addr.c 46.8 KB
Newer Older
1
#include <linux/ceph/ceph_debug.h>
S
Sage Weil 已提交
2 3 4 5 6 7

#include <linux/backing-dev.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/writeback.h>	/* generic_writepages */
8
#include <linux/slab.h>
S
Sage Weil 已提交
9 10 11 12
#include <linux/pagevec.h>
#include <linux/task_io_accounting_ops.h>

#include "super.h"
13
#include "mds_client.h"
14
#include "cache.h"
15
#include <linux/ceph/osd_client.h>
S
Sage Weil 已提交
16 17 18 19 20 21 22 23 24 25 26 27

/*
 * Ceph address space ops.
 *
 * There are a few funny things going on here.
 *
 * The page->private field is used to reference a struct
 * ceph_snap_context for _every_ dirty page.  This indicates which
 * snapshot the page was logically dirtied in, and thus which snap
 * context needs to be associated with the osd write during writeback.
 *
 * Similarly, struct ceph_inode_info maintains a set of counters to
L
Lucas De Marchi 已提交
28
 * count dirty pages on the inode.  In the absence of snapshots,
S
Sage Weil 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
 * i_wrbuffer_ref == i_wrbuffer_ref_head == the dirty page count.
 *
 * When a snapshot is taken (that is, when the client receives
 * notification that a snapshot was taken), each inode with caps and
 * with dirty pages (dirty pages implies there is a cap) gets a new
 * ceph_cap_snap in the i_cap_snaps list (which is sorted in ascending
 * order, new snaps go to the tail).  The i_wrbuffer_ref_head count is
 * moved to capsnap->dirty. (Unless a sync write is currently in
 * progress.  In that case, the capsnap is said to be "pending", new
 * writes cannot start, and the capsnap isn't "finalized" until the
 * write completes (or fails) and a final size/mtime for the inode for
 * that snap can be settled upon.)  i_wrbuffer_ref_head is reset to 0.
 *
 * On writeback, we must submit writes to the osd IN SNAP ORDER.  So,
 * we look for the first capsnap in i_cap_snaps and write out pages in
 * that snap context _only_.  Then we move on to the next capsnap,
 * eventually reaching the "live" or "head" context (i.e., pages that
 * are not yet snapped) and are writing the most recently dirtied
 * pages.
 *
 * Invalidate and so forth must take care to ensure the dirty page
 * accounting is preserved.
 */

Y
Yehuda Sadeh 已提交
53 54 55 56 57
#define CONGESTION_ON_THRESH(congestion_kb) (congestion_kb >> (PAGE_SHIFT-10))
#define CONGESTION_OFF_THRESH(congestion_kb)				\
	(CONGESTION_ON_THRESH(congestion_kb) -				\
	 (CONGESTION_ON_THRESH(congestion_kb) >> 2))

58 59 60 61 62 63
static inline struct ceph_snap_context *page_snap_context(struct page *page)
{
	if (PagePrivate(page))
		return (void *)page->private;
	return NULL;
}
S
Sage Weil 已提交
64 65 66 67 68 69 70 71 72 73 74

/*
 * Dirty a page.  Optimistically adjust accounting, on the assumption
 * that we won't race with invalidate.  If we do, readjust.
 */
static int ceph_set_page_dirty(struct page *page)
{
	struct address_space *mapping = page->mapping;
	struct inode *inode;
	struct ceph_inode_info *ci;
	struct ceph_snap_context *snapc;
75
	int ret;
S
Sage Weil 已提交
76 77 78 79

	if (unlikely(!mapping))
		return !TestSetPageDirty(page);

80
	if (PageDirty(page)) {
S
Sage Weil 已提交
81 82
		dout("%p set_page_dirty %p idx %lu -- already dirty\n",
		     mapping->host, page, page->index);
83
		BUG_ON(!PagePrivate(page));
S
Sage Weil 已提交
84 85 86 87 88 89 90 91 92 93 94 95 96
		return 0;
	}

	inode = mapping->host;
	ci = ceph_inode(inode);

	/*
	 * Note that we're grabbing a snapc ref here without holding
	 * any locks!
	 */
	snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);

	/* dirty the head */
97
	spin_lock(&ci->i_ceph_lock);
98
	if (ci->i_head_snapc == NULL)
S
Sage Weil 已提交
99 100 101
		ci->i_head_snapc = ceph_get_snap_context(snapc);
	++ci->i_wrbuffer_ref_head;
	if (ci->i_wrbuffer_ref == 0)
102
		ihold(inode);
S
Sage Weil 已提交
103 104 105 106 107 108 109
	++ci->i_wrbuffer_ref;
	dout("%p set_page_dirty %p idx %lu head %d/%d -> %d/%d "
	     "snapc %p seq %lld (%d snaps)\n",
	     mapping->host, page, page->index,
	     ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1,
	     ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head,
	     snapc, snapc->seq, snapc->num_snaps);
110
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
111

112 113 114 115 116 117 118
	/*
	 * Reference snap context in page->private.  Also set
	 * PagePrivate so that we get invalidatepage callback.
	 */
	BUG_ON(PagePrivate(page));
	page->private = (unsigned long)snapc;
	SetPagePrivate(page);
S
Sage Weil 已提交
119

120 121 122
	ret = __set_page_dirty_nobuffers(page);
	WARN_ON(!PageLocked(page));
	WARN_ON(!page->mapping);
S
Sage Weil 已提交
123

124
	return ret;
S
Sage Weil 已提交
125 126 127 128 129 130 131
}

/*
 * If we are truncating the full page (i.e. offset == 0), adjust the
 * dirty page counters appropriately.  Only called if there is private
 * data on the page.
 */
132 133
static void ceph_invalidatepage(struct page *page, unsigned int offset,
				unsigned int length)
S
Sage Weil 已提交
134
{
135
	struct inode *inode;
S
Sage Weil 已提交
136
	struct ceph_inode_info *ci;
137
	struct ceph_snap_context *snapc = page_snap_context(page);
S
Sage Weil 已提交
138

139
	inode = page->mapping->host;
140 141 142 143 144 145 146
	ci = ceph_inode(inode);

	if (offset != 0 || length != PAGE_CACHE_SIZE) {
		dout("%p invalidatepage %p idx %lu partial dirty page %u~%u\n",
		     inode, page, page->index, offset, length);
		return;
	}
147

