binfmt_elf.c 61.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
/*
 * linux/fs/binfmt_elf.c
 *
 * These are the functions used to load ELF format executables as used
 * on SVr4 machines.  Information on the format may be found in the book
 * "UNIX SYSTEM V RELEASE 4 Programmers Guide: Ansi C and Programming Support
 * Tools".
 *
 * Copyright 1993, 1994: Eric Youngdale (ericy@cais.com).
 */

#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/mman.h>
#include <linux/errno.h>
#include <linux/signal.h>
#include <linux/binfmts.h>
#include <linux/string.h>
#include <linux/file.h>
#include <linux/slab.h>
#include <linux/personality.h>
#include <linux/elfcore.h>
#include <linux/init.h>
#include <linux/highuid.h>
#include <linux/compiler.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
30
#include <linux/vmalloc.h>
L
Linus Torvalds 已提交
31 32
#include <linux/security.h>
#include <linux/random.h>
33
#include <linux/elf.h>
K
Kees Cook 已提交
34
#include <linux/elf-randomize.h>
35
#include <linux/utsname.h>
36
#include <linux/coredump.h>
37
#include <linux/sched.h>
38
#include <linux/sched/coredump.h>
39
#include <linux/dax.h>
40
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
41 42 43
#include <asm/param.h>
#include <asm/page.h>

44 45 46
#ifndef user_long_t
#define user_long_t long
#endif
47 48 49 50
#ifndef user_siginfo_t
#define user_siginfo_t siginfo_t
#endif

51
static int load_elf_binary(struct linux_binprm *bprm);
52 53
static unsigned long elf_map(struct file *, unsigned long, struct elf_phdr *,
				int, int, unsigned long);
L
Linus Torvalds 已提交
54

55 56 57 58 59 60
#ifdef CONFIG_USELIB
static int load_elf_library(struct file *);
#else
#define load_elf_library NULL
#endif

L
Linus Torvalds 已提交
61 62 63 64
/*
 * If we don't support core dumping, then supply a NULL so we
 * don't even try.
 */
C
Christoph Hellwig 已提交
65
#ifdef CONFIG_ELF_CORE
66
static int elf_core_dump(struct coredump_params *cprm);
L
Linus Torvalds 已提交
67 68 69 70 71
#else
#define elf_core_dump	NULL
#endif

#if ELF_EXEC_PAGESIZE > PAGE_SIZE
72
#define ELF_MIN_ALIGN	ELF_EXEC_PAGESIZE
L
Linus Torvalds 已提交
73
#else
74
#define ELF_MIN_ALIGN	PAGE_SIZE
L
Linus Torvalds 已提交
75 76 77 78 79 80 81 82 83 84 85
#endif

#ifndef ELF_CORE_EFLAGS
#define ELF_CORE_EFLAGS	0
#endif

#define ELF_PAGESTART(_v) ((_v) & ~(unsigned long)(ELF_MIN_ALIGN-1))
#define ELF_PAGEOFFSET(_v) ((_v) & (ELF_MIN_ALIGN-1))
#define ELF_PAGEALIGN(_v) (((_v) + ELF_MIN_ALIGN - 1) & ~(ELF_MIN_ALIGN - 1))

static struct linux_binfmt elf_format = {
M
Mikael Pettersson 已提交
86 87 88 89 90
	.module		= THIS_MODULE,
	.load_binary	= load_elf_binary,
	.load_shlib	= load_elf_library,
	.core_dump	= elf_core_dump,
	.min_coredump	= ELF_EXEC_PAGESIZE,
L
Linus Torvalds 已提交
91 92
};

A
Andrew Morton 已提交
93
#define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE)
L
Linus Torvalds 已提交
94

95
static int set_brk(unsigned long start, unsigned long end, int prot)
L
Linus Torvalds 已提交
96 97 98 99
{
	start = ELF_PAGEALIGN(start);
	end = ELF_PAGEALIGN(end);
	if (end > start) {
100 101 102 103 104 105 106
		/*
		 * Map the last of the bss segment.
		 * If the header is requesting these pages to be
		 * executable, honour that (ppc32 needs this).
		 */
		int error = vm_brk_flags(start, end - start,
				prot & PROT_EXEC ? VM_EXEC : 0);
107 108
		if (error)
			return error;
L
Linus Torvalds 已提交
109 110 111 112 113 114 115 116
	}
	current->mm->start_brk = current->mm->brk = end;
	return 0;
}

/* We need to explicitly zero any fractional pages
   after the data section (i.e. bss).  This would
   contain the junk from the file that should not
117 118
   be in memory
 */
L
Linus Torvalds 已提交
119 120 121 122 123 124 125 126 127 128 129 130 131
static int padzero(unsigned long elf_bss)
{
	unsigned long nbyte;

	nbyte = ELF_PAGEOFFSET(elf_bss);
	if (nbyte) {
		nbyte = ELF_MIN_ALIGN - nbyte;
		if (clear_user((void __user *) elf_bss, nbyte))
			return -EFAULT;
	}
	return 0;
}

O
Ohad Ben-Cohen 已提交
132
/* Let's use some macros to make this stack manipulation a little clearer */
L
Linus Torvalds 已提交
133 134 135 136
#ifdef CONFIG_STACK_GROWSUP
#define STACK_ADD(sp, items) ((elf_addr_t __user *)(sp) + (items))
#define STACK_ROUND(sp, items) \
	((15 + (unsigned long) ((sp) + (items))) &~ 15UL)
137 138 139
#define STACK_ALLOC(sp, len) ({ \
	elf_addr_t __user *old_sp = (elf_addr_t __user *)sp; sp += len; \
	old_sp; })
L
Linus Torvalds 已提交
140 141 142 143 144 145 146
#else
#define STACK_ADD(sp, items) ((elf_addr_t __user *)(sp) - (items))
#define STACK_ROUND(sp, items) \
	(((unsigned long) (sp - items)) &~ 15UL)
#define STACK_ALLOC(sp, len) ({ sp -= len ; sp; })
#endif

147 148 149 150 151 152 153 154 155
#ifndef ELF_BASE_PLATFORM
/*
 * AT_BASE_PLATFORM indicates the "real" hardware/microarchitecture.
 * If the arch defines ELF_BASE_PLATFORM (in asm/elf.h), the value
 * will be copied to the user stack in the same manner as AT_PLATFORM.
 */
#define ELF_BASE_PLATFORM NULL
#endif

L
Linus Torvalds 已提交
156
static int
157
create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
158
		unsigned long load_addr, unsigned long interp_load_addr)
L
Linus Torvalds 已提交
159 160 161 162 163 164 165 166
{
	unsigned long p = bprm->p;
	int argc = bprm->argc;
	int envc = bprm->envc;
	elf_addr_t __user *argv;
	elf_addr_t __user *envp;
	elf_addr_t __user *sp;
	elf_addr_t __user *u_platform;
167
	elf_addr_t __user *u_base_platform;
168
	elf_addr_t __user *u_rand_bytes;
L
Linus Torvalds 已提交
169
	const char *k_platform = ELF_PLATFORM;
170
	const char *k_base_platform = ELF_BASE_PLATFORM;
171
	unsigned char k_rand_bytes[16];
L
Linus Torvalds 已提交
172 173 174
	int items;
	elf_addr_t *elf_info;
	int ei_index = 0;
175
	const struct cred *cred = current_cred();
176
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
177

178 179 180 181 182 183 184 185
	/*
	 * In some cases (e.g. Hyper-Threading), we want to avoid L1
	 * evictions by the processes running on the same package. One
	 * thing we can do is to shuffle the initial stack for them.
	 */

	p = arch_align_stack(p);

L
Linus Torvalds 已提交
186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
	/*
	 * If this architecture has a platform capability string, copy it
	 * to userspace.  In some cases (Sparc), this info is impossible
	 * for userspace to get any other way, in others (i386) it is
	 * merely difficult.
	 */
	u_platform = NULL;
	if (k_platform) {
		size_t len = strlen(k_platform) + 1;

		u_platform = (elf_addr_t __user *)STACK_ALLOC(p, len);
		if (__copy_to_user(u_platform, k_platform, len))
			return -EFAULT;
	}

201 202 203 204 205 206 207 208 209 210 211 212 213
	/*
	 * If this architecture has a "base" platform capability
	 * string, copy it to userspace.
	 */
	u_base_platform = NULL;
	if (k_base_platform) {
		size_t len = strlen(k_base_platform) + 1;

		u_base_platform = (elf_addr_t __user *)STACK_ALLOC(p, len);
		if (__copy_to_user(u_base_platform, k_base_platform, len))
			return -EFAULT;
	}

214 215 216 217 218 219 220 221 222
	/*
	 * Generate 16 random bytes for userspace PRNG seeding.
	 */
	get_random_bytes(k_rand_bytes, sizeof(k_rand_bytes));
	u_rand_bytes = (elf_addr_t __user *)
		       STACK_ALLOC(p, sizeof(k_rand_bytes));
	if (__copy_to_user(u_rand_bytes, k_rand_bytes, sizeof(k_rand_bytes)))
		return -EFAULT;

L
Linus Torvalds 已提交
223
	/* Create the ELF interpreter info */
224
	elf_info = (elf_addr_t *)current->mm->saved_auxv;
225
	/* update AT_VECTOR_SIZE_BASE if the number of NEW_AUX_ENT() changes */
L
Linus Torvalds 已提交
226
#define NEW_AUX_ENT(id, val) \
227
	do { \
228 229
		elf_info[ei_index++] = id; \
		elf_info[ei_index++] = val; \
230
	} while (0)
L
Linus Torvalds 已提交
231 232 233 234 235

#ifdef ARCH_DLINFO
	/* 
	 * ARCH_DLINFO must come first so PPC can do its special alignment of
	 * AUXV.
236 237
	 * update AT_VECTOR_SIZE_ARCH if the number of NEW_AUX_ENT() in
	 * ARCH_DLINFO changes
L
Linus Torvalds 已提交
238 239 240 241 242 243 244
	 */
	ARCH_DLINFO;
#endif
	NEW_AUX_ENT(AT_HWCAP, ELF_HWCAP);
	NEW_AUX_ENT(AT_PAGESZ, ELF_EXEC_PAGESIZE);
	NEW_AUX_ENT(AT_CLKTCK, CLOCKS_PER_SEC);
	NEW_AUX_ENT(AT_PHDR, load_addr + exec->e_phoff);
245
	NEW_AUX_ENT(AT_PHENT, sizeof(struct elf_phdr));
L
Linus Torvalds 已提交
246 247 248 249
	NEW_AUX_ENT(AT_PHNUM, exec->e_phnum);
	NEW_AUX_ENT(AT_BASE, interp_load_addr);
	NEW_AUX_ENT(AT_FLAGS, 0);
	NEW_AUX_ENT(AT_ENTRY, exec->e_entry);
250 251 252 253
	NEW_AUX_ENT(AT_UID, from_kuid_munged(cred->user_ns, cred->uid));
	NEW_AUX_ENT(AT_EUID, from_kuid_munged(cred->user_ns, cred->euid));
	NEW_AUX_ENT(AT_GID, from_kgid_munged(cred->user_ns, cred->gid));
	NEW_AUX_ENT(AT_EGID, from_kgid_munged(cred->user_ns, cred->egid));
254
 	NEW_AUX_ENT(AT_SECURE, security_bprm_secureexec(bprm));
255
	NEW_AUX_ENT(AT_RANDOM, (elf_addr_t)(unsigned long)u_rand_bytes);
M
Michael Neuling 已提交
256 257 258
#ifdef ELF_HWCAP2
	NEW_AUX_ENT(AT_HWCAP2, ELF_HWCAP2);
#endif
259
	NEW_AUX_ENT(AT_EXECFN, bprm->exec);
L
Linus Torvalds 已提交
260
	if (k_platform) {
261
		NEW_AUX_ENT(AT_PLATFORM,
262
			    (elf_addr_t)(unsigned long)u_platform);
L
Linus Torvalds 已提交
263
	}
264 265 266 267
	if (k_base_platform) {
		NEW_AUX_ENT(AT_BASE_PLATFORM,
			    (elf_addr_t)(unsigned long)u_base_platform);
	}
L
Linus Torvalds 已提交
268
	if (bprm->interp_flags & BINPRM_FLAGS_EXECFD) {
269
		NEW_AUX_ENT(AT_EXECFD, bprm->interp_data);
L
Linus Torvalds 已提交
270 271 272 273 274 275 276 277 278 279 280
	}
#undef NEW_AUX_ENT
	/* AT_NULL is zero; clear the rest too */
	memset(&elf_info[ei_index], 0,
	       sizeof current->mm->saved_auxv - ei_index * sizeof elf_info[0]);

