mem.c 20.3 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
/*
 *  linux/drivers/char/mem.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  Added devfs support. 
 *    Jan-11-1998, C. Scott Ananian <cananian@alumni.princeton.edu>
 *  Shared /dev/zero mmaping support, Feb 2000, Kanoj Sarcar <kanoj@sgi.com>
 */

#include <linux/mm.h>
#include <linux/miscdevice.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/mman.h>
#include <linux/random.h>
#include <linux/init.h>
#include <linux/raw.h>
#include <linux/tty.h>
#include <linux/capability.h>
#include <linux/ptrace.h>
#include <linux/device.h>
23 24
#include <linux/highmem.h>
#include <linux/crash_dump.h>
L
Linus Torvalds 已提交
25
#include <linux/backing-dev.h>
26
#include <linux/bootmem.h>
27
#include <linux/splice.h>
28
#include <linux/pfn.h>
29
#include <linux/smp_lock.h>
L
Linus Torvalds 已提交
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

#include <asm/uaccess.h>
#include <asm/io.h>

#ifdef CONFIG_IA64
# include <linux/efi.h>
#endif

/*
 * Architectures vary in how they handle caching for addresses
 * outside of main memory.
 *
 */
static inline int uncached_access(struct file *file, unsigned long addr)
{
45
#if defined(CONFIG_IA64)
L
Linus Torvalds 已提交
46 47 48 49
	/*
	 * On ia64, we ignore O_SYNC because we cannot tolerate memory attribute aliases.
	 */
	return !(efi_mem_attributes(addr) & EFI_MEMORY_WB);
50 51 52 53 54 55 56
#elif defined(CONFIG_MIPS)
	{
		extern int __uncached_access(struct file *file,
					     unsigned long addr);

		return __uncached_access(file, addr);
	}
L
Linus Torvalds 已提交
57 58 59 60 61 62 63 64 65 66 67 68
#else
	/*
	 * Accessing memory above the top the kernel knows about or through a file pointer
	 * that was marked O_SYNC will be done non-cached.
	 */
	if (file->f_flags & O_SYNC)
		return 1;
	return addr >= __pa(high_memory);
#endif
}

#ifndef ARCH_HAS_VALID_PHYS_ADDR_RANGE
69
static inline int valid_phys_addr_range(unsigned long addr, size_t count)
L
Linus Torvalds 已提交
70
{
71
	if (addr + count > __pa(high_memory))
L
Linus Torvalds 已提交
72 73 74 75
		return 0;

	return 1;
}
76

77
static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t size)
78 79 80
{
	return 1;
}
L
Linus Torvalds 已提交
81 82
#endif

83
#ifdef CONFIG_STRICT_DEVMEM
84
static inline int range_is_allowed(unsigned long pfn, unsigned long size)
85
{
86 87 88 89 90 91 92 93
	u64 from = ((u64)pfn) << PAGE_SHIFT;
	u64 to = from + size;
	u64 cursor = from;

	while (cursor < to) {
		if (!devmem_is_allowed(pfn)) {
			printk(KERN_INFO
		"Program %s tried to access /dev/mem between %Lx->%Lx.\n",
94 95 96
				current->comm, from, to);
			return 0;
		}
97 98
		cursor += PAGE_SIZE;
		pfn++;
99 100 101 102
	}
	return 1;
}
#else
103
static inline int range_is_allowed(unsigned long pfn, unsigned long size)
104 105 106 107 108
{
	return 1;
}
#endif

109 110 111 112
void __attribute__((weak)) unxlate_dev_mem_ptr(unsigned long phys, void *addr)
{
}

L
Linus Torvalds 已提交
113 114 115 116 117 118 119 120 121 122 123
/*
 * This funcion reads the *physical* memory. The f_pos points directly to the 
 * memory location. 
 */
static ssize_t read_mem(struct file * file, char __user * buf,
			size_t count, loff_t *ppos)
{
	unsigned long p = *ppos;
	ssize_t read, sz;
	char *ptr;