148 149 150 151 152
	ceph_invalidate_fscache_page(inode, page);

	if (!PagePrivate(page))
		return;

S
Sage Weil 已提交
153 154 155 156 157 158 159 160
	/*
	 * We can get non-dirty pages here due to races between
	 * set_page_dirty and truncate_complete_page; just spit out a
	 * warning, in case we end up with accounting problems later.
	 */
	if (!PageDirty(page))
		pr_err("%p invalidatepage %p page not dirty\n", inode, page);

161
	ClearPageChecked(page);
S
Sage Weil 已提交
162

163 164 165 166 167 168 169
	dout("%p invalidatepage %p idx %lu full dirty page\n",
	     inode, page, page->index);

	ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
	ceph_put_snap_context(snapc);
	page->private = 0;
	ClearPagePrivate(page);
S
Sage Weil 已提交
170 171 172 173 174 175 176
}

static int ceph_releasepage(struct page *page, gfp_t g)
{
	struct inode *inode = page->mapping ? page->mapping->host : NULL;
	dout("%p releasepage %p idx %lu\n", inode, page, page->index);
	WARN_ON(PageDirty(page));
177 178 179 180 181 182

	/* Can we release the page from the cache? */
	if (!ceph_release_fscache_page(page, g))
		return 0;

	return !PagePrivate(page);
S
Sage Weil 已提交
183 184 185 186 187 188 189
}

/*
 * read a single page, without unlocking it.
 */
static int readpage_nounlock(struct file *filp, struct page *page)
{
A
Al Viro 已提交
190
	struct inode *inode = file_inode(filp);
S
Sage Weil 已提交
191
	struct ceph_inode_info *ci = ceph_inode(inode);
192
	struct ceph_osd_client *osdc =
193
		&ceph_inode_to_client(inode)->client->osdc;
S
Sage Weil 已提交
194
	int err = 0;
Y
Yan, Zheng 已提交
195
	u64 off = page_offset(page);
S
Sage Weil 已提交
196 197
	u64 len = PAGE_CACHE_SIZE;

Y
Yan, Zheng 已提交
198
	if (off >= i_size_read(inode)) {
199
		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
Y
Yan, Zheng 已提交
200 201 202
		SetPageUptodate(page);
		return 0;
	}
203

204 205 206 207 208 209 210 211 212 213 214
	if (ci->i_inline_version != CEPH_INLINE_NONE) {
		/*
		 * Uptodate inline data should have been added
		 * into page cache while getting Fcr caps.
		 */
		if (off == 0)
			return -EINVAL;
		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
		SetPageUptodate(page);
		return 0;
	}
Y
Yan, Zheng 已提交
215 216

	err = ceph_readpage_from_fscache(inode, page);
217 218 219
	if (err == 0)
		goto out;

S
Sage Weil 已提交
220 221 222
	dout("readpage inode %p file %p page %p index %lu\n",
	     inode, filp, page, page->index);
	err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
Y
Yan, Zheng 已提交
223
				  off, &len,
S
Sage Weil 已提交
224
				  ci->i_truncate_seq, ci->i_truncate_size,
225
				  &page, 1, 0);
S
Sage Weil 已提交
226 227 228 229
	if (err == -ENOENT)
		err = 0;
	if (err < 0) {
		SetPageError(page);
230
		ceph_fscache_readpage_cancel(inode, page);
S
Sage Weil 已提交
231 232
		goto out;
	}
233 234 235 236 237
	if (err < PAGE_CACHE_SIZE)
		/* zero fill remainder of page */
		zero_user_segment(page, err, PAGE_CACHE_SIZE);
	else
		flush_dcache_page(page);
S
Sage Weil 已提交
238

239 240
	SetPageUptodate(page);
	ceph_readpage_to_fscache(inode, page);
241

S
Sage Weil 已提交
242 243 244 245 246 247 248 249 250 251 252 253
out:
	return err < 0 ? err : 0;
}

static int ceph_readpage(struct file *filp, struct page *page)
{
	int r = readpage_nounlock(filp, page);
	unlock_page(page);
	return r;
}

/*
S
Sage Weil 已提交
254
 * Finish an async read(ahead) op.
S
Sage Weil 已提交
255
 */
S
Sage Weil 已提交
256
static void finish_read(struct ceph_osd_request *req, struct ceph_msg *msg)
S
Sage Weil 已提交
257
{
S
Sage Weil 已提交
258
	struct inode *inode = req->r_inode;
259
	struct ceph_osd_data *osd_data;
260 261
	int rc = req->r_result;
	int bytes = le32_to_cpu(msg->hdr.data_len);
262
	int num_pages;
S
Sage Weil 已提交
263
	int i;
S
Sage Weil 已提交
264

S
Sage Weil 已提交
265 266 267
	dout("finish_read %p req %p rc %d bytes %d\n", inode, req, rc, bytes);

	/* unlock all pages, zeroing any data we didn't read */
268
	osd_data = osd_req_op_extent_osd_data(req, 0);
269 270 271
	BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES);
	num_pages = calc_pages_for((u64)osd_data->alignment,
					(u64)osd_data->length);
272
	for (i = 0; i < num_pages; i++) {
273
		struct page *page = osd_data->pages[i];
S
Sage Weil 已提交
274

275 276
		if (rc < 0)
			goto unlock;
S
Sage Weil 已提交
277 278 279 280
		if (bytes < (int)PAGE_CACHE_SIZE) {
			/* zero (remainder of) page */
			int s = bytes < 0 ? 0 : bytes;
			zero_user_segment(page, s, PAGE_CACHE_SIZE);
S
Sage Weil 已提交
281
		}
S
Sage Weil 已提交
282 283 284 285
 		dout("finish_read %p uptodate %p idx %lu\n", inode, page,
		     page->index);
		flush_dcache_page(page);
		SetPageUptodate(page);
286
		ceph_readpage_to_fscache(inode, page);
287
unlock:
S
Sage Weil 已提交
288 289
		unlock_page(page);
		page_cache_release(page);
290
		bytes -= PAGE_CACHE_SIZE;
S
Sage Weil 已提交
291
	}
292
	kfree(osd_data->pages);
S
Sage Weil 已提交
293 294
}

