file.c 23.8 KB
Newer Older
1
#include <linux/ceph/ceph_debug.h>
S
Sage Weil 已提交
2

3
#include <linux/module.h>
S
Sage Weil 已提交
4
#include <linux/sched.h>
5
#include <linux/slab.h>
S
Sage Weil 已提交
6
#include <linux/file.h>
S
Sage Weil 已提交
7
#include <linux/mount.h>
S
Sage Weil 已提交
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
#include <linux/namei.h>
#include <linux/writeback.h>

#include "super.h"
#include "mds_client.h"

/*
 * Ceph file operations
 *
 * Implement basic open/close functionality, and implement
 * read/write.
 *
 * We implement three modes of file I/O:
 *  - buffered uses the generic_file_aio_{read,write} helpers
 *
 *  - synchronous is used when there is multi-client read/write
 *    sharing, avoids the page cache, and synchronously waits for an
 *    ack from the OSD.
 *
 *  - direct io takes the variant of the sync path that references
 *    user pages directly.
 *
 * fsync() flushes and waits on dirty pages, but just queues metadata
 * for writeback: since the MDS can recover size and mtime there is no
 * need to wait for MDS acknowledgement.
 */


/*
 * Prepare an open request.  Preallocate ceph_cap to avoid an
 * inopportune ENOMEM later.
 */
static struct ceph_mds_request *
prepare_open_request(struct super_block *sb, int flags, int create_mode)
{
43 44
	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
	struct ceph_mds_request *req;
	int want_auth = USE_ANY_MDS;
	int op = (flags & O_CREAT) ? CEPH_MDS_OP_CREATE : CEPH_MDS_OP_OPEN;

	if (flags & (O_WRONLY|O_RDWR|O_CREAT|O_TRUNC))
		want_auth = USE_AUTH_MDS;

	req = ceph_mdsc_create_request(mdsc, op, want_auth);
	if (IS_ERR(req))
		goto out;
	req->r_fmode = ceph_flags_to_mode(flags);
	req->r_args.open.flags = cpu_to_le32(flags);
	req->r_args.open.mode = cpu_to_le32(create_mode);
out:
	return req;
}

/*
 * initialize private struct file data.
 * if we fail, clean up by dropping fmode reference on the ceph_inode
 */
static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
{
	struct ceph_file_info *cf;
	int ret = 0;

	switch (inode->i_mode & S_IFMT) {
	case S_IFREG:
	case S_IFDIR:
		dout("init_file %p %p 0%o (regular)\n", inode, file,
		     inode->i_mode);
		cf = kmem_cache_alloc(ceph_file_cachep, GFP_NOFS | __GFP_ZERO);
		if (cf == NULL) {
			ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
			return -ENOMEM;
		}
		cf->fmode = fmode;
		cf->next_offset = 2;
		file->private_data = cf;
		BUG_ON(inode->i_fop->release != ceph_release);
		break;

	case S_IFLNK:
		dout("init_file %p %p 0%o (symlink)\n", inode, file,
		     inode->i_mode);
		ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
		break;

	default:
		dout("init_file %p %p 0%o (special)\n", inode, file,
		     inode->i_mode);
		/*
		 * we need to drop the open ref now, since we don't
		 * have .release set to ceph_release.
		 */
		ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
		BUG_ON(inode->i_fop->release == ceph_release);

		/* call the proper open fop */
		ret = inode->i_fop->open(inode, file);
	}
	return ret;
}

/*
 * If we already have the requisite capabilities, we can satisfy
 * the open request locally (no need to request new caps from the
 * MDS).  We do, however, need to inform the MDS (asynchronously)
 * if our wanted caps set expands.
 */
int ceph_open(struct inode *inode, struct file *file)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
118 119
	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
120 121
	struct ceph_mds_request *req;
	struct ceph_file_info *cf = file->private_data;
S
Sage Weil 已提交
122
	struct inode *parent_inode = NULL;
S
Sage Weil 已提交
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
	int err;
	int flags, fmode, wanted;

	if (cf) {
		dout("open file %p is already opened\n", file);
		return 0;
	}

	/* filter out O_CREAT|O_EXCL; vfs did that already.  yuck. */
	flags = file->f_flags & ~(O_CREAT|O_EXCL);
	if (S_ISDIR(inode->i_mode))
		flags = O_DIRECTORY;  /* mds likes to know */

	dout("open inode %p ino %llx.%llx file %p flags %d (%d)\n", inode,
	     ceph_vinop(inode), file, flags, file->f_flags);
	fmode = ceph_flags_to_mode(flags);
	wanted = ceph_caps_for_mode(fmode);

