ioctl.c 23.1 KB
Newer Older
H
Herbert Poetzl 已提交
1 2 3 4 5 6 7 8 9
/*
 * linux/fs/ocfs2/ioctl.c
 *
 * Copyright (C) 2006 Herbert Poetzl
 * adapted from Remy Card's ext2/ioctl.c
 */

#include <linux/fs.h>
#include <linux/mount.h>
10
#include <linux/compat.h>
H
Herbert Poetzl 已提交
11 12 13 14 15 16

#include <cluster/masklog.h>

#include "ocfs2.h"
#include "alloc.h"
#include "dlmglue.h"
17
#include "file.h"
H
Herbert Poetzl 已提交
18 19 20 21
#include "inode.h"
#include "journal.h"

#include "ocfs2_fs.h"
22
#include "ioctl.h"
23
#include "resize.h"
T
Tao Ma 已提交
24
#include "refcounttree.h"
25 26 27
#include "sysfile.h"
#include "dir.h"
#include "buffer_head_io.h"
28
#include "suballoc.h"
29
#include "move_extents.h"
30

31 32 33 34 35 36 37 38 39 40
#define o2info_from_user(a, b)	\
		copy_from_user(&(a), (b), sizeof(a))
#define o2info_to_user(a, b)	\
		copy_to_user((typeof(a) __user *)b, &(a), sizeof(a))

/*
 * This call is void because we are already reporting an error that may
 * be -EFAULT.  The error will be returned from the ioctl(2) call.  It's
 * just a best-effort to tell userspace that this request caused the error.
 */
41
static inline void o2info_set_request_error(struct ocfs2_info_request *kreq,
42 43 44 45 46 47
					struct ocfs2_info_request __user *req)
{
	kreq->ir_flags |= OCFS2_INFO_FL_ERROR;
	(void)put_user(kreq->ir_flags, (__u32 __user *)&(req->ir_flags));
}

48
static inline void o2info_set_request_filled(struct ocfs2_info_request *req)
49 50 51 52
{
	req->ir_flags |= OCFS2_INFO_FL_FILLED;
}

53
static inline void o2info_clear_request_filled(struct ocfs2_info_request *req)
54 55 56 57
{
	req->ir_flags &= ~OCFS2_INFO_FL_FILLED;
}

58 59 60 61
static inline int o2info_coherent(struct ocfs2_info_request *req)
{
	return (!(req->ir_flags & OCFS2_INFO_FL_NON_COHERENT));
}
62

H
Herbert Poetzl 已提交
63 64 65 66
static int ocfs2_get_inode_attr(struct inode *inode, unsigned *flags)
{
	int status;

M
Mark Fasheh 已提交
67
	status = ocfs2_inode_lock(inode, NULL, 0);
H
Herbert Poetzl 已提交
68 69 70 71
	if (status < 0) {
		mlog_errno(status);
		return status;
	}
72
	ocfs2_get_inode_flags(OCFS2_I(inode));
H
Herbert Poetzl 已提交
73
	*flags = OCFS2_I(inode)->ip_attr;
M
Mark Fasheh 已提交
74
	ocfs2_inode_unlock(inode, 0);
H
Herbert Poetzl 已提交
75 76 77 78 79 80 81 82 83

	return status;
}

static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags,
				unsigned mask)
{
	struct ocfs2_inode_info *ocfs2_inode = OCFS2_I(inode);
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
84
	handle_t *handle = NULL;
H
Herbert Poetzl 已提交
85 86 87 88 89 90
	struct buffer_head *bh = NULL;
	unsigned oldflags;
	int status;

	mutex_lock(&inode->i_mutex);

M
Mark Fasheh 已提交
91
	status = ocfs2_inode_lock(inode, &bh, 1);
H
Herbert Poetzl 已提交
92 93 94 95 96 97
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

	status = -EACCES;
98
	if (!inode_owner_or_capable(inode))
H
Herbert Poetzl 已提交
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115
		goto bail_unlock;

	if (!S_ISDIR(inode->i_mode))
		flags &= ~OCFS2_DIRSYNC_FL;

	oldflags = ocfs2_inode->ip_attr;
	flags = flags & mask;
	flags |= oldflags & ~mask;