295 296 297 298 299 300 301 302
static void ceph_unlock_page_vector(struct page **pages, int num_pages)
{
	int i;

	for (i = 0; i < num_pages; i++)
		unlock_page(pages[i]);
}

S
Sage Weil 已提交
303
/*
S
Sage Weil 已提交
304 305
 * start an async read(ahead) operation.  return nr_pages we submitted
 * a read for on success, or negative error code.
S
Sage Weil 已提交
306
 */
307
static int start_read(struct inode *inode, struct list_head *page_list, int max)
S
Sage Weil 已提交
308
{
309 310
	struct ceph_osd_client *osdc =
		&ceph_inode_to_client(inode)->client->osdc;
S
Sage Weil 已提交
311 312
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct page *page = list_entry(page_list->prev, struct page, lru);
313
	struct ceph_vino vino;
S
Sage Weil 已提交
314 315
	struct ceph_osd_request *req;
	u64 off;
S
Sage Weil 已提交
316
	u64 len;
S
Sage Weil 已提交
317 318 319 320 321
	int i;
	struct page **pages;
	pgoff_t next_index;
	int nr_pages = 0;
	int ret;
S
Sage Weil 已提交
322

323
	off = (u64) page_offset(page);
S
Sage Weil 已提交
324

S
Sage Weil 已提交
325 326 327 328 329 330 331
	/* count pages */
	next_index = page->index;
	list_for_each_entry_reverse(page, page_list, lru) {
		if (page->index != next_index)
			break;
		nr_pages++;
		next_index++;
332 333
		if (max && nr_pages == max)
			break;
S
Sage Weil 已提交
334
	}
S
Sage Weil 已提交
335
	len = nr_pages << PAGE_CACHE_SHIFT;
S
Sage Weil 已提交
336 337
	dout("start_read %p nr_pages %d is %lld~%lld\n", inode, nr_pages,
	     off, len);
338 339
	vino = ceph_vino(inode);
	req = ceph_osdc_new_request(osdc, &ci->i_layout, vino, off, &len,
340
				    0, 1, CEPH_OSD_OP_READ,
341
				    CEPH_OSD_FLAG_READ, NULL,
S
Sage Weil 已提交
342
				    ci->i_truncate_seq, ci->i_truncate_size,
343
				    false);
344 345
	if (IS_ERR(req))
		return PTR_ERR(req);
S
Sage Weil 已提交
346

S
Sage Weil 已提交
347
	/* build page vector */
348
	nr_pages = calc_pages_for(0, len);
S
Sage Weil 已提交
349 350 351 352 353 354 355
	pages = kmalloc(sizeof(*pages) * nr_pages, GFP_NOFS);
	ret = -ENOMEM;
	if (!pages)
		goto out;
	for (i = 0; i < nr_pages; ++i) {
		page = list_entry(page_list->prev, struct page, lru);
		BUG_ON(PageLocked(page));
S
Sage Weil 已提交
356
		list_del(&page->lru);
357

S
Sage Weil 已提交
358 359 360
 		dout("start_read %p adding %p idx %lu\n", inode, page,
		     page->index);
		if (add_to_page_cache_lru(page, &inode->i_data, page->index,
S
Sage Weil 已提交
361
					  GFP_NOFS)) {
362
			ceph_fscache_uncache_page(inode, page);
S
Sage Weil 已提交
363
			page_cache_release(page);
S
Sage Weil 已提交
364
			dout("start_read %p add_to_page_cache failed %p\n",
S
Sage Weil 已提交
365
			     inode, page);
S
Sage Weil 已提交
366 367
			nr_pages = i;
			goto out_pages;
S
Sage Weil 已提交
368
		}
S
Sage Weil 已提交
369
		pages[i] = page;
S
Sage Weil 已提交
370
	}
371
	osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, false, false);
S
Sage Weil 已提交
372 373 374
	req->r_callback = finish_read;
	req->r_inode = inode;

375
	ceph_osdc_build_request(req, off, NULL, vino.snap, NULL);
376

S
Sage Weil 已提交
377 378 379 380 381 382 383 384
	dout("start_read %p starting %p %lld~%lld\n", inode, req, off, len);
	ret = ceph_osdc_start_request(osdc, req, false);
	if (ret < 0)
		goto out_pages;
	ceph_osdc_put_request(req);
	return nr_pages;

out_pages:
385
	ceph_unlock_page_vector(pages, nr_pages);
S
Sage Weil 已提交
386 387 388 389 390
	ceph_release_page_vector(pages, nr_pages);
out:
	ceph_osdc_put_request(req);
	return ret;
}
S
Sage Weil 已提交
391

S
Sage Weil 已提交
392 393 394 395 396 397 398 399

/*
 * Read multiple pages.  Leave pages we don't read + unlock in page_list;
 * the caller (VM) cleans them up.
 */
static int ceph_readpages(struct file *file, struct address_space *mapping,
			  struct list_head *page_list, unsigned nr_pages)
{
A
Al Viro 已提交
400
	struct inode *inode = file_inode(file);
401
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
S
Sage Weil 已提交
402
	int rc = 0;
403 404
	int max = 0;

Y
Yan, Zheng 已提交
405 406 407
	if (ceph_inode(inode)->i_inline_version != CEPH_INLINE_NONE)
		return -EINVAL;

408 409 410 411 412 413
	rc = ceph_readpages_from_fscache(mapping->host, mapping, page_list,
					 &nr_pages);

	if (rc == 0)
		goto out;

414 415 416
	if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE)
		max = (fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1)
			>> PAGE_SHIFT;
S
Sage Weil 已提交
417

418 419
	dout("readpages %p file %p nr_pages %d max %d\n", inode,
		file, nr_pages,
420
	     max);
S
Sage Weil 已提交
421
	while (!list_empty(page_list)) {
422
		rc = start_read(inode, page_list, max);
S
Sage Weil 已提交
423 424 425 426
		if (rc < 0)
			goto out;
		BUG_ON(rc == 0);
	}
S
Sage Weil 已提交
427
out:
428 429
	ceph_fscache_readpages_cancel(inode, page_list);

S
Sage Weil 已提交
430
	dout("readpages %p file %p ret %d\n", inode, file, rc);
S
Sage Weil 已提交
431 432 433 434 435 436 437
	return rc;
}