124
	if (!valid_phys_addr_range(p, count))
L
Linus Torvalds 已提交
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154
		return -EFAULT;
	read = 0;
#ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED
	/* we don't have page 0 mapped on sparc and m68k.. */
	if (p < PAGE_SIZE) {
		sz = PAGE_SIZE - p;
		if (sz > count) 
			sz = count; 
		if (sz > 0) {
			if (clear_user(buf, sz))
				return -EFAULT;
			buf += sz; 
			p += sz; 
			count -= sz; 
			read += sz; 
		}
	}
#endif

	while (count > 0) {
		/*
		 * Handle first page in case it's not aligned
		 */
		if (-p & (PAGE_SIZE - 1))
			sz = -p & (PAGE_SIZE - 1);
		else
			sz = PAGE_SIZE;

		sz = min_t(unsigned long, sz, count);

155 156 157
		if (!range_is_allowed(p >> PAGE_SHIFT, count))
			return -EPERM;

L
Linus Torvalds 已提交
158 159 160 161 162 163
		/*
		 * On ia64 if a page has been mapped somewhere as
		 * uncached, then it must also be accessed uncached
		 * by the kernel or data corruption may occur
		 */
		ptr = xlate_dev_mem_ptr(p);
164 165
		if (!ptr)
			return -EFAULT;
L
Linus Torvalds 已提交
166

167 168
		if (copy_to_user(buf, ptr, sz)) {
			unxlate_dev_mem_ptr(p, ptr);
L
Linus Torvalds 已提交
169
			return -EFAULT;
170 171 172 173
		}

		unxlate_dev_mem_ptr(p, ptr);

L
Linus Torvalds 已提交
174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
		buf += sz;
		p += sz;
		count -= sz;
		read += sz;
	}

	*ppos += read;
	return read;
}

static ssize_t write_mem(struct file * file, const char __user * buf, 
			 size_t count, loff_t *ppos)
{
	unsigned long p = *ppos;
	ssize_t written, sz;
	unsigned long copied;
	void *ptr;

192
	if (!valid_phys_addr_range(p, count))
L
Linus Torvalds 已提交
193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221
		return -EFAULT;

	written = 0;

#ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED
	/* we don't have page 0 mapped on sparc and m68k.. */
	if (p < PAGE_SIZE) {
		unsigned long sz = PAGE_SIZE - p;
		if (sz > count)
			sz = count;
		/* Hmm. Do something? */
		buf += sz;
		p += sz;
		count -= sz;
		written += sz;
	}
#endif

	while (count > 0) {
		/*
		 * Handle first page in case it's not aligned
		 */
		if (-p & (PAGE_SIZE - 1))
			sz = -p & (PAGE_SIZE - 1);
		else
			sz = PAGE_SIZE;

		sz = min_t(unsigned long, sz, count);

222 223 224
		if (!range_is_allowed(p >> PAGE_SHIFT, sz))
			return -EPERM;

L
Linus Torvalds 已提交
225 226 227 228 229 230
		/*
		 * On ia64 if a page has been mapped somewhere as
		 * uncached, then it must also be accessed uncached
		 * by the kernel or data corruption may occur
		 */
		ptr = xlate_dev_mem_ptr(p);
231 232 233 234 235
		if (!ptr) {
			if (written)
				break;
			return -EFAULT;
		}
L
Linus Torvalds 已提交
236 237 238

		copied = copy_from_user(ptr, buf, sz);
		if (copied) {
239
			written += sz - copied;
240
			unxlate_dev_mem_ptr(p, ptr);
241 242
			if (written)
				break;
L
Linus Torvalds 已提交
243 244
			return -EFAULT;
		}
245 246 247

		unxlate_dev_mem_ptr(p, ptr);

L
Linus Torvalds 已提交
248 249 250 251 252 253 254 255 256 257
		buf += sz;
		p += sz;
		count -= sz;
		written += sz;
	}

	*ppos += written;
	return written;
}

258 259 260 261 262 263
int __attribute__((weak)) phys_mem_access_prot_allowed(struct file *file,
	unsigned long pfn, unsigned long size, pgprot_t *vma_prot)
{
	return 1;
}