	/*
	 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
	 * the relevant capability.
	 */
	status = -EPERM;
	if ((oldflags & OCFS2_IMMUTABLE_FL) || ((flags ^ oldflags) &
		(OCFS2_APPEND_FL | OCFS2_IMMUTABLE_FL))) {
		if (!capable(CAP_LINUX_IMMUTABLE))
116 117 118 119 120 121 122 123
			goto bail_unlock;
	}

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (IS_ERR(handle)) {
		status = PTR_ERR(handle);
		mlog_errno(status);
		goto bail_unlock;
H
Herbert Poetzl 已提交
124 125 126 127 128 129 130 131 132
	}

	ocfs2_inode->ip_attr = flags;
	ocfs2_set_inode_flags(inode);

	status = ocfs2_mark_inode_dirty(handle, inode, bh);
	if (status < 0)
		mlog_errno(status);

133
	ocfs2_commit_trans(osb, handle);
134

H
Herbert Poetzl 已提交
135
bail_unlock:
M
Mark Fasheh 已提交
136
	ocfs2_inode_unlock(inode, 1);
H
Herbert Poetzl 已提交
137 138 139
bail:
	mutex_unlock(&inode->i_mutex);

140
	brelse(bh);
H
Herbert Poetzl 已提交
141 142 143 144

	return status;
}

145 146 147 148 149 150 151 152 153 154
int ocfs2_info_handle_blocksize(struct inode *inode,
				struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_blocksize oib;

	if (o2info_from_user(oib, req))
		goto bail;

	oib.ib_blocksize = inode->i_sb->s_blocksize;
155

156
	o2info_set_request_filled(&oib.ib_req);
157 158 159 160 161 162 163

	if (o2info_to_user(oib, req))
		goto bail;

	status = 0;
bail:
	if (status)
164
		o2info_set_request_error(&oib.ib_req, req);
165 166 167 168 169 170 171 172 173 174 175 176 177 178 179

	return status;
}

int ocfs2_info_handle_clustersize(struct inode *inode,
				  struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_clustersize oic;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oic, req))
		goto bail;

	oic.ic_clustersize = osb->s_clustersize;
180

181
	o2info_set_request_filled(&oic.ic_req);
182 183 184 185 186 187 188

	if (o2info_to_user(oic, req))
		goto bail;

	status = 0;
bail:
	if (status)
189
		o2info_set_request_error(&oic.ic_req, req);
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204

	return status;
}

int ocfs2_info_handle_maxslots(struct inode *inode,
			       struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_maxslots oim;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oim, req))
		goto bail;

	oim.im_max_slots = osb->max_slots;
205

206
	o2info_set_request_filled(&oim.im_req);
207 208 209 210 211 212 213

	if (o2info_to_user(oim, req))
		goto bail;

	status = 0;
bail:
	if (status)
214
		o2info_set_request_error(&oim.im_req, req);
215 216 217 218 219 220 221 222 223 224 225 226 227 228 229

	return status;
}

int ocfs2_info_handle_label(struct inode *inode,
			    struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_label oil;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oil, req))
		goto bail;

	memcpy(oil.il_label, osb->vol_label, OCFS2_MAX_VOL_LABEL_LEN);
230

231
	o2info_set_request_filled(&oil.il_req);
232 233 234 235 236 237 238

	if (o2info_to_user(oil, req))
		goto bail;

	status = 0;
bail:
	if (status)
239
		o2info_set_request_error(&oil.il_req, req);
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254

	return status;
}

int ocfs2_info_handle_uuid(struct inode *inode,
			   struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_uuid oiu;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oiu, req))
		goto bail;

	memcpy(oiu.iu_uuid_str, osb->uuid_str, OCFS2_TEXT_UUID_LEN + 1);
255

256
	o2info_set_request_filled(&oiu.iu_req);
257 258 259 260 261 262 263

	if (o2info_to_user(oiu, req))
		goto bail;

	status = 0;
bail:
	if (status)
264
		o2info_set_request_error(&oiu.iu_req, req);
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281

	return status;
}

int ocfs2_info_handle_fs_features(struct inode *inode,
				  struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_fs_features oif;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oif, req))
		goto bail;

	oif.if_compat_features = osb->s_feature_compat;
	oif.if_incompat_features = osb->s_feature_incompat;
	oif.if_ro_compat_features = osb->s_feature_ro_compat;
282

