[GitHub/exynos8895/android_kernel_samsung_universal8895.git] / arch / arm / kernel / vdso.c

/*
 * Adapted from arm64 version.
 *
 * Copyright (C) 2012 ARM Limited
 * Copyright (C) 2015 Mentor Graphics Corporation.
 *
 * 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/>.
 */

#include <linux/cache.h>
#include <linux/elf.h>
#include <linux/err.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/of.h>
#include <linux/printk.h>
#include <linux/slab.h>
#include <linux/timekeeper_internal.h>
#include <linux/vmalloc.h>
#include <asm/arch_timer.h>
#include <asm/barrier.h>
#include <asm/cacheflush.h>
#include <asm/page.h>
#include <asm/vdso.h>
#include <asm/vdso_datapage.h>
#include <clocksource/arm_arch_timer.h>

#define MAX_SYMNAME	64

static struct page **vdso_text_pagelist;

/* Total number of pages needed for the data and text portions of the VDSO. */
unsigned int vdso_total_pages __ro_after_init;

/*
 * The VDSO data page.
 */
static union vdso_data_store vdso_data_store __page_aligned_data;
static struct vdso_data *vdso_data = &vdso_data_store.data;

static struct page *vdso_data_page __ro_after_init;
static const struct vm_special_mapping vdso_data_mapping = {
	.name = "[vvar]",
	.pages = &vdso_data_page,
};

static struct vm_special_mapping vdso_text_mapping __ro_after_init = {
	.name = "[vdso]",
};

struct elfinfo {
	Elf32_Ehdr	*hdr;		/* ptr to ELF */
	Elf32_Sym	*dynsym;	/* ptr to .dynsym section */
	unsigned long	dynsymsize;	/* size of .dynsym section */
	char		*dynstr;	/* ptr to .dynstr section */
};

/* Cached result of boot-time check for whether the arch timer exists,
 * and if so, whether the virtual counter is useable.
 */
static bool cntvct_ok __ro_after_init;

static bool __init cntvct_functional(void)
{
	struct device_node *np;
	bool ret = false;

	if (!IS_ENABLED(CONFIG_ARM_ARCH_TIMER))
		goto out;

	/* The arm_arch_timer core should export
	 * arch_timer_use_virtual or similar so we don't have to do
	 * this.
	 */
	np = of_find_compatible_node(NULL, NULL, "arm,armv7-timer");
	if (!np)
		goto out_put;

	if (of_property_read_bool(np, "arm,cpu-registers-not-fw-configured"))
		goto out_put;

	ret = true;

out_put:
	of_node_put(np);
out:
	return ret;
}

static void * __init find_section(Elf32_Ehdr *ehdr, const char *name,
				  unsigned long *size)
{
	Elf32_Shdr *sechdrs;
	unsigned int i;
	char *secnames;

	/* Grab section headers and strings so we can tell who is who */
	sechdrs = (void *)ehdr + ehdr->e_shoff;
	secnames = (void *)ehdr + sechdrs[ehdr->e_shstrndx].sh_offset;

	/* Find the section they want */
	for (i = 1; i < ehdr->e_shnum; i++) {
		if (strcmp(secnames + sechdrs[i].sh_name, name) == 0) {
			if (size)
				*size = sechdrs[i].sh_size;
			return (void *)ehdr + sechdrs[i].sh_offset;
		}
	}

	if (size)
		*size = 0;
	return NULL;
}

static Elf32_Sym * __init find_symbol(struct elfinfo *lib, const char *symname)
{
	unsigned int i;

	for (i = 0; i < (lib->dynsymsize / sizeof(Elf32_Sym)); i++) {
		char name[MAX_SYMNAME], *c;

		if (lib->dynsym[i].st_name == 0)
			continue;
		strlcpy(name, lib->dynstr + lib->dynsym[i].st_name,
			MAX_SYMNAME);
		c = strchr(name, '@');
		if (c)
			*c = 0;
		if (strcmp(symname, name) == 0)
			return &lib->dynsym[i];
	}
	return NULL;
}