264 265 266 267 268 269 270 271 272 273 274 275 276 277
#ifndef __HAVE_PHYS_MEM_ACCESS_PROT
static pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
				     unsigned long size, pgprot_t vma_prot)
{
#ifdef pgprot_noncached
	unsigned long offset = pfn << PAGE_SHIFT;

	if (uncached_access(file, offset))
		return pgprot_noncached(vma_prot);
#endif
	return vma_prot;
}
#endif

278 279 280 281 282 283 284 285 286
#ifndef CONFIG_MMU
static unsigned long get_unmapped_area_mem(struct file *file,
					   unsigned long addr,
					   unsigned long len,
					   unsigned long pgoff,
					   unsigned long flags)
{
	if (!valid_mmap_phys_addr_range(pgoff, len))
		return (unsigned long) -EINVAL;
287
	return pgoff << PAGE_SHIFT;
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
}

/* can't do an in-place private mapping if there's no MMU */
static inline int private_mapping_ok(struct vm_area_struct *vma)
{
	return vma->vm_flags & VM_MAYSHARE;
}
#else
#define get_unmapped_area_mem	NULL

static inline int private_mapping_ok(struct vm_area_struct *vma)
{
	return 1;
}
#endif

304
static struct vm_operations_struct mmap_mem_ops = {
305 306 307
#ifdef CONFIG_HAVE_IOREMAP_PROT
	.access = generic_access_phys
#endif
308 309
};

L
Linus Torvalds 已提交
310 311
static int mmap_mem(struct file * file, struct vm_area_struct * vma)
{
312 313
	size_t size = vma->vm_end - vma->vm_start;

314
	if (!valid_mmap_phys_addr_range(vma->vm_pgoff, size))
315 316
		return -EINVAL;

317 318 319
	if (!private_mapping_ok(vma))
		return -ENOSYS;

320 321 322
	if (!range_is_allowed(vma->vm_pgoff, size))
		return -EPERM;

323 324 325 326
	if (!phys_mem_access_prot_allowed(file, vma->vm_pgoff, size,
						&vma->vm_page_prot))
		return -EINVAL;

327
	vma->vm_page_prot = phys_mem_access_prot(file, vma->vm_pgoff,
328
						 size,
L
Linus Torvalds 已提交
329 330
						 vma->vm_page_prot);

331 332
	vma->vm_ops = &mmap_mem_ops;

L
Linus Torvalds 已提交
333 334 335 336
	/* Remap-pfn-range will mark the range VM_IO and VM_RESERVED */
	if (remap_pfn_range(vma,
			    vma->vm_start,
			    vma->vm_pgoff,
337
			    size,
338
			    vma->vm_page_prot)) {
L
Linus Torvalds 已提交
339
		return -EAGAIN;
340
	}
L
Linus Torvalds 已提交
341 342 343
	return 0;
}

344
#ifdef CONFIG_DEVKMEM
L
Linus Torvalds 已提交
345 346
static int mmap_kmem(struct file * file, struct vm_area_struct * vma)
{
L
Linus Torvalds 已提交
347 348
	unsigned long pfn;

349 350
	/* Turn a kernel-virtual address into a physical page frame */
	pfn = __pa((u64)vma->vm_pgoff << PAGE_SHIFT) >> PAGE_SHIFT;
L
Linus Torvalds 已提交
351

L
Linus Torvalds 已提交
352 353 354 355 356 357 358
	/*
	 * RED-PEN: on some architectures there is more mapped memory
	 * than available in mem_map which pfn_valid checks
	 * for. Perhaps should add a new macro here.
	 *
	 * RED-PEN: vmalloc is not supported right now.
	 */
L
Linus Torvalds 已提交
359
	if (!pfn_valid(pfn))
L
Linus Torvalds 已提交
360
		return -EIO;
L
Linus Torvalds 已提交
361 362

	vma->vm_pgoff = pfn;
L
Linus Torvalds 已提交
363 364
	return mmap_mem(file, vma);
}
365
#endif
L
Linus Torvalds 已提交
366