/*
 * Get ref for the oldest snapc for an inode with dirty data... that is, the
 * only snap context we are allowed to write back.
 */
438 439
static struct ceph_snap_context *get_oldest_context(struct inode *inode,
						    u64 *snap_size)
S
Sage Weil 已提交
440 441 442 443 444
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_snap_context *snapc = NULL;
	struct ceph_cap_snap *capsnap = NULL;

445
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
446 447 448 449 450 451 452 453 454 455
	list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
		dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
		     capsnap->context, capsnap->dirty_pages);
		if (capsnap->dirty_pages) {
			snapc = ceph_get_snap_context(capsnap->context);
			if (snap_size)
				*snap_size = capsnap->size;
			break;
		}
	}
456
	if (!snapc && ci->i_wrbuffer_ref_head) {
457
		snapc = ceph_get_snap_context(ci->i_head_snapc);
S
Sage Weil 已提交
458 459 460
		dout(" head snapc %p has %d dirty pages\n",
		     snapc, ci->i_wrbuffer_ref_head);
	}
461
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
462 463 464 465 466 467 468 469 470 471 472 473 474
	return snapc;
}

/*
 * Write a single page, but leave the page locked.
 *
 * If we get a write error, set the page error bit, but still adjust the
 * dirty page accounting (i.e., page is no longer dirty).
 */
static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
{
	struct inode *inode;
	struct ceph_inode_info *ci;
475
	struct ceph_fs_client *fsc;
S
Sage Weil 已提交
476
	struct ceph_osd_client *osdc;
477
	struct ceph_snap_context *snapc, *oldest;
478
	loff_t page_off = page_offset(page);
Y
Yehuda Sadeh 已提交
479
	long writeback_stat;
480 481 482
	u64 truncate_size, snap_size = 0;
	u32 truncate_seq;
	int err = 0, len = PAGE_CACHE_SIZE;
S
Sage Weil 已提交
483 484 485 486 487 488 489 490 491

	dout("writepage %p idx %lu\n", page, page->index);

	if (!page->mapping || !page->mapping->host) {
		dout("writepage %p - no mapping\n", page);
		return -EFAULT;
	}
	inode = page->mapping->host;
	ci = ceph_inode(inode);
492 493
	fsc = ceph_inode_to_client(inode);
	osdc = &fsc->client->osdc;
S
Sage Weil 已提交
494 495

	/* verify this is a writeable snap context */
496
	snapc = page_snap_context(page);
S
Sage Weil 已提交
497 498 499 500
	if (snapc == NULL) {
		dout("writepage %p page %p not dirty?\n", inode, page);
		goto out;
	}
501 502
	oldest = get_oldest_context(inode, &snap_size);
	if (snapc->seq > oldest->seq) {
S
Sage Weil 已提交
503
		dout("writepage %p page %p snapc %p not writeable - noop\n",
504
		     inode, page, snapc);
S
Sage Weil 已提交
505 506
		/* we should only noop if called by kswapd */
		WARN_ON((current->flags & PF_MEMALLOC) == 0);
507
		ceph_put_snap_context(oldest);
S
Sage Weil 已提交
508 509
		goto out;
	}
510
	ceph_put_snap_context(oldest);
S
Sage Weil 已提交
511

512 513 514 515 516 517 518
	spin_lock(&ci->i_ceph_lock);
	truncate_seq = ci->i_truncate_seq;
	truncate_size = ci->i_truncate_size;
	if (!snap_size)
		snap_size = i_size_read(inode);
	spin_unlock(&ci->i_ceph_lock);

S
Sage Weil 已提交
519
	/* is this a partial page at end of file? */
520 521 522 523 524 525
	if (page_off >= snap_size) {
		dout("%p page eof %llu\n", page, snap_size);
		goto out;
	}
	if (snap_size < page_off + len)
		len = snap_size - page_off;
S
Sage Weil 已提交
526

S
Sage Weil 已提交
527 528
	dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
	     inode, page, page->index, page_off, len, snapc);
S
Sage Weil 已提交
529

530
	writeback_stat = atomic_long_inc_return(&fsc->writeback_count);
Y
Yehuda Sadeh 已提交
531
	if (writeback_stat >
532 533
	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
		set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
Y
Yehuda Sadeh 已提交
534

535 536
	ceph_readpage_to_fscache(inode, page);

S
Sage Weil 已提交
537 538 539 540
	set_page_writeback(page);
	err = ceph_osdc_writepages(osdc, ceph_vino(inode),
				   &ci->i_layout, snapc,
				   page_off, len,
541
				   truncate_seq, truncate_size,
542
				   &inode->i_mtime, &page, 1);
S
Sage Weil 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556
	if (err < 0) {
		dout("writepage setting page/mapping error %d %p\n", err, page);
		SetPageError(page);
		mapping_set_error(&inode->i_data, err);
		if (wbc)
			wbc->pages_skipped++;
	} else {
		dout("writepage cleaned page %p\n", page);
		err = 0;  /* vfs expects us to return 0 */
	}
	page->private = 0;
	ClearPagePrivate(page);
	end_page_writeback(page);
	ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
557
	ceph_put_snap_context(snapc);  /* page's reference */
S
Sage Weil 已提交
558 559 560 561 562 563
out:
	return err;
}

static int ceph_writepage(struct page *page, struct writeback_control *wbc)
{
564 565 566
	int err;
	struct inode *inode = page->mapping->host;
	BUG_ON(!inode);
567
	ihold(inode);
568
	err = writepage_nounlock(page, wbc);
S
Sage Weil 已提交
569
	unlock_page(page);
570
	iput(inode);
S
Sage Weil 已提交
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
	return err;
}


/*
 * lame release_pages helper.  release_pages() isn't exported to
 * modules.
 */
static void ceph_release_pages(struct page **pages, int num)
{
	struct pagevec pvec;
	int i;

	pagevec_init(&pvec, 0);
	for (i = 0; i < num; i++) {
		if (pagevec_add(&pvec, pages[i]) == 0)
			pagevec_release(&pvec);
	}
	pagevec_release(&pvec);
}

