[linux-2.6-block.git] / arch / arm64 / kernel / vdso.c

/*
 * VDSO implementation for AArch64 and vector page setup for AArch32.
 *
 * Copyright (C) 2012 ARM Limited
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 *
 * Author: Will Deacon <will.deacon@arm.com>
 */

#include <linux/cache.h>
#include <linux/clocksource.h>
#include <linux/elf.h>
#include <linux/err.h>
#include <linux/errno.h>
#include <linux/gfp.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/signal.h>
#include <linux/slab.h>
#include <linux/timekeeper_internal.h>
#include <linux/vmalloc.h>

#include <asm/cacheflush.h>
#include <asm/signal32.h>
#include <asm/vdso.h>
#include <asm/vdso_datapage.h>

extern char vdso_start[], vdso_end[];
static unsigned long vdso_pages __ro_after_init;

/*
 * The vDSO data page.
 */
static union {
	struct vdso_data	data;
	u8			page[PAGE_SIZE];
} vdso_data_store __page_aligned_data;
struct vdso_data *vdso_data = &vdso_data_store.data;

#ifdef CONFIG_COMPAT
/*
 * Create and map the vectors page for AArch32 tasks.
 */
static struct page *vectors_page[1] __ro_after_init;

static int __init alloc_vectors_page(void)
{
	extern char __kuser_helper_start[], __kuser_helper_end[];
	extern char __aarch32_sigret_code_start[], __aarch32_sigret_code_end[];

	int kuser_sz = __kuser_helper_end - __kuser_helper_start;
	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
	unsigned long vpage;

	vpage = get_zeroed_page(GFP_ATOMIC);

	if (!vpage)
		return -ENOMEM;

	/* kuser helpers */
	memcpy((void *)vpage + 0x1000 - kuser_sz, __kuser_helper_start,
		kuser_sz);

	/* sigreturn code */
	memcpy((void *)vpage + AARCH32_KERN_SIGRET_CODE_OFFSET,
               __aarch32_sigret_code_start, sigret_sz);

	flush_icache_range(vpage, vpage + PAGE_SIZE);
	vectors_page[0] = virt_to_page(vpage);

	return 0;
}
arch_initcall(alloc_vectors_page);

int aarch32_setup_vectors_page(struct linux_binprm *bprm, int uses_interp)
{
	struct mm_struct *mm = current->mm;
	unsigned long addr = AARCH32_VECTORS_BASE;
	static const struct vm_special_mapping spec = {
		.name	= "[vectors]",
		.pages	= vectors_page,

	};
	void *ret;

	if (down_write_killable(&mm->mmap_sem))
		return -EINTR;
	current->mm->context.vdso = (void *)addr;

	/* Map vectors page at the high address. */
	ret = _install_special_mapping(mm, addr, PAGE_SIZE,
				       VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC,
				       &spec);

	up_write(&mm->mmap_sem);

	return PTR_ERR_OR_ZERO(ret);
}
#endif /* CONFIG_COMPAT */

static int vdso_mremap(const struct vm_special_mapping *sm,
		struct vm_area_struct *new_vma)
{
	unsigned long new_size = new_vma->vm_end - new_vma->vm_start;
	unsigned long vdso_size = vdso_end - vdso_start;

	if (vdso_size != new_size)
		return -EINVAL;

	current->mm->context.vdso = (void *)new_vma->vm_start;

	return 0;
}

static struct vm_special_mapping vdso_spec[2] __ro_after_init = {
	{
		.name	= "[vvar]",
	},
	{
		.name	= "[vdso]",
		.mremap = vdso_mremap,
	},
};

static int __init vdso_init(void)
{
	int i;
	struct page **vdso_pagelist;
	unsigned long pfn;

	if (memcmp(vdso_start, "\177ELF", 4)) {
		pr_err("vDSO is not a valid ELF object!\n");
		return -EINVAL;
	}

	vdso_pages = (vdso_end - vdso_start) >> PAGE_SHIFT;
	pr_info("vdso: %ld pages (%ld code @ %p, %ld data @ %p)\n",
		vdso_pages + 1, vdso_pages, vdso_start, 1L, vdso_data);