367 368 369 370
#ifdef CONFIG_CRASH_DUMP
/*
 * Read memory corresponding to the old kernel.
 */
371
static ssize_t read_oldmem(struct file *file, char __user *buf,
372 373
				size_t count, loff_t *ppos)
{
374 375 376
	unsigned long pfn, offset;
	size_t read = 0, csize;
	int rc = 0;
377

M
Maneesh Soni 已提交
378
	while (count) {
379
		pfn = *ppos / PAGE_SIZE;
380 381
		if (pfn > saved_max_pfn)
			return read;
382

383 384 385 386 387
		offset = (unsigned long)(*ppos % PAGE_SIZE);
		if (count > PAGE_SIZE - offset)
			csize = PAGE_SIZE - offset;
		else
			csize = count;
388

389 390 391
		rc = copy_oldmem_page(pfn, buf, csize, offset, 1);
		if (rc < 0)
			return rc;
392 393 394 395 396 397 398 399
		buf += csize;
		*ppos += csize;
		read += csize;
		count -= csize;
	}
	return read;
}
#endif
L
Linus Torvalds 已提交
400

401
#ifdef CONFIG_DEVKMEM
L
Linus Torvalds 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531
/*
 * This function reads the *virtual* memory as seen by the kernel.
 */
static ssize_t read_kmem(struct file *file, char __user *buf, 
			 size_t count, loff_t *ppos)
{
	unsigned long p = *ppos;
	ssize_t low_count, read, sz;
	char * kbuf; /* k-addr because vread() takes vmlist_lock rwlock */

	read = 0;
	if (p < (unsigned long) high_memory) {
		low_count = count;
		if (count > (unsigned long) high_memory - p)
			low_count = (unsigned long) high_memory - p;

#ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED
		/* we don't have page 0 mapped on sparc and m68k.. */
		if (p < PAGE_SIZE && low_count > 0) {
			size_t tmp = PAGE_SIZE - p;
			if (tmp > low_count) tmp = low_count;
			if (clear_user(buf, tmp))
				return -EFAULT;
			buf += tmp;
			p += tmp;
			read += tmp;
			low_count -= tmp;
			count -= tmp;
		}
#endif
		while (low_count > 0) {
			/*
			 * Handle first page in case it's not aligned
			 */
			if (-p & (PAGE_SIZE - 1))
				sz = -p & (PAGE_SIZE - 1);
			else
				sz = PAGE_SIZE;

			sz = min_t(unsigned long, sz, low_count);

			/*
			 * On ia64 if a page has been mapped somewhere as
			 * uncached, then it must also be accessed uncached
			 * by the kernel or data corruption may occur
			 */
			kbuf = xlate_dev_kmem_ptr((char *)p);

			if (copy_to_user(buf, kbuf, sz))
				return -EFAULT;
			buf += sz;
			p += sz;
			read += sz;
			low_count -= sz;
			count -= sz;
		}
	}

	if (count > 0) {
		kbuf = (char *)__get_free_page(GFP_KERNEL);
		if (!kbuf)
			return -ENOMEM;
		while (count > 0) {
			int len = count;

			if (len > PAGE_SIZE)
				len = PAGE_SIZE;
			len = vread(kbuf, (char *)p, len);
			if (!len)
				break;
			if (copy_to_user(buf, kbuf, len)) {
				free_page((unsigned long)kbuf);
				return -EFAULT;
			}
			count -= len;
			buf += len;
			read += len;
			p += len;
		}
		free_page((unsigned long)kbuf);
	}
 	*ppos = p;
 	return read;
}


static inline ssize_t
do_write_kmem(void *p, unsigned long realp, const char __user * buf,
	      size_t count, loff_t *ppos)
{
	ssize_t written, sz;
	unsigned long copied;

	written = 0;
#ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED
	/* we don't have page 0 mapped on sparc and m68k.. */
	if (realp < PAGE_SIZE) {
		unsigned long sz = PAGE_SIZE - realp;
		if (sz > count)
			sz = count;
		/* Hmm. Do something? */
		buf += sz;
		p += sz;
		realp += sz;
		count -= sz;
		written += sz;
	}
#endif

	while (count > 0) {
		char *ptr;
		/*
		 * Handle first page in case it's not aligned
		 */
		if (-realp & (PAGE_SIZE - 1))
			sz = -realp & (PAGE_SIZE - 1);
		else
			sz = PAGE_SIZE;

		sz = min_t(unsigned long, sz, count);