283
	o2info_set_request_filled(&oif.if_req);
284 285 286 287 288 289 290

	if (o2info_to_user(oif, req))
		goto bail;

	status = 0;
bail:
	if (status)
291
		o2info_set_request_error(&oif.if_req, req);
292 293 294 295 296 297 298 299 300 301 302 303 304 305

	return status;
}

int ocfs2_info_handle_journal_size(struct inode *inode,
				   struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_journal_size oij;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (o2info_from_user(oij, req))
		goto bail;

306
	oij.ij_journal_size = i_size_read(osb->journal->j_inode);
307

308
	o2info_set_request_filled(&oij.ij_req);
309 310 311 312 313 314 315

	if (o2info_to_user(oij, req))
		goto bail;

	status = 0;
bail:
	if (status)
316
		o2info_set_request_error(&oij.ij_req, req);
317 318 319 320

	return status;
}

321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
int ocfs2_info_scan_inode_alloc(struct ocfs2_super *osb,
				struct inode *inode_alloc, u64 blkno,
				struct ocfs2_info_freeinode *fi, u32 slot)
{
	int status = 0, unlock = 0;

	struct buffer_head *bh = NULL;
	struct ocfs2_dinode *dinode_alloc = NULL;

	if (inode_alloc)
		mutex_lock(&inode_alloc->i_mutex);

	if (o2info_coherent(&fi->ifi_req)) {
		status = ocfs2_inode_lock(inode_alloc, &bh, 0);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
		unlock = 1;
	} else {
		status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
	}

	dinode_alloc = (struct ocfs2_dinode *)bh->b_data;

	fi->ifi_stat[slot].lfi_total =
		le32_to_cpu(dinode_alloc->id1.bitmap1.i_total);
	fi->ifi_stat[slot].lfi_free =
		le32_to_cpu(dinode_alloc->id1.bitmap1.i_total) -
		le32_to_cpu(dinode_alloc->id1.bitmap1.i_used);

bail:
	if (unlock)
		ocfs2_inode_unlock(inode_alloc, 0);

	if (inode_alloc)
		mutex_unlock(&inode_alloc->i_mutex);

	brelse(bh);

	return status;
}

int ocfs2_info_handle_freeinode(struct inode *inode,
				struct ocfs2_info_request __user *req)
{
	u32 i;
	u64 blkno = -1;
	char namebuf[40];
	int status = -EFAULT, type = INODE_ALLOC_SYSTEM_INODE;
	struct ocfs2_info_freeinode *oifi = NULL;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	struct inode *inode_alloc = NULL;

	oifi = kzalloc(sizeof(struct ocfs2_info_freeinode), GFP_KERNEL);
	if (!oifi) {
		status = -ENOMEM;
		mlog_errno(status);
383
		goto out_err;
384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
	}

	if (o2info_from_user(*oifi, req))
		goto bail;

	oifi->ifi_slotnum = osb->max_slots;

	for (i = 0; i < oifi->ifi_slotnum; i++) {
		if (o2info_coherent(&oifi->ifi_req)) {
			inode_alloc = ocfs2_get_system_file_inode(osb, type, i);
			if (!inode_alloc) {
				mlog(ML_ERROR, "unable to get alloc inode in "
				    "slot %u\n", i);
				status = -EIO;
				goto bail;
			}
		} else {
			ocfs2_sprintf_system_inode_name(namebuf,
							sizeof(namebuf),
							type, i);
			status = ocfs2_lookup_ino_from_name(osb->sys_root_inode,
							    namebuf,
							    strlen(namebuf),
							    &blkno);
			if (status < 0) {
				status = -ENOENT;
				goto bail;
			}
		}

		status = ocfs2_info_scan_inode_alloc(osb, inode_alloc, blkno, oifi, i);
		if (status < 0)
			goto bail;

		iput(inode_alloc);
		inode_alloc = NULL;
	}

	o2info_set_request_filled(&oifi->ifi_req);

	if (o2info_to_user(*oifi, req))
		goto bail;

	status = 0;
bail:
	if (status)
		o2info_set_request_error(&oifi->ifi_req, req);

