[linux-2.6-block.git] / arch / x86 / vdso / vdso32-setup.c

/*
 * (C) Copyright 2002 Linus Torvalds
 * Portions based on the vdso-randomization code from exec-shield:
 * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar
 *
 * This file contains the needed initializations to support sysenter.
 */

#include <linux/init.h>
#include <linux/smp.h>
#include <linux/thread_info.h>
#include <linux/sched.h>
#include <linux/gfp.h>
#include <linux/string.h>
#include <linux/elf.h>
#include <linux/mm.h>
#include <linux/err.h>
#include <linux/module.h>

#include <asm/cpufeature.h>
#include <asm/msr.h>
#include <asm/pgtable.h>
#include <asm/unistd.h>
#include <asm/elf.h>
#include <asm/tlbflush.h>
#include <asm/vdso.h>
#include <asm/proto.h>

enum {
	VDSO_DISABLED = 0,
	VDSO_ENABLED = 1,
	VDSO_COMPAT = 2,
};

#ifdef CONFIG_COMPAT_VDSO
#define VDSO_DEFAULT	VDSO_COMPAT
#else
#define VDSO_DEFAULT	VDSO_ENABLED
#endif

#ifdef CONFIG_X86_64
#define vdso_enabled			sysctl_vsyscall32
#define arch_setup_additional_pages	syscall32_setup_pages
#endif

/*
 * This is the difference between the prelinked addresses in the vDSO images
 * and the VDSO_HIGH_BASE address where CONFIG_COMPAT_VDSO places the vDSO
 * in the user address space.
 */
#define VDSO_ADDR_ADJUST	(VDSO_HIGH_BASE - (unsigned long)VDSO32_PRELINK)

/*
 * Should the kernel map a VDSO page into processes and pass its
 * address down to glibc upon exec()?
 */
unsigned int __read_mostly vdso_enabled = VDSO_DEFAULT;

static int __init vdso_setup(char *s)
{
	vdso_enabled = simple_strtoul(s, NULL, 0);

	return 1;
}

/*
 * For consistency, the argument vdso32=[012] affects the 32-bit vDSO
 * behavior on both 64-bit and 32-bit kernels.
 * On 32-bit kernels, vdso=[012] means the same thing.
 */
__setup("vdso32=", vdso_setup);

#ifdef CONFIG_X86_32
__setup_param("vdso=", vdso32_setup, vdso_setup, 0);

EXPORT_SYMBOL_GPL(vdso_enabled);
#endif

static __init void reloc_symtab(Elf32_Ehdr *ehdr,
				unsigned offset, unsigned size)
{
	Elf32_Sym *sym = (void *)ehdr + offset;
	unsigned nsym = size / sizeof(*sym);
	unsigned i;

	for(i = 0; i < nsym; i++, sym++) {
		if (sym->st_shndx == SHN_UNDEF ||
		    sym->st_shndx == SHN_ABS)
			continue;  /* skip */

		if (sym->st_shndx > SHN_LORESERVE) {
			printk(KERN_INFO "VDSO: unexpected st_shndx %x\n",
			       sym->st_shndx);
			continue;
		}

		switch(ELF_ST_TYPE(sym->st_info)) {
		case STT_OBJECT:
		case STT_FUNC:
		case STT_SECTION:
		case STT_FILE:
			sym->st_value += VDSO_ADDR_ADJUST;
		}
	}
}

static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset)
{
	Elf32_Dyn *dyn = (void *)ehdr + offset;

	for(; dyn->d_tag != DT_NULL; dyn++)
		switch(dyn->d_tag) {
		case DT_PLTGOT:
		case DT_HASH:
		case DT_STRTAB:
		case DT_SYMTAB:
		case DT_RELA:
		case DT_INIT:
		case DT_FINI:
		case DT_REL:
		case DT_DEBUG:
		case DT_JMPREL:
		case DT_VERSYM:
		case DT_VERDEF:
		case DT_VERNEED:
		case DT_ADDRRNGLO ... DT_ADDRRNGHI:
			/* definitely pointers needing relocation */
			dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
			break;

		case DT_ENCODING ... OLD_DT_LOOS-1:
		case DT_LOOS ... DT_HIOS-1:
			/* Tags above DT_ENCODING are pointers if
			   they're even */
			if (dyn->d_tag >= DT_ENCODING &&
			    (dyn->d_tag & 1) == 0)
				dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
			break;