/*
 * async writeback completion handler.
 *
 * If we get an error, set the mapping error bit, but not the individual
 * page error bits.
 */
static void writepages_finish(struct ceph_osd_request *req,
			      struct ceph_msg *msg)
{
	struct inode *inode = req->r_inode;
	struct ceph_inode_info *ci = ceph_inode(inode);
603
	struct ceph_osd_data *osd_data;
S
Sage Weil 已提交
604 605
	unsigned wrote;
	struct page *page;
606
	int num_pages;
S
Sage Weil 已提交
607 608 609
	int i;
	struct ceph_snap_context *snapc = req->r_snapc;
	struct address_space *mapping = inode->i_mapping;
610
	int rc = req->r_result;
611
	u64 bytes = req->r_ops[0].extent.length;
612
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
Y
Yehuda Sadeh 已提交
613
	long writeback_stat;
S
Sage Weil 已提交
614
	unsigned issued = ceph_caps_issued(ci);
S
Sage Weil 已提交
615

616
	osd_data = osd_req_op_extent_osd_data(req, 0);
617 618 619
	BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES);
	num_pages = calc_pages_for((u64)osd_data->alignment,
					(u64)osd_data->length);
S
Sage Weil 已提交
620
	if (rc >= 0) {
621 622 623 624 625 626
		/*
		 * Assume we wrote the pages we originally sent.  The
		 * osd might reply with fewer pages if our writeback
		 * raced with a truncation and was adjusted at the osd,
		 * so don't believe the reply.
		 */
627
		wrote = num_pages;
S
Sage Weil 已提交
628 629 630 631 632 633 634 635
	} else {
		wrote = 0;
		mapping_set_error(mapping, rc);
	}
	dout("writepages_finish %p rc %d bytes %llu wrote %d (pages)\n",
	     inode, rc, bytes, wrote);

	/* clean all pages */
636
	for (i = 0; i < num_pages; i++) {
637
		page = osd_data->pages[i];
S
Sage Weil 已提交
638 639 640
		BUG_ON(!page);
		WARN_ON(!PageUptodate(page));

Y
Yehuda Sadeh 已提交
641
		writeback_stat =
642
			atomic_long_dec_return(&fsc->writeback_count);
Y
Yehuda Sadeh 已提交
643
		if (writeback_stat <
644 645
		    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
			clear_bdi_congested(&fsc->backing_dev_info,
Y
Yehuda Sadeh 已提交
646 647
					    BLK_RW_ASYNC);

648
		ceph_put_snap_context(page_snap_context(page));
S
Sage Weil 已提交
649 650 651 652
		page->private = 0;
		ClearPagePrivate(page);
		dout("unlocking %d %p\n", i, page);
		end_page_writeback(page);
653 654 655 656 657 658 659

		/*
		 * We lost the cache cap, need to truncate the page before
		 * it is unlocked, otherwise we'd truncate it later in the
		 * page truncation thread, possibly losing some data that
		 * raced its way in
		 */
660
		if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0)
661 662
			generic_error_remove_page(inode->i_mapping, page);

S
Sage Weil 已提交
663 664 665
		unlock_page(page);
	}
	dout("%p wrote+cleaned %d pages\n", inode, wrote);
666
	ceph_put_wrbuffer_cap_refs(ci, num_pages, snapc);
S
Sage Weil 已提交
667

668 669 670
	ceph_release_pages(osd_data->pages, num_pages);
	if (osd_data->pages_from_pool)
		mempool_free(osd_data->pages,
671
			     ceph_sb_to_client(inode->i_sb)->wb_pagevec_pool);
S
Sage Weil 已提交
672
	else
673
		kfree(osd_data->pages);
S
Sage Weil 已提交
674 675 676 677 678 679 680 681 682 683 684
	ceph_osdc_put_request(req);
}

/*
 * initiate async writeback
 */
static int ceph_writepages_start(struct address_space *mapping,
				 struct writeback_control *wbc)
{
	struct inode *inode = mapping->host;
	struct ceph_inode_info *ci = ceph_inode(inode);
685 686
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
	struct ceph_vino vino = ceph_vino(inode);
S
Sage Weil 已提交
687 688 689 690
	pgoff_t index, start, end;
	int range_whole = 0;
	int should_loop = 1;
	pgoff_t max_pages = 0, max_pages_ever = 0;
691
	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
S
Sage Weil 已提交
692 693 694 695 696
	struct pagevec pvec;
	int done = 0;
	int rc = 0;
	unsigned wsize = 1 << inode->i_blkbits;
	struct ceph_osd_request *req = NULL;
D
Dan Carpenter 已提交
697
	int do_sync = 0;
698 699
	u64 truncate_size, snap_size;
	u32 truncate_seq;
S
Sage Weil 已提交
700 701 702 703 704 705

	/*
	 * Include a 'sync' in the OSD request if this is a data
	 * integrity write (e.g., O_SYNC write or fsync()), or if our
	 * cap is being revoked.
	 */
706 707
	if ((wbc->sync_mode == WB_SYNC_ALL) ||
		ceph_caps_revoking(ci, CEPH_CAP_FILE_BUFFER))
S
Sage Weil 已提交
708 709 710 711 712 713
		do_sync = 1;
	dout("writepages_start %p dosync=%d (mode=%s)\n",
	     inode, do_sync,
	     wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));

714
	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
715
		pr_warn("writepage_start %p on forced umount\n", inode);
S
Sage Weil 已提交
716 717
		return -EIO; /* we're in a forced umount, don't write! */
	}
718 719
	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
		wsize = fsc->mount_options->wsize;
S
Sage Weil 已提交
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743
	if (wsize < PAGE_CACHE_SIZE)
		wsize = PAGE_CACHE_SIZE;
	max_pages_ever = wsize >> PAGE_CACHE_SHIFT;

	pagevec_init(&pvec, 0);

	/* where to start/end? */
	if (wbc->range_cyclic) {
		start = mapping->writeback_index; /* Start from prev offset */
		end = -1;
		dout(" cyclic, start at %lu\n", start);
	} else {
		start = wbc->range_start >> PAGE_CACHE_SHIFT;
		end = wbc->range_end >> PAGE_CACHE_SHIFT;
		if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
			range_whole = 1;
		should_loop = 0;
		dout(" not cyclic, %lu to %lu\n", start, end);
	}
	index = start;

retry:
	/* find oldest snap context with dirty data */
	ceph_put_snap_context(snapc);
744
	snap_size = 0;
S
Sage Weil 已提交
745 746 747 748 749 750 751
	snapc = get_oldest_context(inode, &snap_size);
	if (!snapc) {
		/* hmm, why does writepages get called when there
		   is no dirty data? */
		dout(" no snap context with dirty data?\n");
		goto out;
	}
752 753
	if (snap_size == 0)
		snap_size = i_size_read(inode);
S
Sage Weil 已提交
754 755
	dout(" oldest snapc is %p seq %lld (%d snaps)\n",
	     snapc, snapc->seq, snapc->num_snaps);
756 757 758 759 760 761 762 763

	spin_lock(&ci->i_ceph_lock);
	truncate_seq = ci->i_truncate_seq;
	truncate_size = ci->i_truncate_size;
	if (!snap_size)
		snap_size = i_size_read(inode);
	spin_unlock(&ci->i_ceph_lock);