		/*
		 * On ia64 if a page has been mapped somewhere as
		 * uncached, then it must also be accessed uncached
		 * by the kernel or data corruption may occur
		 */
		ptr = xlate_dev_kmem_ptr(p);

		copied = copy_from_user(ptr, buf, sz);
		if (copied) {
532 533 534
			written += sz - copied;
			if (written)
				break;
L
Linus Torvalds 已提交
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
			return -EFAULT;
		}
		buf += sz;
		p += sz;
		realp += sz;
		count -= sz;
		written += sz;
	}

	*ppos += written;
	return written;
}


/*
 * This function writes to the *virtual* memory as seen by the kernel.
 */
static ssize_t write_kmem(struct file * file, const char __user * buf, 
			  size_t count, loff_t *ppos)
{
	unsigned long p = *ppos;
	ssize_t wrote = 0;
	ssize_t virtr = 0;
	ssize_t written;
	char * kbuf; /* k-addr because vwrite() takes vmlist_lock rwlock */

	if (p < (unsigned long) high_memory) {

		wrote = count;
		if (count > (unsigned long) high_memory - p)
			wrote = (unsigned long) high_memory - p;

		written = do_write_kmem((void*)p, p, buf, wrote, ppos);
		if (written != wrote)
			return written;
		wrote = written;
		p += wrote;
		buf += wrote;
		count -= wrote;
	}

	if (count > 0) {
		kbuf = (char *)__get_free_page(GFP_KERNEL);
		if (!kbuf)
			return wrote ? wrote : -ENOMEM;
		while (count > 0) {
			int len = count;

			if (len > PAGE_SIZE)
				len = PAGE_SIZE;
			if (len) {
				written = copy_from_user(kbuf, buf, len);
				if (written) {
588 589
					if (wrote + virtr)
						break;
L
Linus Torvalds 已提交
590
					free_page((unsigned long)kbuf);
591
					return -EFAULT;
L
Linus Torvalds 已提交
592 593 594 595 596 597 598 599 600 601 602 603 604 605
				}
			}
			len = vwrite(kbuf, (char *)p, len);
			count -= len;
			buf += len;
			virtr += len;
			p += len;
		}
		free_page((unsigned long)kbuf);
	}

 	*ppos = p;
 	return virtr + wrote;
}
606
#endif
L
Linus Torvalds 已提交
607

608
#ifdef CONFIG_DEVPORT
L
Linus Torvalds 已提交
609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636
static ssize_t read_port(struct file * file, char __user * buf,
			 size_t count, loff_t *ppos)
{
	unsigned long i = *ppos;
	char __user *tmp = buf;

	if (!access_ok(VERIFY_WRITE, buf, count))
		return -EFAULT; 
	while (count-- > 0 && i < 65536) {
		if (__put_user(inb(i),tmp) < 0) 
			return -EFAULT;  
		i++;
		tmp++;
	}
	*ppos = i;
	return tmp-buf;
}

static ssize_t write_port(struct file * file, const char __user * buf,
			  size_t count, loff_t *ppos)
{
	unsigned long i = *ppos;
	const char __user * tmp = buf;

	if (!access_ok(VERIFY_READ,buf,count))
		return -EFAULT;
	while (count-- > 0 && i < 65536) {
		char c;
637 638 639
		if (__get_user(c, tmp)) {
			if (tmp > buf)
				break;
L
Linus Torvalds 已提交
640
			return -EFAULT; 
641
		}
L
Linus Torvalds 已提交
642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662
		outb(c,i);
		i++;
		tmp++;
	}
	*ppos = i;
	return tmp-buf;
}
#endif

static ssize_t read_null(struct file * file, char __user * buf,
			 size_t count, loff_t *ppos)
{
	return 0;
}

static ssize_t write_null(struct file * file, const char __user * buf,
			  size_t count, loff_t *ppos)
{
	return count;
}

