end > low))
	85	return -ETXTBSY;
	86	}
	87	}
	88
	89	/* We also should not overwrite the tce tables */
	90	for (node = of_find_node_by_type(NULL, "pci"); node != NULL;
	91	node = of_find_node_by_type(node, "pci")) {
	92	basep = (unsigned long *)get_property(node, "linux,tce-base",
	93	NULL);
	94	sizep = (unsigned int *)get_property(node, "linux,tce-size",
	95	NULL);
	96	if (basep == NULL \|\| sizep == NULL)
	97	continue;
	98
	99	low = *basep;
	100	high = low + (*sizep);
	101
72414d3f	102	for (i = 0; i < image->nr_segments; i++) {
fce0d574 S	103	begin = image->segment[i].mem;
	104	end = begin + image->segment[i].memsz;
	105
	106	if ((begin < high) && (end > low))
	107	return -ETXTBSY;
	108	}
	109	}
	110
	111	return 0;
	112	}
	113
	114	void machine_kexec_cleanup(struct kimage *image)
	115	{
	116	/* we do nothing in prepare that needs to be undone */
	117	}
	118
	119	#define IND_FLAGS (IND_DESTINATION \| IND_INDIRECTION \| IND_DONE \| IND_SOURCE)
	120
	121	static void copy_segments(unsigned long ind)
	122	{
	123	unsigned long entry;
	124	unsigned long *ptr;
	125	void *dest;
	126	void *addr;
	127
	128	/*
	129	* We rely on kexec_load to create a lists that properly
	130	* initializes these pointers before they are used.
	131	* We will still crash if the list is wrong, but at least
	132	* the compiler will be quiet.
	133	*/
	134	ptr = NULL;
	135	dest = NULL;
	136
	137	for (entry = ind; !(entry & IND_DONE); entry = *ptr++) {
	138	addr = __va(entry & PAGE_MASK);
	139
	140	switch (entry & IND_FLAGS) {
	141	case IND_DESTINATION:
	142	dest = addr;
	143	break;
	144	case IND_INDIRECTION:
	145	ptr = addr;
	146	break;
	147	case IND_SOURCE:
	148	copy_page(dest, addr);
	149	dest += PAGE_SIZE;
	150	}
	151	}
	152	}
	153
	154	void kexec_copy_flush(struct kimage *image)
	155	{
	156	long i, nr_segments = image->nr_segments;
	157	struct kexec_segment ranges[KEXEC_SEGMENT_MAX];
	158
	159	/* save the ranges on the stack to efficiently flush the icache */
	160	memcpy(ranges, image->segment, sizeof(ranges));
	161
	162	/*
	163	* After this call we may not use anything allocated in dynamic
	164	* memory, including *image.
	165	*
	166	* Only globals and the stack are allowed.
167	*/
168	copy_segments(image->head);
169
170	/*
171	* we need to clear the icache for all dest pages sometime,
172	* including ones that were in place on the original copy
173	*/
174	for (i = 0; i < nr_segments; i++)
175	flush_icache_range(ranges[i].mem + KERNELBASE,
176	ranges[i].mem + KERNELBASE +
177	ranges[i].memsz);
178	}
179
180	#ifdef CONFIG_SMP
181
182	/* FIXME: we should schedule this function to be called on all cpus based
183	* on calling the interrupts, but we would like to call it off irq level
184	* so that the interrupt controller is clean.
185	*/
186	void kexec_smp_down(void *arg)
187	{
c5e24354 ME	188	if (ppc_md.kexec_cpu_down)
c5e24354 ME	189	ppc_md.kexec_cpu_down(0, 1);
fce0d574 S	190
	191	local_irq_disable();
	192	kexec_smp_wait();
	193	/* NOTREACHED */
	194	}
	195
	196	static void kexec_prepare_cpus(void)
	197	{
	198	int my_cpu, i, notified=-1;
	199
	200	smp_call_function(kexec_smp_down, NULL, 0, /* wait */0);
	201	my_cpu = get_cpu();
	202
	203	/* check the others cpus are now down (via paca hw cpu id == -1) */
	204	for (i=0; i < NR_CPUS; i++) {
	205	if (i == my_cpu)
	206	continue;
	207
	208	while (paca[i].hw_cpu_id != -1) {
b3ca8093	209	barrier();
fce0d574 S	210	if (!cpu_possible(i)) {
	211	printk("kexec: cpu %d hw_cpu_id %d is not"
	212	" possible, ignoring\n",
	213	i, paca[i].hw_cpu_id);
	214	break;
	215	}
	216	if (!cpu_online(i)) {
	217	/* Fixme: this can be spinning in
	218	* pSeries_secondary_wait with a paca
	219	* waiting for it to go online.