	/* And advance past the AT_NULL entry.  */
	ei_index += 2;

	sp = STACK_ADD(p, ei_index);

281
	items = (argc + 1) + (envc + 1) + 1;
L
Linus Torvalds 已提交
282 283 284 285 286
	bprm->p = STACK_ROUND(sp, items);

	/* Point sp at the lowest address on the stack */
#ifdef CONFIG_STACK_GROWSUP
	sp = (elf_addr_t __user *)bprm->p - items - ei_index;
287
	bprm->exec = (unsigned long)sp; /* XXX: PARISC HACK */
L
Linus Torvalds 已提交
288 289 290 291
#else
	sp = (elf_addr_t __user *)bprm->p;
#endif

292 293 294 295 296 297 298 299 300

	/*
	 * Grow the stack manually; some architectures have a limit on how
	 * far ahead a user-space access may be in order to grow the stack.
	 */
	vma = find_extend_vma(current->mm, bprm->p);
	if (!vma)
		return -EFAULT;

L
Linus Torvalds 已提交
301 302 303
	/* Now, let's put argc (and argv, envp if appropriate) on the stack */
	if (__put_user(argc, sp++))
		return -EFAULT;
304 305
	argv = sp;
	envp = argv + argc + 1;
L
Linus Torvalds 已提交
306 307

	/* Populate argv and envp */
308
	p = current->mm->arg_end = current->mm->arg_start;
L
Linus Torvalds 已提交
309 310
	while (argc-- > 0) {
		size_t len;
311 312
		if (__put_user((elf_addr_t)p, argv++))
			return -EFAULT;
313 314
		len = strnlen_user((void __user *)p, MAX_ARG_STRLEN);
		if (!len || len > MAX_ARG_STRLEN)
315
			return -EINVAL;
L
Linus Torvalds 已提交
316 317 318 319 320 321 322
		p += len;
	}
	if (__put_user(0, argv))
		return -EFAULT;
	current->mm->arg_end = current->mm->env_start = p;
	while (envc-- > 0) {
		size_t len;
323 324
		if (__put_user((elf_addr_t)p, envp++))
			return -EFAULT;
325 326
		len = strnlen_user((void __user *)p, MAX_ARG_STRLEN);
		if (!len || len > MAX_ARG_STRLEN)
327
			return -EINVAL;
L
Linus Torvalds 已提交
328 329 330 331 332 333 334 335 336 337 338 339 340
		p += len;
	}
	if (__put_user(0, envp))
		return -EFAULT;
	current->mm->env_end = p;

	/* Put the elf_info on the stack in the right place.  */
	sp = (elf_addr_t __user *)envp + 1;
	if (copy_to_user(sp, elf_info, ei_index * sizeof(elf_addr_t)))
		return -EFAULT;
	return 0;
}

341 342
#ifndef elf_map

L
Linus Torvalds 已提交
343
static unsigned long elf_map(struct file *filep, unsigned long addr,
J
Jiri Kosina 已提交
344 345
		struct elf_phdr *eppnt, int prot, int type,
		unsigned long total_size)
L
Linus Torvalds 已提交
346 347
{
	unsigned long map_addr;
J
Jiri Kosina 已提交
348 349 350 351
	unsigned long size = eppnt->p_filesz + ELF_PAGEOFFSET(eppnt->p_vaddr);
	unsigned long off = eppnt->p_offset - ELF_PAGEOFFSET(eppnt->p_vaddr);
	addr = ELF_PAGESTART(addr);
	size = ELF_PAGEALIGN(size);
L
Linus Torvalds 已提交
352

353 354
	/* mmap() will return -EINVAL if given a zero size, but a
	 * segment with zero filesize is perfectly valid */
J
Jiri Kosina 已提交
355 356 357 358 359 360 361 362 363 364 365 366 367
	if (!size)
		return addr;

	/*
	* total_size is the size of the ELF (interpreter) image.
	* The _first_ mmap needs to know the full size, otherwise
	* randomization might put this image into an overlapping
	* position with the ELF binary image. (since size < total_size)
	* So we first map the 'big' image - and unmap the remainder at
	* the end. (which unmap is needed for ELF images with holes.)
	*/
	if (total_size) {
		total_size = ELF_PAGEALIGN(total_size);
368
		map_addr = vm_mmap(filep, addr, total_size, prot, type, off);
J
Jiri Kosina 已提交
369
		if (!BAD_ADDR(map_addr))
370
			vm_munmap(map_addr+size, total_size-size);
J
Jiri Kosina 已提交
371
	} else
372
		map_addr = vm_mmap(filep, addr, size, prot, type, off);
J
Jiri Kosina 已提交
373

L
Linus Torvalds 已提交
374 375 376
	return(map_addr);
}

377 378
#endif /* !elf_map */

J
Jiri Kosina 已提交
379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
static unsigned long total_mapping_size(struct elf_phdr *cmds, int nr)
{
	int i, first_idx = -1, last_idx = -1;

	for (i = 0; i < nr; i++) {
		if (cmds[i].p_type == PT_LOAD) {
			last_idx = i;
			if (first_idx == -1)
				first_idx = i;
		}
	}
	if (first_idx == -1)
		return 0;

	return cmds[last_idx].p_vaddr + cmds[last_idx].p_memsz -
				ELF_PAGESTART(cmds[first_idx].p_vaddr);
}

397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
/**
 * load_elf_phdrs() - load ELF program headers
 * @elf_ex:   ELF header of the binary whose program headers should be loaded
 * @elf_file: the opened ELF binary file
 *
 * Loads ELF program headers from the binary file elf_file, which has the ELF
 * header pointed to by elf_ex, into a newly allocated array. The caller is
 * responsible for freeing the allocated data. Returns an ERR_PTR upon failure.
 */
static struct elf_phdr *load_elf_phdrs(struct elfhdr *elf_ex,
				       struct file *elf_file)
{
	struct elf_phdr *elf_phdata = NULL;
	int retval, size, err = -1;

	/*
	 * If the size of this structure has changed, then punt, since
	 * we will be doing the wrong thing.
	 */
	if (elf_ex->e_phentsize != sizeof(struct elf_phdr))
		goto out;

	/* Sanity check the number of program headers... */
	if (elf_ex->e_phnum < 1 ||
		elf_ex->e_phnum > 65536U / sizeof(struct elf_phdr))
		goto out;

	/* ...and their total size. */
	size = sizeof(struct elf_phdr) * elf_ex->e_phnum;
	if (size > ELF_MIN_ALIGN)
		goto out;

	elf_phdata = kmalloc(size, GFP_KERNEL);
	if (!elf_phdata)
		goto out;

	/* Read in the program headers */
	retval = kernel_read(elf_file, elf_ex->e_phoff,
			     (char *)elf_phdata, size);
	if (retval != size) {
		err = (retval < 0) ? retval : -EIO;
		goto out;
	}

	/* Success! */
	err = 0;
out:
	if (err) {
		kfree(elf_phdata);
		elf_phdata = NULL;
	}
	return elf_phdata;
}
J
Jiri Kosina 已提交
450

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
#ifndef CONFIG_ARCH_BINFMT_ELF_STATE

/**
 * struct arch_elf_state - arch-specific ELF loading state
 *
 * This structure is used to preserve architecture specific data during
 * the loading of an ELF file, throughout the checking of architecture
 * specific ELF headers & through to the point where the ELF load is
 * known to be proceeding (ie. SET_PERSONALITY).
 *
 * This implementation is a dummy for architectures which require no
 * specific state.
 */
struct arch_elf_state {
};

#define INIT_ARCH_ELF_STATE {}

/**
 * arch_elf_pt_proc() - check a PT_LOPROC..PT_HIPROC ELF program header
 * @ehdr:	The main ELF header
 * @phdr:	The program header to check
 * @elf:	The open ELF file
 * @is_interp:	True if the phdr is from the interpreter of the ELF being
 *		loaded, else false.
 * @state:	Architecture-specific state preserved throughout the process
 *		of loading the ELF.
 *
 * Inspects the program header phdr to validate its correctness and/or
 * suitability for the system. Called once per ELF program header in the
 * range PT_LOPROC to PT_HIPROC, for both the ELF being loaded and its
 * interpreter.
 *
 * Return: Zero to proceed with the ELF load, non-zero to fail the ELF load
 *         with that return code.
 */
static inline int arch_elf_pt_proc(struct elfhdr *ehdr,
				   struct elf_phdr *phdr,
				   struct file *elf, bool is_interp,
				   struct arch_elf_state *state)
{
	/* Dummy implementation, always proceed */
	return 0;
}

/**
497
 * arch_check_elf() - check an ELF executable
498 499
 * @ehdr:	The main ELF header
 * @has_interp:	True if the ELF has an interpreter, else false.
500
 * @interp_ehdr: The interpreter's ELF header
501 502 503 504 505 506 507 508 509 510 511
 * @state:	Architecture-specific state preserved throughout the process
 *		of loading the ELF.
 *
 * Provides a final opportunity for architecture code to reject the loading
 * of the ELF & cause an exec syscall to return an error. This is called after
 * all program headers to be checked by arch_elf_pt_proc have been.
 *
 * Return: Zero to proceed with the ELF load, non-zero to fail the ELF load
 *         with that return code.
 */
static inline int arch_check_elf(struct elfhdr *ehdr, bool has_interp,
512
				 struct elfhdr *interp_ehdr,
513 514 515 516 517 518 519
				 struct arch_elf_state *state)
{
	/* Dummy implementation, always proceed */
	return 0;
}

#endif /* !CONFIG_ARCH_BINFMT_ELF_STATE */
J
Jiri Kosina 已提交
520

L
Linus Torvalds 已提交
521 522 523 524 525
/* This is much more generalized than the library routine read function,
   so we keep this separate.  Technically the library read function
   is only provided so that we can read a.out libraries that have
   an ELF header */

526
static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
J
Jiri Kosina 已提交
527
		struct file *interpreter, unsigned long *interp_map_addr,
528
		unsigned long no_base, struct elf_phdr *interp_elf_phdata)
L
Linus Torvalds 已提交
529 530 531 532 533
{
	struct elf_phdr *eppnt;
	unsigned long load_addr = 0;
	int load_addr_set = 0;
	unsigned long last_bss = 0, elf_bss = 0;
534
	int bss_prot = 0;
L
Linus Torvalds 已提交
535
	unsigned long error = ~0UL;
J
Jiri Kosina 已提交
536
	unsigned long total_size;
537
	int i;
L
Linus Torvalds 已提交
538 539 540 541 542 543 544

	/* First of all, some simple consistency checks */
	if (interp_elf_ex->e_type != ET_EXEC &&
	    interp_elf_ex->e_type != ET_DYN)
		goto out;
	if (!elf_check_arch(interp_elf_ex))
		goto out;
A
Al Viro 已提交
545
	if (!interpreter->f_op->mmap)
L
Linus Torvalds 已提交
546 547
		goto out;

548 549
	total_size = total_mapping_size(interp_elf_phdata,
					interp_elf_ex->e_phnum);
J
Jiri Kosina 已提交
550 551
	if (!total_size) {
		error = -EINVAL;
552
		goto out;
J
Jiri Kosina 已提交
553 554
	}

555
	eppnt = interp_elf_phdata;
556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571
	for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) {
		if (eppnt->p_type == PT_LOAD) {
			int elf_type = MAP_PRIVATE | MAP_DENYWRITE;
			int elf_prot = 0;
			unsigned long vaddr = 0;
			unsigned long k, map_addr;

			if (eppnt->p_flags & PF_R)
		    		elf_prot = PROT_READ;
			if (eppnt->p_flags & PF_W)
				elf_prot |= PROT_WRITE;
			if (eppnt->p_flags & PF_X)
				elf_prot |= PROT_EXEC;
			vaddr = eppnt->p_vaddr;
			if (interp_elf_ex->e_type == ET_EXEC || load_addr_set)
				elf_type |= MAP_FIXED;
J
Jiri Kosina 已提交
572 573
			else if (no_base && interp_elf_ex->e_type == ET_DYN)
				load_addr = -vaddr;
574 575

			map_addr = elf_map(interpreter, load_addr + vaddr,
576
					eppnt, elf_prot, elf_type, total_size);
J
Jiri Kosina 已提交
577 578 579
			total_size = 0;
			if (!*interp_map_addr)
				*interp_map_addr = map_addr;
580 581
			error = map_addr;
			if (BAD_ADDR(map_addr))
582
				goto out;
583 584 585 586 587 588 589 590 591 592 593 594 595

			if (!load_addr_set &&
			    interp_elf_ex->e_type == ET_DYN) {
				load_addr = map_addr - ELF_PAGESTART(vaddr);
				load_addr_set = 1;
			}