663 664 665 666 667 668 669 670 671 672 673 674
static int pipe_to_null(struct pipe_inode_info *info, struct pipe_buffer *buf,
			struct splice_desc *sd)
{
	return sd->len;
}

static ssize_t splice_write_null(struct pipe_inode_info *pipe,struct file *out,
				 loff_t *ppos, size_t len, unsigned int flags)
{
	return splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_null);
}

L
Linus Torvalds 已提交
675 676 677
static ssize_t read_zero(struct file * file, char __user * buf, 
			 size_t count, loff_t *ppos)
{
N
Nick Piggin 已提交
678
	size_t written;
L
Linus Torvalds 已提交
679 680 681 682 683 684 685

	if (!count)
		return 0;

	if (!access_ok(VERIFY_WRITE, buf, count))
		return -EFAULT;

N
Nick Piggin 已提交
686 687 688 689
	written = 0;
	while (count) {
		unsigned long unwritten;
		size_t chunk = count;
L
Linus Torvalds 已提交
690

N
Nick Piggin 已提交
691 692 693 694
		if (chunk > PAGE_SIZE)
			chunk = PAGE_SIZE;	/* Just for latency reasons */
		unwritten = clear_user(buf, chunk);
		written += chunk - unwritten;
L
Linus Torvalds 已提交
695
		if (unwritten)
N
Nick Piggin 已提交
696
			break;
697 698
		if (signal_pending(current))
			return written ? written : -ERESTARTSYS;
L
Linus Torvalds 已提交
699
		buf += chunk;
N
Nick Piggin 已提交
700
		count -= chunk;
L
Linus Torvalds 已提交
701 702
		cond_resched();
	}
N
Nick Piggin 已提交
703
	return written ? written : -EFAULT;
L
Linus Torvalds 已提交
704 705 706 707
}

static int mmap_zero(struct file * file, struct vm_area_struct * vma)
{
N
Nick Piggin 已提交
708
#ifndef CONFIG_MMU
L
Linus Torvalds 已提交
709
	return -ENOSYS;
N
Nick Piggin 已提交
710 711 712 713
#endif
	if (vma->vm_flags & VM_SHARED)
		return shmem_zero_setup(vma);
	return 0;
L
Linus Torvalds 已提交
714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
}

static ssize_t write_full(struct file * file, const char __user * buf,
			  size_t count, loff_t *ppos)
{
	return -ENOSPC;
}

/*
 * Special lseek() function for /dev/null and /dev/zero.  Most notably, you
 * can fopen() both devices with "a" now.  This was previously impossible.
 * -- SRB.
 */

static loff_t null_lseek(struct file * file, loff_t offset, int orig)
{
	return file->f_pos = 0;
}

/*
 * The memory devices use the full 32/64 bits of the offset, and so we cannot
 * check against negative addresses: they are ok. The return value is weird,
 * though, in that case (0).
 *
 * also note that seeking relative to the "end of file" isn't supported:
 * it has no meaning, so it returns -EINVAL.
 */
static loff_t memory_lseek(struct file * file, loff_t offset, int orig)
{
	loff_t ret;

745
	mutex_lock(&file->f_path.dentry->d_inode->i_mutex);
L
Linus Torvalds 已提交
746 747 748 749 750 751 752 753 754 755 756 757 758 759
	switch (orig) {
		case 0:
			file->f_pos = offset;
			ret = file->f_pos;
			force_successful_syscall_return();
			break;
		case 1:
			file->f_pos += offset;
			ret = file->f_pos;
			force_successful_syscall_return();
			break;
		default:
			ret = -EINVAL;
	}
760
	mutex_unlock(&file->f_path.dentry->d_inode->i_mutex);
L
Linus Torvalds 已提交
761 762 763 764 765 766 767 768 769 770 771 772 773 774
	return ret;
}

static int open_port(struct inode * inode, struct file * filp)
{
	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
}