	kfree(oifi);
433
out_err:
434 435 436
	return status;
}

437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667
static void o2ffg_update_histogram(struct ocfs2_info_free_chunk_list *hist,
				   unsigned int chunksize)
{
	int index;

	index = __ilog2_u32(chunksize);
	if (index >= OCFS2_INFO_MAX_HIST)
		index = OCFS2_INFO_MAX_HIST - 1;

	hist->fc_chunks[index]++;
	hist->fc_clusters[index] += chunksize;
}

static void o2ffg_update_stats(struct ocfs2_info_freefrag_stats *stats,
			       unsigned int chunksize)
{
	if (chunksize > stats->ffs_max)
		stats->ffs_max = chunksize;

	if (chunksize < stats->ffs_min)
		stats->ffs_min = chunksize;

	stats->ffs_avg += chunksize;
	stats->ffs_free_chunks_real++;
}

void ocfs2_info_update_ffg(struct ocfs2_info_freefrag *ffg,
			   unsigned int chunksize)
{
	o2ffg_update_histogram(&(ffg->iff_ffs.ffs_fc_hist), chunksize);
	o2ffg_update_stats(&(ffg->iff_ffs), chunksize);
}

int ocfs2_info_freefrag_scan_chain(struct ocfs2_super *osb,
				   struct inode *gb_inode,
				   struct ocfs2_dinode *gb_dinode,
				   struct ocfs2_chain_rec *rec,
				   struct ocfs2_info_freefrag *ffg,
				   u32 chunks_in_group)
{
	int status = 0, used;
	u64 blkno;

	struct buffer_head *bh = NULL;
	struct ocfs2_group_desc *bg = NULL;

	unsigned int max_bits, num_clusters;
	unsigned int offset = 0, cluster, chunk;
	unsigned int chunk_free, last_chunksize = 0;

	if (!le32_to_cpu(rec->c_free))
		goto bail;

	do {
		if (!bg)
			blkno = le64_to_cpu(rec->c_blkno);
		else
			blkno = le64_to_cpu(bg->bg_next_group);

		if (bh) {
			brelse(bh);
			bh = NULL;
		}

		if (o2info_coherent(&ffg->iff_req))
			status = ocfs2_read_group_descriptor(gb_inode,
							     gb_dinode,
							     blkno, &bh);
		else
			status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh);

		if (status < 0) {
			mlog(ML_ERROR, "Can't read the group descriptor # "
			     "%llu from device.", (unsigned long long)blkno);
			status = -EIO;
			goto bail;
		}

		bg = (struct ocfs2_group_desc *)bh->b_data;

		if (!le16_to_cpu(bg->bg_free_bits_count))
			continue;

		max_bits = le16_to_cpu(bg->bg_bits);
		offset = 0;

		for (chunk = 0; chunk < chunks_in_group; chunk++) {
			/*
			 * last chunk may be not an entire one.
			 */
			if ((offset + ffg->iff_chunksize) > max_bits)
				num_clusters = max_bits - offset;
			else
				num_clusters = ffg->iff_chunksize;

			chunk_free = 0;
			for (cluster = 0; cluster < num_clusters; cluster++) {
				used = ocfs2_test_bit(offset,
						(unsigned long *)bg->bg_bitmap);
				/*
				 * - chunk_free counts free clusters in #N chunk.
				 * - last_chunksize records the size(in) clusters
				 *   for the last real free chunk being counted.
				 */
				if (!used) {
					last_chunksize++;
					chunk_free++;
				}

				if (used && last_chunksize) {
					ocfs2_info_update_ffg(ffg,
							      last_chunksize);
					last_chunksize = 0;
				}

				offset++;
			}

			if (chunk_free == ffg->iff_chunksize)
				ffg->iff_ffs.ffs_free_chunks++;
		}

		/*
		 * need to update the info for last free chunk.
		 */
		if (last_chunksize)
			ocfs2_info_update_ffg(ffg, last_chunksize);