static void __init vdso_nullpatch_one(struct elfinfo *lib, const char *symname)
{
	Elf32_Sym *sym;

	sym = find_symbol(lib, symname);
	if (!sym)
		return;

	sym->st_name = 0;
}

static void __init patch_vdso(void *ehdr)
{
	struct elfinfo einfo;

	einfo = (struct elfinfo) {
		.hdr = ehdr,
	};

	einfo.dynsym = find_section(einfo.hdr, ".dynsym", &einfo.dynsymsize);
	einfo.dynstr = find_section(einfo.hdr, ".dynstr", NULL);

	/* If the virtual counter is absent or non-functional we don't
	 * want programs to incur the slight additional overhead of
	 * dispatching through the VDSO only to fall back to syscalls.
	 */
	if (!cntvct_ok) {
		vdso_nullpatch_one(&einfo, "__vdso_gettimeofday");
		vdso_nullpatch_one(&einfo, "__vdso_clock_gettime");
	}
}

static int __init vdso_init(void)
{
	unsigned int text_pages;
	int i;

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

	text_pages = (&vdso_end - &vdso_start) >> PAGE_SHIFT;
	pr_debug("vdso: %i text pages at base %p\n", text_pages, &vdso_start);

	/* Allocate the VDSO text pagelist */
	vdso_text_pagelist = kcalloc(text_pages, sizeof(struct page *),
				     GFP_KERNEL);
	if (vdso_text_pagelist == NULL)
		return -ENOMEM;

	/* Grab the VDSO data page. */
	vdso_data_page = virt_to_page(vdso_data);

	/* Grab the VDSO text pages. */
	for (i = 0; i < text_pages; i++) {
		struct page *page;

		page = virt_to_page(&vdso_start + i * PAGE_SIZE);
		vdso_text_pagelist[i] = page;
	}

	vdso_text_mapping.pages = vdso_text_pagelist;

	vdso_total_pages = 1; /* for the data/vvar page */
	vdso_total_pages += text_pages;

	cntvct_ok = cntvct_functional();

	patch_vdso(&vdso_start);

	return 0;
}
arch_initcall(vdso_init);

static int install_vvar(struct mm_struct *mm, unsigned long addr)
{
	struct vm_area_struct *vma;

	vma = _install_special_mapping(mm, addr, PAGE_SIZE,
				       VM_READ | VM_MAYREAD,
				       &vdso_data_mapping);

	return PTR_ERR_OR_ZERO(vma);
}

/* assumes mmap_sem is write-locked */
void arm_install_vdso(struct mm_struct *mm, unsigned long addr)
{
	struct vm_area_struct *vma;
	unsigned long len;

	mm->context.vdso = 0;

	if (vdso_text_pagelist == NULL)
		return;

	if (install_vvar(mm, addr))
		return;

	/* Account for vvar page. */
	addr += PAGE_SIZE;
	len = (vdso_total_pages - 1) << PAGE_SHIFT;

	vma = _install_special_mapping(mm, addr, len,
		VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
		&vdso_text_mapping);

	if (!IS_ERR(vma))
		mm->context.vdso = addr;
}

static void vdso_write_begin(struct vdso_data *vdata)
{
	++vdso_data->tb_seq_count;
	smp_wmb(); /* Pairs with smp_rmb in vdso_read_retry */
}

static void vdso_write_end(struct vdso_data *vdata)
{
	smp_wmb(); /* Pairs with smp_rmb in vdso_read_begin */
	++vdso_data->tb_seq_count;
}

static bool tk_is_cntvct(const struct timekeeper *tk)
{
	if (!IS_ENABLED(CONFIG_ARM_ARCH_TIMER))
		return false;

	if (strcmp(tk->tkr_mono.clock->name, "arch_sys_counter") != 0)
		return false;

