[linux-2.6-block.git] / arch / x86 / include / asm / kexec.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _ASM_X86_KEXEC_H
#define _ASM_X86_KEXEC_H

#ifdef CONFIG_X86_32
# define PA_CONTROL_PAGE	0
# define VA_CONTROL_PAGE	1
# define PA_PGD			2
# define PA_SWAP_PAGE		3
# define PAGES_NR		4
#else
# define PA_CONTROL_PAGE	0
# define VA_CONTROL_PAGE	1
# define PA_TABLE_PAGE		2
# define PA_SWAP_PAGE		3
# define PAGES_NR		4
#endif

# define KEXEC_CONTROL_CODE_MAX_SIZE	2048

#ifndef __ASSEMBLY__

#include <linux/string.h>

#include <asm/page.h>
#include <asm/ptrace.h>
#include <asm/bootparam.h>

struct kimage;

/*
 * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
 * I.e. Maximum page that is mapped directly into kernel memory,
 * and kmap is not required.
 *
 * So far x86_64 is limited to 40 physical address bits.
 */
#ifdef CONFIG_X86_32
/* Maximum physical address we can use pages from */
# define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
/* Maximum address we can reach in physical address mode */
# define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
/* Maximum address we can use for the control code buffer */
# define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE

# define KEXEC_CONTROL_PAGE_SIZE	4096

/* The native architecture */
# define KEXEC_ARCH KEXEC_ARCH_386

/* We can also handle crash dumps from 64 bit kernel. */
# define vmcore_elf_check_arch_cross(x) ((x)->e_machine == EM_X86_64)
#else
/* Maximum physical address we can use pages from */
# define KEXEC_SOURCE_MEMORY_LIMIT      (MAXMEM-1)
/* Maximum address we can reach in physical address mode */
# define KEXEC_DESTINATION_MEMORY_LIMIT (MAXMEM-1)
/* Maximum address we can use for the control pages */
# define KEXEC_CONTROL_MEMORY_LIMIT     (MAXMEM-1)

/* Allocate one page for the pdp and the second for the code */
# define KEXEC_CONTROL_PAGE_SIZE  (4096UL + 4096UL)

/* The native architecture */
# define KEXEC_ARCH KEXEC_ARCH_X86_64
#endif

/* Memory to backup during crash kdump */
#define KEXEC_BACKUP_SRC_START	(0UL)
#define KEXEC_BACKUP_SRC_END	(640 * 1024UL)	/* 640K */

/*
 * CPU does not save ss and sp on stack if execution is already
 * running in kernel mode at the time of NMI occurrence. This code
 * fixes it.
 */
static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
				      struct pt_regs *oldregs)
{
#ifdef CONFIG_X86_32
	newregs->sp = (unsigned long)&(oldregs->sp);
	asm volatile("xorl %%eax, %%eax\n\t"
		     "movw %%ss, %%ax\n\t"
		     :"=a"(newregs->ss));
#endif
}