	} while (le64_to_cpu(bg->bg_next_group));

bail:
	brelse(bh);

	return status;
}

int ocfs2_info_freefrag_scan_bitmap(struct ocfs2_super *osb,
				    struct inode *gb_inode, u64 blkno,
				    struct ocfs2_info_freefrag *ffg)
{
	u32 chunks_in_group;
	int status = 0, unlock = 0, i;

	struct buffer_head *bh = NULL;
	struct ocfs2_chain_list *cl = NULL;
	struct ocfs2_chain_rec *rec = NULL;
	struct ocfs2_dinode *gb_dinode = NULL;

	if (gb_inode)
		mutex_lock(&gb_inode->i_mutex);

	if (o2info_coherent(&ffg->iff_req)) {
		status = ocfs2_inode_lock(gb_inode, &bh, 0);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
		unlock = 1;
	} else {
		status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
	}

	gb_dinode = (struct ocfs2_dinode *)bh->b_data;
	cl = &(gb_dinode->id2.i_chain);

	/*
	 * Chunksize(in) clusters from userspace should be
	 * less than clusters in a group.
	 */
	if (ffg->iff_chunksize > le16_to_cpu(cl->cl_cpg)) {
		status = -EINVAL;
		goto bail;
	}

	memset(&ffg->iff_ffs, 0, sizeof(struct ocfs2_info_freefrag_stats));

	ffg->iff_ffs.ffs_min = ~0U;
	ffg->iff_ffs.ffs_clusters =
			le32_to_cpu(gb_dinode->id1.bitmap1.i_total);
	ffg->iff_ffs.ffs_free_clusters = ffg->iff_ffs.ffs_clusters -
			le32_to_cpu(gb_dinode->id1.bitmap1.i_used);

	chunks_in_group = le16_to_cpu(cl->cl_cpg) / ffg->iff_chunksize + 1;

	for (i = 0; i < le16_to_cpu(cl->cl_next_free_rec); i++) {
		rec = &(cl->cl_recs[i]);
		status = ocfs2_info_freefrag_scan_chain(osb, gb_inode,
							gb_dinode,
							rec, ffg,
							chunks_in_group);
		if (status)
			goto bail;
	}

	if (ffg->iff_ffs.ffs_free_chunks_real)
		ffg->iff_ffs.ffs_avg = (ffg->iff_ffs.ffs_avg /
					ffg->iff_ffs.ffs_free_chunks_real);
bail:
	if (unlock)
		ocfs2_inode_unlock(gb_inode, 0);

	if (gb_inode)
		mutex_unlock(&gb_inode->i_mutex);

	if (gb_inode)
		iput(gb_inode);

	brelse(bh);

	return status;
}

int ocfs2_info_handle_freefrag(struct inode *inode,
			       struct ocfs2_info_request __user *req)
{
	u64 blkno = -1;
	char namebuf[40];
	int status = -EFAULT, type = GLOBAL_BITMAP_SYSTEM_INODE;

	struct ocfs2_info_freefrag *oiff;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	struct inode *gb_inode = NULL;

	oiff = kzalloc(sizeof(struct ocfs2_info_freefrag), GFP_KERNEL);
	if (!oiff) {
		status = -ENOMEM;
		mlog_errno(status);
668
		goto out_err;
669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
	}

	if (o2info_from_user(*oiff, req))
		goto bail;
	/*
	 * chunksize from userspace should be power of 2.
	 */
	if ((oiff->iff_chunksize & (oiff->iff_chunksize - 1)) ||
	    (!oiff->iff_chunksize)) {
		status = -EINVAL;
		goto bail;
	}

	if (o2info_coherent(&oiff->iff_req)) {
		gb_inode = ocfs2_get_system_file_inode(osb, type,
						       OCFS2_INVALID_SLOT);
		if (!gb_inode) {
			mlog(ML_ERROR, "unable to get global_bitmap inode\n");
			status = -EIO;
			goto bail;
		}
	} else {
		ocfs2_sprintf_system_inode_name(namebuf, sizeof(namebuf), type,
						OCFS2_INVALID_SLOT);
		status = ocfs2_lookup_ino_from_name(osb->sys_root_inode,
						    namebuf,
						    strlen(namebuf),
						    &blkno);
		if (status < 0) {
			status = -ENOENT;
			goto bail;
		}
	}

	status = ocfs2_info_freefrag_scan_bitmap(osb, gb_inode, blkno, oiff);
	if (status < 0)
		goto bail;

	o2info_set_request_filled(&oiff->iff_req);