	return true;
}

/**
 * update_vsyscall - update the vdso data page
 *
 * Increment the sequence counter, making it odd, indicating to
 * userspace that an update is in progress.  Update the fields used
 * for coarse clocks and, if the architected system timer is in use,
 * the fields used for high precision clocks.  Increment the sequence
 * counter again, making it even, indicating to userspace that the
 * update is finished.
 *
 * Userspace is expected to sample tb_seq_count before reading any other
 * fields from the data page.  If tb_seq_count is odd, userspace is
 * expected to wait until it becomes even.  After copying data from
 * the page, userspace must sample tb_seq_count again; if it has changed
 * from its previous value, userspace must retry the whole sequence.
 *
 * Calls to update_vsyscall are serialized by the timekeeping core.
 */
void update_vsyscall(struct timekeeper *tk)
{
	struct timespec64 *wtm = &tk->wall_to_monotonic;

	if (!cntvct_ok) {
		/* The entry points have been zeroed, so there is no
		 * point in updating the data page.
		 */
		return;
	}

	vdso_write_begin(vdso_data);

	vdso_data->use_syscall			= !tk_is_cntvct(tk);
	vdso_data->xtime_coarse_sec		= tk->xtime_sec;
	vdso_data->xtime_coarse_nsec		= (u32)(tk->tkr_mono.xtime_nsec >>
							tk->tkr_mono.shift);
	vdso_data->wtm_clock_sec		= wtm->tv_sec;
	vdso_data->wtm_clock_nsec		= wtm->tv_nsec;

	if (!vdso_data->use_syscall) {
		vdso_data->cs_cycle_last	= tk->tkr_mono.cycle_last;
		vdso_data->xtime_clock_sec	= tk->xtime_sec;
		vdso_data->xtime_clock_snsec	= tk->tkr_mono.xtime_nsec;
		vdso_data->cs_mono_mult		= tk->tkr_mono.mult;
		/* tkr_mono.shift == tkr_raw.shift */
		vdso_data->cs_shift		= tk->tkr_mono.shift;
		vdso_data->cs_mask		= tk->tkr_mono.mask;
	}

	vdso_write_end(vdso_data);