			/*
			 * Check to see if the section's size will overflow the
			 * allowed task size. Note that p_filesz must always be
			 * <= p_memsize so it's only necessary to check p_memsz.
			 */
			k = load_addr + eppnt->p_vaddr;
596
			if (BAD_ADDR(k) ||
597 598 599 600
			    eppnt->p_filesz > eppnt->p_memsz ||
			    eppnt->p_memsz > TASK_SIZE ||
			    TASK_SIZE - eppnt->p_memsz < k) {
				error = -ENOMEM;
601
				goto out;
602 603 604 605 606 607 608 609 610 611 612 613 614 615
			}

			/*
			 * Find the end of the file mapping for this phdr, and
			 * keep track of the largest address we see for this.
			 */
			k = load_addr + eppnt->p_vaddr + eppnt->p_filesz;
			if (k > elf_bss)
				elf_bss = k;

			/*
			 * Do the same thing for the memory mapping - between
			 * elf_bss and last_bss is the bss section.
			 */
616
			k = load_addr + eppnt->p_vaddr + eppnt->p_memsz;
617
			if (k > last_bss) {
618
				last_bss = k;
619 620
				bss_prot = elf_prot;
			}
621
		}
L
Linus Torvalds 已提交
622 623
	}

624 625 626 627 628 629 630 631 632 633 634 635
	/*
	 * Now fill out the bss section: first pad the last page from
	 * the file up to the page boundary, and zero it from elf_bss
	 * up to the end of the page.
	 */
	if (padzero(elf_bss)) {
		error = -EFAULT;
		goto out;
	}
	/*
	 * Next, align both the file and mem bss up to the page size,
	 * since this is where elf_bss was just zeroed up to, and where
636
	 * last_bss will end after the vm_brk_flags() below.
637 638 639 640
	 */
	elf_bss = ELF_PAGEALIGN(elf_bss);
	last_bss = ELF_PAGEALIGN(last_bss);
	/* Finally, if there is still more bss to allocate, do it. */
641
	if (last_bss > elf_bss) {
642 643
		error = vm_brk_flags(elf_bss, last_bss - elf_bss,
				bss_prot & PROT_EXEC ? VM_EXEC : 0);
644
		if (error)
645
			goto out;
L
Linus Torvalds 已提交
646 647
	}

J
Jiri Kosina 已提交
648
	error = load_addr;
L
Linus Torvalds 已提交
649 650 651 652 653 654 655 656 657
out:
	return error;
}

/*
 * These are the functions used to load ELF style executables and shared
 * libraries.  There is no binary dependent code anywhere else.
 */

658
#ifndef STACK_RND_MASK
659
#define STACK_RND_MASK (0x7ff >> (PAGE_SHIFT - 12))	/* 8MB of VA */
660
#endif
L
Linus Torvalds 已提交
661 662 663

static unsigned long randomize_stack_top(unsigned long stack_top)
{
664
	unsigned long random_variable = 0;
L
Linus Torvalds 已提交
665

666 667
	if ((current->flags & PF_RANDOMIZE) &&
		!(current->personality & ADDR_NO_RANDOMIZE)) {
D
Daniel Cashman 已提交
668
		random_variable = get_random_long();
669
		random_variable &= STACK_RND_MASK;
670 671
		random_variable <<= PAGE_SHIFT;
	}
L
Linus Torvalds 已提交
672
#ifdef CONFIG_STACK_GROWSUP
673
	return PAGE_ALIGN(stack_top) + random_variable;
L
Linus Torvalds 已提交
674
#else
675
	return PAGE_ALIGN(stack_top) - random_variable;
L
Linus Torvalds 已提交
676 677 678
#endif
}

679
static int load_elf_binary(struct linux_binprm *bprm)
L
Linus Torvalds 已提交
680 681 682 683 684 685
{
	struct file *interpreter = NULL; /* to shut gcc up */
 	unsigned long load_addr = 0, load_bias = 0;
	int load_addr_set = 0;
	char * elf_interpreter = NULL;
	unsigned long error;
686
	struct elf_phdr *elf_ppnt, *elf_phdata, *interp_elf_phdata = NULL;
L
Linus Torvalds 已提交
687
	unsigned long elf_bss, elf_brk;
688
	int bss_prot = 0;
L
Linus Torvalds 已提交
689
	int retval, i;
J
Jiri Kosina 已提交
690 691
	unsigned long elf_entry;
	unsigned long interp_load_addr = 0;
L
Linus Torvalds 已提交
692
	unsigned long start_code, end_code, start_data, end_data;
693
	unsigned long reloc_func_desc __maybe_unused = 0;
694
	int executable_stack = EXSTACK_DEFAULT;
695
	struct pt_regs *regs = current_pt_regs();
L
Linus Torvalds 已提交
696 697 698 699
	struct {
		struct elfhdr elf_ex;
		struct elfhdr interp_elf_ex;
	} *loc;
700
	struct arch_elf_state arch_state = INIT_ARCH_ELF_STATE;
L
Linus Torvalds 已提交
701 702 703 704 705 706 707 708

	loc = kmalloc(sizeof(*loc), GFP_KERNEL);
	if (!loc) {
		retval = -ENOMEM;
		goto out_ret;
	}
	
	/* Get the exec-header */
709
	loc->elf_ex = *((struct elfhdr *)bprm->buf);
L
Linus Torvalds 已提交
710 711 712 713 714 715 716 717 718 719

	retval = -ENOEXEC;
	/* First of all, some simple consistency checks */
	if (memcmp(loc->elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
		goto out;

	if (loc->elf_ex.e_type != ET_EXEC && loc->elf_ex.e_type != ET_DYN)
		goto out;
	if (!elf_check_arch(&loc->elf_ex))
		goto out;
A
Al Viro 已提交
720
	if (!bprm->file->f_op->mmap)
L
Linus Torvalds 已提交
721 722
		goto out;

723
	elf_phdata = load_elf_phdrs(&loc->elf_ex, bprm->file);
L
Linus Torvalds 已提交
724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
	if (!elf_phdata)
		goto out;

	elf_ppnt = elf_phdata;
	elf_bss = 0;
	elf_brk = 0;

	start_code = ~0UL;
	end_code = 0;
	start_data = 0;
	end_data = 0;

	for (i = 0; i < loc->elf_ex.e_phnum; i++) {
		if (elf_ppnt->p_type == PT_INTERP) {
			/* This is the program interpreter used for
			 * shared libraries - for now assume that this
			 * is an a.out format binary
			 */
			retval = -ENOEXEC;
			if (elf_ppnt->p_filesz > PATH_MAX || 
			    elf_ppnt->p_filesz < 2)
745
				goto out_free_ph;
L
Linus Torvalds 已提交
746 747

			retval = -ENOMEM;
748
			elf_interpreter = kmalloc(elf_ppnt->p_filesz,
749
						  GFP_KERNEL);
L
Linus Torvalds 已提交
750
			if (!elf_interpreter)
751
				goto out_free_ph;
L
Linus Torvalds 已提交
752 753

			retval = kernel_read(bprm->file, elf_ppnt->p_offset,
754 755
					     elf_interpreter,
					     elf_ppnt->p_filesz);
L
Linus Torvalds 已提交
756 757 758 759 760 761 762 763 764 765 766 767 768 769
			if (retval != elf_ppnt->p_filesz) {
				if (retval >= 0)
					retval = -EIO;
				goto out_free_interp;
			}
			/* make sure path is NULL terminated */
			retval = -ENOEXEC;
			if (elf_interpreter[elf_ppnt->p_filesz - 1] != '\0')
				goto out_free_interp;

			interpreter = open_exec(elf_interpreter);
			retval = PTR_ERR(interpreter);
			if (IS_ERR(interpreter))
				goto out_free_interp;
770 771 772 773 774 775

			/*
			 * If the binary is not readable then enforce
			 * mm->dumpable = 0 regardless of the interpreter's
			 * permissions.
			 */
776
			would_dump(bprm, interpreter);
777

778 779 780 781 782
			/* Get the exec headers */
			retval = kernel_read(interpreter, 0,
					     (void *)&loc->interp_elf_ex,
					     sizeof(loc->interp_elf_ex));
			if (retval != sizeof(loc->interp_elf_ex)) {
L
Linus Torvalds 已提交
783 784 785 786 787 788 789 790 791 792 793 794
				if (retval >= 0)
					retval = -EIO;
				goto out_free_dentry;
			}

			break;
		}
		elf_ppnt++;
	}

	elf_ppnt = elf_phdata;
	for (i = 0; i < loc->elf_ex.e_phnum; i++, elf_ppnt++)
795 796
		switch (elf_ppnt->p_type) {
		case PT_GNU_STACK:
L
Linus Torvalds 已提交
797 798 799 800 801
			if (elf_ppnt->p_flags & PF_X)
				executable_stack = EXSTACK_ENABLE_X;
			else
				executable_stack = EXSTACK_DISABLE_X;
			break;
802 803 804 805 806 807 808 809

		case PT_LOPROC ... PT_HIPROC:
			retval = arch_elf_pt_proc(&loc->elf_ex, elf_ppnt,
						  bprm->file, false,
						  &arch_state);
			if (retval)
				goto out_free_dentry;
			break;
L
Linus Torvalds 已提交
810 811 812 813 814
		}

	/* Some simple consistency checks for the interpreter */
	if (elf_interpreter) {
		retval = -ELIBBAD;
815 816
		/* Not an ELF interpreter */
		if (memcmp(loc->interp_elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
L
Linus Torvalds 已提交
817 818
			goto out_free_dentry;
		/* Verify the interpreter has a valid arch */
819
		if (!elf_check_arch(&loc->interp_elf_ex))
L
Linus Torvalds 已提交
820
			goto out_free_dentry;
821 822 823 824 825 826

		/* Load the interpreter program headers */
		interp_elf_phdata = load_elf_phdrs(&loc->interp_elf_ex,
						   interpreter);
		if (!interp_elf_phdata)
			goto out_free_dentry;
827 828 829 830 831 832 833 834 835 836 837 838 839

		/* Pass PT_LOPROC..PT_HIPROC headers to arch code */
		elf_ppnt = interp_elf_phdata;
		for (i = 0; i < loc->interp_elf_ex.e_phnum; i++, elf_ppnt++)
			switch (elf_ppnt->p_type) {
			case PT_LOPROC ... PT_HIPROC:
				retval = arch_elf_pt_proc(&loc->interp_elf_ex,
							  elf_ppnt, interpreter,
							  true, &arch_state);
				if (retval)
					goto out_free_dentry;
				break;
			}
L
Linus Torvalds 已提交
840 841
	}

842 843 844 845 846
	/*
	 * Allow arch code to reject the ELF at this point, whilst it's
	 * still possible to return an error to the code that invoked
	 * the exec syscall.
	 */
847 848 849
	retval = arch_check_elf(&loc->elf_ex,
				!!interpreter, &loc->interp_elf_ex,
				&arch_state);
850 851 852
	if (retval)
		goto out_free_dentry;

L
Linus Torvalds 已提交
853 854 855 856 857 858 859
	/* Flush all traces of the currently running executable */
	retval = flush_old_exec(bprm);
	if (retval)
		goto out_free_dentry;

	/* Do this immediately, since STACK_TOP as used in setup_arg_pages
	   may depend on the personality.  */
860
	SET_PERSONALITY2(loc->elf_ex, &arch_state);
L
Linus Torvalds 已提交
861 862 863
	if (elf_read_implies_exec(loc->elf_ex, executable_stack))
		current->personality |= READ_IMPLIES_EXEC;

864
	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
L
Linus Torvalds 已提交
865
		current->flags |= PF_RANDOMIZE;
866 867

	setup_new_exec(bprm);
868
	install_exec_creds(bprm);
L
Linus Torvalds 已提交
869 870 871 872 873