	/* Allocate the vDSO pagelist, plus a page for the data. */
	vdso_pagelist = kcalloc(vdso_pages + 1, sizeof(struct page *),
				GFP_KERNEL);
	if (vdso_pagelist == NULL)
		return -ENOMEM;

	/* Grab the vDSO data page. */
	vdso_pagelist[0] = phys_to_page(__pa_symbol(vdso_data));


	/* Grab the vDSO code pages. */
	pfn = sym_to_pfn(vdso_start);

	for (i = 0; i < vdso_pages; i++)
		vdso_pagelist[i + 1] = pfn_to_page(pfn + i);

	vdso_spec[0].pages = &vdso_pagelist[0];
	vdso_spec[1].pages = &vdso_pagelist[1];

	return 0;
}
arch_initcall(vdso_init);

int arch_setup_additional_pages(struct linux_binprm *bprm,
				int uses_interp)
{
	struct mm_struct *mm = current->mm;
	unsigned long vdso_base, vdso_text_len, vdso_mapping_len;
	void *ret;

	vdso_text_len = vdso_pages << PAGE_SHIFT;
	/* Be sure to map the data page */
	vdso_mapping_len = vdso_text_len + PAGE_SIZE;

	if (down_write_killable(&mm->mmap_sem))
		return -EINTR;
	vdso_base = get_unmapped_area(NULL, 0, vdso_mapping_len, 0, 0);
	if (IS_ERR_VALUE(vdso_base)) {
		ret = ERR_PTR(vdso_base);
		goto up_fail;
	}
	ret = _install_special_mapping(mm, vdso_base, PAGE_SIZE,
				       VM_READ|VM_MAYREAD,
				       &vdso_spec[0]);
	if (IS_ERR(ret))
		goto up_fail;

	vdso_base += PAGE_SIZE;
	mm->context.vdso = (void *)vdso_base;
	ret = _install_special_mapping(mm, vdso_base, vdso_text_len,
				       VM_READ|VM_EXEC|
				       VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC,
				       &vdso_spec[1]);
	if (IS_ERR(ret))
		goto up_fail;


	up_write(&mm->mmap_sem);
	return 0;

up_fail:
	mm->context.vdso = NULL;
	up_write(&mm->mmap_sem);
	return PTR_ERR(ret);
}

/*
 * Update the vDSO data page to keep in sync with kernel timekeeping.
 */
void update_vsyscall(struct timekeeper *tk)
{
	u32 use_syscall = !tk->tkr_mono.clock->archdata.vdso_direct;

	++vdso_data->tb_seq_count;
	smp_wmb();

	vdso_data->use_syscall			= use_syscall;
	vdso_data->xtime_coarse_sec		= tk->xtime_sec;
	vdso_data->xtime_coarse_nsec		= tk->tkr_mono.xtime_nsec >>
							tk->tkr_mono.shift;
	vdso_data->wtm_clock_sec		= tk->wall_to_monotonic.tv_sec;
	vdso_data->wtm_clock_nsec		= tk->wall_to_monotonic.tv_nsec;

	if (!use_syscall) {
		/* tkr_mono.cycle_last == tkr_raw.cycle_last */
		vdso_data->cs_cycle_last	= tk->tkr_mono.cycle_last;
		vdso_data->raw_time_sec         = tk->raw_sec;
		vdso_data->raw_time_nsec        = tk->tkr_raw.xtime_nsec;
		vdso_data->xtime_clock_sec	= tk->xtime_sec;
		vdso_data->xtime_clock_nsec	= tk->tkr_mono.xtime_nsec;
		vdso_data->cs_mono_mult		= tk->tkr_mono.mult;
		vdso_data->cs_raw_mult		= tk->tkr_raw.mult;
		/* tkr_mono.shift == tkr_raw.shift */
		vdso_data->cs_shift		= tk->tkr_mono.shift;
	}