709 710
	if (o2info_to_user(*oiff, req)) {
		status = -EFAULT;
711
		goto bail;
712
	}
713 714 715 716 717 718 719

	status = 0;
bail:
	if (status)
		o2info_set_request_error(&oiff->iff_req, req);

	kfree(oiff);
720
out_err:
721 722 723 724 725 726 727 728 729 730 731 732
	return status;
}

int ocfs2_info_handle_unknown(struct inode *inode,
			      struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_request oir;

	if (o2info_from_user(oir, req))
		goto bail;

733
	o2info_clear_request_filled(&oir);
734 735 736 737 738 739 740

	if (o2info_to_user(oir, req))
		goto bail;

	status = 0;
bail:
	if (status)
741
		o2info_set_request_error(&oir, req);
742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794

	return status;
}

/*
 * Validate and distinguish OCFS2_IOC_INFO requests.
 *
 * - validate the magic number.
 * - distinguish different requests.
 * - validate size of different requests.
 */
int ocfs2_info_handle_request(struct inode *inode,
			      struct ocfs2_info_request __user *req)
{
	int status = -EFAULT;
	struct ocfs2_info_request oir;

	if (o2info_from_user(oir, req))
		goto bail;

	status = -EINVAL;
	if (oir.ir_magic != OCFS2_INFO_MAGIC)
		goto bail;

	switch (oir.ir_code) {
	case OCFS2_INFO_BLOCKSIZE:
		if (oir.ir_size == sizeof(struct ocfs2_info_blocksize))
			status = ocfs2_info_handle_blocksize(inode, req);
		break;
	case OCFS2_INFO_CLUSTERSIZE:
		if (oir.ir_size == sizeof(struct ocfs2_info_clustersize))
			status = ocfs2_info_handle_clustersize(inode, req);
		break;
	case OCFS2_INFO_MAXSLOTS:
		if (oir.ir_size == sizeof(struct ocfs2_info_maxslots))
			status = ocfs2_info_handle_maxslots(inode, req);
		break;
	case OCFS2_INFO_LABEL:
		if (oir.ir_size == sizeof(struct ocfs2_info_label))
			status = ocfs2_info_handle_label(inode, req);
		break;
	case OCFS2_INFO_UUID:
		if (oir.ir_size == sizeof(struct ocfs2_info_uuid))
			status = ocfs2_info_handle_uuid(inode, req);
		break;
	case OCFS2_INFO_FS_FEATURES:
		if (oir.ir_size == sizeof(struct ocfs2_info_fs_features))
			status = ocfs2_info_handle_fs_features(inode, req);
		break;
	case OCFS2_INFO_JOURNAL_SIZE:
		if (oir.ir_size == sizeof(struct ocfs2_info_journal_size))
			status = ocfs2_info_handle_journal_size(inode, req);
		break;
795 796 797 798
	case OCFS2_INFO_FREEINODE:
		if (oir.ir_size == sizeof(struct ocfs2_info_freeinode))
			status = ocfs2_info_handle_freeinode(inode, req);
		break;
799 800 801 802
	case OCFS2_INFO_FREEFRAG:
		if (oir.ir_size == sizeof(struct ocfs2_info_freefrag))
			status = ocfs2_info_handle_freefrag(inode, req);
		break;
803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
	default:
		status = ocfs2_info_handle_unknown(inode, req);
		break;
	}

bail:
	return status;
}

int ocfs2_get_request_ptr(struct ocfs2_info *info, int idx,
			  u64 *req_addr, int compat_flag)
{
	int status = -EFAULT;
	u64 __user *bp = NULL;

	if (compat_flag) {
#ifdef CONFIG_COMPAT
		/*
		 * pointer bp stores the base address of a pointers array,
		 * which collects all addresses of separate request.
		 */
		bp = (u64 __user *)(unsigned long)compat_ptr(info->oi_requests);
#else
		BUG();
#endif
	} else
		bp = (u64 __user *)(unsigned long)(info->oi_requests);

	if (o2info_from_user(*req_addr, bp + idx))
		goto bail;