	/* Do this so that we can load the interpreter, if need be.  We will
	   change some of these later */
	retval = setup_arg_pages(bprm, randomize_stack_top(STACK_TOP),
				 executable_stack);
874
	if (retval < 0)
L
Linus Torvalds 已提交
875 876 877 878
		goto out_free_dentry;
	
	current->mm->start_stack = bprm->p;

879
	/* Now we do a little grungy work by mmapping the ELF image into
J
Jiri Kosina 已提交
880
	   the correct location in memory. */
881 882
	for(i = 0, elf_ppnt = elf_phdata;
	    i < loc->elf_ex.e_phnum; i++, elf_ppnt++) {
L
Linus Torvalds 已提交
883 884
		int elf_prot = 0, elf_flags;
		unsigned long k, vaddr;
885
		unsigned long total_size = 0;
L
Linus Torvalds 已提交
886 887 888 889 890 891 892 893 894 895

		if (elf_ppnt->p_type != PT_LOAD)
			continue;

		if (unlikely (elf_brk > elf_bss)) {
			unsigned long nbyte;
	            
			/* There was a PT_LOAD segment with p_memsz > p_filesz
			   before this one. Map anonymous pages, if needed,
			   and clear the area.  */
M
Mikael Pettersson 已提交
896
			retval = set_brk(elf_bss + load_bias,
897 898
					 elf_brk + load_bias,
					 bss_prot);
899
			if (retval)
L
Linus Torvalds 已提交
900 901 902 903 904 905 906 907 908 909
				goto out_free_dentry;
			nbyte = ELF_PAGEOFFSET(elf_bss);
			if (nbyte) {
				nbyte = ELF_MIN_ALIGN - nbyte;
				if (nbyte > elf_brk - elf_bss)
					nbyte = elf_brk - elf_bss;
				if (clear_user((void __user *)elf_bss +
							load_bias, nbyte)) {
					/*
					 * This bss-zeroing can fail if the ELF
910
					 * file specifies odd protections. So
L
Linus Torvalds 已提交
911 912 913 914 915 916
					 * we don't check the return value
					 */
				}
			}
		}

917 918 919 920 921 922
		if (elf_ppnt->p_flags & PF_R)
			elf_prot |= PROT_READ;
		if (elf_ppnt->p_flags & PF_W)
			elf_prot |= PROT_WRITE;
		if (elf_ppnt->p_flags & PF_X)
			elf_prot |= PROT_EXEC;
L
Linus Torvalds 已提交
923

924
		elf_flags = MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE;
L
Linus Torvalds 已提交
925 926 927 928 929

		vaddr = elf_ppnt->p_vaddr;
		if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) {
			elf_flags |= MAP_FIXED;
		} else if (loc->elf_ex.e_type == ET_DYN) {
930 931 932 933
			/* Try and get dynamic programs out of the way of the
			 * default mmap base, as well as whatever program they
			 * might try to exec.  This is because the brk will
			 * follow the loader, and is not movable.  */
K
Kees Cook 已提交
934
			load_bias = ELF_ET_DYN_BASE - vaddr;
935
			if (current->flags & PF_RANDOMIZE)
K
Kees Cook 已提交
936 937
				load_bias += arch_mmap_rnd();
			load_bias = ELF_PAGESTART(load_bias);
938 939 940
			total_size = total_mapping_size(elf_phdata,
							loc->elf_ex.e_phnum);
			if (!total_size) {
941
				retval = -EINVAL;
942 943
				goto out_free_dentry;
			}
L
Linus Torvalds 已提交
944 945
		}

946
		error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt,
947
				elf_prot, elf_flags, total_size);
L
Linus Torvalds 已提交
948
		if (BAD_ADDR(error)) {
949 950
			retval = IS_ERR((void *)error) ?
				PTR_ERR((void*)error) : -EINVAL;
L
Linus Torvalds 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964
			goto out_free_dentry;
		}

		if (!load_addr_set) {
			load_addr_set = 1;
			load_addr = (elf_ppnt->p_vaddr - elf_ppnt->p_offset);
			if (loc->elf_ex.e_type == ET_DYN) {
				load_bias += error -
				             ELF_PAGESTART(load_bias + vaddr);
				load_addr += load_bias;
				reloc_func_desc = load_bias;
			}
		}
		k = elf_ppnt->p_vaddr;
965 966 967 968
		if (k < start_code)
			start_code = k;
		if (start_data < k)
			start_data = k;
L
Linus Torvalds 已提交
969 970 971 972 973 974

		/*
		 * Check to see if the section's size will overflow the
		 * allowed task size. Note that p_filesz must always be
		 * <= p_memsz so it is only necessary to check p_memsz.
		 */
975
		if (BAD_ADDR(k) || elf_ppnt->p_filesz > elf_ppnt->p_memsz ||
L
Linus Torvalds 已提交
976 977
		    elf_ppnt->p_memsz > TASK_SIZE ||
		    TASK_SIZE - elf_ppnt->p_memsz < k) {
978
			/* set_brk can never work. Avoid overflows. */
979
			retval = -EINVAL;
L
Linus Torvalds 已提交
980 981 982 983 984 985 986 987 988 989 990 991
			goto out_free_dentry;
		}

		k = elf_ppnt->p_vaddr + elf_ppnt->p_filesz;

		if (k > elf_bss)
			elf_bss = k;
		if ((elf_ppnt->p_flags & PF_X) && end_code < k)
			end_code = k;
		if (end_data < k)
			end_data = k;
		k = elf_ppnt->p_vaddr + elf_ppnt->p_memsz;
992 993
		if (k > elf_brk) {
			bss_prot = elf_prot;
L
Linus Torvalds 已提交
994
			elf_brk = k;
995
		}
L
Linus Torvalds 已提交
996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
	}

	loc->elf_ex.e_entry += load_bias;
	elf_bss += load_bias;
	elf_brk += load_bias;
	start_code += load_bias;
	end_code += load_bias;
	start_data += load_bias;
	end_data += load_bias;

	/* Calling set_brk effectively mmaps the pages that we need
	 * for the bss and break sections.  We must do this before
	 * mapping in the interpreter, to make sure it doesn't wind
	 * up getting placed where the bss needs to go.
	 */
1011
	retval = set_brk(elf_bss, elf_brk, bss_prot);
1012
	if (retval)
L
Linus Torvalds 已提交
1013
		goto out_free_dentry;
A
akpm@osdl.org 已提交
1014
	if (likely(elf_bss != elf_brk) && unlikely(padzero(elf_bss))) {
L
Linus Torvalds 已提交
1015 1016 1017 1018 1019
		retval = -EFAULT; /* Nobody gets to see this, but.. */
		goto out_free_dentry;
	}

	if (elf_interpreter) {
A
Alan Cox 已提交
1020
		unsigned long interp_map_addr = 0;
1021 1022 1023 1024

		elf_entry = load_elf_interp(&loc->interp_elf_ex,
					    interpreter,
					    &interp_map_addr,
1025
					    load_bias, interp_elf_phdata);
1026 1027 1028 1029 1030 1031 1032
		if (!IS_ERR((void *)elf_entry)) {
			/*
			 * load_elf_interp() returns relocation
			 * adjustment
			 */
			interp_load_addr = elf_entry;
			elf_entry += loc->interp_elf_ex.e_entry;
J
Jiri Kosina 已提交
1033
		}
L
Linus Torvalds 已提交
1034
		if (BAD_ADDR(elf_entry)) {
1035 1036
			retval = IS_ERR((void *)elf_entry) ?
					(int)elf_entry : -EINVAL;
L
Linus Torvalds 已提交
1037 1038 1039 1040 1041 1042 1043 1044 1045
			goto out_free_dentry;
		}
		reloc_func_desc = interp_load_addr;

		allow_write_access(interpreter);
		fput(interpreter);
		kfree(elf_interpreter);
	} else {
		elf_entry = loc->elf_ex.e_entry;
1046
		if (BAD_ADDR(elf_entry)) {
1047
			retval = -EINVAL;
1048 1049
			goto out_free_dentry;
		}
L
Linus Torvalds 已提交
1050 1051
	}

1052
	kfree(interp_elf_phdata);
L
Linus Torvalds 已提交
1053 1054 1055 1056
	kfree(elf_phdata);

	set_binfmt(&elf_format);

1057
#ifdef ARCH_HAS_SETUP_ADDITIONAL_PAGES
1058
	retval = arch_setup_additional_pages(bprm, !!elf_interpreter);
1059
	if (retval < 0)
1060
		goto out;
1061 1062
#endif /* ARCH_HAS_SETUP_ADDITIONAL_PAGES */

1063
	retval = create_elf_tables(bprm, &loc->elf_ex,
1064
			  load_addr, interp_load_addr);
1065
	if (retval < 0)
1066
		goto out;
L
Linus Torvalds 已提交
1067 1068 1069 1070 1071 1072 1073
	/* N.B. passed_fileno might not be initialized? */
	current->mm->end_code = end_code;
	current->mm->start_code = start_code;
	current->mm->start_data = start_data;
	current->mm->end_data = end_data;
	current->mm->start_stack = bprm->p;

1074
	if ((current->flags & PF_RANDOMIZE) && (randomize_va_space > 1)) {
J
Jiri Kosina 已提交
1075 1076
		current->mm->brk = current->mm->start_brk =
			arch_randomize_brk(current->mm);
1077
#ifdef compat_brk_randomized
1078 1079 1080
		current->brk_randomized = 1;
#endif
	}
J
Jiri Kosina 已提交
1081

L
Linus Torvalds 已提交
1082 1083 1084 1085
	if (current->personality & MMAP_PAGE_ZERO) {
		/* Why this, you ask???  Well SVr4 maps page 0 as read-only,
		   and some applications "depend" upon this behavior.
		   Since we do not have the power to recompile these, we
1086
		   emulate the SVr4 behavior. Sigh. */
1087
		error = vm_mmap(NULL, 0, PAGE_SIZE, PROT_READ | PROT_EXEC,
L
Linus Torvalds 已提交
1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113
				MAP_FIXED | MAP_PRIVATE, 0);
	}

#ifdef ELF_PLAT_INIT
	/*
	 * The ABI may specify that certain registers be set up in special
	 * ways (on i386 %edx is the address of a DT_FINI function, for
	 * example.  In addition, it may also specify (eg, PowerPC64 ELF)
	 * that the e_entry field is the address of the function descriptor
	 * for the startup routine, rather than the address of the startup
	 * routine itself.  This macro performs whatever initialization to
	 * the regs structure is required as well as any relocations to the
	 * function descriptor entries when executing dynamically links apps.
	 */
	ELF_PLAT_INIT(regs, reloc_func_desc);
#endif

	start_thread(regs, elf_entry, bprm->p);
	retval = 0;
out:
	kfree(loc);
out_ret:
	return retval;

	/* error cleanup */
out_free_dentry:
1114
	kfree(interp_elf_phdata);
L
Linus Torvalds 已提交
1115 1116 1117 1118
	allow_write_access(interpreter);
	if (interpreter)
		fput(interpreter);
out_free_interp:
J
Jesper Juhl 已提交
1119
	kfree(elf_interpreter);
L
Linus Torvalds 已提交
1120 1121 1122 1123 1124
out_free_ph:
	kfree(elf_phdata);
	goto out;
}

1125
#ifdef CONFIG_USELIB
L
Linus Torvalds 已提交
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136
/* This is really simpleminded and specialized - we are loading an
   a.out library that is given an ELF header. */
static int load_elf_library(struct file *file)
{
	struct elf_phdr *elf_phdata;
	struct elf_phdr *eppnt;
	unsigned long elf_bss, bss, len;
	int retval, error, i, j;
	struct elfhdr elf_ex;

	error = -ENOEXEC;
1137
	retval = kernel_read(file, 0, (char *)&elf_ex, sizeof(elf_ex));
L
Linus Torvalds 已提交
1138 1139 1140 1141 1142 1143 1144 1145
	if (retval != sizeof(elf_ex))
		goto out;

	if (memcmp(elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
		goto out;

	/* First of all, some simple consistency checks */
	if (elf_ex.e_type != ET_EXEC || elf_ex.e_phnum > 2 ||
A
Al Viro 已提交
1146
	    !elf_check_arch(&elf_ex) || !file->f_op->mmap)
L
Linus Torvalds 已提交
1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174
		goto out;