/*
 * This function is responsible for capturing register states if coming
 * via panic otherwise just fix up the ss and sp if coming via kernel
 * mode exception.
 */
static inline void crash_setup_regs(struct pt_regs *newregs,
				    struct pt_regs *oldregs)
{
	if (oldregs) {
		memcpy(newregs, oldregs, sizeof(*newregs));
		crash_fixup_ss_esp(newregs, oldregs);
	} else {
#ifdef CONFIG_X86_32
		asm volatile("movl %%ebx,%0" : "=m"(newregs->bx));
		asm volatile("movl %%ecx,%0" : "=m"(newregs->cx));
		asm volatile("movl %%edx,%0" : "=m"(newregs->dx));
		asm volatile("movl %%esi,%0" : "=m"(newregs->si));
		asm volatile("movl %%edi,%0" : "=m"(newregs->di));
		asm volatile("movl %%ebp,%0" : "=m"(newregs->bp));
		asm volatile("movl %%eax,%0" : "=m"(newregs->ax));
		asm volatile("movl %%esp,%0" : "=m"(newregs->sp));
		asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
		asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
		asm volatile("movl %%ds, %%eax;" :"=a"(newregs->ds));
		asm volatile("movl %%es, %%eax;" :"=a"(newregs->es));
		asm volatile("pushfl; popl %0" :"=m"(newregs->flags));
#else
		asm volatile("movq %%rbx,%0" : "=m"(newregs->bx));
		asm volatile("movq %%rcx,%0" : "=m"(newregs->cx));
		asm volatile("movq %%rdx,%0" : "=m"(newregs->dx));
		asm volatile("movq %%rsi,%0" : "=m"(newregs->si));
		asm volatile("movq %%rdi,%0" : "=m"(newregs->di));
		asm volatile("movq %%rbp,%0" : "=m"(newregs->bp));
		asm volatile("movq %%rax,%0" : "=m"(newregs->ax));
		asm volatile("movq %%rsp,%0" : "=m"(newregs->sp));
		asm volatile("movq %%r8,%0" : "=m"(newregs->r8));
		asm volatile("movq %%r9,%0" : "=m"(newregs->r9));
		asm volatile("movq %%r10,%0" : "=m"(newregs->r10));
		asm volatile("movq %%r11,%0" : "=m"(newregs->r11));
		asm volatile("movq %%r12,%0" : "=m"(newregs->r12));
		asm volatile("movq %%r13,%0" : "=m"(newregs->r13));
		asm volatile("movq %%r14,%0" : "=m"(newregs->r14));
		asm volatile("movq %%r15,%0" : "=m"(newregs->r15));
		asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
		asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
		asm volatile("pushfq; popq %0" :"=m"(newregs->flags));
#endif
		newregs->ip = (unsigned long)current_text_addr();
	}
}

#ifdef CONFIG_X86_32
asmlinkage unsigned long
relocate_kernel(unsigned long indirection_page,
		unsigned long control_page,
		unsigned long start_address,
		unsigned int has_pae,
		unsigned int preserve_context);
#else
unsigned long
relocate_kernel(unsigned long indirection_page,
		unsigned long page_list,
		unsigned long start_address,
		unsigned int preserve_context,
		unsigned int sme_active);
#endif

#define ARCH_HAS_KIMAGE_ARCH

#ifdef CONFIG_X86_32
struct kimage_arch {
	pgd_t *pgd;
#ifdef CONFIG_X86_PAE
	pmd_t *pmd0;
	pmd_t *pmd1;
#endif
	pte_t *pte0;
	pte_t *pte1;
};
#else
struct kimage_arch {
	p4d_t *p4d;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
	/* Details of backup region */
	unsigned long backup_src_start;
	unsigned long backup_src_sz;

	/* Physical address of backup segment */
	unsigned long backup_load_addr;

	/* Core ELF header buffer */
	void *elf_headers;
	unsigned long elf_headers_sz;
	unsigned long elf_load_addr;
};
#endif /* CONFIG_X86_32 */

#ifdef CONFIG_X86_64
/*
 * Number of elements and order of elements in this structure should match
 * with the ones in arch/x86/purgatory/entry64.S. If you make a change here
 * make an appropriate change in purgatory too.
 */
struct kexec_entry64_regs {
	uint64_t rax;
	uint64_t rcx;
	uint64_t rdx;
	uint64_t rbx;
	uint64_t rsp;
	uint64_t rbp;
	uint64_t rsi;
	uint64_t rdi;
	uint64_t r8;
	uint64_t r9;
	uint64_t r10;
	uint64_t r11;
	uint64_t r12;
	uint64_t r13;
	uint64_t r14;
	uint64_t r15;
	uint64_t rip;
};

extern int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages,
				       gfp_t gfp);
#define arch_kexec_post_alloc_pages arch_kexec_post_alloc_pages

extern void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages);
#define arch_kexec_pre_free_pages arch_kexec_pre_free_pages