	/* snapped files are read-only */
	if (ceph_snap(inode) != CEPH_NOSNAP && (file->f_mode & FMODE_WRITE))
		return -EROFS;

	/* trivially open snapdir */
	if (ceph_snap(inode) == CEPH_SNAPDIR) {
147
		spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
148
		__ceph_get_fmode(ci, fmode);
149
		spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
150 151 152 153
		return ceph_init_file(inode, file, fmode);
	}

	/*
154 155
	 * No need to block if we have caps on the auth MDS (for
	 * write) or any MDS (for read).  Update wanted set
S
Sage Weil 已提交
156 157
	 * asynchronously.
	 */
158
	spin_lock(&ci->i_ceph_lock);
159 160
	if (__ceph_is_any_real_caps(ci) &&
	    (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) {
S
Sage Weil 已提交
161 162 163 164 165 166 167
		int mds_wanted = __ceph_caps_mds_wanted(ci);
		int issued = __ceph_caps_issued(ci, NULL);

		dout("open %p fmode %d want %s issued %s using existing\n",
		     inode, fmode, ceph_cap_string(wanted),
		     ceph_cap_string(issued));
		__ceph_get_fmode(ci, fmode);
168
		spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
169 170 171 172 173 174 175 176 177 178 179

		/* adjust wanted? */
		if ((issued & wanted) != wanted &&
		    (mds_wanted & wanted) != wanted &&
		    ceph_snap(inode) != CEPH_SNAPDIR)
			ceph_check_caps(ci, 0, NULL);

		return ceph_init_file(inode, file, fmode);
	} else if (ceph_snap(inode) != CEPH_NOSNAP &&
		   (ci->i_snap_caps & wanted) == wanted) {
		__ceph_get_fmode(ci, fmode);
180
		spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
181 182
		return ceph_init_file(inode, file, fmode);
	}
183
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
184 185 186 187 188 189 190

	dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted));
	req = prepare_open_request(inode->i_sb, flags, 0);
	if (IS_ERR(req)) {
		err = PTR_ERR(req);
		goto out;
	}
191 192
	req->r_inode = inode;
	ihold(inode);
S
Sage Weil 已提交
193
	req->r_num_caps = 1;
S
Sage Weil 已提交
194 195
	if (flags & (O_CREAT|O_TRUNC))
		parent_inode = ceph_get_dentry_parent_inode(file->f_dentry);
S
Sage Weil 已提交
196
	err = ceph_mdsc_do_request(mdsc, parent_inode, req);
S
Sage Weil 已提交
197
	iput(parent_inode);
S
Sage Weil 已提交
198 199 200 201 202 203 204 205 206 207
	if (!err)
		err = ceph_init_file(inode, file, req->r_fmode);
	ceph_mdsc_put_request(req);
	dout("open result=%d on %llx.%llx\n", err, ceph_vinop(inode));
out:
	return err;
}


/*
S
Sage Weil 已提交
208 209
 * Do a lookup + open with a single request.  If we get a non-existent
 * file or symlink, return 1 so the VFS can retry.
S
Sage Weil 已提交
210
 */
S
Sage Weil 已提交
211
int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
A
Al Viro 已提交
212
		     struct file *file, unsigned flags, umode_t mode,
A
Al Viro 已提交
213
		     int *opened)
S
Sage Weil 已提交
214
{
215 216
	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
217
	struct ceph_mds_request *req;
S
Sage Weil 已提交
218
	struct dentry *dn;
S
Sage Weil 已提交
219 220
	int err;