	/* Now read in all of the header information */

	j = sizeof(struct elf_phdr) * elf_ex.e_phnum;
	/* j < ELF_MIN_ALIGN because elf_ex.e_phnum <= 2 */

	error = -ENOMEM;
	elf_phdata = kmalloc(j, GFP_KERNEL);
	if (!elf_phdata)
		goto out;

	eppnt = elf_phdata;
	error = -ENOEXEC;
	retval = kernel_read(file, elf_ex.e_phoff, (char *)eppnt, j);
	if (retval != j)
		goto out_free_ph;

	for (j = 0, i = 0; i<elf_ex.e_phnum; i++)
		if ((eppnt + i)->p_type == PT_LOAD)
			j++;
	if (j != 1)
		goto out_free_ph;

	while (eppnt->p_type != PT_LOAD)
		eppnt++;

	/* Now use mmap to map the library into memory. */
1175
	error = vm_mmap(file,
L
Linus Torvalds 已提交
1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191
			ELF_PAGESTART(eppnt->p_vaddr),
			(eppnt->p_filesz +
			 ELF_PAGEOFFSET(eppnt->p_vaddr)),
			PROT_READ | PROT_WRITE | PROT_EXEC,
			MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE,
			(eppnt->p_offset -
			 ELF_PAGEOFFSET(eppnt->p_vaddr)));
	if (error != ELF_PAGESTART(eppnt->p_vaddr))
		goto out_free_ph;

	elf_bss = eppnt->p_vaddr + eppnt->p_filesz;
	if (padzero(elf_bss)) {
		error = -EFAULT;
		goto out_free_ph;
	}

1192 1193
	len = ELF_PAGESTART(eppnt->p_filesz + eppnt->p_vaddr +
			    ELF_MIN_ALIGN - 1);
L
Linus Torvalds 已提交
1194
	bss = eppnt->p_memsz + eppnt->p_vaddr;
M
Michal Hocko 已提交
1195 1196
	if (bss > len) {
		error = vm_brk(len, bss - len);
1197
		if (error)
M
Michal Hocko 已提交
1198 1199
			goto out_free_ph;
	}
L
Linus Torvalds 已提交
1200 1201 1202 1203 1204 1205 1206
	error = 0;

out_free_ph:
	kfree(elf_phdata);
out:
	return error;
}
1207
#endif /* #ifdef CONFIG_USELIB */
L
Linus Torvalds 已提交
1208

C
Christoph Hellwig 已提交
1209
#ifdef CONFIG_ELF_CORE
L
Linus Torvalds 已提交
1210 1211 1212 1213 1214 1215 1216
/*
 * ELF core dumper
 *
 * Modelled on fs/exec.c:aout_core_dump()
 * Jeremy Fitzhardinge <jeremy@sw.oz.au>
 */

1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
/*
 * The purpose of always_dump_vma() is to make sure that special kernel mappings
 * that are useful for post-mortem analysis are included in every core dump.
 * In that way we ensure that the core dump is fully interpretable later
 * without matching up the same kernel and hardware config to see what PC values
 * meant. These special mappings include - vDSO, vsyscall, and other
 * architecture specific mappings
 */
static bool always_dump_vma(struct vm_area_struct *vma)
{
	/* Any vsyscall mappings? */
	if (vma == get_gate_vma(vma->vm_mm))
		return true;
1230 1231 1232 1233 1234 1235 1236 1237

	/*
	 * Assume that all vmas with a .name op should always be dumped.
	 * If this changes, a new vm_ops field can easily be added.
	 */
	if (vma->vm_ops && vma->vm_ops->name && vma->vm_ops->name(vma))
		return true;

1238 1239 1240 1241 1242 1243 1244 1245 1246 1247
	/*
	 * arch_vma_name() returns non-NULL for special architecture mappings,
	 * such as vDSO sections.
	 */
	if (arch_vma_name(vma))
		return true;

	return false;
}

L
Linus Torvalds 已提交
1248
/*
R
Roland McGrath 已提交
1249
 * Decide what to dump of a segment, part, all or none.
L
Linus Torvalds 已提交
1250
 */
R
Roland McGrath 已提交
1251 1252
static unsigned long vma_dump_size(struct vm_area_struct *vma,
				   unsigned long mm_flags)
L
Linus Torvalds 已提交
1253
{
1254 1255
#define FILTER(type)	(mm_flags & (1UL << MMF_DUMP_##type))

1256 1257
	/* always dump the vdso and vsyscall sections */
	if (always_dump_vma(vma))
R
Roland McGrath 已提交
1258
		goto whole;
R
Roland McGrath 已提交
1259

1260
	if (vma->vm_flags & VM_DONTDUMP)
1261 1262
		return 0;

1263 1264 1265 1266 1267 1268 1269 1270 1271
	/* support for DAX */
	if (vma_is_dax(vma)) {
		if ((vma->vm_flags & VM_SHARED) && FILTER(DAX_SHARED))
			goto whole;
		if (!(vma->vm_flags & VM_SHARED) && FILTER(DAX_PRIVATE))
			goto whole;
		return 0;
	}

1272 1273 1274 1275 1276 1277
	/* Hugetlb memory check */
	if (vma->vm_flags & VM_HUGETLB) {
		if ((vma->vm_flags & VM_SHARED) && FILTER(HUGETLB_SHARED))
			goto whole;
		if (!(vma->vm_flags & VM_SHARED) && FILTER(HUGETLB_PRIVATE))
			goto whole;
1278
		return 0;
1279 1280
	}

L
Linus Torvalds 已提交
1281
	/* Do not dump I/O mapped devices or special mappings */
1282
	if (vma->vm_flags & VM_IO)
L
Linus Torvalds 已提交
1283 1284
		return 0;

1285 1286
	/* By default, dump shared memory if mapped from an anonymous file. */
	if (vma->vm_flags & VM_SHARED) {
A
Al Viro 已提交
1287
		if (file_inode(vma->vm_file)->i_nlink == 0 ?
R
Roland McGrath 已提交
1288 1289 1290
		    FILTER(ANON_SHARED) : FILTER(MAPPED_SHARED))
			goto whole;
		return 0;
1291
	}
L
Linus Torvalds 已提交
1292

R
Roland McGrath 已提交
1293 1294 1295 1296 1297
	/* Dump segments that have been written to.  */
	if (vma->anon_vma && FILTER(ANON_PRIVATE))
		goto whole;
	if (vma->vm_file == NULL)
		return 0;
L
Linus Torvalds 已提交
1298

R
Roland McGrath 已提交
1299 1300 1301 1302 1303 1304 1305 1306
	if (FILTER(MAPPED_PRIVATE))
		goto whole;

	/*
	 * If this looks like the beginning of a DSO or executable mapping,
	 * check for an ELF header.  If we find one, dump the first page to
	 * aid in determining what was mapped here.
	 */
R
Roland McGrath 已提交
1307 1308
	if (FILTER(ELF_HEADERS) &&
	    vma->vm_pgoff == 0 && (vma->vm_flags & VM_READ)) {
R
Roland McGrath 已提交
1309 1310
		u32 __user *header = (u32 __user *) vma->vm_start;
		u32 word;
R
Roland McGrath 已提交
1311
		mm_segment_t fs = get_fs();
R
Roland McGrath 已提交
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323
		/*
		 * Doing it this way gets the constant folded by GCC.
		 */
		union {
			u32 cmp;
			char elfmag[SELFMAG];
		} magic;
		BUILD_BUG_ON(SELFMAG != sizeof word);
		magic.elfmag[EI_MAG0] = ELFMAG0;
		magic.elfmag[EI_MAG1] = ELFMAG1;
		magic.elfmag[EI_MAG2] = ELFMAG2;
		magic.elfmag[EI_MAG3] = ELFMAG3;
R
Roland McGrath 已提交
1324 1325 1326 1327 1328 1329 1330 1331 1332
		/*
		 * Switch to the user "segment" for get_user(),
		 * then put back what elf_core_dump() had in place.
		 */
		set_fs(USER_DS);
		if (unlikely(get_user(word, header)))
			word = 0;
		set_fs(fs);
		if (word == magic.cmp)
R
Roland McGrath 已提交
1333 1334 1335 1336 1337 1338 1339 1340 1341
			return PAGE_SIZE;
	}

#undef	FILTER

	return 0;

whole:
	return vma->vm_end - vma->vm_start;
L
Linus Torvalds 已提交
1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363
}

/* An ELF note in memory */
struct memelfnote
{
	const char *name;
	int type;
	unsigned int datasz;
	void *data;
};

static int notesize(struct memelfnote *en)
{
	int sz;

	sz = sizeof(struct elf_note);
	sz += roundup(strlen(en->name) + 1, 4);
	sz += roundup(en->datasz, 4);

	return sz;
}

A
Al Viro 已提交
1364
static int writenote(struct memelfnote *men, struct coredump_params *cprm)
1365 1366
{
	struct elf_note en;
L
Linus Torvalds 已提交
1367 1368 1369 1370
	en.n_namesz = strlen(men->name) + 1;
	en.n_descsz = men->datasz;
	en.n_type = men->type;

A
Al Viro 已提交
1371
	return dump_emit(cprm, &en, sizeof(en)) &&
A
Al Viro 已提交
1372 1373
	    dump_emit(cprm, men->name, en.n_namesz) && dump_align(cprm, 4) &&
	    dump_emit(cprm, men->data, men->datasz) && dump_align(cprm, 4);
L
Linus Torvalds 已提交
1374 1375
}

R
Roland McGrath 已提交
1376
static void fill_elf_header(struct elfhdr *elf, int segs,
1377
			    u16 machine, u32 flags)
L
Linus Torvalds 已提交
1378
{
1379 1380
	memset(elf, 0, sizeof(*elf));

L
Linus Torvalds 已提交
1381 1382 1383 1384 1385 1386 1387
	memcpy(elf->e_ident, ELFMAG, SELFMAG);
	elf->e_ident[EI_CLASS] = ELF_CLASS;
	elf->e_ident[EI_DATA] = ELF_DATA;
	elf->e_ident[EI_VERSION] = EV_CURRENT;
	elf->e_ident[EI_OSABI] = ELF_OSABI;

	elf->e_type = ET_CORE;
R
Roland McGrath 已提交
1388
	elf->e_machine = machine;
L
Linus Torvalds 已提交
1389 1390
	elf->e_version = EV_CURRENT;
	elf->e_phoff = sizeof(struct elfhdr);
R
Roland McGrath 已提交
1391
	elf->e_flags = flags;
L
Linus Torvalds 已提交
1392 1393 1394
	elf->e_ehsize = sizeof(struct elfhdr);
	elf->e_phentsize = sizeof(struct elf_phdr);
	elf->e_phnum = segs;
1395

L
Linus Torvalds 已提交
1396 1397 1398
	return;
}

1399
static void fill_elf_note_phdr(struct elf_phdr *phdr, int sz, loff_t offset)
L
Linus Torvalds 已提交
1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422
{
	phdr->p_type = PT_NOTE;
	phdr->p_offset = offset;
	phdr->p_vaddr = 0;
	phdr->p_paddr = 0;
	phdr->p_filesz = sz;
	phdr->p_memsz = 0;
	phdr->p_flags = 0;
	phdr->p_align = 0;
	return;
}

static void fill_note(struct memelfnote *note, const char *name, int type, 
		unsigned int sz, void *data)
{
	note->name = name;
	note->type = type;
	note->datasz = sz;
	note->data = data;
	return;
}

/*
1423 1424
 * fill up all the fields in prstatus from the given task struct, except
 * registers which need to be filled up separately.
L
Linus Torvalds 已提交
1425 1426
 */
static void fill_prstatus(struct elf_prstatus *prstatus,
1427
		struct task_struct *p, long signr)
L
Linus Torvalds 已提交
1428 1429 1430 1431
{
	prstatus->pr_info.si_signo = prstatus->pr_cursig = signr;
	prstatus->pr_sigpend = p->pending.signal.sig[0];
	prstatus->pr_sighold = p->blocked.sig[0];
1432 1433 1434
	rcu_read_lock();
	prstatus->pr_ppid = task_pid_vnr(rcu_dereference(p->real_parent));
	rcu_read_unlock();
1435 1436 1437
	prstatus->pr_pid = task_pid_vnr(p);
	prstatus->pr_pgrp = task_pgrp_vnr(p);
	prstatus->pr_sid = task_session_vnr(p);
L
Linus Torvalds 已提交
1438
	if (thread_group_leader(p)) {
1439
		struct task_cputime cputime;
1440