	smp_wmb();
	++vdso_data->tb_seq_count;
}

void update_vsyscall_tz(void)
{
	vdso_data->tz_minuteswest	= sys_tz.tz_minuteswest;
	vdso_data->tz_dsttime		= sys_tz.tz_dsttime;
}
Commit	Line	Data
9031fefd WD	1	/*
	2	* VDSO implementation for AArch64 and vector page setup for AArch32.
	3	*
	4	* Copyright (C) 2012 ARM Limited
	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License version 2 as
	8	* published by the Free Software Foundation.
	9	*
	10	* This program is distributed in the hope that it will be useful,
	11	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	12	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	13	* GNU General Public License for more details.
	14	*
	15	* You should have received a copy of the GNU General Public License
	16	* along with this program. If not, see <http://www.gnu.org/licenses/>.
	17	*
	18	* Author: Will Deacon <will.deacon@arm.com>
	19	*/
	20
5a9e3e15	21	#include <linux/cache.h>
9031fefd WD	22	#include <linux/clocksource.h>
	23	#include <linux/elf.h>
	24	#include <linux/err.h>
	25	#include <linux/errno.h>
	26	#include <linux/gfp.h>
5a9e3e15	27	#include <linux/kernel.h>
9031fefd WD	28	#include <linux/mm.h>
	29	#include <linux/sched.h>
	30	#include <linux/signal.h>
	31	#include <linux/slab.h>
c60b0c28	32	#include <linux/timekeeper_internal.h>
9031fefd WD	33	#include <linux/vmalloc.h>
	34
	35	#include <asm/cacheflush.h>
	36	#include <asm/signal32.h>
	37	#include <asm/vdso.h>
	38	#include <asm/vdso_datapage.h>
	39
dbbb08f5	40	extern char vdso_start[], vdso_end[];
5a9e3e15	41	static unsigned long vdso_pages __ro_after_init;
9031fefd WD	42
	43	/*
	44	* The vDSO data page.
	45	*/
	46	static union {
	47	struct vdso_data data;
	48	u8 page[PAGE_SIZE];
	49	} vdso_data_store __page_aligned_data;
	50	struct vdso_data *vdso_data = &vdso_data_store.data;
	51
	52	#ifdef CONFIG_COMPAT
	53	/*
	54	* Create and map the vectors page for AArch32 tasks.
	55	*/
5a9e3e15	56	static struct page *vectors_page[1] __ro_after_init;
9031fefd	57
1aed28f9	58	static int __init alloc_vectors_page(void)
9031fefd WD	59	{
9031fefd WD	60	extern char __kuser_helper_start[], __kuser_helper_end[];
a1d5ebaf ML	61	extern char __aarch32_sigret_code_start[], __aarch32_sigret_code_end[];
a1d5ebaf ML	62
9031fefd	63	int kuser_sz = __kuser_helper_end - __kuser_helper_start;
a1d5ebaf	64	int sigret_sz = __aarch32_sigret_code_end - __aarch32_sigret_code_start;
9031fefd WD	65	unsigned long vpage;
	66
	67	vpage = get_zeroed_page(GFP_ATOMIC);
	68
	69	if (!vpage)
	70	return -ENOMEM;
	71
	72	/* kuser helpers */
	73	memcpy((void *)vpage + 0x1000 - kuser_sz, __kuser_helper_start,
	74	kuser_sz);
	75
	76	/* sigreturn code */
	77	memcpy((void *)vpage + AARCH32_KERN_SIGRET_CODE_OFFSET,
a1d5ebaf	78	__aarch32_sigret_code_start, sigret_sz);
9031fefd WD	79
	80	flush_icache_range(vpage, vpage + PAGE_SIZE);
	81	vectors_page[0] = virt_to_page(vpage);
	82
	83	return 0;
	84	}
	85	arch_initcall(alloc_vectors_page);