S
Sage Weil 已提交
221 222 223 224 225 226 227 228 229 230
	dout("atomic_open %p dentry %p '%.*s' %s flags %d mode 0%o\n",
	     dir, dentry, dentry->d_name.len, dentry->d_name.name,
	     d_unhashed(dentry) ? "unhashed" : "hashed", flags, mode);

	if (dentry->d_name.len > NAME_MAX)
		return -ENAMETOOLONG;

	err = ceph_init_dentry(dentry);
	if (err < 0)
		return err;
S
Sage Weil 已提交
231 232 233 234

	/* do the open */
	req = prepare_open_request(dir->i_sb, flags, mode);
	if (IS_ERR(req))
A
Al Viro 已提交
235
		return PTR_ERR(req);
S
Sage Weil 已提交
236 237 238 239 240 241 242
	req->r_dentry = dget(dentry);
	req->r_num_caps = 2;
	if (flags & O_CREAT) {
		req->r_dentry_drop = CEPH_CAP_FILE_SHARED;
		req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
	}
	req->r_locked_dir = dir;           /* caller holds dir->i_mutex */
243 244 245
	err = ceph_mdsc_do_request(mdsc,
				   (flags & (O_CREAT|O_TRUNC)) ? dir : NULL,
				   req);
246
	err = ceph_handle_snapdir(req, dentry, err);
S
Sage Weil 已提交
247
	if (err == 0 && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry)
S
Sage Weil 已提交
248
		err = ceph_handle_notrace_create(dir, dentry);
249

S
Sage Weil 已提交
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
	if (d_unhashed(dentry)) {
		dn = ceph_finish_lookup(req, dentry, err);
		if (IS_ERR(dn))
			err = PTR_ERR(dn);
	} else {
		/* we were given a hashed negative dentry */
		dn = NULL;
	}
	if (err)
		goto out_err;
	if (dn || dentry->d_inode == NULL || S_ISLNK(dentry->d_inode->i_mode)) {
		/* make vfs retry on splice, ENOENT, or symlink */
		dout("atomic_open finish_no_open on dn %p\n", dn);
		err = finish_no_open(file, dn);
	} else {
		dout("atomic_open finish_open on dn %p\n", dn);
		err = finish_open(file, dentry, ceph_open, opened);
	}
268

S
Sage Weil 已提交
269 270 271
out_err:
	ceph_mdsc_put_request(req);
	dout("atomic_open result=%d\n", err);
A
Al Viro 已提交
272
	return err;
S
Sage Weil 已提交
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
}

int ceph_release(struct inode *inode, struct file *file)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_file_info *cf = file->private_data;

	dout("release inode %p file %p\n", inode, file);
	ceph_put_fmode(ci, cf->fmode);
	if (cf->last_readdir)
		ceph_mdsc_put_request(cf->last_readdir);
	kfree(cf->last_name);
	kfree(cf->dir_info);
	dput(cf->dentry);
	kmem_cache_free(ceph_file_cachep, cf);
288 289

	/* wake up anyone waiting for caps on this inode */
290
	wake_up_all(&ci->i_cap_wq);
S
Sage Weil 已提交
291 292 293 294 295 296 297 298 299 300 301 302
	return 0;
}

/*
 * Read a range of bytes striped over one or more objects.  Iterate over
 * objects we stripe over.  (That's not atomic, but good enough for now.)
 *
 * If we get a short result from the OSD, check against i_size; we need to
 * only return a short read to the caller if we hit EOF.
 */
static int striped_read(struct inode *inode,
			u64 off, u64 len,
303
			struct page **pages, int num_pages,
304
			int *checkeof, bool o_direct,
305
			unsigned long buf_align)
S
Sage Weil 已提交
306
{
307
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
S
Sage Weil 已提交
308 309
	struct ceph_inode_info *ci = ceph_inode(inode);
	u64 pos, this_len;
310
	int io_align, page_align;
S
Sage Weil 已提交
311 312 313 314 315 316 317 318 319 320 321 322 323 324
	int left, pages_left;
	int read;
	struct page **page_pos;
	int ret;
	bool hit_stripe, was_short;