L
Linus Torvalds 已提交
1441
		/*
1442 1443
		 * This is the record for the group leader.  It shows the
		 * group-wide total, not its individual thread total.
L
Linus Torvalds 已提交
1444
		 */
1445 1446 1447
		thread_group_cputime(p, &cputime);
		prstatus->pr_utime = ns_to_timeval(cputime.utime);
		prstatus->pr_stime = ns_to_timeval(cputime.stime);
L
Linus Torvalds 已提交
1448
	} else {
1449
		u64 utime, stime;
1450

1451 1452 1453
		task_cputime(p, &utime, &stime);
		prstatus->pr_utime = ns_to_timeval(utime);
		prstatus->pr_stime = ns_to_timeval(stime);
L
Linus Torvalds 已提交
1454
	}
1455

1456 1457
	prstatus->pr_cutime = ns_to_timeval(p->signal->cutime);
	prstatus->pr_cstime = ns_to_timeval(p->signal->cstime);
L
Linus Torvalds 已提交
1458 1459 1460 1461 1462
}

static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p,
		       struct mm_struct *mm)
{
1463
	const struct cred *cred;
1464
	unsigned int i, len;
L
Linus Torvalds 已提交
1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479
	
	/* first copy the parameters from user space */
	memset(psinfo, 0, sizeof(struct elf_prpsinfo));

	len = mm->arg_end - mm->arg_start;
	if (len >= ELF_PRARGSZ)
		len = ELF_PRARGSZ-1;
	if (copy_from_user(&psinfo->pr_psargs,
		           (const char __user *)mm->arg_start, len))
		return -EFAULT;
	for(i = 0; i < len; i++)
		if (psinfo->pr_psargs[i] == 0)
			psinfo->pr_psargs[i] = ' ';
	psinfo->pr_psargs[len] = 0;

1480 1481 1482
	rcu_read_lock();
	psinfo->pr_ppid = task_pid_vnr(rcu_dereference(p->real_parent));
	rcu_read_unlock();
1483 1484 1485
	psinfo->pr_pid = task_pid_vnr(p);
	psinfo->pr_pgrp = task_pgrp_vnr(p);
	psinfo->pr_sid = task_session_vnr(p);
L
Linus Torvalds 已提交
1486 1487 1488

	i = p->state ? ffz(~p->state) + 1 : 0;
	psinfo->pr_state = i;
1489
	psinfo->pr_sname = (i > 5) ? '.' : "RSDTZW"[i];
L
Linus Torvalds 已提交
1490 1491 1492
	psinfo->pr_zomb = psinfo->pr_sname == 'Z';
	psinfo->pr_nice = task_nice(p);
	psinfo->pr_flag = p->flags;
1493 1494
	rcu_read_lock();
	cred = __task_cred(p);
1495 1496
	SET_UID(psinfo->pr_uid, from_kuid_munged(cred->user_ns, cred->uid));
	SET_GID(psinfo->pr_gid, from_kgid_munged(cred->user_ns, cred->gid));
1497
	rcu_read_unlock();
L
Linus Torvalds 已提交
1498 1499 1500 1501 1502
	strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname));
	
	return 0;
}

R
Roland McGrath 已提交
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512
static void fill_auxv_note(struct memelfnote *note, struct mm_struct *mm)
{
	elf_addr_t *auxv = (elf_addr_t *) mm->saved_auxv;
	int i = 0;
	do
		i += 2;
	while (auxv[i - 2] != AT_NULL);
	fill_note(note, "CORE", NT_AUXV, i * sizeof(elf_addr_t), auxv);
}

1513
static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata,
A
Al Viro 已提交
1514
		const siginfo_t *siginfo)
1515 1516 1517 1518 1519 1520 1521 1522
{
	mm_segment_t old_fs = get_fs();
	set_fs(KERNEL_DS);
	copy_siginfo_to_user((user_siginfo_t __user *) csigdata, siginfo);
	set_fs(old_fs);
	fill_note(note, "CORE", NT_SIGINFO, sizeof(*csigdata), csigdata);
}

1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534
#define MAX_FILE_NOTE_SIZE (4*1024*1024)
/*
 * Format of NT_FILE note:
 *
 * long count     -- how many files are mapped
 * long page_size -- units for file_ofs
 * array of [COUNT] elements of
 *   long start
 *   long end
 *   long file_ofs
 * followed by COUNT filenames in ASCII: "FILE1" NUL "FILE2" NUL...
 */
1535
static int fill_files_note(struct memelfnote *note)
1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549
{
	struct vm_area_struct *vma;
	unsigned count, size, names_ofs, remaining, n;
	user_long_t *data;
	user_long_t *start_end_ofs;
	char *name_base, *name_curpos;

	/* *Estimated* file count and total data size needed */
	count = current->mm->map_count;
	size = count * 64;

	names_ofs = (2 + 3 * count) * sizeof(data[0]);
 alloc:
	if (size >= MAX_FILE_NOTE_SIZE) /* paranoia check */
1550
		return -EINVAL;
1551 1552 1553
	size = round_up(size, PAGE_SIZE);
	data = vmalloc(size);
	if (!data)
1554
		return -ENOMEM;
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566

	start_end_ofs = data + 2;
	name_base = name_curpos = ((char *)data) + names_ofs;
	remaining = size - names_ofs;
	count = 0;
	for (vma = current->mm->mmap; vma != NULL; vma = vma->vm_next) {
		struct file *file;
		const char *filename;

		file = vma->vm_file;
		if (!file)
			continue;
M
Miklos Szeredi 已提交
1567
		filename = file_path(file, name_curpos, remaining);
1568 1569 1570 1571 1572 1573 1574 1575 1576
		if (IS_ERR(filename)) {
			if (PTR_ERR(filename) == -ENAMETOOLONG) {
				vfree(data);
				size = size * 5 / 4;
				goto alloc;
			}
			continue;
		}

M
Miklos Szeredi 已提交
1577
		/* file_path() fills at the end, move name down */
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606
		/* n = strlen(filename) + 1: */
		n = (name_curpos + remaining) - filename;
		remaining = filename - name_curpos;
		memmove(name_curpos, filename, n);
		name_curpos += n;

		*start_end_ofs++ = vma->vm_start;
		*start_end_ofs++ = vma->vm_end;
		*start_end_ofs++ = vma->vm_pgoff;
		count++;
	}

	/* Now we know exact count of files, can store it */
	data[0] = count;
	data[1] = PAGE_SIZE;
	/*
	 * Count usually is less than current->mm->map_count,
	 * we need to move filenames down.
	 */
	n = current->mm->map_count - count;
	if (n != 0) {
		unsigned shift_bytes = n * 3 * sizeof(data[0]);
		memmove(name_base - shift_bytes, name_base,
			name_curpos - name_base);
		name_curpos -= shift_bytes;
	}

	size = name_curpos - (char *)data;
	fill_note(note, "CORE", NT_FILE, size, data);
1607
	return 0;
1608 1609
}

1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622
#ifdef CORE_DUMP_USE_REGSET
#include <linux/regset.h>

struct elf_thread_core_info {
	struct elf_thread_core_info *next;
	struct task_struct *task;
	struct elf_prstatus prstatus;
	struct memelfnote notes[0];
};

struct elf_note_info {
	struct elf_thread_core_info *thread;
	struct memelfnote psinfo;
1623
	struct memelfnote signote;
1624
	struct memelfnote auxv;
1625
	struct memelfnote files;
1626
	user_siginfo_t csigdata;
1627 1628 1629 1630
	size_t size;
	int thread_notes;
};

1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642
/*
 * When a regset has a writeback hook, we call it on each thread before
 * dumping user memory.  On register window machines, this makes sure the
 * user memory backing the register data is up to date before we read it.
 */
static void do_thread_regset_writeback(struct task_struct *task,
				       const struct user_regset *regset)
{
	if (regset->writeback)
		regset->writeback(task, regset, 1);
}

1643
#ifndef PRSTATUS_SIZE
1644
#define PRSTATUS_SIZE(S, R) sizeof(S)
1645 1646 1647
#endif

#ifndef SET_PR_FPVALID
1648
#define SET_PR_FPVALID(S, V, R) ((S)->pr_fpvalid = (V))
1649 1650
#endif

1651 1652 1653 1654 1655
static int fill_thread_core_info(struct elf_thread_core_info *t,
				 const struct user_regset_view *view,
				 long signr, size_t *total)
{
	unsigned int i;
1656
	unsigned int regset_size = view->regsets[0].n * view->regsets[0].size;
1657 1658 1659 1660 1661 1662 1663 1664

	/*
	 * NT_PRSTATUS is the one special case, because the regset data
	 * goes into the pr_reg field inside the note contents, rather
	 * than being the whole note contents.  We fill the reset in here.
	 * We assume that regset 0 is NT_PRSTATUS.
	 */
	fill_prstatus(&t->prstatus, t->task, signr);
1665 1666
	(void) view->regsets[0].get(t->task, &view->regsets[0], 0, regset_size,
				    &t->prstatus.pr_reg, NULL);
1667 1668

	fill_note(&t->notes[0], "CORE", NT_PRSTATUS,
1669
		  PRSTATUS_SIZE(t->prstatus, regset_size), &t->prstatus);
1670 1671
	*total += notesize(&t->notes[0]);

1672 1673
	do_thread_regset_writeback(t->task, &view->regsets[0]);

1674 1675 1676 1677 1678 1679 1680
	/*
	 * Each other regset might generate a note too.  For each regset
	 * that has no core_note_type or is inactive, we leave t->notes[i]
	 * all zero and we'll know to skip writing it later.
	 */
	for (i = 1; i < view->n; ++i) {
		const struct user_regset *regset = &view->regsets[i];
1681
		do_thread_regset_writeback(t->task, regset);
1682
		if (regset->core_note_type && regset->get &&
1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698
		    (!regset->active || regset->active(t->task, regset))) {
			int ret;
			size_t size = regset->n * regset->size;
			void *data = kmalloc(size, GFP_KERNEL);
			if (unlikely(!data))
				return 0;
			ret = regset->get(t->task, regset,
					  0, size, data, NULL);
			if (unlikely(ret))
				kfree(data);
			else {
				if (regset->core_note_type != NT_PRFPREG)
					fill_note(&t->notes[i], "LINUX",
						  regset->core_note_type,
						  size, data);
				else {
1699 1700
					SET_PR_FPVALID(&t->prstatus,
							1, regset_size);
1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
					fill_note(&t->notes[i], "CORE",
						  NT_PRFPREG, size, data);
				}
				*total += notesize(&t->notes[i]);
			}
		}
	}

	return 1;
}

static int fill_note_info(struct elfhdr *elf, int phdrs,
			  struct elf_note_info *info,
A
Al Viro 已提交
1714
			  const siginfo_t *siginfo, struct pt_regs *regs)
1715 1716 1717 1718 1719
{
	struct task_struct *dump_task = current;
	const struct user_regset_view *view = task_user_regset_view(dump_task);
	struct elf_thread_core_info *t;
	struct elf_prpsinfo *psinfo;
1720
	struct core_thread *ct;
1721 1722 1723 1724 1725 1726
	unsigned int i;

	info->size = 0;
	info->thread = NULL;

	psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL);
1727 1728
	if (psinfo == NULL) {
		info->psinfo.data = NULL; /* So we don't free this wrongly */
1729
		return 0;
1730
	}
1731

A
Amerigo Wang 已提交
1732 1733
	fill_note(&info->psinfo, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo);

1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755
	/*
	 * Figure out how many notes we're going to need for each thread.
	 */
	info->thread_notes = 0;
	for (i = 0; i < view->n; ++i)
		if (view->regsets[i].core_note_type != 0)
			++info->thread_notes;

	/*
	 * Sanity check.  We rely on regset 0 being in NT_PRSTATUS,
	 * since it is our one special case.
	 */
	if (unlikely(info->thread_notes == 0) ||
	    unlikely(view->regsets[0].core_note_type != NT_PRSTATUS)) {
		WARN_ON(1);
		return 0;
	}

	/*
	 * Initialize the ELF file header.
	 */
	fill_elf_header(elf, phdrs,
1756
			view->e_machine, view->e_flags);
1757 1758 1759 1760