	flush_dcache_page(virt_to_page(vdso_data));
}

void update_vsyscall_tz(void)
{
	vdso_data->tz_minuteswest	= sys_tz.tz_minuteswest;
	vdso_data->tz_dsttime		= sys_tz.tz_dsttime;
	flush_dcache_page(virt_to_page(vdso_data));
}
Commit	Line	Data
ecf99a43 NL	1	/*
	2	* Adapted from arm64 version.
	3	*
	4	* Copyright (C) 2012 ARM Limited
	5	* Copyright (C) 2015 Mentor Graphics Corporation.
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License version 2 as
	9	* published by the Free Software Foundation.
	10	*
	11	* This program is distributed in the hope that it will be useful,
	12	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	13	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	14	* GNU General Public License for more details.
	15	*
	16	* You should have received a copy of the GNU General Public License
	17	* along with this program. If not, see <http://www.gnu.org/licenses/>.
	18	*/
	19
b4b7a246	20	#include <linux/cache.h>
ecf99a43 NL	21	#include <linux/elf.h>
	22	#include <linux/err.h>
	23	#include <linux/kernel.h>
	24	#include <linux/mm.h>
	25	#include <linux/of.h>
	26	#include <linux/printk.h>
	27	#include <linux/slab.h>
	28	#include <linux/timekeeper_internal.h>
	29	#include <linux/vmalloc.h>
	30	#include <asm/arch_timer.h>
	31	#include <asm/barrier.h>
	32	#include <asm/cacheflush.h>
	33	#include <asm/page.h>
	34	#include <asm/vdso.h>
	35	#include <asm/vdso_datapage.h>
	36	#include <clocksource/arm_arch_timer.h>
	37
	38	#define MAX_SYMNAME 64
	39
	40	static struct page **vdso_text_pagelist;
	41
	42	/* Total number of pages needed for the data and text portions of the VDSO. */
b4b7a246	43	unsigned int vdso_total_pages __ro_after_init;
ecf99a43 NL	44
	45	/*
	46	* The VDSO data page.
	47	*/
	48	static union vdso_data_store vdso_data_store __page_aligned_data;
	49	static struct vdso_data *vdso_data = &vdso_data_store.data;
	50
b4b7a246 JZ	51	static struct page *vdso_data_page __ro_after_init;
b4b7a246 JZ	52	static const struct vm_special_mapping vdso_data_mapping = {
ecf99a43 NL	53	.name = "[vvar]",
	54	.pages = &vdso_data_page,
	55	};
	56
b4b7a246	57	static struct vm_special_mapping vdso_text_mapping __ro_after_init = {
ecf99a43 NL	58	.name = "[vdso]",
	59	};
	60
	61	struct elfinfo {
	62	Elf32_Ehdr hdr; / ptr to ELF */
	63	Elf32_Sym dynsym; / ptr to .dynsym section */
	64	unsigned long dynsymsize; /* size of .dynsym section */
	65	char dynstr; / ptr to .dynstr section */
	66	};
	67
	68	/* Cached result of boot-time check for whether the arch timer exists,
	69	* and if so, whether the virtual counter is useable.
	70	*/
b4b7a246	71	static bool cntvct_ok __ro_after_init;
ecf99a43 NL	72
	73	static bool __init cntvct_functional(void)
	74	{
	75	struct device_node *np;
	76	bool ret = false;
	77
	78	if (!IS_ENABLED(CONFIG_ARM_ARCH_TIMER))
	79	goto out;
	80
	81	/* The arm_arch_timer core should export
	82	* arch_timer_use_virtual or similar so we don't have to do
	83	* this.
	84	*/
	85	np = of_find_compatible_node(NULL, NULL, "arm,armv7-timer");
	86	if (!np)
	87	goto out_put;
	88
	89	if (of_property_read_bool(np, "arm,cpu-registers-not-fw-configured"))
	90	goto out_put;
	91
	92	ret = true;
	93
	94	out_put:
	95	of_node_put(np);
	96	out:
	97	return ret;
	98	}
	99
	100	static void * __init find_section(Elf32_Ehdr ehdr, const char name,
	101	unsigned long *size)
	102	{
	103	Elf32_Shdr *sechdrs;
	104	unsigned int i;
	105	char *secnames;
	106
	107	/* Grab section headers and strings so we can tell who is who */