		case DT_VERDEFNUM:
		case DT_VERNEEDNUM:
		case DT_FLAGS_1:
		case DT_RELACOUNT:
		case DT_RELCOUNT:
		case DT_VALRNGLO ... DT_VALRNGHI:
			/* definitely not pointers */
			break;

		case OLD_DT_LOOS ... DT_LOOS-1:
		case DT_HIOS ... DT_VALRNGLO-1:
		default:
			if (dyn->d_tag > DT_ENCODING)
				printk(KERN_INFO "VDSO: unexpected DT_tag %x\n",
				       dyn->d_tag);
			break;
		}
}

static __init void relocate_vdso(Elf32_Ehdr *ehdr)
{
	Elf32_Phdr *phdr;
	Elf32_Shdr *shdr;
	int i;

	BUG_ON(memcmp(ehdr->e_ident, ELFMAG, SELFMAG) != 0 ||
	       !elf_check_arch_ia32(ehdr) ||
	       ehdr->e_type != ET_DYN);

	ehdr->e_entry += VDSO_ADDR_ADJUST;

	/* rebase phdrs */
	phdr = (void *)ehdr + ehdr->e_phoff;
	for (i = 0; i < ehdr->e_phnum; i++) {
		phdr[i].p_vaddr += VDSO_ADDR_ADJUST;

		/* relocate dynamic stuff */
		if (phdr[i].p_type == PT_DYNAMIC)
			reloc_dyn(ehdr, phdr[i].p_offset);
	}

	/* rebase sections */
	shdr = (void *)ehdr + ehdr->e_shoff;
	for(i = 0; i < ehdr->e_shnum; i++) {
		if (!(shdr[i].sh_flags & SHF_ALLOC))
			continue;

		shdr[i].sh_addr += VDSO_ADDR_ADJUST;

		if (shdr[i].sh_type == SHT_SYMTAB ||
		    shdr[i].sh_type == SHT_DYNSYM)
			reloc_symtab(ehdr, shdr[i].sh_offset,
				     shdr[i].sh_size);
	}
}

static struct page *vdso32_pages[1];

#ifdef CONFIG_X86_64

#define	vdso32_sysenter()	(boot_cpu_has(X86_FEATURE_SYSENTER32))
#define	vdso32_syscall()	(boot_cpu_has(X86_FEATURE_SYSCALL32))

/* May not be __init: called during resume */
void syscall32_cpu_init(void)
{
	/* Load these always in case some future AMD CPU supports
	   SYSENTER from compat mode too. */
	checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
	checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
	checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);

	wrmsrl(MSR_CSTAR, ia32_cstar_target);
}

#define compat_uses_vma		1

static inline void map_compat_vdso(int map)
{
}

#else  /* CONFIG_X86_32 */

#define vdso32_sysenter()	(boot_cpu_has(X86_FEATURE_SEP))
#define vdso32_syscall()	(0)

void enable_sep_cpu(void)
{
	int cpu = get_cpu();
	struct tss_struct *tss = &per_cpu(init_tss, cpu);

	if (!boot_cpu_has(X86_FEATURE_SEP)) {
		put_cpu();
		return;
	}

	tss->x86_tss.ss1 = __KERNEL_CS;
	tss->x86_tss.sp1 = sizeof(struct tss_struct) + (unsigned long) tss;
	wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
	wrmsr(MSR_IA32_SYSENTER_ESP, tss->x86_tss.sp1, 0);
	wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long) ia32_sysenter_target, 0);
	put_cpu();	
}

static struct vm_area_struct gate_vma;

static int __init gate_vma_init(void)
{
	gate_vma.vm_mm = NULL;
	gate_vma.vm_start = FIXADDR_USER_START;
	gate_vma.vm_end = FIXADDR_USER_END;
	gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC;
	gate_vma.vm_page_prot = __P101;
	/*
	 * Make sure the vDSO gets into every core dump.
	 * Dumping its contents makes post-mortem fully interpretable later
	 * without matching up the same kernel and hardware config to see
	 * what PC values meant.
	 */
	gate_vma.vm_flags |= VM_ALWAYSDUMP;
	return 0;
}

#define compat_uses_vma		0

static void map_compat_vdso(int map)
{
	static int vdso_mapped;

	if (map == vdso_mapped)
		return;

	vdso_mapped = map;