	/*
	 * we may need to do multiple reads.  not atomic, unfortunately.
	 */
	pos = off;
	left = len;
	page_pos = pages;
	pages_left = num_pages;
	read = 0;
325
	io_align = off & ~PAGE_MASK;
S
Sage Weil 已提交
326 327

more:
328
	if (o_direct)
329
		page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
330 331
	else
		page_align = pos & ~PAGE_MASK;
S
Sage Weil 已提交
332
	this_len = left;
333
	ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
S
Sage Weil 已提交
334 335 336
				  &ci->i_layout, pos, &this_len,
				  ci->i_truncate_seq,
				  ci->i_truncate_size,
337
				  page_pos, pages_left, page_align);
S
Sage Weil 已提交
338 339
	if (ret == -ENOENT)
		ret = 0;
340 341
	hit_stripe = this_len < left;
	was_short = ret >= 0 && ret < this_len;
S
Sage Weil 已提交
342 343 344 345
	dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read,
	     ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : "");

	if (ret > 0) {
S
Sage Weil 已提交
346
		int didpages = (page_align + ret) >> PAGE_CACHE_SHIFT;
S
Sage Weil 已提交
347 348 349

		if (read < pos - off) {
			dout(" zero gap %llu to %llu\n", off + read, pos);
S
Sage Weil 已提交
350
			ceph_zero_page_vector_range(page_align + read,
351
						    pos - off - read, pages);
S
Sage Weil 已提交
352 353 354 355 356 357 358 359 360 361 362 363 364
		}
		pos += ret;
		read = pos - off;
		left -= ret;
		page_pos += didpages;
		pages_left -= didpages;

		/* hit stripe? */
		if (left && hit_stripe)
			goto more;
	}

	if (was_short) {
365 366 367 368 369 370 371 372 373 374
		/* did we bounce off eof? */
		if (pos + left > inode->i_size)
			*checkeof = 1;

		/* zero trailing bytes (inside i_size) */
		if (left > 0 && pos < inode->i_size) {
			if (pos + left > inode->i_size)
				left = inode->i_size - pos;

			dout("zero tail %d\n", left);
S
Sage Weil 已提交
375
			ceph_zero_page_vector_range(page_align + read, left,
376
						    pages);
377
			read += left;
S
Sage Weil 已提交
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
		}
	}

	if (ret >= 0)
		ret = read;
	dout("striped_read returns %d\n", ret);
	return ret;
}

/*
 * Completely synchronous read and write methods.  Direct from __user
 * buffer to osd, or directly to user pages (if O_DIRECT).
 *
 * If the read spans object boundary, just do multiple reads.
 */
static ssize_t ceph_sync_read(struct file *file, char __user *data,
394
			      unsigned len, loff_t *poff, int *checkeof)
S
Sage Weil 已提交
395 396 397 398
{
	struct inode *inode = file->f_dentry->d_inode;
	struct page **pages;
	u64 off = *poff;
399
	int num_pages, ret;
S
Sage Weil 已提交
400 401 402 403

	dout("sync_read on file %p %llu~%u %s\n", file, off, len,
	     (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");

404 405
	if (file->f_flags & O_DIRECT) {
		num_pages = calc_pages_for((unsigned long)data, len);
406
		pages = ceph_get_direct_page_vector(data, num_pages, true);
407 408
	} else {
		num_pages = calc_pages_for(off, len);
409
		pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
410
	}
S
Sage Weil 已提交
411 412 413
	if (IS_ERR(pages))
		return PTR_ERR(pages);

414 415 416 417 418 419
	/*
	 * flush any page cache pages in this range.  this
	 * will make concurrent normal and sync io slow,
	 * but it will at least behave sensibly when they are
	 * in sequence.
	 */
420 421 422 423
	ret = filemap_write_and_wait(inode->i_mapping);
	if (ret < 0)
		goto done;

424
	ret = striped_read(inode, off, len, pages, num_pages, checkeof,
425 426
			   file->f_flags & O_DIRECT,
			   (unsigned long)data & ~PAGE_MASK);
S
Sage Weil 已提交
427 428

	if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
429
		ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
S
Sage Weil 已提交
430 431 432
	if (ret >= 0)
		*poff = off + ret;

433
done:
S
Sage Weil 已提交
434
	if (file->f_flags & O_DIRECT)
435
		ceph_put_page_vector(pages, num_pages, true);
S
Sage Weil 已提交
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
	else
		ceph_release_page_vector(pages, num_pages);
	dout("sync_read result %d\n", ret);
	return ret;
}