	status = 0;
bail:
	return status;
}

/*
 * OCFS2_IOC_INFO handles an array of requests passed from userspace.
 *
 * ocfs2_info_handle() recevies a large info aggregation, grab and
 * validate the request count from header, then break it into small
 * pieces, later specific handlers can handle them one by one.
 *
 * Idea here is to make each separate request small enough to ensure
 * a better backward&forward compatibility, since a small piece of
 * request will be less likely to be broken if disk layout get changed.
 */
int ocfs2_info_handle(struct inode *inode, struct ocfs2_info *info,
		      int compat_flag)
{
	int i, status = 0;
	u64 req_addr;
	struct ocfs2_info_request __user *reqp;

	if ((info->oi_count > OCFS2_INFO_MAX_REQUEST) ||
	    (!info->oi_requests)) {
		status = -EINVAL;
		goto bail;
	}

	for (i = 0; i < info->oi_count; i++) {

		status = ocfs2_get_request_ptr(info, i, &req_addr, compat_flag);
		if (status)
			break;

A
Al Viro 已提交
869
		reqp = (struct ocfs2_info_request __user *)(unsigned long)req_addr;
870 871 872 873 874 875 876 877 878 879 880 881 882 883
		if (!reqp) {
			status = -EINVAL;
			goto bail;
		}

		status = ocfs2_info_handle_request(inode, reqp);
		if (status)
			break;
	}

bail:
	return status;
}

884
long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
H
Herbert Poetzl 已提交
885
{
A
Al Viro 已提交
886
	struct inode *inode = file_inode(filp);
H
Herbert Poetzl 已提交
887
	unsigned int flags;
888
	int new_clusters;
H
Herbert Poetzl 已提交
889
	int status;
890
	struct ocfs2_space_resv sr;
891
	struct ocfs2_new_group_input input;
T
Tao Ma 已提交
892
	struct reflink_arguments args;
A
Al Viro 已提交
893 894
	const char __user *old_path;
	const char __user *new_path;
T
Tao Ma 已提交
895
	bool preserve;
896
	struct ocfs2_info info;
A
Al Viro 已提交
897
	void __user *argp = (void __user *)arg;
H
Herbert Poetzl 已提交
898 899 900 901 902 903 904 905 906 907 908 909 910

	switch (cmd) {
	case OCFS2_IOC_GETFLAGS:
		status = ocfs2_get_inode_attr(inode, &flags);
		if (status < 0)
			return status;

		flags &= OCFS2_FL_VISIBLE;
		return put_user(flags, (int __user *) arg);
	case OCFS2_IOC_SETFLAGS:
		if (get_user(flags, (int __user *) arg))
			return -EFAULT;

911
		status = mnt_want_write_file(filp);
912 913 914
		if (status)
			return status;
		status = ocfs2_set_inode_attr(inode, flags,
H
Herbert Poetzl 已提交
915
			OCFS2_FL_MODIFIABLE);
A
Al Viro 已提交
916
		mnt_drop_write_file(filp);
917
		return status;
918 919 920 921 922 923 924 925
	case OCFS2_IOC_RESVSP:
	case OCFS2_IOC_RESVSP64:
	case OCFS2_IOC_UNRESVSP:
	case OCFS2_IOC_UNRESVSP64:
		if (copy_from_user(&sr, (int __user *) arg, sizeof(sr)))
			return -EFAULT;

		return ocfs2_change_file_space(filp, cmd, &sr);
926
	case OCFS2_IOC_GROUP_EXTEND:
927 928 929
		if (!capable(CAP_SYS_RESOURCE))
			return -EPERM;

930 931 932
		if (get_user(new_clusters, (int __user *)arg))
			return -EFAULT;

933 934 935 936 937 938
		status = mnt_want_write_file(filp);
		if (status)
			return status;
		status = ocfs2_group_extend(inode, new_clusters);
		mnt_drop_write_file(filp);
		return status;
939 940
	case OCFS2_IOC_GROUP_ADD:
	case OCFS2_IOC_GROUP_ADD64:
941 942 943
		if (!capable(CAP_SYS_RESOURCE))
			return -EPERM;

944 945 946
		if (copy_from_user(&input, (int __user *) arg, sizeof(input)))
			return -EFAULT;