	__set_fixmap(FIX_VDSO, page_to_pfn(vdso32_pages[0]) << PAGE_SHIFT,
		     map ? PAGE_READONLY_EXEC : PAGE_NONE);

	/* flush stray tlbs */
	flush_tlb_all();
}

#endif	/* CONFIG_X86_64 */

int __init sysenter_setup(void)
{
	void *syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
	const void *vsyscall;
	size_t vsyscall_len;

	vdso32_pages[0] = virt_to_page(syscall_page);

#ifdef CONFIG_X86_32
	gate_vma_init();
#endif

	if (vdso32_syscall()) {
		vsyscall = &vdso32_syscall_start;
		vsyscall_len = &vdso32_syscall_end - &vdso32_syscall_start;
	} else if (vdso32_sysenter()){
		vsyscall = &vdso32_sysenter_start;
		vsyscall_len = &vdso32_sysenter_end - &vdso32_sysenter_start;
	} else {
		vsyscall = &vdso32_int80_start;
		vsyscall_len = &vdso32_int80_end - &vdso32_int80_start;
	}

	memcpy(syscall_page, vsyscall, vsyscall_len);
	relocate_vdso(syscall_page);

	return 0;
}

/* Setup a VMA at program startup for the vsyscall page */
int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
{
	struct mm_struct *mm = current->mm;
	unsigned long addr;
	int ret = 0;
	bool compat;

	if (vdso_enabled == VDSO_DISABLED)
		return 0;

	down_write(&mm->mmap_sem);

	/* Test compat mode once here, in case someone
	   changes it via sysctl */
	compat = (vdso_enabled == VDSO_COMPAT);

	map_compat_vdso(compat);

	if (compat)
		addr = VDSO_HIGH_BASE;
	else {
		addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
		if (IS_ERR_VALUE(addr)) {
			ret = addr;
			goto up_fail;
		}
	}

	if (compat_uses_vma || !compat) {
		/*
		 * MAYWRITE to allow gdb to COW and set breakpoints
		 *
		 * Make sure the vDSO gets into every core dump.
		 * Dumping its contents makes post-mortem fully
		 * interpretable later without matching up the same
		 * kernel and hardware config to see what PC values
		 * meant.
		 */
		ret = install_special_mapping(mm, addr, PAGE_SIZE,
					      VM_READ|VM_EXEC|
					      VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC|
					      VM_ALWAYSDUMP,
					      vdso32_pages);

		if (ret)
			goto up_fail;
	}

	current->mm->context.vdso = (void *)addr;
	current_thread_info()->sysenter_return =
		VDSO32_SYMBOL(addr, SYSENTER_RETURN);

  up_fail:
	up_write(&mm->mmap_sem);

	return ret;
}

#ifdef CONFIG_X86_64

__initcall(sysenter_setup);

#ifdef CONFIG_SYSCTL
/* Register vsyscall32 into the ABI table */
#include <linux/sysctl.h>

static ctl_table abi_table2[] = {
	{
		.procname	= "vsyscall32",
		.data		= &sysctl_vsyscall32,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec
	},
	{}
};

static ctl_table abi_root_table2[] = {
	{
		.ctl_name = CTL_ABI,
		.procname = "abi",
		.mode = 0555,
		.child = abi_table2
	},
	{}
};

static __init int ia32_binfmt_init(void)
{
	register_sysctl_table(abi_root_table2);
	return 0;
}
__initcall(ia32_binfmt_init);
#endif

#else  /* CONFIG_X86_32 */

const char *arch_vma_name(struct vm_area_struct *vma)
{
	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
		return "[vdso]";
	return NULL;
}

struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
{
	struct mm_struct *mm = tsk->mm;

	/* Check to see if this task was created in compat vdso mode */
	if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE)
		return &gate_vma;
	return NULL;
}

int in_gate_area(struct task_struct *task, unsigned long addr)
{
	const struct vm_area_struct *vma = get_gate_vma(task);