	220	*/
	221	printk("kexec: cpu %d hw_cpu_id %d is not"
	222	" online, ignoring\n",
	223	i, paca[i].hw_cpu_id);
	224	break;
	225	}
	226	if (i != notified) {
	227	printk( "kexec: waiting for cpu %d (physical"
	228	" %d) to go down\n",
	229	i, paca[i].hw_cpu_id);
	230	notified = i;
	231	}
	232	}
	233	}
	234
	235	/* after we tell the others to go down */
c5e24354 ME	236	if (ppc_md.kexec_cpu_down)
c5e24354 ME	237	ppc_md.kexec_cpu_down(0, 0);
fce0d574 S	238
	239	put_cpu();
	240
	241	local_irq_disable();
	242	}
	243
	244	#else /* ! SMP */
	245
	246	static void kexec_prepare_cpus(void)
	247	{
	248	/*
	249	* move the secondarys to us so that we can copy
	250	* the new kernel 0-0x100 safely
	251	*
	252	* do this if kexec in setup.c ?
75eedfed OJ	253	*
	254	* We need to release the cpus if we are ever going from an
	255	* UP to an SMP kernel.
fce0d574	256	*/
75eedfed	257	smp_release_cpus();
c5e24354 ME	258	if (ppc_md.kexec_cpu_down)
c5e24354 ME	259	ppc_md.kexec_cpu_down(0, 0);
fce0d574 S	260	local_irq_disable();
	261	}
	262
	263	#endif /* SMP */
	264
	265	/*
	266	* kexec thread structure and stack.
	267	*
	268	* We need to make sure that this is 16384-byte aligned due to the
	269	* way process stacks are handled. It also must be statically allocated
	270	* or allocated as part of the kimage, because everything else may be
	271	* overwritten when we copy the kexec image. We piggyback on the
	272	* "init_task" linker section here to statically allocate a stack.
	273	*
	274	* We could use a smaller stack if we don't care about anything using
	275	* current, but that audit has not been performed.
	276	*/
	277	union thread_union kexec_stack
	278	__attribute__((__section__(".data.init_task"))) = { };
	279
	280	/* Our assembly helper, in kexec_stub.S */
	281	extern NORET_TYPE void kexec_sequence(void *newstack, unsigned long start,
72414d3f MS	282	void image, void control,
72414d3f MS	283	void (*clear_all)(void)) ATTRIB_NORET;
fce0d574 S	284
	285	/* too late to fail here */
	286	void machine_kexec(struct kimage *image)
	287	{
	288
	289	/* prepare control code if any */
	290
	291	/* shutdown other cpus into our wait loop and quiesce interrupts */
	292	kexec_prepare_cpus();
	293
	294	/* switch to a staticly allocated stack. Based on irq stack code.
	295	* XXX: the task struct will likely be invalid once we do the copy!
	296	*/
	297	kexec_stack.thread_info.task = current_thread_info()->task;
	298	kexec_stack.thread_info.flags = 0;
	299
	300	/* Some things are best done in assembly. Finding globals with
	301	* a toc is easier in C, so pass in what we can.
	302	*/
	303	kexec_sequence(&kexec_stack, image->start, image,
	304	page_address(image->control_code_page),
	305	ppc_md.hpte_clear_all);
	306	/* NOTREACHED */
	307	}
593e537b ME	308
	309	/* Values we need to export to the second kernel via the device tree. */
	310	static unsigned long htab_base, htab_size, kernel_end;
	311
	312	static struct property htab_base_prop = {
	313	.name = "linux,htab-base",
	314	.length = sizeof(unsigned long),
	315	.value = (unsigned char *)&htab_base,
	316	};
	317
	318	static struct property htab_size_prop = {
	319	.name = "linux,htab-size",
	320	.length = sizeof(unsigned long),
	321	.value = (unsigned char *)&htab_size,
	322	};
	323
	324	static struct property kernel_end_prop = {
	325	.name = "linux,kernel-end",
	326	.length = sizeof(unsigned long),
	327	.value = (unsigned char *)&kernel_end,
	328	};
	329
	330	static void __init export_htab_values(void)
	331	{
	332	struct device_node *node;
	333
	334	node = of_find_node_by_path("/chosen");
	335	if (!node)
	336	return;
	337
	338	kernel_end = __pa(_end);
	339	prom_add_property(node, &kernel_end_prop);
	340
	341	/* On machines with no htab htab_address is NULL */
	342	if (NULL == htab_address)
	343	goto out;
	344
	345	htab_base = __pa(htab_address);
	346	prom_add_property(node, &htab_base_prop);
	347
	348	htab_size = 1UL << ppc64_pft_size;
	349	prom_add_property(node, &htab_size_prop);
	350
	351	out:
	352	of_node_put(node);
	353	}
	354
	355	void __init kexec_setup(void)
	356	{
	357	export_htab_values();
	358	}