S
Sage Weil 已提交
764 765 766 767 768 769 770 771 772 773 774 775 776 777
	if (last_snapc && snapc != last_snapc) {
		/* if we switched to a newer snapc, restart our scan at the
		 * start of the original file range. */
		dout("  snapc differs from last pass, restarting at %lu\n",
		     index);
		index = start;
	}
	last_snapc = snapc;

	while (!done && index <= end) {
		unsigned i;
		int first;
		pgoff_t next;
		int pvec_pages, locked_pages;
778 779
		struct page **pages = NULL;
		mempool_t *pool = NULL;	/* Becomes non-null if mempool used */
S
Sage Weil 已提交
780 781 782
		struct page *page;
		int want;
		u64 offset, len;
Y
Yehuda Sadeh 已提交
783
		long writeback_stat;
S
Sage Weil 已提交
784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830

		next = 0;
		locked_pages = 0;
		max_pages = max_pages_ever;

get_more_pages:
		first = -1;
		want = min(end - index,
			   min((pgoff_t)PAGEVEC_SIZE,
			       max_pages - (pgoff_t)locked_pages) - 1)
			+ 1;
		pvec_pages = pagevec_lookup_tag(&pvec, mapping, &index,
						PAGECACHE_TAG_DIRTY,
						want);
		dout("pagevec_lookup_tag got %d\n", pvec_pages);
		if (!pvec_pages && !locked_pages)
			break;
		for (i = 0; i < pvec_pages && locked_pages < max_pages; i++) {
			page = pvec.pages[i];
			dout("? %p idx %lu\n", page, page->index);
			if (locked_pages == 0)
				lock_page(page);  /* first page */
			else if (!trylock_page(page))
				break;

			/* only dirty pages, or our accounting breaks */
			if (unlikely(!PageDirty(page)) ||
			    unlikely(page->mapping != mapping)) {
				dout("!dirty or !mapping %p\n", page);
				unlock_page(page);
				break;
			}
			if (!wbc->range_cyclic && page->index > end) {
				dout("end of range %p\n", page);
				done = 1;
				unlock_page(page);
				break;
			}
			if (next && (page->index != next)) {
				dout("not consecutive %p\n", page);
				unlock_page(page);
				break;
			}
			if (wbc->sync_mode != WB_SYNC_NONE) {
				dout("waiting on writeback %p\n", page);
				wait_on_page_writeback(page);
			}
831 832
			if (page_offset(page) >= snap_size) {
				dout("%p page eof %llu\n", page, snap_size);
S
Sage Weil 已提交
833 834 835 836 837 838 839 840 841 842 843
				done = 1;
				unlock_page(page);
				break;
			}
			if (PageWriteback(page)) {
				dout("%p under writeback\n", page);
				unlock_page(page);
				break;
			}

			/* only if matching snap context */
844
			pgsnapc = page_snap_context(page);
845 846 847
			if (pgsnapc->seq > snapc->seq) {
				dout("page snapc %p %lld > oldest %p %lld\n",
				     pgsnapc, pgsnapc->seq, snapc, snapc->seq);
S
Sage Weil 已提交
848 849 850 851 852 853 854 855 856 857 858 859
				unlock_page(page);
				if (!locked_pages)
					continue; /* keep looking for snap */
				break;
			}

			if (!clear_page_dirty_for_io(page)) {
				dout("%p !clear_page_dirty_for_io\n", page);
				unlock_page(page);
				break;
			}

860 861 862 863 864 865
			/*
			 * We have something to write.  If this is
			 * the first locked page this time through,
			 * allocate an osd request and a page array
			 * that it will use.
			 */
S
Sage Weil 已提交
866
			if (locked_pages == 0) {
867
				BUG_ON(pages);
S
Sage Weil 已提交
868
				/* prepare async write request */
869
				offset = (u64)page_offset(page);
S
Sage Weil 已提交
870
				len = wsize;
871 872
				req = ceph_osdc_new_request(&fsc->client->osdc,
							&ci->i_layout, vino,
873 874
							offset, &len, 0,
							do_sync ? 2 : 1,
875 876 877 878 879
							CEPH_OSD_OP_WRITE,
							CEPH_OSD_FLAG_WRITE |
							CEPH_OSD_FLAG_ONDISK,
							snapc, truncate_seq,
							truncate_size, true);
880 881
				if (IS_ERR(req)) {
					rc = PTR_ERR(req);
882 883 884 885
					unlock_page(page);
					break;
				}

886
				if (do_sync)
887 888
					osd_req_op_init(req, 1,
							CEPH_OSD_OP_STARTSYNC, 0);
889

A
Alex Elder 已提交
890 891 892 893
				req->r_callback = writepages_finish;
				req->r_inode = inode;

				max_pages = calc_pages_for(0, (u64)len);
894 895
				pages = kmalloc(max_pages * sizeof (*pages),
						GFP_NOFS);
A
Alex Elder 已提交
896 897 898
				if (!pages) {
					pool = fsc->wb_pagevec_pool;
					pages = mempool_alloc(pool, GFP_NOFS);
899
					BUG_ON(!pages);
A
Alex Elder 已提交
900
				}
S
Sage Weil 已提交
901 902 903 904 905 906 907
			}

			/* note position of first page in pvec */
			if (first < 0)
				first = i;
			dout("%p will write page %p idx %lu\n",
			     inode, page, page->index);
Y
Yehuda Sadeh 已提交
908