947 948 949 950 951 952
		status = mnt_want_write_file(filp);
		if (status)
			return status;
		status = ocfs2_group_add(inode, &input);
		mnt_drop_write_file(filp);
		return status;
T
Tao Ma 已提交
953
	case OCFS2_IOC_REFLINK:
A
Al Viro 已提交
954
		if (copy_from_user(&args, argp, sizeof(args)))
T
Tao Ma 已提交
955
			return -EFAULT;
A
Al Viro 已提交
956 957
		old_path = (const char __user *)(unsigned long)args.old_path;
		new_path = (const char __user *)(unsigned long)args.new_path;
T
Tao Ma 已提交
958 959 960
		preserve = (args.preserve != 0);

		return ocfs2_reflink_ioctl(inode, old_path, new_path, preserve);
961
	case OCFS2_IOC_INFO:
A
Al Viro 已提交
962
		if (copy_from_user(&info, argp, sizeof(struct ocfs2_info)))
963 964 965
			return -EFAULT;

		return ocfs2_info_handle(inode, &info, 0);
T
Tao Ma 已提交
966 967 968 969 970 971 972 973 974
	case FITRIM:
	{
		struct super_block *sb = inode->i_sb;
		struct fstrim_range range;
		int ret = 0;

		if (!capable(CAP_SYS_ADMIN))
			return -EPERM;

A
Al Viro 已提交
975
		if (copy_from_user(&range, argp, sizeof(range)))
T
Tao Ma 已提交
976 977 978 979 980 981
			return -EFAULT;

		ret = ocfs2_trim_fs(sb, &range);
		if (ret < 0)
			return ret;

A
Al Viro 已提交
982
		if (copy_to_user(argp, &range, sizeof(range)))
T
Tao Ma 已提交
983 984 985 986
			return -EFAULT;

		return 0;
	}
987
	case OCFS2_IOC_MOVE_EXT:
A
Al Viro 已提交
988
		return ocfs2_ioctl_move_extents(filp, argp);
H
Herbert Poetzl 已提交
989 990 991 992 993
	default:
		return -ENOTTY;
	}
}

M
Mark Fasheh 已提交
994 995 996
#ifdef CONFIG_COMPAT
long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
{
997 998
	bool preserve;
	struct reflink_arguments args;
A
Al Viro 已提交
999
	struct inode *inode = file_inode(file);
1000
	struct ocfs2_info info;
A
Al Viro 已提交
1001
	void __user *argp = (void __user *)arg;
1002

M
Mark Fasheh 已提交
1003 1004 1005 1006 1007 1008 1009
	switch (cmd) {
	case OCFS2_IOC32_GETFLAGS:
		cmd = OCFS2_IOC_GETFLAGS;
		break;
	case OCFS2_IOC32_SETFLAGS:
		cmd = OCFS2_IOC_SETFLAGS;
		break;
1010 1011 1012 1013
	case OCFS2_IOC_RESVSP:
	case OCFS2_IOC_RESVSP64:
	case OCFS2_IOC_UNRESVSP:
	case OCFS2_IOC_UNRESVSP64:
1014
	case OCFS2_IOC_GROUP_EXTEND:
1015 1016
	case OCFS2_IOC_GROUP_ADD:
	case OCFS2_IOC_GROUP_ADD64:
T
Tao Ma 已提交
1017
	case FITRIM:
1018
		break;
1019
	case OCFS2_IOC_REFLINK:
A
Al Viro 已提交
1020
		if (copy_from_user(&args, argp, sizeof(args)))
1021 1022 1023 1024 1025
			return -EFAULT;
		preserve = (args.preserve != 0);

		return ocfs2_reflink_ioctl(inode, compat_ptr(args.old_path),
					   compat_ptr(args.new_path), preserve);
1026
	case OCFS2_IOC_INFO:
A
Al Viro 已提交
1027
		if (copy_from_user(&info, argp, sizeof(struct ocfs2_info)))
1028 1029 1030
			return -EFAULT;

		return ocfs2_info_handle(inode, &info, 1);
1031 1032
	case OCFS2_IOC_MOVE_EXT:
		break;
M
Mark Fasheh 已提交
1033 1034 1035 1036
	default:
		return -ENOIOCTLCMD;
	}

1037
	return ocfs2_ioctl(file, cmd, arg);
M
Mark Fasheh 已提交
1038 1039
}
#endif