#endif

typedef void crash_vmclear_fn(void);
extern crash_vmclear_fn __rcu *crash_vmclear_loaded_vmcss;
extern void kdump_nmi_shootdown_cpus(void);

#endif /* __ASSEMBLY__ */

#endif /* _ASM_X86_KEXEC_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
1965aae3 PA	2	#ifndef _ASM_X86_KEXEC_H
1965aae3 PA	3	#define _ASM_X86_KEXEC_H
3c233d13	4
96a388de	5	#ifdef CONFIG_X86_32
3c233d13 HH	6	# define PA_CONTROL_PAGE 0
	7	# define VA_CONTROL_PAGE 1
	8	# define PA_PGD 2
9868ee63 HY	9	# define PA_SWAP_PAGE 3
9868ee63 HY	10	# define PAGES_NR 4
96a388de	11	#else
3c233d13	12	# define PA_CONTROL_PAGE 0
fee7b0d8 HY	13	# define VA_CONTROL_PAGE 1
	14	# define PA_TABLE_PAGE 2
	15	# define PA_SWAP_PAGE 3
	16	# define PAGES_NR 4
96a388de	17	#endif
3c233d13	18
fb45daa6	19	# define KEXEC_CONTROL_CODE_MAX_SIZE 2048
fb45daa6	20
3c233d13 HH	21	#ifndef __ASSEMBLY__
	22
	23	#include <linux/string.h>
	24
	25	#include <asm/page.h>
	26	#include <asm/ptrace.h>
27f48d3e	27	#include <asm/bootparam.h>
3c233d13	28
dd5f7260 VG	29	struct kimage;
dd5f7260 VG	30
3c233d13 HH	31	/*
	32	* KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
	33	* I.e. Maximum page that is mapped directly into kernel memory,
	34	* and kmap is not required.
	35	*
	36	* So far x86_64 is limited to 40 physical address bits.
	37	*/
	38	#ifdef CONFIG_X86_32
	39	/* Maximum physical address we can use pages from */
	40	# define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
	41	/* Maximum address we can reach in physical address mode */
	42	# define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
	43	/* Maximum address we can use for the control code buffer */
	44	# define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE
	45
163f6876	46	# define KEXEC_CONTROL_PAGE_SIZE 4096
3c233d13 HH	47
	48	/* The native architecture */
	49	# define KEXEC_ARCH KEXEC_ARCH_386
	50
	51	/* We can also handle crash dumps from 64 bit kernel. */
	52	# define vmcore_elf_check_arch_cross(x) ((x)->e_machine == EM_X86_64)
	53	#else
	54	/* Maximum physical address we can use pages from */
577af55d	55	# define KEXEC_SOURCE_MEMORY_LIMIT (MAXMEM-1)
3c233d13	56	/* Maximum address we can reach in physical address mode */
577af55d	57	# define KEXEC_DESTINATION_MEMORY_LIMIT (MAXMEM-1)
3c233d13	58	/* Maximum address we can use for the control pages */
577af55d	59	# define KEXEC_CONTROL_MEMORY_LIMIT (MAXMEM-1)
3c233d13 HH	60
3c233d13 HH	61	/* Allocate one page for the pdp and the second for the code */
163f6876	62	# define KEXEC_CONTROL_PAGE_SIZE (4096UL + 4096UL)
3c233d13 HH	63
	64	/* The native architecture */
	65	# define KEXEC_ARCH KEXEC_ARCH_X86_64
	66	#endif
	67
dd5f7260 VG	68	/* Memory to backup during crash kdump */
	69	#define KEXEC_BACKUP_SRC_START (0UL)
	70	#define KEXEC_BACKUP_SRC_END (640 * 1024UL) /* 640K */
	71
3c233d13 HH	72	/*