/*
 * Write commit callback, called if we requested both an ACK and
 * ONDISK commit reply from the OSD.
 */
static void sync_write_commit(struct ceph_osd_request *req,
			      struct ceph_msg *msg)
{
	struct ceph_inode_info *ci = ceph_inode(req->r_inode);

	dout("sync_write_commit %p tid %llu\n", req, req->r_tid);
	spin_lock(&ci->i_unsafe_lock);
	list_del_init(&req->r_unsafe_item);
	spin_unlock(&ci->i_unsafe_lock);
	ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR);
}

/*
 * Synchronous write, straight from __user pointer or user pages (if
 * O_DIRECT).
 *
 * If write spans object boundary, just do multiple writes.  (For a
 * correct atomic write, we should e.g. take write locks on all
 * objects, rollback on failure, etc.)
 */
static ssize_t ceph_sync_write(struct file *file, const char __user *data,
			       size_t left, loff_t *offset)
{
	struct inode *inode = file->f_dentry->d_inode;
	struct ceph_inode_info *ci = ceph_inode(inode);
471
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
S
Sage Weil 已提交
472 473 474 475 476 477 478 479 480
	struct ceph_osd_request *req;
	struct page **pages;
	int num_pages;
	long long unsigned pos;
	u64 len;
	int written = 0;
	int flags;
	int do_sync = 0;
	int check_caps = 0;
481
	int page_align, io_align;
482
	unsigned long buf_align;
S
Sage Weil 已提交
483 484 485 486 487 488 489 490 491 492 493 494 495 496
	int ret;
	struct timespec mtime = CURRENT_TIME;

	if (ceph_snap(file->f_dentry->d_inode) != CEPH_NOSNAP)
		return -EROFS;

	dout("sync_write on file %p %lld~%u %s\n", file, *offset,
	     (unsigned)left, (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");

	if (file->f_flags & O_APPEND)
		pos = i_size_read(inode);
	else
		pos = *offset;

497 498 499 500 501 502 503 504 505 506
	ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left);
	if (ret < 0)
		return ret;

	ret = invalidate_inode_pages2_range(inode->i_mapping,
					    pos >> PAGE_CACHE_SHIFT,
					    (pos + left) >> PAGE_CACHE_SHIFT);
	if (ret < 0)
		dout("invalidate_inode_pages2_range returned %d\n", ret);

S
Sage Weil 已提交
507 508 509 510 511 512 513 514 515 516 517 518 519
	flags = CEPH_OSD_FLAG_ORDERSNAP |
		CEPH_OSD_FLAG_ONDISK |
		CEPH_OSD_FLAG_WRITE;
	if ((file->f_flags & (O_SYNC|O_DIRECT)) == 0)
		flags |= CEPH_OSD_FLAG_ACK;
	else
		do_sync = 1;

	/*
	 * we may need to do multiple writes here if we span an object
	 * boundary.  this isn't atomic, unfortunately.  :(
	 */
more:
520 521
	io_align = pos & ~PAGE_MASK;
	buf_align = (unsigned long)data & ~PAGE_MASK;
S
Sage Weil 已提交
522
	len = left;
523
	if (file->f_flags & O_DIRECT) {
524 525
		/* write from beginning of first page, regardless of
		   io alignment */
526 527 528
		page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
		num_pages = calc_pages_for((unsigned long)data, len);
	} else {
529
		page_align = pos & ~PAGE_MASK;
530 531
		num_pages = calc_pages_for(pos, len);
	}
532
	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
S
Sage Weil 已提交
533 534 535 536 537
				    ceph_vino(inode), pos, &len,
				    CEPH_OSD_OP_WRITE, flags,
				    ci->i_snap_realm->cached_context,
				    do_sync,
				    ci->i_truncate_seq, ci->i_truncate_size,
538
				    &mtime, false, 2, page_align);
539 540
	if (IS_ERR(req))
		return PTR_ERR(req);
S
Sage Weil 已提交
541 542

	if (file->f_flags & O_DIRECT) {
543
		pages = ceph_get_direct_page_vector(data, num_pages, false);
S
Sage Weil 已提交
544 545 546 547 548 549 550 551 552
		if (IS_ERR(pages)) {
			ret = PTR_ERR(pages);
			goto out;
		}