#define zero_lseek	null_lseek
#define full_lseek      null_lseek
#define write_zero	write_null
#define read_full       read_zero
#define open_mem	open_port
#define open_kmem	open_mem
775
#define open_oldmem	open_mem
L
Linus Torvalds 已提交
776

777
static const struct file_operations mem_fops = {
L
Linus Torvalds 已提交
778 779 780 781 782
	.llseek		= memory_lseek,
	.read		= read_mem,
	.write		= write_mem,
	.mmap		= mmap_mem,
	.open		= open_mem,
783
	.get_unmapped_area = get_unmapped_area_mem,
L
Linus Torvalds 已提交
784 785
};

786
#ifdef CONFIG_DEVKMEM
787
static const struct file_operations kmem_fops = {
L
Linus Torvalds 已提交
788 789 790 791 792
	.llseek		= memory_lseek,
	.read		= read_kmem,
	.write		= write_kmem,
	.mmap		= mmap_kmem,
	.open		= open_kmem,
793
	.get_unmapped_area = get_unmapped_area_mem,
L
Linus Torvalds 已提交
794
};
795
#endif
L
Linus Torvalds 已提交
796

797
static const struct file_operations null_fops = {
L
Linus Torvalds 已提交
798 799 800
	.llseek		= null_lseek,
	.read		= read_null,
	.write		= write_null,
801
	.splice_write	= splice_write_null,
L
Linus Torvalds 已提交
802 803
};

804
#ifdef CONFIG_DEVPORT
805
static const struct file_operations port_fops = {
L
Linus Torvalds 已提交
806 807 808 809 810 811 812
	.llseek		= memory_lseek,
	.read		= read_port,
	.write		= write_port,
	.open		= open_port,
};
#endif

813
static const struct file_operations zero_fops = {
L
Linus Torvalds 已提交
814 815 816 817 818 819
	.llseek		= zero_lseek,
	.read		= read_zero,
	.write		= write_zero,
	.mmap		= mmap_zero,
};

820 821 822 823
/*
 * capabilities for /dev/zero
 * - permits private mappings, "copies" are taken of the source of zeros
 */
L
Linus Torvalds 已提交
824 825 826 827
static struct backing_dev_info zero_bdi = {
	.capabilities	= BDI_CAP_MAP_COPY,
};

828
static const struct file_operations full_fops = {
L
Linus Torvalds 已提交
829 830 831 832 833
	.llseek		= full_lseek,
	.read		= read_full,
	.write		= write_full,
};

834
#ifdef CONFIG_CRASH_DUMP
835
static const struct file_operations oldmem_fops = {
836 837 838 839 840
	.read	= read_oldmem,
	.open	= open_oldmem,
};
#endif

L
Linus Torvalds 已提交
841 842 843 844
static ssize_t kmsg_write(struct file * file, const char __user * buf,
			  size_t count, loff_t *ppos)
{
	char *tmp;
845
	ssize_t ret;
L
Linus Torvalds 已提交
846 847 848 849 850 851 852 853

	tmp = kmalloc(count + 1, GFP_KERNEL);
	if (tmp == NULL)
		return -ENOMEM;
	ret = -EFAULT;
	if (!copy_from_user(tmp, buf, count)) {
		tmp[count] = 0;
		ret = printk("%s", tmp);
854 855 856
		if (ret > count)
			/* printk can add a prefix */
			ret = count;
L
Linus Torvalds 已提交
857 858 859 860 861
	}
	kfree(tmp);
	return ret;
}

862
static const struct file_operations kmsg_fops = {
L
Linus Torvalds 已提交
863 864 865 866 867
	.write =	kmsg_write,
};

static int memory_open(struct inode * inode, struct file * filp)
{
868 869 870
	int ret = 0;