S
Sage Weil 已提交
909
			writeback_stat =
910
			       atomic_long_inc_return(&fsc->writeback_count);
S
Sage Weil 已提交
911
			if (writeback_stat > CONGESTION_ON_THRESH(
912 913
				    fsc->mount_options->congestion_kb)) {
				set_bdi_congested(&fsc->backing_dev_info,
S
Sage Weil 已提交
914
						  BLK_RW_ASYNC);
Y
Yehuda Sadeh 已提交
915 916
			}

S
Sage Weil 已提交
917
			set_page_writeback(page);
918
			pages[locked_pages] = page;
S
Sage Weil 已提交
919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946
			locked_pages++;
			next = page->index + 1;
		}

		/* did we get anything? */
		if (!locked_pages)
			goto release_pvec_pages;
		if (i) {
			int j;
			BUG_ON(!locked_pages || first < 0);

			if (pvec_pages && i == pvec_pages &&
			    locked_pages < max_pages) {
				dout("reached end pvec, trying for more\n");
				pagevec_reinit(&pvec);
				goto get_more_pages;
			}

			/* shift unused pages over in the pvec...  we
			 * will need to release them below. */
			for (j = i; j < pvec_pages; j++) {
				dout(" pvec leftover page %p\n",
				     pvec.pages[j]);
				pvec.pages[j-i+first] = pvec.pages[j];
			}
			pvec.nr -= i-first;
		}

947 948 949
		/* Format the osd request message and submit the write */

		offset = page_offset(pages[0]);
950
		len = min(snap_size - offset,
S
Sage Weil 已提交
951 952 953 954
			  (u64)locked_pages << PAGE_CACHE_SHIFT);
		dout("writepages got %d pages at %llu~%llu\n",
		     locked_pages, offset, len);

955
		osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0,
956
							!!pool, false);
957 958 959 960 961 962

		pages = NULL;	/* request message now owns the pages array */
		pool = NULL;

		/* Update the write op length in case we changed it */

963
		osd_req_op_extent_update(req, 0, len);
964 965

		vino = ceph_vino(inode);
966 967
		ceph_osdc_build_request(req, offset, snapc, vino.snap,
					&inode->i_mtime);
S
Sage Weil 已提交
968

969 970
		rc = ceph_osdc_start_request(&fsc->client->osdc, req, true);
		BUG_ON(rc);
S
Sage Weil 已提交
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015
		req = NULL;

		/* continue? */
		index = next;
		wbc->nr_to_write -= locked_pages;
		if (wbc->nr_to_write <= 0)
			done = 1;

release_pvec_pages:
		dout("pagevec_release on %d pages (%p)\n", (int)pvec.nr,
		     pvec.nr ? pvec.pages[0] : NULL);
		pagevec_release(&pvec);

		if (locked_pages && !done)
			goto retry;
	}

	if (should_loop && !done) {
		/* more to do; loop back to beginning of file */
		dout("writepages looping back to beginning of file\n");
		should_loop = 0;
		index = 0;
		goto retry;
	}

	if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
		mapping->writeback_index = index;

out:
	if (req)
		ceph_osdc_put_request(req);
	ceph_put_snap_context(snapc);
	dout("writepages done, rc = %d\n", rc);
	return rc;
}



/*
 * See if a given @snapc is either writeable, or already written.
 */
static int context_is_writeable_or_written(struct inode *inode,
					   struct ceph_snap_context *snapc)
{
	struct ceph_snap_context *oldest = get_oldest_context(inode, NULL);
1016 1017 1018 1019
	int ret = !oldest || snapc->seq <= oldest->seq;

	ceph_put_snap_context(oldest);
	return ret;
S
Sage Weil 已提交
1020 1021 1022 1023 1024
}

/*
 * We are only allowed to write into/dirty the page if the page is
 * clean, or already dirty within the same snap context.
1025 1026 1027 1028
 *
 * called with page locked.
 * return success with page locked,
 * or any failure (incl -EAGAIN) with page unlocked.
S
Sage Weil 已提交
1029
 */
1030 1031 1032
static int ceph_update_writeable_page(struct file *file,
			    loff_t pos, unsigned len,
			    struct page *page)
S
Sage Weil 已提交
1033
{
A
Al Viro 已提交
1034
	struct inode *inode = file_inode(file);
S
Sage Weil 已提交
1035
	struct ceph_inode_info *ci = ceph_inode(inode);
1036
	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
S
Sage Weil 已提交
1037 1038 1039 1040 1041
	loff_t page_off = pos & PAGE_CACHE_MASK;
	int pos_in_page = pos & ~PAGE_CACHE_MASK;
	int end_in_page = pos_in_page + len;
	loff_t i_size;
	int r;
1042
	struct ceph_snap_context *snapc, *oldest;
S
Sage Weil 已提交
1043 1044 1045 1046 1047 1048 1049 1050 1051

retry_locked:
	/* writepages currently holds page lock, but if we change that later, */
	wait_on_page_writeback(page);

	/* check snap context */
	BUG_ON(!ci->i_snap_realm);
	down_read(&mdsc->snap_rwsem);
	BUG_ON(!ci->i_snap_realm->cached_context);
1052
	snapc = page_snap_context(page);
1053
	if (snapc && snapc != ci->i_head_snapc) {
S
Sage Weil 已提交
1054 1055 1056 1057
		/*
		 * this page is already dirty in another (older) snap
		 * context!  is it writeable now?
		 */
1058
		oldest = get_oldest_context(inode, NULL);
S
Sage Weil 已提交
1059 1060
		up_read(&mdsc->snap_rwsem);

1061
		if (snapc->seq > oldest->seq) {
1062
			ceph_put_snap_context(oldest);
S
Sage Weil 已提交
1063
			dout(" page %p snapc %p not current or oldest\n",
1064
			     page, snapc);
S
Sage Weil 已提交
1065 1066 1067 1068
			/*
			 * queue for writeback, and wait for snapc to
			 * be writeable or written
			 */
1069
			snapc = ceph_get_snap_context(snapc);
S
Sage Weil 已提交
1070
			unlock_page(page);
1071
			ceph_queue_writeback(inode);
1072
			r = wait_event_interruptible(ci->i_cap_wq,
S
Sage Weil 已提交
1073 1074
			       context_is_writeable_or_written(inode, snapc));
			ceph_put_snap_context(snapc);
1075 1076
			if (r == -ERESTARTSYS)
				return r;
1077
			return -EAGAIN;
S
Sage Weil 已提交
1078
		}
1079
		ceph_put_snap_context(oldest);
S
Sage Weil 已提交
1080 1081 1082 1083 1084 1085 1086 1087 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 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125