		/*
		 * throw out any page cache pages in this range. this
		 * may block.
		 */
S
Sage Weil 已提交
553
		truncate_inode_pages_range(inode->i_mapping, pos,
S
Sage Weil 已提交
554
					   (pos+len) | (PAGE_CACHE_SIZE-1));
S
Sage Weil 已提交
555
	} else {
556
		pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
S
Sage Weil 已提交
557 558 559 560
		if (IS_ERR(pages)) {
			ret = PTR_ERR(pages);
			goto out;
		}
561
		ret = ceph_copy_user_to_page_vector(pages, data, pos, len);
S
Sage Weil 已提交
562 563 564 565 566 567 568 569 570 571 572 573 574 575 576
		if (ret < 0) {
			ceph_release_page_vector(pages, num_pages);
			goto out;
		}

		if ((file->f_flags & O_SYNC) == 0) {
			/* get a second commit callback */
			req->r_safe_callback = sync_write_commit;
			req->r_own_pages = 1;
		}
	}
	req->r_pages = pages;
	req->r_num_pages = num_pages;
	req->r_inode = inode;

577
	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
S
Sage Weil 已提交
578 579 580 581 582 583 584
	if (!ret) {
		if (req->r_safe_callback) {
			/*
			 * Add to inode unsafe list only after we
			 * start_request so that a tid has been assigned.
			 */
			spin_lock(&ci->i_unsafe_lock);
585 586
			list_add_tail(&req->r_unsafe_item,
				      &ci->i_unsafe_writes);
S
Sage Weil 已提交
587 588 589
			spin_unlock(&ci->i_unsafe_lock);
			ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
		}
590
		
591
		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
592 593 594 595 596 597
		if (ret < 0 && req->r_safe_callback) {
			spin_lock(&ci->i_unsafe_lock);
			list_del_init(&req->r_unsafe_item);
			spin_unlock(&ci->i_unsafe_lock);
			ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR);
		}
S
Sage Weil 已提交
598 599 600
	}

	if (file->f_flags & O_DIRECT)
601
		ceph_put_page_vector(pages, num_pages, false);
S
Sage Weil 已提交
602 603 604 605 606 607 608 609 610
	else if (file->f_flags & O_SYNC)
		ceph_release_page_vector(pages, num_pages);

out:
	ceph_osdc_put_request(req);
	if (ret == 0) {
		pos += len;
		written += len;
		left -= len;
611
		data += written;
S
Sage Weil 已提交
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
		if (left)
			goto more;

		ret = written;
		*offset = pos;
		if (pos > i_size_read(inode))
			check_caps = ceph_inode_set_size(inode, pos);
		if (check_caps)
			ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY,
					NULL);
	}
	return ret;
}

/*
 * Wrap generic_file_aio_read with checks for cap bits on the inode.
 * Atomically grab references, so that those bits are not released
 * back to the MDS mid-read.
 *
 * Hmm, the sync read case isn't actually async... should it be?
 */
static ssize_t ceph_aio_read(struct kiocb *iocb, const struct iovec *iov,
			     unsigned long nr_segs, loff_t pos)
{
	struct file *filp = iocb->ki_filp;
637
	struct ceph_file_info *fi = filp->private_data;
S
Sage Weil 已提交
638 639 640 641
	loff_t *ppos = &iocb->ki_pos;
	size_t len = iov->iov_len;
	struct inode *inode = filp->f_dentry->d_inode;
	struct ceph_inode_info *ci = ceph_inode(inode);
Y
Yehuda Sadeh 已提交
642
	void __user *base = iov->iov_base;
S
Sage Weil 已提交
643
	ssize_t ret;
644
	int want, got = 0;
645
	int checkeof = 0, read = 0;
S
Sage Weil 已提交
646 647 648

	dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n",
	     inode, ceph_vinop(inode), pos, (unsigned)len, inode);
649
again:
S
Sage Weil 已提交
650
	__ceph_do_pending_vmtruncate(inode);
651 652 653 654 655
	if (fi->fmode & CEPH_FILE_MODE_LAZY)
		want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO;
	else
		want = CEPH_CAP_FILE_CACHE;
	ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, &got, -1);
S
Sage Weil 已提交
656 657 658 659 660 661
	if (ret < 0)
		goto out;
	dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n",
	     inode, ceph_vinop(inode), pos, (unsigned)len,
	     ceph_cap_string(got));