	return vma && addr >= vma->vm_start && addr < vma->vm_end;
}

int in_gate_area_no_task(unsigned long addr)
{
	return 0;
}

#endif	/* CONFIG_X86_64 */
Commit	Line	Data
1da177e4	1	/*
1da177e4	2	* (C) Copyright 2002 Linus Torvalds
e6e5494c IM	3	* Portions based on the vdso-randomization code from exec-shield:
e6e5494c IM	4	* Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar
1da177e4 LT	5	*
	6	* This file contains the needed initializations to support sysenter.
	7	*/
	8
	9	#include <linux/init.h>
	10	#include <linux/smp.h>
	11	#include <linux/thread_info.h>
	12	#include <linux/sched.h>
	13	#include <linux/gfp.h>
	14	#include <linux/string.h>
	15	#include <linux/elf.h>
e6e5494c	16	#include <linux/mm.h>
4e950f6f	17	#include <linux/err.h>
e6e5494c	18	#include <linux/module.h>
1da177e4 LT	19
	20	#include <asm/cpufeature.h>
	21	#include <asm/msr.h>
	22	#include <asm/pgtable.h>
	23	#include <asm/unistd.h>
d4f7a2c1	24	#include <asm/elf.h>
1dbf527c	25	#include <asm/tlbflush.h>
6c3652ef	26	#include <asm/vdso.h>
af65d648	27	#include <asm/proto.h>
1dbf527c JF	28
	29	enum {
	30	VDSO_DISABLED = 0,
	31	VDSO_ENABLED = 1,
	32	VDSO_COMPAT = 2,
	33	};
	34
	35	#ifdef CONFIG_COMPAT_VDSO
	36	#define VDSO_DEFAULT VDSO_COMPAT
	37	#else
	38	#define VDSO_DEFAULT VDSO_ENABLED
	39	#endif
1da177e4	40
af65d648 RM	41	#ifdef CONFIG_X86_64
	42	#define vdso_enabled sysctl_vsyscall32
	43	#define arch_setup_additional_pages syscall32_setup_pages
	44	#endif
	45
	46	/*
	47	* This is the difference between the prelinked addresses in the vDSO images
	48	* and the VDSO_HIGH_BASE address where CONFIG_COMPAT_VDSO places the vDSO
	49	* in the user address space.
	50	*/
	51	#define VDSO_ADDR_ADJUST (VDSO_HIGH_BASE - (unsigned long)VDSO32_PRELINK)
	52
e6e5494c IM	53	/*
	54	* Should the kernel map a VDSO page into processes and pass its
	55	* address down to glibc upon exec()?
	56	*/
1dbf527c	57	unsigned int __read_mostly vdso_enabled = VDSO_DEFAULT;
e6e5494c	58
e6e5494c IM	59	static int __init vdso_setup(char *s)
	60	{
	61	vdso_enabled = simple_strtoul(s, NULL, 0);
	62
	63	return 1;
	64	}
	65
af65d648 RM	66	/*
	67	* For consistency, the argument vdso32=[012] affects the 32-bit vDSO
	68	* behavior on both 64-bit and 32-bit kernels.
	69	* On 32-bit kernels, vdso=[012] means the same thing.
	70	*/
	71	__setup("vdso32=", vdso_setup);
e6e5494c	72
af65d648 RM	73	#ifdef CONFIG_X86_32
	74	__setup_param("vdso=", vdso32_setup, vdso_setup, 0);
	75
	76	EXPORT_SYMBOL_GPL(vdso_enabled);
	77	#endif
1da177e4	78
d4f7a2c1 JF	79	static __init void reloc_symtab(Elf32_Ehdr *ehdr,
	80	unsigned offset, unsigned size)
	81	{
	82	Elf32_Sym sym = (void )ehdr + offset;
	83	unsigned nsym = size / sizeof(*sym);
	84	unsigned i;
	85
	86	for(i = 0; i < nsym; i++, sym++) {
	87	if (sym->st_shndx == SHN_UNDEF \|\|
	88	sym->st_shndx == SHN_ABS)
	89	continue; /* skip */
	90
	91	if (sym->st_shndx > SHN_LORESERVE) {
	92	printk(KERN_INFO "VDSO: unexpected st_shndx %x\n",
	93	sym->st_shndx);
	94	continue;
	95	}
	96
	97	switch(ELF_ST_TYPE(sym->st_info)) {
	98	case STT_OBJECT:
	99	case STT_FUNC:
	100	case STT_SECTION:
	101	case STT_FILE:
af65d648	102	sym->st_value += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	103	}
	104	}
	105	}
	106
