machine_kexec_64.c 7.71 KB
Newer Older
1
/*
Dave Jones's avatar
Dave Jones committed
2
 * handle transition of Linux booting another kernel
3 4 5 6 7 8 9 10 11
 * Copyright (C) 2002-2005 Eric Biederman  <ebiederm@xmission.com>
 *
 * This source code is licensed under the GNU General Public License,
 * Version 2.  See the file COPYING for more details.
 */

#include <linux/mm.h>
#include <linux/kexec.h>
#include <linux/string.h>
12
#include <linux/gfp.h>
13
#include <linux/reboot.h>
Ken'ichi Ohmichi's avatar
Ken'ichi Ohmichi committed
14
#include <linux/numa.h>
Ingo Molnar's avatar
Ingo Molnar committed
15
#include <linux/ftrace.h>
16
#include <linux/io.h>
17
#include <linux/suspend.h>
Ingo Molnar's avatar
Ingo Molnar committed
18

19
#include <asm/init.h>
20 21 22
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>
23
#include <asm/debugreg.h>
24

25 26 27 28
static struct kexec_file_ops *kexec_file_loaders[] = {
		NULL,
};

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
static void free_transition_pgtable(struct kimage *image)
{
	free_page((unsigned long)image->arch.pud);
	free_page((unsigned long)image->arch.pmd);
	free_page((unsigned long)image->arch.pte);
}

static int init_transition_pgtable(struct kimage *image, pgd_t *pgd)
{
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
	unsigned long vaddr, paddr;
	int result = -ENOMEM;

	vaddr = (unsigned long)relocate_kernel;
	paddr = __pa(page_address(image->control_code_page)+PAGE_SIZE);
	pgd += pgd_index(vaddr);
	if (!pgd_present(*pgd)) {
		pud = (pud_t *)get_zeroed_page(GFP_KERNEL);
		if (!pud)
			goto err;
		image->arch.pud = pud;
		set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE));
	}
	pud = pud_offset(pgd, vaddr);
	if (!pud_present(*pud)) {
		pmd = (pmd_t *)get_zeroed_page(GFP_KERNEL);
		if (!pmd)
			goto err;
		image->arch.pmd = pmd;
		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE));
	}
	pmd = pmd_offset(pud, vaddr);
	if (!pmd_present(*pmd)) {
		pte = (pte_t *)get_zeroed_page(GFP_KERNEL);
		if (!pte)
			goto err;
		image->arch.pte = pte;
		set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE));
	}
	pte = pte_offset_kernel(pmd, vaddr);
	set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC));
	return 0;
err:
	free_transition_pgtable(image);
	return result;
}

78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
static void *alloc_pgt_page(void *data)
{
	struct kimage *image = (struct kimage *)data;
	struct page *page;
	void *p = NULL;

	page = kimage_alloc_control_pages(image, 0);
	if (page) {
		p = page_address(page);
		clear_page(p);
	}

	return p;
}

93 94
static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
{
95 96 97 98 99
	struct x86_mapping_info info = {
		.alloc_pgt_page	= alloc_pgt_page,
		.context	= image,
		.pmd_flag	= __PAGE_KERNEL_LARGE_EXEC,
	};
100
	unsigned long mstart, mend;
101
	pgd_t *level4p;
102
	int result;
103 104
	int i;

105
	level4p = (pgd_t *)__va(start_pgtable);
106
	clear_page(level4p);
107 108 109 110 111 112 113 114 115
	for (i = 0; i < nr_pfn_mapped; i++) {
		mstart = pfn_mapped[i].start << PAGE_SHIFT;
		mend   = pfn_mapped[i].end << PAGE_SHIFT;

		result = kernel_ident_mapping_init(&info,
						 level4p, mstart, mend);
		if (result)
			return result;
	}
116

117
	/*
118 119 120 121
	 * segments's mem ranges could be outside 0 ~ max_pfn,
	 * for example when jump back to original kernel from kexeced kernel.
	 * or first kernel is booted with user mem map, and second kernel
	 * could be loaded out of that range.
122
	 */
123 124 125 126
	for (i = 0; i < image->nr_segments; i++) {
		mstart = image->segment[i].mem;
		mend   = mstart + image->segment[i].memsz;

127 128
		result = kernel_ident_mapping_init(&info,
						 level4p, mstart, mend);
129 130 131 132 133

		if (result)
			return result;
	}

134
	return init_transition_pgtable(image, level4p);
135 136 137 138
}

static void set_idt(void *newidt, u16 limit)
{
139
	struct desc_ptr curidt;
140 141

	/* x86-64 supports unaliged loads & stores */
142 143
	curidt.size    = limit;
	curidt.address = (unsigned long)newidt;
144 145

	__asm__ __volatile__ (
146 147
		"lidtq %0\n"
		: : "m" (curidt)
148 149 150 151 152 153
		);
};


static void set_gdt(void *newgdt, u16 limit)
{
154
	struct desc_ptr curgdt;
155 156

	/* x86-64 supports unaligned loads & stores */
157 158
	curgdt.size    = limit;
	curgdt.address = (unsigned long)newgdt;
159 160

	__asm__ __volatile__ (
161 162
		"lgdtq %0\n"
		: : "m" (curgdt)
163 164 165 166 167 168
		);
};

static void load_segments(void)
{
	__asm__ __volatile__ (
169 170 171 172 173
		"\tmovl %0,%%ds\n"
		"\tmovl %0,%%es\n"
		"\tmovl %0,%%ss\n"
		"\tmovl %0,%%fs\n"
		"\tmovl %0,%%gs\n"
Michael Matz's avatar
Michael Matz committed
174
		: : "a" (__KERNEL_DS) : "memory"
175 176 177 178 179
		);
}

int machine_kexec_prepare(struct kimage *image)
{
180
	unsigned long start_pgtable;
181 182 183
	int result;