662
	if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 ||
S
Sage Weil 已提交
663
	    (iocb->ki_filp->f_flags & O_DIRECT) ||
664 665
	    (inode->i_sb->s_flags & MS_SYNCHRONOUS) ||
	    (fi->flags & CEPH_F_SYNC))
S
Sage Weil 已提交
666
		/* hmm, this isn't really async... */
667
		ret = ceph_sync_read(filp, base, len, ppos, &checkeof);
S
Sage Weil 已提交
668 669 670 671 672 673 674
	else
		ret = generic_file_aio_read(iocb, iov, nr_segs, pos);

out:
	dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n",
	     inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret);
	ceph_put_cap_refs(ci, got);
675 676 677 678 679 680

	if (checkeof && ret >= 0) {
		int statret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);

		/* hit EOF or hole? */
		if (statret == 0 && *ppos < inode->i_size) {
681
			dout("aio_read sync_read hit hole, ppos %lld < size %lld, reading more\n", *ppos, inode->i_size);
682 683 684 685 686 687 688 689 690 691
			read += ret;
			base += ret;
			len -= ret;
			checkeof = 0;
			goto again;
		}
	}
	if (ret >= 0)
		ret += read;

S
Sage Weil 已提交
692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
	return ret;
}

/*
 * Take cap references to avoid releasing caps to MDS mid-write.
 *
 * If we are synchronous, and write with an old snap context, the OSD
 * may return EOLDSNAPC.  In that case, retry the write.. _after_
 * dropping our cap refs and allowing the pending snap to logically
 * complete _before_ this write occurs.
 *
 * If we are near ENOSPC, write synchronously.
 */
static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
		       unsigned long nr_segs, loff_t pos)
{
	struct file *file = iocb->ki_filp;
709
	struct ceph_file_info *fi = file->private_data;
S
Sage Weil 已提交
710 711
	struct inode *inode = file->f_dentry->d_inode;
	struct ceph_inode_info *ci = ceph_inode(inode);
712 713
	struct ceph_osd_client *osdc =
		&ceph_sb_to_client(inode->i_sb)->client->osdc;
S
Sage Weil 已提交
714
	loff_t endoff = pos + iov->iov_len;
715
	int want, got = 0;
716
	int ret, err;
S
Sage Weil 已提交
717 718 719 720 721 722 723 724 725 726 727

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return -EROFS;

retry_snap:
	if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL))
		return -ENOSPC;
	__ceph_do_pending_vmtruncate(inode);
	dout("aio_write %p %llx.%llx %llu~%u getting caps. i_size %llu\n",
	     inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
	     inode->i_size);
728 729 730 731 732
	if (fi->fmode & CEPH_FILE_MODE_LAZY)
		want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO;
	else
		want = CEPH_CAP_FILE_BUFFER;
	ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff);
S
Sage Weil 已提交
733
	if (ret < 0)
734
		goto out_put;
S
Sage Weil 已提交
735 736 737 738 739

	dout("aio_write %p %llx.%llx %llu~%u  got cap refs on %s\n",
	     inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
	     ceph_cap_string(got));

740
	if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 ||
S
Sage Weil 已提交
741
	    (iocb->ki_filp->f_flags & O_DIRECT) ||
742 743
	    (inode->i_sb->s_flags & MS_SYNCHRONOUS) ||
	    (fi->flags & CEPH_F_SYNC)) {
S
Sage Weil 已提交
744 745 746
		ret = ceph_sync_write(file, iov->iov_base, iov->iov_len,
			&iocb->ki_pos);
	} else {
747 748 749 750 751
		/*
		 * buffered write; drop Fw early to avoid slow
		 * revocation if we get stuck on balance_dirty_pages
		 */
		int dirty;
S
Sage Weil 已提交
752