	107	static __init void reloc_dyn(Elf32_Ehdr *ehdr, unsigned offset)
	108	{
	109	Elf32_Dyn dyn = (void )ehdr + offset;
	110
	111	for(; dyn->d_tag != DT_NULL; dyn++)
	112	switch(dyn->d_tag) {
	113	case DT_PLTGOT:
	114	case DT_HASH:
	115	case DT_STRTAB:
	116	case DT_SYMTAB:
	117	case DT_RELA:
	118	case DT_INIT:
	119	case DT_FINI:
	120	case DT_REL:
	121	case DT_DEBUG:
	122	case DT_JMPREL:
	123	case DT_VERSYM:
	124	case DT_VERDEF:
	125	case DT_VERNEED:
	126	case DT_ADDRRNGLO ... DT_ADDRRNGHI:
	127	/* definitely pointers needing relocation */
af65d648	128	dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	129	break;
	130
	131	case DT_ENCODING ... OLD_DT_LOOS-1:
	132	case DT_LOOS ... DT_HIOS-1:
	133	/* Tags above DT_ENCODING are pointers if
	134	they're even */
	135	if (dyn->d_tag >= DT_ENCODING &&
	136	(dyn->d_tag & 1) == 0)
af65d648	137	dyn->d_un.d_ptr += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	138	break;
	139
	140	case DT_VERDEFNUM:
	141	case DT_VERNEEDNUM:
	142	case DT_FLAGS_1:
	143	case DT_RELACOUNT:
	144	case DT_RELCOUNT:
	145	case DT_VALRNGLO ... DT_VALRNGHI:
	146	/* definitely not pointers */
	147	break;
	148
	149	case OLD_DT_LOOS ... DT_LOOS-1:
	150	case DT_HIOS ... DT_VALRNGLO-1:
	151	default:
	152	if (dyn->d_tag > DT_ENCODING)
	153	printk(KERN_INFO "VDSO: unexpected DT_tag %x\n",
	154	dyn->d_tag);
	155	break;
	156	}
	157	}
	158
	159	static __init void relocate_vdso(Elf32_Ehdr *ehdr)
	160	{
	161	Elf32_Phdr *phdr;
	162	Elf32_Shdr *shdr;
	163	int i;
	164
ecb783ea	165	BUG_ON(memcmp(ehdr->e_ident, ELFMAG, SELFMAG) != 0 \|\|
af65d648	166	!elf_check_arch_ia32(ehdr) \|\|
d4f7a2c1 JF	167	ehdr->e_type != ET_DYN);
d4f7a2c1 JF	168
af65d648	169	ehdr->e_entry += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	170
	171	/* rebase phdrs */
	172	phdr = (void *)ehdr + ehdr->e_phoff;
	173	for (i = 0; i < ehdr->e_phnum; i++) {
af65d648	174	phdr[i].p_vaddr += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	175
	176	/* relocate dynamic stuff */
	177	if (phdr[i].p_type == PT_DYNAMIC)
	178	reloc_dyn(ehdr, phdr[i].p_offset);
	179	}
	180
	181	/* rebase sections */
	182	shdr = (void *)ehdr + ehdr->e_shoff;
	183	for(i = 0; i < ehdr->e_shnum; i++) {
	184	if (!(shdr[i].sh_flags & SHF_ALLOC))
	185	continue;
	186
af65d648	187	shdr[i].sh_addr += VDSO_ADDR_ADJUST;
d4f7a2c1 JF	188
	189	if (shdr[i].sh_type == SHT_SYMTAB \|\|
	190	shdr[i].sh_type == SHT_DYNSYM)
	191	reloc_symtab(ehdr, shdr[i].sh_offset,
	192	shdr[i].sh_size);
	193	}
	194	}
d4f7a2c1	195
af65d648 RM	196	static struct page *vdso32_pages[1];
	197
	198	#ifdef CONFIG_X86_64
	199
b6ad92d4	200	#define vdso32_sysenter() (boot_cpu_has(X86_FEATURE_SYSENTER32))
6a52e4b1	201	#define vdso32_syscall() (boot_cpu_has(X86_FEATURE_SYSCALL32))
af65d648 RM	202
	203	/* May not be __init: called during resume */
	204	void syscall32_cpu_init(void)
	205	{
af65d648 RM	206	/* Load these always in case some future AMD CPU supports
	207	SYSENTER from compat mode too. */
	208	checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
	209	checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
	210	checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
	211
	212	wrmsrl(MSR_CSTAR, ia32_cstar_target);
	213	}
	214
	215	#define compat_uses_vma 1