	/*
	 * Allocate a structure for each thread.
	 */
1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778
	for (ct = &dump_task->mm->core_state->dumper; ct; ct = ct->next) {
		t = kzalloc(offsetof(struct elf_thread_core_info,
				     notes[info->thread_notes]),
			    GFP_KERNEL);
		if (unlikely(!t))
			return 0;

		t->task = ct->task;
		if (ct->task == dump_task || !info->thread) {
			t->next = info->thread;
			info->thread = t;
		} else {
			/*
			 * Make sure to keep the original task at
			 * the head of the list.
			 */
			t->next = info->thread->next;
			info->thread->next = t;
1779
		}
1780
	}
1781 1782 1783 1784 1785

	/*
	 * Now fill in each thread's information.
	 */
	for (t = info->thread; t != NULL; t = t->next)
1786
		if (!fill_thread_core_info(t, view, siginfo->si_signo, &info->size))
1787 1788 1789 1790 1791 1792 1793 1794
			return 0;

	/*
	 * Fill in the two process-wide notes.
	 */
	fill_psinfo(psinfo, dump_task->group_leader, dump_task->mm);
	info->size += notesize(&info->psinfo);

1795 1796 1797
	fill_siginfo_note(&info->signote, &info->csigdata, siginfo);
	info->size += notesize(&info->signote);

1798 1799 1800
	fill_auxv_note(&info->auxv, current->mm);
	info->size += notesize(&info->auxv);

1801 1802
	if (fill_files_note(&info->files) == 0)
		info->size += notesize(&info->files);
1803

1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816
	return 1;
}

static size_t get_note_info_size(struct elf_note_info *info)
{
	return info->size;
}

/*
 * Write all the notes for each thread.  When writing the first thread, the
 * process-wide notes are interleaved after the first thread-specific note.
 */
static int write_note_info(struct elf_note_info *info,
A
Al Viro 已提交
1817
			   struct coredump_params *cprm)
1818
{
1819
	bool first = true;
1820 1821 1822 1823 1824
	struct elf_thread_core_info *t = info->thread;

	do {
		int i;

A
Al Viro 已提交
1825
		if (!writenote(&t->notes[0], cprm))
1826 1827
			return 0;

A
Al Viro 已提交
1828
		if (first && !writenote(&info->psinfo, cprm))
1829
			return 0;
A
Al Viro 已提交
1830
		if (first && !writenote(&info->signote, cprm))
1831
			return 0;
A
Al Viro 已提交
1832
		if (first && !writenote(&info->auxv, cprm))
1833
			return 0;
1834
		if (first && info->files.data &&
A
Al Viro 已提交
1835
				!writenote(&info->files, cprm))
1836
			return 0;
1837 1838 1839

		for (i = 1; i < info->thread_notes; ++i)
			if (t->notes[i].data &&
A
Al Viro 已提交
1840
			    !writenote(&t->notes[i], cprm))
1841 1842
				return 0;

1843
		first = false;
1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862
		t = t->next;
	} while (t);

	return 1;
}

static void free_note_info(struct elf_note_info *info)
{
	struct elf_thread_core_info *threads = info->thread;
	while (threads) {
		unsigned int i;
		struct elf_thread_core_info *t = threads;
		threads = t->next;
		WARN_ON(t->notes[0].data && t->notes[0].data != &t->prstatus);
		for (i = 1; i < info->thread_notes; ++i)
			kfree(t->notes[i].data);
		kfree(t);
	}
	kfree(info->psinfo.data);
1863
	vfree(info->files.data);
1864 1865 1866 1867
}

#else

L
Linus Torvalds 已提交
1868 1869 1870 1871 1872 1873 1874 1875
/* Here is the structure in which status of each thread is captured. */
struct elf_thread_status
{
	struct list_head list;
	struct elf_prstatus prstatus;	/* NT_PRSTATUS */
	elf_fpregset_t fpu;		/* NT_PRFPREG */
	struct task_struct *thread;
#ifdef ELF_CORE_COPY_XFPREGS
1876
	elf_fpxregset_t xfpu;		/* ELF_CORE_XFPREG_TYPE */
L
Linus Torvalds 已提交
1877 1878 1879 1880 1881 1882 1883
#endif
	struct memelfnote notes[3];
	int num_notes;
};

/*
 * In order to add the specific thread information for the elf file format,
1884 1885
 * we need to keep a linked list of every threads pr_status and then create
 * a single section for them in the final core file.
L
Linus Torvalds 已提交
1886 1887 1888 1889 1890 1891 1892 1893 1894 1895
 */
static int elf_dump_thread_status(long signr, struct elf_thread_status *t)
{
	int sz = 0;
	struct task_struct *p = t->thread;
	t->num_notes = 0;

	fill_prstatus(&t->prstatus, p, signr);
	elf_core_copy_task_regs(p, &t->prstatus.pr_reg);	
	
1896 1897
	fill_note(&t->notes[0], "CORE", NT_PRSTATUS, sizeof(t->prstatus),
		  &(t->prstatus));
L
Linus Torvalds 已提交
1898 1899 1900
	t->num_notes++;
	sz += notesize(&t->notes[0]);

1901 1902 1903 1904
	if ((t->prstatus.pr_fpvalid = elf_core_copy_task_fpregs(p, NULL,
								&t->fpu))) {
		fill_note(&t->notes[1], "CORE", NT_PRFPREG, sizeof(t->fpu),
			  &(t->fpu));
L
Linus Torvalds 已提交
1905 1906 1907 1908 1909 1910
		t->num_notes++;
		sz += notesize(&t->notes[1]);
	}

#ifdef ELF_CORE_COPY_XFPREGS
	if (elf_core_copy_task_xfpregs(p, &t->xfpu)) {
1911 1912
		fill_note(&t->notes[2], "LINUX", ELF_CORE_XFPREG_TYPE,
			  sizeof(t->xfpu), &t->xfpu);
L
Linus Torvalds 已提交
1913 1914 1915 1916 1917 1918 1919
		t->num_notes++;
		sz += notesize(&t->notes[2]);
	}
#endif	
	return sz;
}

R
Roland McGrath 已提交
1920 1921
struct elf_note_info {
	struct memelfnote *notes;
1922
	struct memelfnote *notes_files;
R
Roland McGrath 已提交
1923 1924 1925 1926 1927 1928 1929
	struct elf_prstatus *prstatus;	/* NT_PRSTATUS */
	struct elf_prpsinfo *psinfo;	/* NT_PRPSINFO */
	struct list_head thread_list;
	elf_fpregset_t *fpu;
#ifdef ELF_CORE_COPY_XFPREGS
	elf_fpxregset_t *xfpu;
#endif
1930
	user_siginfo_t csigdata;
R
Roland McGrath 已提交
1931 1932 1933 1934
	int thread_status_size;
	int numnote;
};

A
Amerigo Wang 已提交
1935
static int elf_note_info_init(struct elf_note_info *info)
R
Roland McGrath 已提交
1936
{
A
Amerigo Wang 已提交
1937
	memset(info, 0, sizeof(*info));
R
Roland McGrath 已提交
1938 1939
	INIT_LIST_HEAD(&info->thread_list);

1940
	/* Allocate space for ELF notes */
1941
	info->notes = kmalloc(8 * sizeof(struct memelfnote), GFP_KERNEL);
R
Roland McGrath 已提交
1942 1943 1944 1945
	if (!info->notes)
		return 0;
	info->psinfo = kmalloc(sizeof(*info->psinfo), GFP_KERNEL);
	if (!info->psinfo)
1946
		return 0;
R
Roland McGrath 已提交
1947 1948
	info->prstatus = kmalloc(sizeof(*info->prstatus), GFP_KERNEL);
	if (!info->prstatus)
1949
		return 0;
R
Roland McGrath 已提交
1950 1951
	info->fpu = kmalloc(sizeof(*info->fpu), GFP_KERNEL);
	if (!info->fpu)
1952
		return 0;
R
Roland McGrath 已提交
1953 1954 1955
#ifdef ELF_CORE_COPY_XFPREGS
	info->xfpu = kmalloc(sizeof(*info->xfpu), GFP_KERNEL);
	if (!info->xfpu)
1956
		return 0;
R
Roland McGrath 已提交
1957
#endif
A
Amerigo Wang 已提交
1958 1959 1960 1961 1962
	return 1;
}

static int fill_note_info(struct elfhdr *elf, int phdrs,
			  struct elf_note_info *info,
A
Al Viro 已提交
1963
			  const siginfo_t *siginfo, struct pt_regs *regs)
A
Amerigo Wang 已提交
1964 1965
{
	struct list_head *t;
1966 1967
	struct core_thread *ct;
	struct elf_thread_status *ets;
A
Amerigo Wang 已提交
1968 1969 1970

	if (!elf_note_info_init(info))
		return 0;
R
Roland McGrath 已提交
1971

1972 1973 1974 1975 1976
	for (ct = current->mm->core_state->dumper.next;
					ct; ct = ct->next) {
		ets = kzalloc(sizeof(*ets), GFP_KERNEL);
		if (!ets)
			return 0;
1977

1978 1979 1980
		ets->thread = ct->task;
		list_add(&ets->list, &info->thread_list);
	}
1981

1982 1983
	list_for_each(t, &info->thread_list) {
		int sz;
R
Roland McGrath 已提交
1984

1985 1986 1987
		ets = list_entry(t, struct elf_thread_status, list);
		sz = elf_dump_thread_status(siginfo->si_signo, ets);
		info->thread_status_size += sz;
R
Roland McGrath 已提交
1988 1989 1990
	}
	/* now collect the dump for the current */
	memset(info->prstatus, 0, sizeof(*info->prstatus));
1991
	fill_prstatus(info->prstatus, current, siginfo->si_signo);
R
Roland McGrath 已提交
1992 1993 1994
	elf_core_copy_regs(&info->prstatus->pr_reg, regs);

	/* Set up header */
1995
	fill_elf_header(elf, phdrs, ELF_ARCH, ELF_CORE_EFLAGS);
R
Roland McGrath 已提交
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

	/*
	 * Set up the notes in similar form to SVR4 core dumps made
	 * with info from their /proc.
	 */

	fill_note(info->notes + 0, "CORE", NT_PRSTATUS,
		  sizeof(*info->prstatus), info->prstatus);
	fill_psinfo(info->psinfo, current->group_leader, current->mm);
	fill_note(info->notes + 1, "CORE", NT_PRPSINFO,
		  sizeof(*info->psinfo), info->psinfo);

2008 2009
	fill_siginfo_note(info->notes + 2, &info->csigdata, siginfo);
	fill_auxv_note(info->notes + 3, current->mm);
2010
	info->numnote = 4;
R
Roland McGrath 已提交
2011

2012 2013 2014 2015
	if (fill_files_note(info->notes + info->numnote) == 0) {
		info->notes_files = info->notes + info->numnote;
		info->numnote++;
	}
R
Roland McGrath 已提交
2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046

	/* Try to dump the FPU. */
	info->prstatus->pr_fpvalid = elf_core_copy_task_fpregs(current, regs,
							       info->fpu);
	if (info->prstatus->pr_fpvalid)
		fill_note(info->notes + info->numnote++,
			  "CORE", NT_PRFPREG, sizeof(*info->fpu), info->fpu);
#ifdef ELF_CORE_COPY_XFPREGS
	if (elf_core_copy_task_xfpregs(current, info->xfpu))
		fill_note(info->notes + info->numnote++,
			  "LINUX", ELF_CORE_XFPREG_TYPE,
			  sizeof(*info->xfpu), info->xfpu);
#endif

	return 1;
}

static size_t get_note_info_size(struct elf_note_info *info)
{
	int sz = 0;
	int i;

	for (i = 0; i < info->numnote; i++)
		sz += notesize(info->notes + i);

	sz += info->thread_status_size;

	return sz;
}

static int write_note_info(struct elf_note_info *info,
A
Al Viro 已提交
2047
			   struct coredump_params *cprm)
R
Roland McGrath 已提交
2048 2049 2050 2051 2052
{
	int i;
	struct list_head *t;

	for (i = 0; i < info->numnote; i++)
A
Al Viro 已提交
2053
		if (!writenote(info->notes + i, cprm))
R
Roland McGrath 已提交
2054 2055 2056 2057 2058 2059 2060 2061
			return 0;