	86
	87	int aarch32_setup_vectors_page(struct linux_binprm *bprm, int uses_interp)
	88	{
	89	struct mm_struct *mm = current->mm;
	90	unsigned long addr = AARCH32_VECTORS_BASE;
b6d081bd	91	static const struct vm_special_mapping spec = {
2fea7f6c WD	92	.name = "[vectors]",
	93	.pages = vectors_page,
	94
	95	};
	96	void *ret;
9031fefd	97
69048176 MH	98	if (down_write_killable(&mm->mmap_sem))
69048176 MH	99	return -EINTR;
9031fefd WD	100	current->mm->context.vdso = (void *)addr;
	101
	102	/* Map vectors page at the high address. */
2fea7f6c WD	103	ret = _install_special_mapping(mm, addr, PAGE_SIZE,
	104	VM_READ\|VM_EXEC\|VM_MAYREAD\|VM_MAYEXEC,
	105	&spec);
9031fefd WD	106
	107	up_write(&mm->mmap_sem);
	108
2fea7f6c	109	return PTR_ERR_OR_ZERO(ret);
9031fefd WD	110	}
	111	#endif /* CONFIG_COMPAT */
	112
73958695 DS	113	static int vdso_mremap(const struct vm_special_mapping *sm,
	114	struct vm_area_struct *new_vma)
	115	{
	116	unsigned long new_size = new_vma->vm_end - new_vma->vm_start;
	117	unsigned long vdso_size = vdso_end - vdso_start;
	118
	119	if (vdso_size != new_size)
	120	return -EINVAL;
	121
	122	current->mm->context.vdso = (void *)new_vma->vm_start;
	123
	124	return 0;
	125	}
	126
5a9e3e15 JZ	127	static struct vm_special_mapping vdso_spec[2] __ro_after_init = {
	128	{
	129	.name = "[vvar]",
	130	},
	131	{
	132	.name = "[vdso]",
73958695	133	.mremap = vdso_mremap,
5a9e3e15 JZ	134	},
5a9e3e15 JZ	135	};
2fea7f6c	136
9031fefd WD	137	static int __init vdso_init(void)
9031fefd WD	138	{
16fb1a9b	139	int i;
5a9e3e15	140	struct page **vdso_pagelist;
2077be67	141	unsigned long pfn;
16fb1a9b	142
dbbb08f5	143	if (memcmp(vdso_start, "\177ELF", 4)) {
16fb1a9b NL	144	pr_err("vDSO is not a valid ELF object!\n");
	145	return -EINVAL;
	146	}
9031fefd	147
dbbb08f5	148	vdso_pages = (vdso_end - vdso_start) >> PAGE_SHIFT;
601255ae	149	pr_info("vdso: %ld pages (%ld code @ %p, %ld data @ %p)\n",
dbbb08f5	150	vdso_pages + 1, vdso_pages, vdso_start, 1L, vdso_data);
9031fefd WD	151
9031fefd WD	152	/* Allocate the vDSO pagelist, plus a page for the data. */
16fb1a9b	153	vdso_pagelist = kcalloc(vdso_pages + 1, sizeof(struct page *),
9031fefd	154	GFP_KERNEL);
16fb1a9b	155	if (vdso_pagelist == NULL)
9031fefd	156	return -ENOMEM;
9031fefd	157
601255ae	158	/* Grab the vDSO data page. */
2077be67 LA	159	vdso_pagelist[0] = phys_to_page(__pa_symbol(vdso_data));
2077be67 LA	160
601255ae	161
9031fefd	162	/* Grab the vDSO code pages. */
dbbb08f5	163	pfn = sym_to_pfn(vdso_start);
2077be67	164
16fb1a9b	165	for (i = 0; i < vdso_pages; i++)
2077be67	166	vdso_pagelist[i + 1] = pfn_to_page(pfn + i);
9031fefd	167
5a9e3e15 JZ	168	vdso_spec[0].pages = &vdso_pagelist[0];
5a9e3e15 JZ	169	vdso_spec[1].pages = &vdso_pagelist[1];
2fea7f6c	170
16fb1a9b	171	return 0;
9031fefd WD	172	}
	173	arch_initcall(vdso_init);
	174
	175	int arch_setup_additional_pages(struct linux_binprm *bprm,
	176	int uses_interp)
	177	{
	178	struct mm_struct *mm = current->mm;