	73	* CPU does not save ss and sp on stack if execution is already
	74	* running in kernel mode at the time of NMI occurrence. This code
	75	* fixes it.
	76	*/
	77	static inline void crash_fixup_ss_esp(struct pt_regs *newregs,
	78	struct pt_regs *oldregs)
	79	{
	80	#ifdef CONFIG_X86_32
	81	newregs->sp = (unsigned long)&(oldregs->sp);
b69a3f9d JP	82	asm volatile("xorl %%eax, %%eax\n\t"
	83	"movw %%ss, %%ax\n\t"
	84	:"=a"(newregs->ss));
3c233d13 HH	85	#endif
	86	}
	87
	88	/*
	89	* This function is responsible for capturing register states if coming
	90	* via panic otherwise just fix up the ss and sp if coming via kernel
	91	* mode exception.
	92	*/
	93	static inline void crash_setup_regs(struct pt_regs *newregs,
	94	struct pt_regs *oldregs)
	95	{
	96	if (oldregs) {
	97	memcpy(newregs, oldregs, sizeof(*newregs));
	98	crash_fixup_ss_esp(newregs, oldregs);
	99	} else {
	100	#ifdef CONFIG_X86_32
b69a3f9d JP	101	asm volatile("movl %%ebx,%0" : "=m"(newregs->bx));
	102	asm volatile("movl %%ecx,%0" : "=m"(newregs->cx));
	103	asm volatile("movl %%edx,%0" : "=m"(newregs->dx));
	104	asm volatile("movl %%esi,%0" : "=m"(newregs->si));
	105	asm volatile("movl %%edi,%0" : "=m"(newregs->di));
	106	asm volatile("movl %%ebp,%0" : "=m"(newregs->bp));
	107	asm volatile("movl %%eax,%0" : "=m"(newregs->ax));
	108	asm volatile("movl %%esp,%0" : "=m"(newregs->sp));
	109	asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
	110	asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
	111	asm volatile("movl %%ds, %%eax;" :"=a"(newregs->ds));
	112	asm volatile("movl %%es, %%eax;" :"=a"(newregs->es));
	113	asm volatile("pushfl; popl %0" :"=m"(newregs->flags));
3c233d13	114	#else
b69a3f9d JP	115	asm volatile("movq %%rbx,%0" : "=m"(newregs->bx));
	116	asm volatile("movq %%rcx,%0" : "=m"(newregs->cx));
	117	asm volatile("movq %%rdx,%0" : "=m"(newregs->dx));
	118	asm volatile("movq %%rsi,%0" : "=m"(newregs->si));
	119	asm volatile("movq %%rdi,%0" : "=m"(newregs->di));
	120	asm volatile("movq %%rbp,%0" : "=m"(newregs->bp));
	121	asm volatile("movq %%rax,%0" : "=m"(newregs->ax));
	122	asm volatile("movq %%rsp,%0" : "=m"(newregs->sp));
	123	asm volatile("movq %%r8,%0" : "=m"(newregs->r8));
	124	asm volatile("movq %%r9,%0" : "=m"(newregs->r9));
	125	asm volatile("movq %%r10,%0" : "=m"(newregs->r10));
	126	asm volatile("movq %%r11,%0" : "=m"(newregs->r11));
	127	asm volatile("movq %%r12,%0" : "=m"(newregs->r12));
	128	asm volatile("movq %%r13,%0" : "=m"(newregs->r13));
	129	asm volatile("movq %%r14,%0" : "=m"(newregs->r14));
	130	asm volatile("movq %%r15,%0" : "=m"(newregs->r15));
	131	asm volatile("movl %%ss, %%eax;" :"=a"(newregs->ss));
	132	asm volatile("movl %%cs, %%eax;" :"=a"(newregs->cs));
	133	asm volatile("pushfq; popq %0" :"=m"(newregs->flags));
3c233d13 HH	134	#endif
	135	newregs->ip = (unsigned long)current_text_addr();
	136	}