	108	sechdrs = (void *)ehdr + ehdr->e_shoff;
	109	secnames = (void *)ehdr + sechdrs[ehdr->e_shstrndx].sh_offset;
	110
	111	/* Find the section they want */
	112	for (i = 1; i < ehdr->e_shnum; i++) {
	113	if (strcmp(secnames + sechdrs[i].sh_name, name) == 0) {
	114	if (size)
	115	*size = sechdrs[i].sh_size;
	116	return (void *)ehdr + sechdrs[i].sh_offset;
	117	}
	118	}
	119
	120	if (size)
	121	*size = 0;
	122	return NULL;
	123	}
	124
	125	static Elf32_Sym * __init find_symbol(struct elfinfo lib, const char symname)
	126	{
	127	unsigned int i;
	128
	129	for (i = 0; i < (lib->dynsymsize / sizeof(Elf32_Sym)); i++) {
	130	char name[MAX_SYMNAME], *c;
	131
	132	if (lib->dynsym[i].st_name == 0)
	133	continue;
	134	strlcpy(name, lib->dynstr + lib->dynsym[i].st_name,
	135	MAX_SYMNAME);
136	c = strchr(name, '@');
137	if (c)
138	*c = 0;
139	if (strcmp(symname, name) == 0)
140	return &lib->dynsym[i];
141	}
142	return NULL;
143	}
144
145	static void __init vdso_nullpatch_one(struct elfinfo lib, const char symname)
146	{
147	Elf32_Sym *sym;
148
149	sym = find_symbol(lib, symname);
150	if (!sym)
151	return;
152
153	sym->st_name = 0;
154	}
155
156	static void __init patch_vdso(void *ehdr)
157	{
158	struct elfinfo einfo;
159
160	einfo = (struct elfinfo) {
161	.hdr = ehdr,
162	};
163
164	einfo.dynsym = find_section(einfo.hdr, ".dynsym", &einfo.dynsymsize);
165	einfo.dynstr = find_section(einfo.hdr, ".dynstr", NULL);
166
167	/* If the virtual counter is absent or non-functional we don't
168	* want programs to incur the slight additional overhead of
169	* dispatching through the VDSO only to fall back to syscalls.
170	*/
171	if (!cntvct_ok) {
172	vdso_nullpatch_one(&einfo, "__vdso_gettimeofday");
173	vdso_nullpatch_one(&einfo, "__vdso_clock_gettime");
174	}
175	}
176
177	static int __init vdso_init(void)
178	{
179	unsigned int text_pages;
180	int i;
181
182	if (memcmp(&vdso_start, "\177ELF", 4)) {
183	pr_err("VDSO is not a valid ELF object!\n");
184	return -ENOEXEC;
185	}
186
187	text_pages = (&vdso_end - &vdso_start) >> PAGE_SHIFT;
188	pr_debug("vdso: %i text pages at base %p\n", text_pages, &vdso_start);
189
190	/* Allocate the VDSO text pagelist */
191	vdso_text_pagelist = kcalloc(text_pages, sizeof(struct page *),
192	GFP_KERNEL);
193	if (vdso_text_pagelist == NULL)
194	return -ENOMEM;
195
196	/* Grab the VDSO data page. */
197	vdso_data_page = virt_to_page(vdso_data);
198
199	/* Grab the VDSO text pages. */
200	for (i = 0; i < text_pages; i++) {
201	struct page *page;
202
203	page = virt_to_page(&vdso_start + i * PAGE_SIZE);
204	vdso_text_pagelist[i] = page;
205	}
206
207	vdso_text_mapping.pages = vdso_text_pagelist;
208
209	vdso_total_pages = 1; /* for the data/vvar page */
210	vdso_total_pages += text_pages;
211
212	cntvct_ok = cntvct_functional();
213
214	patch_vdso(&vdso_start);
215
216	return 0;
217	}
218	arch_initcall(vdso_init);
219
220	static int install_vvar(struct mm_struct *mm, unsigned long addr)
221	{
222	struct vm_area_struct *vma;
223
224	vma = _install_special_mapping(mm, addr, PAGE_SIZE,
225	VM_READ \| VM_MAYREAD,
226	&vdso_data_mapping);
227
e1feee06	228	return PTR_ERR_OR_ZERO(vma);
ecf99a43 NL	229	}
	230
	231	/* assumes mmap_sem is write-locked */
	232	void arm_install_vdso(struct mm_struct *mm, unsigned long addr)
	233	{
	234	struct vm_area_struct *vma;
	235	unsigned long len;
	236
	237	mm->context.vdso = 0;
	238
	239	if (vdso_text_pagelist == NULL)