	lock_kernel();
L
Linus Torvalds 已提交
871 872 873
	switch (iminor(inode)) {
		case 1:
			filp->f_op = &mem_fops;
874 875
			filp->f_mapping->backing_dev_info =
				&directly_mappable_cdev_bdi;
L
Linus Torvalds 已提交
876
			break;
877
#ifdef CONFIG_DEVKMEM
L
Linus Torvalds 已提交
878 879
		case 2:
			filp->f_op = &kmem_fops;
880 881
			filp->f_mapping->backing_dev_info =
				&directly_mappable_cdev_bdi;
L
Linus Torvalds 已提交
882
			break;
883
#endif
L
Linus Torvalds 已提交
884 885 886
		case 3:
			filp->f_op = &null_fops;
			break;
887
#ifdef CONFIG_DEVPORT
L
Linus Torvalds 已提交
888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907
		case 4:
			filp->f_op = &port_fops;
			break;
#endif
		case 5:
			filp->f_mapping->backing_dev_info = &zero_bdi;
			filp->f_op = &zero_fops;
			break;
		case 7:
			filp->f_op = &full_fops;
			break;
		case 8:
			filp->f_op = &random_fops;
			break;
		case 9:
			filp->f_op = &urandom_fops;
			break;
		case 11:
			filp->f_op = &kmsg_fops;
			break;
908 909 910 911 912
#ifdef CONFIG_CRASH_DUMP
		case 12:
			filp->f_op = &oldmem_fops;
			break;
#endif
L
Linus Torvalds 已提交
913
		default:
914
			unlock_kernel();
L
Linus Torvalds 已提交
915 916 917
			return -ENXIO;
	}
	if (filp->f_op && filp->f_op->open)
918 919 920
		ret = filp->f_op->open(inode,filp);
	unlock_kernel();
	return ret;
L
Linus Torvalds 已提交
921 922
}

923
static const struct file_operations memory_fops = {
L
Linus Torvalds 已提交
924 925 926 927 928 929 930
	.open		= memory_open,	/* just a selector for the real open */
};

static const struct {
	unsigned int		minor;
	char			*name;
	umode_t			mode;
931
	const struct file_operations	*fops;
L
Linus Torvalds 已提交
932 933
} devlist[] = { /* list of minor devices */
	{1, "mem",     S_IRUSR | S_IWUSR | S_IRGRP, &mem_fops},
934
#ifdef CONFIG_DEVKMEM
L
Linus Torvalds 已提交
935
	{2, "kmem",    S_IRUSR | S_IWUSR | S_IRGRP, &kmem_fops},
936
#endif
L
Linus Torvalds 已提交
937
	{3, "null",    S_IRUGO | S_IWUGO,           &null_fops},
938
#ifdef CONFIG_DEVPORT
L
Linus Torvalds 已提交
939 940 941 942 943 944 945
	{4, "port",    S_IRUSR | S_IWUSR | S_IRGRP, &port_fops},
#endif
	{5, "zero",    S_IRUGO | S_IWUGO,           &zero_fops},
	{7, "full",    S_IRUGO | S_IWUGO,           &full_fops},
	{8, "random",  S_IRUGO | S_IWUSR,           &random_fops},
	{9, "urandom", S_IRUGO | S_IWUSR,           &urandom_fops},
	{11,"kmsg",    S_IRUGO | S_IWUSR,           &kmsg_fops},
946 947 948
#ifdef CONFIG_CRASH_DUMP
	{12,"oldmem",    S_IRUSR | S_IWUSR | S_IRGRP, &oldmem_fops},
#endif
L
Linus Torvalds 已提交
949 950
};

951
static struct class *mem_class;
L
Linus Torvalds 已提交
952 953 954 955

static int __init chr_dev_init(void)
{
	int i;
P
Peter Zijlstra 已提交
956 957 958 959 960
	int err;

	err = bdi_init(&zero_bdi);
	if (err)
		return err;
L
Linus Torvalds 已提交
961 962 963 964

	if (register_chrdev(MEM_MAJOR,"mem",&memory_fops))
		printk("unable to get major %d for memory devs\n", MEM_MAJOR);

965
	mem_class = class_create(THIS_MODULE, "mem");
966
	for (i = 0; i < ARRAY_SIZE(devlist); i++)
967 968 969
		device_create(mem_class, NULL,
			      MKDEV(MEM_MAJOR, devlist[i].minor), NULL,
			      devlist[i].name);
970

L
Linus Torvalds 已提交
971 972 973 974
	return 0;
}

fs_initcall(chr_dev_init);