	137	}
	138
	139	#ifdef CONFIG_X86_32
3ab83521	140	asmlinkage unsigned long
3c233d13 HH	141	relocate_kernel(unsigned long indirection_page,
	142	unsigned long control_page,
	143	unsigned long start_address,
3ab83521 HY	144	unsigned int has_pae,
3ab83521 HY	145	unsigned int preserve_context);
3c233d13	146	#else
fee7b0d8	147	unsigned long
3c233d13 HH	148	relocate_kernel(unsigned long indirection_page,
3c233d13 HH	149	unsigned long page_list,
fee7b0d8	150	unsigned long start_address,
4e237903 TL	151	unsigned int preserve_context,
4e237903 TL	152	unsigned int sme_active);
3c233d13 HH	153	#endif
3c233d13 HH	154
92be3d6b HY	155	#define ARCH_HAS_KIMAGE_ARCH
92be3d6b HY	156
f5deb796	157	#ifdef CONFIG_X86_32
92be3d6b HY	158	struct kimage_arch {
	159	pgd_t *pgd;
	160	#ifdef CONFIG_X86_PAE
	161	pmd_t *pmd0;
	162	pmd_t *pmd1;
	163	#endif
	164	pte_t *pte0;
	165	pte_t *pte1;
	166	};
f5deb796 HY	167	#else
f5deb796 HY	168	struct kimage_arch {
7f689041	169	p4d_t *p4d;
f5deb796 HY	170	pud_t *pud;
	171	pmd_t *pmd;
	172	pte_t *pte;
dd5f7260 VG	173	/* Details of backup region */
	174	unsigned long backup_src_start;
	175	unsigned long backup_src_sz;
	176
	177	/* Physical address of backup segment */
	178	unsigned long backup_load_addr;
	179
	180	/* Core ELF header buffer */
	181	void *elf_headers;
	182	unsigned long elf_headers_sz;
	183	unsigned long elf_load_addr;
f5deb796	184	};
dd5f7260	185	#endif /* CONFIG_X86_32 */
27f48d3e	186
dd5f7260 VG	187	#ifdef CONFIG_X86_64
	188	/*
	189	* Number of elements and order of elements in this structure should match
	190	* with the ones in arch/x86/purgatory/entry64.S. If you make a change here
	191	* make an appropriate change in purgatory too.
	192	*/
27f48d3e VG	193	struct kexec_entry64_regs {
27f48d3e VG	194	uint64_t rax;
27f48d3e VG	195	uint64_t rcx;
27f48d3e VG	196	uint64_t rdx;
dd5f7260	197	uint64_t rbx;
27f48d3e VG	198	uint64_t rsp;
27f48d3e VG	199	uint64_t rbp;
dd5f7260 VG	200	uint64_t rsi;
dd5f7260 VG	201	uint64_t rdi;
27f48d3e VG	202	uint64_t r8;
	203	uint64_t r9;
	204	uint64_t r10;
	205	uint64_t r11;
	206	uint64_t r12;
	207	uint64_t r13;
	208	uint64_t r14;
	209	uint64_t r15;
	210	uint64_t rip;
	211	};
bba4ed01 TL	212
	213	extern int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages,
	214	gfp_t gfp);
	215	#define arch_kexec_post_alloc_pages arch_kexec_post_alloc_pages
	216
	217	extern void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages);
	218	#define arch_kexec_pre_free_pages arch_kexec_pre_free_pages
	219
92be3d6b HY	220	#endif
92be3d6b HY	221
0ca0d818 ZY	222	typedef void crash_vmclear_fn(void);
0ca0d818 ZY	223	extern crash_vmclear_fn __rcu *crash_vmclear_loaded_vmcss;
0ee59413	224	extern void kdump_nmi_shootdown_cpus(void);
f23d1f4a	225
3c233d13 HH	226	#endif /* __ASSEMBLY__ */
3c233d13 HH	227
1965aae3	228	#endif /* _ASM_X86_KEXEC_H */