	240	return;
	241
	242	if (install_vvar(mm, addr))
	243	return;
	244
	245	/* Account for vvar page. */
	246	addr += PAGE_SIZE;
	247	len = (vdso_total_pages - 1) << PAGE_SHIFT;
	248
	249	vma = _install_special_mapping(mm, addr, len,
	250	VM_READ \| VM_EXEC \| VM_MAYREAD \| VM_MAYWRITE \| VM_MAYEXEC,
	251	&vdso_text_mapping);
	252
	253	if (!IS_ERR(vma))
	254	mm->context.vdso = addr;
	255	}
	256
	257	static void vdso_write_begin(struct vdso_data *vdata)
	258	{
ba88f57f	259	++vdso_data->tb_seq_count;
ecf99a43 NL	260	smp_wmb(); /* Pairs with smp_rmb in vdso_read_retry */
	261	}
	262
	263	static void vdso_write_end(struct vdso_data *vdata)
	264	{
	265	smp_wmb(); /* Pairs with smp_rmb in vdso_read_begin */
ba88f57f	266	++vdso_data->tb_seq_count;
ecf99a43 NL	267	}
	268
	269	static bool tk_is_cntvct(const struct timekeeper *tk)
	270	{
	271	if (!IS_ENABLED(CONFIG_ARM_ARCH_TIMER))
	272	return false;
	273
bb0fd7ab	274	if (strcmp(tk->tkr_mono.clock->name, "arch_sys_counter") != 0)
ecf99a43 NL	275	return false;
	276
	277	return true;
	278	}
	279
	280	/**
	281	* update_vsyscall - update the vdso data page
	282	*
	283	* Increment the sequence counter, making it odd, indicating to
	284	* userspace that an update is in progress. Update the fields used
	285	* for coarse clocks and, if the architected system timer is in use,
	286	* the fields used for high precision clocks. Increment the sequence
	287	* counter again, making it even, indicating to userspace that the
	288	* update is finished.
	289	*
ba88f57f MS	290	* Userspace is expected to sample tb_seq_count before reading any other
ba88f57f MS	291	* fields from the data page. If tb_seq_count is odd, userspace is
ecf99a43	292	* expected to wait until it becomes even. After copying data from
ba88f57f	293	* the page, userspace must sample tb_seq_count again; if it has changed
ecf99a43 NL	294	* from its previous value, userspace must retry the whole sequence.
	295	*
	296	* Calls to update_vsyscall are serialized by the timekeeping core.
	297	*/
	298	void update_vsyscall(struct timekeeper *tk)
	299	{
ecf99a43 NL	300	struct timespec64 *wtm = &tk->wall_to_monotonic;
	301
	302	if (!cntvct_ok) {
	303	/* The entry points have been zeroed, so there is no
	304	* point in updating the data page.
	305	*/
	306	return;
	307	}
	308
	309	vdso_write_begin(vdso_data);
	310
ba88f57f	311	vdso_data->use_syscall = !tk_is_cntvct(tk);
09edea4f NL	312	vdso_data->xtime_coarse_sec = tk->xtime_sec;
	313	vdso_data->xtime_coarse_nsec = (u32)(tk->tkr_mono.xtime_nsec >>
	314	tk->tkr_mono.shift);
ecf99a43 NL	315	vdso_data->wtm_clock_sec = wtm->tv_sec;
	316	vdso_data->wtm_clock_nsec = wtm->tv_nsec;
	317
ba88f57f	318	if (!vdso_data->use_syscall) {
bb0fd7ab	319	vdso_data->cs_cycle_last = tk->tkr_mono.cycle_last;
ecf99a43	320	vdso_data->xtime_clock_sec = tk->xtime_sec;
bb0fd7ab	321	vdso_data->xtime_clock_snsec = tk->tkr_mono.xtime_nsec;
ba88f57f MS	322	vdso_data->cs_mono_mult = tk->tkr_mono.mult;
ba88f57f MS	323	/* tkr_mono.shift == tkr_raw.shift */
bb0fd7ab LT	324	vdso_data->cs_shift = tk->tkr_mono.shift;
bb0fd7ab LT	325	vdso_data->cs_mask = tk->tkr_mono.mask;
ecf99a43 NL	326	}
	327
	328	vdso_write_end(vdso_data);
	329
	330	flush_dcache_page(virt_to_page(vdso_data));
	331	}
	332
	333	void update_vsyscall_tz(void)
	334	{
	335	vdso_data->tz_minuteswest = sys_tz.tz_minuteswest;
	336	vdso_data->tz_dsttime = sys_tz.tz_dsttime;
	337	flush_dcache_page(virt_to_page(vdso_data));
	338	}