	216
	217	static inline void map_compat_vdso(int map)
	218	{
	219	}
	220
	221	#else /* CONFIG_X86_32 */
	222
	223	#define vdso32_sysenter() (boot_cpu_has(X86_FEATURE_SEP))
6a52e4b1	224	#define vdso32_syscall() (0)
af65d648	225
6fe940d6	226	void enable_sep_cpu(void)
1da177e4 LT	227	{
	228	int cpu = get_cpu();
	229	struct tss_struct *tss = &per_cpu(init_tss, cpu);
	230
6fe940d6 LS	231	if (!boot_cpu_has(X86_FEATURE_SEP)) {
	232	put_cpu();
	233	return;
	234	}
	235
a75c54f9	236	tss->x86_tss.ss1 = __KERNEL_CS;
faca6227	237	tss->x86_tss.sp1 = sizeof(struct tss_struct) + (unsigned long) tss;
1da177e4	238	wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
faca6227	239	wrmsr(MSR_IA32_SYSENTER_ESP, tss->x86_tss.sp1, 0);
0aa97fb2	240	wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long) ia32_sysenter_target, 0);
1da177e4 LT	241	put_cpu();
	242	}
	243
1dbf527c JF	244	static struct vm_area_struct gate_vma;
	245
	246	static int __init gate_vma_init(void)
	247	{
	248	gate_vma.vm_mm = NULL;
	249	gate_vma.vm_start = FIXADDR_USER_START;
	250	gate_vma.vm_end = FIXADDR_USER_END;
	251	gate_vma.vm_flags = VM_READ \| VM_MAYREAD \| VM_EXEC \| VM_MAYEXEC;
	252	gate_vma.vm_page_prot = __P101;
	253	/*
	254	* Make sure the vDSO gets into every core dump.
	255	* Dumping its contents makes post-mortem fully interpretable later
	256	* without matching up the same kernel and hardware config to see
	257	* what PC values meant.
	258	*/
	259	gate_vma.vm_flags \|= VM_ALWAYSDUMP;
	260	return 0;
	261	}
	262
af65d648	263	#define compat_uses_vma 0
1da177e4	264
1dbf527c JF	265	static void map_compat_vdso(int map)
	266	{
	267	static int vdso_mapped;
	268
	269	if (map == vdso_mapped)
	270	return;
	271
	272	vdso_mapped = map;
	273
af65d648	274	__set_fixmap(FIX_VDSO, page_to_pfn(vdso32_pages[0]) << PAGE_SHIFT,
1dbf527c JF	275	map ? PAGE_READONLY_EXEC : PAGE_NONE);
	276
	277	/* flush stray tlbs */
	278	flush_tlb_all();
	279	}
	280
af65d648 RM	281	#endif /* CONFIG_X86_64 */
af65d648 RM	282
a6c4e076	283	int __init sysenter_setup(void)
1da177e4	284	{
7d91d531	285	void syscall_page = (void )get_zeroed_page(GFP_ATOMIC);
d4f7a2c1 JF	286	const void *vsyscall;
	287	size_t vsyscall_len;
	288
af65d648	289	vdso32_pages[0] = virt_to_page(syscall_page);
1da177e4	290
af65d648	291	#ifdef CONFIG_X86_32
1dbf527c	292	gate_vma_init();
af65d648	293	#endif
1da177e4	294
6a52e4b1 JF	295	if (vdso32_syscall()) {
	296	vsyscall = &vdso32_syscall_start;
	297	vsyscall_len = &vdso32_syscall_end - &vdso32_syscall_start;
	298	} else if (vdso32_sysenter()){
af65d648 RM	299	vsyscall = &vdso32_sysenter_start;
af65d648 RM	300	vsyscall_len = &vdso32_sysenter_end - &vdso32_sysenter_start;
6a52e4b1 JF	301	} else {
	302	vsyscall = &vdso32_int80_start;
	303	vsyscall_len = &vdso32_int80_end - &vdso32_int80_start;
1da177e4 LT	304	}
1da177e4 LT	305
d4f7a2c1 JF	306	memcpy(syscall_page, vsyscall, vsyscall_len);
d4f7a2c1 JF	307	relocate_vdso(syscall_page);
1da177e4	308
1da177e4 LT	309	return 0;
1da177e4 LT	310	}
e6e5494c	311
e6e5494c	312	/* Setup a VMA at program startup for the vsyscall page */
fc5243d9	313	int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
e6e5494c	314	{
e6e5494c IM	315	struct mm_struct *mm = current->mm;