753
		spin_lock(&ci->i_ceph_lock);
754
		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
755
		spin_unlock(&ci->i_ceph_lock);
756
		ceph_put_cap_refs(ci, got);
S
Sage Weil 已提交
757

758
		ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
S
Sage Weil 已提交
759 760
		if ((ret >= 0 || ret == -EIOCBQUEUED) &&
		    ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host)
761
		     || ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) {
762
			err = vfs_fsync_range(file, pos, pos + ret - 1, 1);
763 764 765
			if (err < 0)
				ret = err;
		}
766 767 768 769

		if (dirty)
			__mark_inode_dirty(inode, dirty);
		goto out;
S
Sage Weil 已提交
770
	}
771

S
Sage Weil 已提交
772
	if (ret >= 0) {
773
		int dirty;
774
		spin_lock(&ci->i_ceph_lock);
775
		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
776
		spin_unlock(&ci->i_ceph_lock);
777 778
		if (dirty)
			__mark_inode_dirty(inode, dirty);
S
Sage Weil 已提交
779 780
	}

781
out_put:
S
Sage Weil 已提交
782 783 784 785 786
	dout("aio_write %p %llx.%llx %llu~%u  dropping cap refs on %s\n",
	     inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
	     ceph_cap_string(got));
	ceph_put_cap_refs(ci, got);

787
out:
S
Sage Weil 已提交
788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
	if (ret == -EOLDSNAPC) {
		dout("aio_write %p %llx.%llx %llu~%u got EOLDSNAPC, retrying\n",
		     inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len);
		goto retry_snap;
	}

	return ret;
}

/*
 * llseek.  be sure to verify file size on SEEK_END.
 */
static loff_t ceph_llseek(struct file *file, loff_t offset, int origin)
{
	struct inode *inode = file->f_mapping->host;
	int ret;

	mutex_lock(&inode->i_mutex);
	__ceph_do_pending_vmtruncate(inode);
807 808

	if (origin == SEEK_END || origin == SEEK_DATA || origin == SEEK_HOLE) {
S
Sage Weil 已提交
809 810 811 812 813
		ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
		if (ret < 0) {
			offset = ret;
			goto out;
		}
814 815 816 817
	}

	switch (origin) {
	case SEEK_END:
S
Sage Weil 已提交
818 819 820 821 822 823 824 825 826 827 828 829 830 831 832
		offset += inode->i_size;
		break;
	case SEEK_CUR:
		/*
		 * Here we special-case the lseek(fd, 0, SEEK_CUR)
		 * position-querying operation.  Avoid rewriting the "same"
		 * f_pos value back to the file because a concurrent read(),
		 * write() or lseek() might have altered it
		 */
		if (offset == 0) {
			offset = file->f_pos;
			goto out;
		}
		offset += file->f_pos;
		break;
833 834 835 836 837 838 839 840 841 842 843 844 845
	case SEEK_DATA:
		if (offset >= inode->i_size) {
			ret = -ENXIO;
			goto out;
		}
		break;
	case SEEK_HOLE:
		if (offset >= inode->i_size) {
			ret = -ENXIO;
			goto out;
		}
		offset = inode->i_size;
		break;
S
Sage Weil 已提交
846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873
	}

	if (offset < 0 || offset > inode->i_sb->s_maxbytes) {
		offset = -EINVAL;
		goto out;
	}

	/* Special lock needed here? */
	if (offset != file->f_pos) {
		file->f_pos = offset;
		file->f_version = 0;
	}

out:
	mutex_unlock(&inode->i_mutex);
	return offset;
}

const struct file_operations ceph_file_fops = {
	.open = ceph_open,
	.release = ceph_release,
	.llseek = ceph_llseek,
	.read = do_sync_read,
	.write = do_sync_write,
	.aio_read = ceph_aio_read,
	.aio_write = ceph_aio_write,
	.mmap = ceph_mmap,
	.fsync = ceph_fsync,
G
Greg Farnum 已提交
874 875
	.lock = ceph_lock,
	.flock = ceph_flock,
S
Sage Weil 已提交
876 877 878 879 880 881
	.splice_read = generic_file_splice_read,
	.splice_write = generic_file_splice_write,
	.unlocked_ioctl = ceph_ioctl,
	.compat_ioctl	= ceph_ioctl,
};