87154938	179	unsigned long vdso_base, vdso_text_len, vdso_mapping_len;
2fea7f6c	180	void *ret;
9031fefd	181
87154938	182	vdso_text_len = vdso_pages << PAGE_SHIFT;
9031fefd	183	/* Be sure to map the data page */
87154938	184	vdso_mapping_len = vdso_text_len + PAGE_SIZE;
9031fefd	185
69048176 MH	186	if (down_write_killable(&mm->mmap_sem))
69048176 MH	187	return -EINTR;
9031fefd WD	188	vdso_base = get_unmapped_area(NULL, 0, vdso_mapping_len, 0, 0);
9031fefd WD	189	if (IS_ERR_VALUE(vdso_base)) {
2fea7f6c	190	ret = ERR_PTR(vdso_base);
9031fefd WD	191	goto up_fail;
9031fefd WD	192	}
601255ae WD	193	ret = _install_special_mapping(mm, vdso_base, PAGE_SIZE,
601255ae WD	194	VM_READ\|VM_MAYREAD,
2fea7f6c WD	195	&vdso_spec[0]);
2fea7f6c WD	196	if (IS_ERR(ret))
87154938 WD	197	goto up_fail;
87154938 WD	198
601255ae WD	199	vdso_base += PAGE_SIZE;
	200	mm->context.vdso = (void *)vdso_base;
	201	ret = _install_special_mapping(mm, vdso_base, vdso_text_len,
	202	VM_READ\|VM_EXEC\|
	203	VM_MAYREAD\|VM_MAYWRITE\|VM_MAYEXEC,
2fea7f6c WD	204	&vdso_spec[1]);
2fea7f6c WD	205	if (IS_ERR(ret))
9031fefd	206	goto up_fail;
9031fefd	207
601255ae	208
9031fefd	209	up_write(&mm->mmap_sem);
87154938	210	return 0;
9031fefd	211
87154938 WD	212	up_fail:
	213	mm->context.vdso = NULL;
	214	up_write(&mm->mmap_sem);
2fea7f6c	215	return PTR_ERR(ret);
9031fefd WD	216	}
9031fefd WD	217
9031fefd WD	218	/*
	219	* Update the vDSO data page to keep in sync with kernel timekeeping.
	220	*/
c60b0c28	221	void update_vsyscall(struct timekeeper *tk)
9031fefd	222	{
1d8f51d4	223	u32 use_syscall = !tk->tkr_mono.clock->archdata.vdso_direct;
9031fefd WD	224
	225	++vdso_data->tb_seq_count;
	226	smp_wmb();
	227
9031fefd	228	vdso_data->use_syscall = use_syscall;
878854a3 NL	229	vdso_data->xtime_coarse_sec = tk->xtime_sec;
	230	vdso_data->xtime_coarse_nsec = tk->tkr_mono.xtime_nsec >>
	231	tk->tkr_mono.shift;
d4022a33 NL	232	vdso_data->wtm_clock_sec = tk->wall_to_monotonic.tv_sec;
d4022a33 NL	233	vdso_data->wtm_clock_nsec = tk->wall_to_monotonic.tv_nsec;
9031fefd WD	234
9031fefd WD	235	if (!use_syscall) {
49eea433	236	/* tkr_mono.cycle_last == tkr_raw.cycle_last */
876e7881	237	vdso_data->cs_cycle_last = tk->tkr_mono.cycle_last;
fc6eead7 JS	238	vdso_data->raw_time_sec = tk->raw_sec;
fc6eead7 JS	239	vdso_data->raw_time_nsec = tk->tkr_raw.xtime_nsec;
c60b0c28	240	vdso_data->xtime_clock_sec = tk->xtime_sec;
876e7881	241	vdso_data->xtime_clock_nsec = tk->tkr_mono.xtime_nsec;
49eea433 KB	242	vdso_data->cs_mono_mult = tk->tkr_mono.mult;
	243	vdso_data->cs_raw_mult = tk->tkr_raw.mult;
	244	/* tkr_mono.shift == tkr_raw.shift */
876e7881	245	vdso_data->cs_shift = tk->tkr_mono.shift;
9031fefd WD	246	}
	247
	248	smp_wmb();
	249	++vdso_data->tb_seq_count;
	250	}
	251
	252	void update_vsyscall_tz(void)
	253	{
9031fefd WD	254	vdso_data->tz_minuteswest = sys_tz.tz_minuteswest;
9031fefd WD	255	vdso_data->tz_dsttime = sys_tz.tz_dsttime;
9031fefd	256	}