e6e5494c IM	316	unsigned long addr;
752783c0	317	int ret = 0;
1dbf527c	318	bool compat;
e6e5494c	319
5de253cc RM	320	if (vdso_enabled == VDSO_DISABLED)
	321	return 0;
	322
e6e5494c	323	down_write(&mm->mmap_sem);
e6e5494c	324
1dbf527c JF	325	/* Test compat mode once here, in case someone
	326	changes it via sysctl */
	327	compat = (vdso_enabled == VDSO_COMPAT);
	328
	329	map_compat_vdso(compat);
	330
	331	if (compat)
	332	addr = VDSO_HIGH_BASE;
	333	else {
	334	addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
	335	if (IS_ERR_VALUE(addr)) {
	336	ret = addr;
	337	goto up_fail;
	338	}
af65d648	339	}
1dbf527c	340
af65d648	341	if (compat_uses_vma \|\| !compat) {
1dbf527c JF	342	/*
	343	* MAYWRITE to allow gdb to COW and set breakpoints
	344	*
	345	* Make sure the vDSO gets into every core dump.
	346	* Dumping its contents makes post-mortem fully
	347	* interpretable later without matching up the same
	348	* kernel and hardware config to see what PC values
	349	* meant.
	350	*/
	351	ret = install_special_mapping(mm, addr, PAGE_SIZE,
	352	VM_READ\|VM_EXEC\|
	353	VM_MAYREAD\|VM_MAYWRITE\|VM_MAYEXEC\|
	354	VM_ALWAYSDUMP,
af65d648	355	vdso32_pages);
1dbf527c JF	356
	357	if (ret)
	358	goto up_fail;
	359	}
e6e5494c IM	360
	361	current->mm->context.vdso = (void *)addr;
	362	current_thread_info()->sysenter_return =
6c3652ef	363	VDSO32_SYMBOL(addr, SYSENTER_RETURN);
1dbf527c JF	364
1dbf527c JF	365	up_fail:
e6e5494c	366	up_write(&mm->mmap_sem);
1dbf527c	367
e6e5494c	368	return ret;
e6e5494c IM	369	}
e6e5494c IM	370
af65d648 RM	371	#ifdef CONFIG_X86_64
	372
	373	__initcall(sysenter_setup);
	374
a97f52e6 RM	375	#ifdef CONFIG_SYSCTL
	376	/* Register vsyscall32 into the ABI table */
	377	#include <linux/sysctl.h>
	378
	379	static ctl_table abi_table2[] = {
	380	{
	381	.procname = "vsyscall32",
	382	.data = &sysctl_vsyscall32,
	383	.maxlen = sizeof(int),
	384	.mode = 0644,
	385	.proc_handler = proc_dointvec
	386	},
	387	{}
	388	};
	389
	390	static ctl_table abi_root_table2[] = {
	391	{
	392	.ctl_name = CTL_ABI,
	393	.procname = "abi",
	394	.mode = 0555,
	395	.child = abi_table2
	396	},
	397	{}
	398	};
	399
	400	static __init int ia32_binfmt_init(void)
	401	{
	402	register_sysctl_table(abi_root_table2);
	403	return 0;
	404	}
	405	__initcall(ia32_binfmt_init);
	406	#endif
	407
af65d648 RM	408	#else /* CONFIG_X86_32 */
af65d648 RM	409
e6e5494c IM	410	const char arch_vma_name(struct vm_area_struct vma)
	411	{
	412	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
	413	return "[vdso]";
	414	return NULL;
	415	}
	416
	417	struct vm_area_struct get_gate_vma(struct task_struct tsk)
	418	{
1dbf527c JF	419	struct mm_struct *mm = tsk->mm;
	420
	421	/* Check to see if this task was created in compat vdso mode */
	422	if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE)
	423	return &gate_vma;
e6e5494c IM	424	return NULL;
	425	}
	426
	427	int in_gate_area(struct task_struct *task, unsigned long addr)
	428	{
2f30c007 JB	429	const struct vm_area_struct *vma = get_gate_vma(task);
	430
	431	return vma && addr >= vma->vm_start && addr < vma->vm_end;
e6e5494c IM	432	}
	433
	434	int in_gate_area_no_task(unsigned long addr)
	435	{
	436	return 0;
	437	}
af65d648 RM	438
af65d648 RM	439	#endif /* CONFIG_X86_64 */