	/* write out the thread status notes section */
	list_for_each(t, &info->thread_list) {
		struct elf_thread_status *tmp =
				list_entry(t, struct elf_thread_status, list);

		for (i = 0; i < tmp->num_notes; i++)
A
Al Viro 已提交
2062
			if (!writenote(&tmp->notes[i], cprm))
R
Roland McGrath 已提交
2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076
				return 0;
	}

	return 1;
}

static void free_note_info(struct elf_note_info *info)
{
	while (!list_empty(&info->thread_list)) {
		struct list_head *tmp = info->thread_list.next;
		list_del(tmp);
		kfree(list_entry(tmp, struct elf_thread_status, list));
	}

2077 2078 2079
	/* Free data possibly allocated by fill_files_note(): */
	if (info->notes_files)
		vfree(info->notes_files->data);
2080

R
Roland McGrath 已提交
2081 2082 2083 2084 2085 2086 2087 2088 2089
	kfree(info->prstatus);
	kfree(info->psinfo);
	kfree(info->notes);
	kfree(info->fpu);
#ifdef ELF_CORE_COPY_XFPREGS
	kfree(info->xfpu);
#endif
}

2090 2091
#endif

2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117
static struct vm_area_struct *first_vma(struct task_struct *tsk,
					struct vm_area_struct *gate_vma)
{
	struct vm_area_struct *ret = tsk->mm->mmap;

	if (ret)
		return ret;
	return gate_vma;
}
/*
 * Helper function for iterating across a vma list.  It ensures that the caller
 * will visit `gate_vma' prior to terminating the search.
 */
static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma,
					struct vm_area_struct *gate_vma)
{
	struct vm_area_struct *ret;

	ret = this_vma->vm_next;
	if (ret)
		return ret;
	if (this_vma == gate_vma)
		return NULL;
	return gate_vma;
}

2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133
static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum,
			     elf_addr_t e_shoff, int segs)
{
	elf->e_shoff = e_shoff;
	elf->e_shentsize = sizeof(*shdr4extnum);
	elf->e_shnum = 1;
	elf->e_shstrndx = SHN_UNDEF;

	memset(shdr4extnum, 0, sizeof(*shdr4extnum));

	shdr4extnum->sh_type = SHT_NULL;
	shdr4extnum->sh_size = elf->e_shnum;
	shdr4extnum->sh_link = elf->e_shstrndx;
	shdr4extnum->sh_info = segs;
}

L
Linus Torvalds 已提交
2134 2135 2136 2137 2138 2139 2140
/*
 * Actual dumper
 *
 * This is a two-pass process; first we find the offsets of the bits,
 * and then they are actually written out.  If we run out of core limit
 * we just truncate.
 */
2141
static int elf_core_dump(struct coredump_params *cprm)
L
Linus Torvalds 已提交
2142 2143 2144
{
	int has_dumped = 0;
	mm_segment_t fs;
2145 2146
	int segs, i;
	size_t vma_data_size = 0;
2147
	struct vm_area_struct *vma, *gate_vma;
L
Linus Torvalds 已提交
2148
	struct elfhdr *elf = NULL;
2149
	loff_t offset = 0, dataoff;
2150
	struct elf_note_info info = { };
2151
	struct elf_phdr *phdr4note = NULL;
2152 2153 2154
	struct elf_shdr *shdr4extnum = NULL;
	Elf_Half e_phnum;
	elf_addr_t e_shoff;
2155
	elf_addr_t *vma_filesz = NULL;
L
Linus Torvalds 已提交
2156 2157 2158 2159

	/*
	 * We no longer stop all VM operations.
	 * 
2160 2161 2162
	 * This is because those proceses that could possibly change map_count
	 * or the mmap / vma pages are now blocked in do_exit on current
	 * finishing this core dump.
L
Linus Torvalds 已提交
2163 2164
	 *
	 * Only ptrace can touch these memory addresses, but it doesn't change
2165
	 * the map_count or the pages allocated. So no possibility of crashing
L
Linus Torvalds 已提交
2166 2167 2168 2169 2170 2171
	 * exists while dumping the mm->vm_next areas to the core file.
	 */
  
	/* alloc memory for large data structures: too large to be on stack */
	elf = kmalloc(sizeof(*elf), GFP_KERNEL);
	if (!elf)
2172
		goto out;
2173 2174 2175 2176
	/*
	 * The number of segs are recored into ELF header as 16bit value.
	 * Please check DEFAULT_MAX_MAP_COUNT definition when you modify here.
	 */
L
Linus Torvalds 已提交
2177
	segs = current->mm->map_count;
2178
	segs += elf_core_extra_phdrs();
L
Linus Torvalds 已提交
2179

2180
	gate_vma = get_gate_vma(current->mm);
2181 2182 2183
	if (gate_vma != NULL)
		segs++;

2184 2185 2186 2187 2188 2189 2190 2191
	/* for notes section */
	segs++;

	/* If segs > PN_XNUM(0xffff), then e_phnum overflows. To avoid
	 * this, kernel supports extended numbering. Have a look at
	 * include/linux/elf.h for further information. */
	e_phnum = segs > PN_XNUM ? PN_XNUM : segs;

L
Linus Torvalds 已提交
2192
	/*
R
Roland McGrath 已提交
2193 2194
	 * Collect all the non-memory information about the process for the
	 * notes.  This also sets up the file header.
L
Linus Torvalds 已提交
2195
	 */
2196
	if (!fill_note_info(elf, e_phnum, &info, cprm->siginfo, cprm->regs))
R
Roland McGrath 已提交
2197
		goto cleanup;
L
Linus Torvalds 已提交
2198

R
Roland McGrath 已提交
2199
	has_dumped = 1;
2200

L
Linus Torvalds 已提交
2201 2202 2203 2204
	fs = get_fs();
	set_fs(KERNEL_DS);

	offset += sizeof(*elf);				/* Elf header */
2205
	offset += segs * sizeof(struct elf_phdr);	/* Program headers */
L
Linus Torvalds 已提交
2206 2207 2208

	/* Write notes phdr entry */
	{
R
Roland McGrath 已提交
2209
		size_t sz = get_note_info_size(&info);
L
Linus Torvalds 已提交
2210

2211
		sz += elf_coredump_extra_notes_size();
2212

2213 2214
		phdr4note = kmalloc(sizeof(*phdr4note), GFP_KERNEL);
		if (!phdr4note)
2215
			goto end_coredump;
2216 2217 2218

		fill_elf_note_phdr(phdr4note, sz, offset);
		offset += sz;
L
Linus Torvalds 已提交
2219 2220 2221 2222
	}

	dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE);

2223 2224 2225
	if (segs - 1 > ULONG_MAX / sizeof(*vma_filesz))
		goto end_coredump;
	vma_filesz = vmalloc((segs - 1) * sizeof(*vma_filesz));
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238
	if (!vma_filesz)
		goto end_coredump;

	for (i = 0, vma = first_vma(current, gate_vma); vma != NULL;
			vma = next_vma(vma, gate_vma)) {
		unsigned long dump_size;

		dump_size = vma_dump_size(vma, cprm->mm_flags);
		vma_filesz[i++] = dump_size;
		vma_data_size += dump_size;
	}

	offset += vma_data_size;
2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250
	offset += elf_core_extra_data_size();
	e_shoff = offset;

	if (e_phnum == PN_XNUM) {
		shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL);
		if (!shdr4extnum)
			goto end_coredump;
		fill_extnum_info(elf, shdr4extnum, e_shoff, segs);
	}

	offset = dataoff;

A
Al Viro 已提交
2251
	if (!dump_emit(cprm, elf, sizeof(*elf)))
2252 2253
		goto end_coredump;

A
Al Viro 已提交
2254
	if (!dump_emit(cprm, phdr4note, sizeof(*phdr4note)))
2255 2256
		goto end_coredump;

L
Linus Torvalds 已提交
2257
	/* Write program headers for segments dump */
2258
	for (i = 0, vma = first_vma(current, gate_vma); vma != NULL;
2259
			vma = next_vma(vma, gate_vma)) {
L
Linus Torvalds 已提交
2260 2261 2262 2263 2264 2265
		struct elf_phdr phdr;

		phdr.p_type = PT_LOAD;
		phdr.p_offset = offset;
		phdr.p_vaddr = vma->vm_start;
		phdr.p_paddr = 0;
2266
		phdr.p_filesz = vma_filesz[i++];
R
Roland McGrath 已提交
2267
		phdr.p_memsz = vma->vm_end - vma->vm_start;
L
Linus Torvalds 已提交
2268 2269
		offset += phdr.p_filesz;
		phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0;
2270 2271 2272 2273
		if (vma->vm_flags & VM_WRITE)
			phdr.p_flags |= PF_W;
		if (vma->vm_flags & VM_EXEC)
			phdr.p_flags |= PF_X;
L
Linus Torvalds 已提交
2274 2275
		phdr.p_align = ELF_EXEC_PAGESIZE;

A
Al Viro 已提交
2276
		if (!dump_emit(cprm, &phdr, sizeof(phdr)))
2277
			goto end_coredump;
L
Linus Torvalds 已提交
2278 2279
	}

2280
	if (!elf_core_write_extra_phdrs(cprm, offset))
2281
		goto end_coredump;
L
Linus Torvalds 已提交
2282 2283

 	/* write out the notes section */
A
Al Viro 已提交
2284
	if (!write_note_info(&info, cprm))
R
Roland McGrath 已提交
2285
		goto end_coredump;
L
Linus Torvalds 已提交
2286

2287
	if (elf_coredump_extra_notes_write(cprm))
2288
		goto end_coredump;
2289

2290
	/* Align to page */
2291
	if (!dump_skip(cprm, dataoff - cprm->pos))
H
Hugh Dickins 已提交
2292
		goto end_coredump;
L
Linus Torvalds 已提交
2293

2294
	for (i = 0, vma = first_vma(current, gate_vma); vma != NULL;
2295
			vma = next_vma(vma, gate_vma)) {
L
Linus Torvalds 已提交
2296
		unsigned long addr;
R
Roland McGrath 已提交
2297
		unsigned long end;
L
Linus Torvalds 已提交
2298

2299
		end = vma->vm_start + vma_filesz[i++];
L
Linus Torvalds 已提交
2300

R
Roland McGrath 已提交
2301
		for (addr = vma->vm_start; addr < end; addr += PAGE_SIZE) {
2302
			struct page *page;
H
Hugh Dickins 已提交
2303 2304 2305 2306 2307
			int stop;

			page = get_dump_page(addr);
			if (page) {
				void *kaddr = kmap(page);
2308
				stop = !dump_emit(cprm, kaddr, PAGE_SIZE);
H
Hugh Dickins 已提交
2309
				kunmap(page);
2310
				put_page(page);
H
Hugh Dickins 已提交
2311
			} else
2312
				stop = !dump_skip(cprm, PAGE_SIZE);
H
Hugh Dickins 已提交
2313 2314
			if (stop)
				goto end_coredump;
L
Linus Torvalds 已提交
2315 2316
		}
	}
2317
	dump_truncate(cprm);
L
Linus Torvalds 已提交
2318

2319
	if (!elf_core_write_extra_data(cprm))
2320
		goto end_coredump;
L
Linus Torvalds 已提交
2321

2322
	if (e_phnum == PN_XNUM) {
2323
		if (!dump_emit(cprm, shdr4extnum, sizeof(*shdr4extnum)))
2324 2325 2326
			goto end_coredump;
	}

L
Linus Torvalds 已提交
2327 2328 2329 2330
end_coredump:
	set_fs(fs);

cleanup:
R
Roland McGrath 已提交
2331
	free_note_info(&info);
2332
	kfree(shdr4extnum);
2333
	vfree(vma_filesz);
2334
	kfree(phdr4note);
2335 2336
	kfree(elf);
out:
L
Linus Torvalds 已提交
2337 2338 2339
	return has_dumped;
}

C
Christoph Hellwig 已提交
2340
#endif		/* CONFIG_ELF_CORE */
L
Linus Torvalds 已提交
2341 2342 2343

static int __init init_elf_binfmt(void)
{
A
Al Viro 已提交
2344 2345
	register_binfmt(&elf_format);
	return 0;
L
Linus Torvalds 已提交
2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356
}

static void __exit exit_elf_binfmt(void)
{
	/* Remove the COFF and ELF loaders. */
	unregister_binfmt(&elf_format);
}

core_initcall(init_elf_binfmt);
module_exit(exit_elf_binfmt);
MODULE_LICENSE("GPL");