	/* Calculate the offsets */
Maneesh Soni's avatar
Maneesh Soni committed
184
	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
185 186 187

	/* Setup the identity mapped 64bit page table */
	result = init_pgtable(image, start_pgtable);
Maneesh Soni's avatar
Maneesh Soni committed
188
	if (result)
189 190 191 192 193 194 195
		return result;

	return 0;
}

void machine_kexec_cleanup(struct kimage *image)
{
196
	free_transition_pgtable(image);
197 198 199 200 201 202
}

/*
 * Do not allocate memory (or fail in any way) in machine_kexec().
 * We are past the point of no return, committed to rebooting now.
 */
Huang Ying's avatar
Huang Ying committed
203
void machine_kexec(struct kimage *image)
204
{
205 206
	unsigned long page_list[PAGES_NR];
	void *control_page;
207
	int save_ftrace_enabled;
208

209
#ifdef CONFIG_KEXEC_JUMP
210
	if (image->preserve_context)
211 212 213 214
		save_processor_state();
#endif

	save_ftrace_enabled = __ftrace_enabled_save();
Ingo Molnar's avatar
Ingo Molnar committed
215

216 217
	/* Interrupts aren't acceptable while we reboot */
	local_irq_disable();
218
	hw_breakpoint_disable();
219

220 221 222 223 224 225 226 227 228 229 230 231 232
	if (image->preserve_context) {
#ifdef CONFIG_X86_IO_APIC
		/*
		 * We need to put APICs in legacy mode so that we can
		 * get timer interrupts in second kernel. kexec/kdump
		 * paths already have calls to disable_IO_APIC() in
		 * one form or other. kexec jump path also need
		 * one.
		 */
		disable_IO_APIC();
#endif
	}

233
	control_page = page_address(image->control_code_page) + PAGE_SIZE;
234
	memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE);
235

236
	page_list[PA_CONTROL_PAGE] = virt_to_phys(control_page);
237
	page_list[VA_CONTROL_PAGE] = (unsigned long)control_page;
238 239
	page_list[PA_TABLE_PAGE] =
	  (unsigned long)__pa(page_address(image->control_code_page));
240

241 242 243 244
	if (image->type == KEXEC_TYPE_DEFAULT)
		page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page)
						<< PAGE_SHIFT);

245 246
	/*
	 * The segment registers are funny things, they have both a
247 248 249 250
	 * visible and an invisible part.  Whenever the visible part is
	 * set to a specific selector, the invisible part is loaded
	 * with from a table in memory.  At no other time is the
	 * descriptor table in memory accessed.
251 252 253 254 255
	 *
	 * I take advantage of this here by force loading the
	 * segments, before I zap the gdt with an invalid value.
	 */
	load_segments();
256 257
	/*
	 * The gdt & idt are now invalid.
258 259
	 * If you want to load them you must set up your own idt & gdt.
	 */
260 261
	set_gdt(phys_to_virt(0), 0);
	set_idt(phys_to_virt(0), 0);
262

263
	/* now call it */
264 265 266 267 268 269
	image->start = relocate_kernel((unsigned long)image->head,
				       (unsigned long)page_list,
				       image->start,
				       image->preserve_context);

#ifdef CONFIG_KEXEC_JUMP
270
	if (image->preserve_context)
271 272 273 274
		restore_processor_state();
#endif

	__ftrace_enabled_restore(save_ftrace_enabled);
275
}
276

Ken'ichi Ohmichi's avatar
Ken'ichi Ohmichi committed
277 278
void arch_crash_save_vmcoreinfo(void)
{
279
	VMCOREINFO_SYMBOL(phys_base);
280
	VMCOREINFO_SYMBOL(init_level4_pgt);
281 282 283 284 285

#ifdef CONFIG_NUMA
	VMCOREINFO_SYMBOL(node_data);
	VMCOREINFO_LENGTH(node_data, MAX_NUMNODES);
#endif
286 287
	vmcoreinfo_append_str("KERNELOFFSET=%lx\n",
			      (unsigned long)&_text - __START_KERNEL);
Ken'ichi Ohmichi's avatar
Ken'ichi Ohmichi committed
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
/* arch-dependent functionality related to kexec file-based syscall */

int arch_kexec_kernel_image_probe(struct kimage *image, void *buf,
				  unsigned long buf_len)
{
	int i, ret = -ENOEXEC;
	struct kexec_file_ops *fops;

	for (i = 0; i < ARRAY_SIZE(kexec_file_loaders); i++) {
		fops = kexec_file_loaders[i];
		if (!fops || !fops->probe)
			continue;

		ret = fops->probe(buf, buf_len);
		if (!ret) {
			image->fops = fops;
			return ret;
		}
	}

	return ret;
}

void *arch_kexec_kernel_image_load(struct kimage *image)
{
	if (!image->fops || !image->fops->load)
		return ERR_PTR(-ENOEXEC);

	return image->fops->load(image, image->kernel_buf,
				 image->kernel_buf_len, image->initrd_buf,
				 image->initrd_buf_len, image->cmdline_buf,
				 image->cmdline_buf_len);
}

int arch_kimage_file_post_load_cleanup(struct kimage *image)
{
	if (!image->fops || !image->fops->cleanup)
		return 0;

	return image->fops->cleanup(image);
}