		/* yay, writeable, do it now (without dropping page lock) */
		dout(" page %p snapc %p not current, but oldest\n",
		     page, snapc);
		if (!clear_page_dirty_for_io(page))
			goto retry_locked;
		r = writepage_nounlock(page, NULL);
		if (r < 0)
			goto fail_nosnap;
		goto retry_locked;
	}

	if (PageUptodate(page)) {
		dout(" page %p already uptodate\n", page);
		return 0;
	}

	/* full page? */
	if (pos_in_page == 0 && len == PAGE_CACHE_SIZE)
		return 0;

	/* past end of file? */
	i_size = inode->i_size;   /* caller holds i_mutex */

	if (page_off >= i_size ||
	    (pos_in_page == 0 && (pos+len) >= i_size &&
	     end_in_page - pos_in_page != PAGE_CACHE_SIZE)) {
		dout(" zeroing %p 0 - %d and %d - %d\n",
		     page, pos_in_page, end_in_page, (int)PAGE_CACHE_SIZE);
		zero_user_segments(page,
				   0, pos_in_page,
				   end_in_page, PAGE_CACHE_SIZE);
		return 0;
	}

	/* we need to read it. */
	up_read(&mdsc->snap_rwsem);
	r = readpage_nounlock(file, page);
	if (r < 0)
		goto fail_nosnap;
	goto retry_locked;
fail_nosnap:
	unlock_page(page);
	return r;
}

1126 1127 1128 1129 1130 1131 1132 1133
/*
 * We are only allowed to write into/dirty the page if the page is
 * clean, or already dirty within the same snap context.
 */
static int ceph_write_begin(struct file *file, struct address_space *mapping,
			    loff_t pos, unsigned len, unsigned flags,
			    struct page **pagep, void **fsdata)
{
A
Al Viro 已提交
1134
	struct inode *inode = file_inode(file);
1135 1136
	struct page *page;
	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
S
Sage Weil 已提交
1137
	int r;
1138 1139

	do {
1140
		/* get a page */
1141
		page = grab_cache_page_write_begin(mapping, index, 0);
S
Sage Weil 已提交
1142 1143 1144
		if (!page)
			return -ENOMEM;
		*pagep = page;
1145 1146

		dout("write_begin file %p inode %p page %p %d~%d\n", file,
S
Sage Weil 已提交
1147
		     inode, page, (int)pos, (int)len);
1148 1149

		r = ceph_update_writeable_page(file, pos, len, page);
1150 1151 1152 1153
		if (r < 0)
			page_cache_release(page);
		else
			*pagep = page;
1154 1155 1156 1157 1158
	} while (r == -EAGAIN);

	return r;
}

S
Sage Weil 已提交
1159 1160 1161 1162 1163 1164 1165 1166 1167
/*
 * we don't do anything in here that simple_write_end doesn't do
 * except adjust dirty page accounting and drop read lock on
 * mdsc->snap_rwsem.
 */
static int ceph_write_end(struct file *file, struct address_space *mapping,
			  loff_t pos, unsigned len, unsigned copied,
			  struct page *page, void *fsdata)
{
A
Al Viro 已提交
1168
	struct inode *inode = file_inode(file);
1169 1170
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205
	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
	int check_cap = 0;

	dout("write_end file %p inode %p page %p %d~%d (%d)\n", file,
	     inode, page, (int)pos, (int)copied, (int)len);

	/* zero the stale part of the page if we did a short copy */
	if (copied < len)
		zero_user_segment(page, from+copied, len);

	/* did file size increase? */
	/* (no need for i_size_read(); we caller holds i_mutex */
	if (pos+copied > inode->i_size)
		check_cap = ceph_inode_set_size(inode, pos+copied);

	if (!PageUptodate(page))
		SetPageUptodate(page);

	set_page_dirty(page);

	unlock_page(page);
	up_read(&mdsc->snap_rwsem);
	page_cache_release(page);

	if (check_cap)
		ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY, NULL);

	return copied;
}

/*
 * we set .direct_IO to indicate direct io is supported, but since we
 * intercept O_DIRECT reads and writes early, this function should
 * never get called.
 */
1206
static ssize_t ceph_direct_io(struct kiocb *iocb, struct iov_iter *iter,
A
Al Viro 已提交
1207
			      loff_t pos)
S
Sage Weil 已提交
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
{
	WARN_ON(1);
	return -EINVAL;
}

const struct address_space_operations ceph_aops = {
	.readpage = ceph_readpage,
	.readpages = ceph_readpages,
	.writepage = ceph_writepage,
	.writepages = ceph_writepages_start,
	.write_begin = ceph_write_begin,
	.write_end = ceph_write_end,
	.set_page_dirty = ceph_set_page_dirty,
	.invalidatepage = ceph_invalidatepage,
	.releasepage = ceph_releasepage,
	.direct_IO = ceph_direct_io,
};


/*
 * vm ops
 */
1230 1231 1232 1233 1234
static int ceph_filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
{
	struct inode *inode = file_inode(vma->vm_file);
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_file_info *fi = vma->vm_file->private_data;
1235
	struct page *pinned_page = NULL;
1236 1237 1238 1239
	loff_t off = vmf->pgoff << PAGE_CACHE_SHIFT;
	int want, got, ret;

	dout("filemap_fault %p %llx.%llx %llu~%zd trying to get caps\n",
1240
	     inode, ceph_vinop(inode), off, (size_t)PAGE_CACHE_SIZE);
1241 1242 1243 1244 1245 1246
	if (fi->fmode & CEPH_FILE_MODE_LAZY)
		want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO;
	else
		want = CEPH_CAP_FILE_CACHE;
	while (1) {
		got = 0;
Y
Yan, Zheng 已提交
1247 1248
		ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want,
				    -1, &got, &pinned_page);
1249 1250 1251 1252 1253 1254 1255 1256
		if (ret == 0)
			break;
		if (ret != -ERESTARTSYS) {
			WARN_ON(1);
			return VM_FAULT_SIGBUS;
		}
	}
	dout("filemap_fault %p %llu~%zd got cap refs on %s\n",
1257
	     inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got));
1258

Y
Yan, Zheng 已提交
1259 1260 1261 1262 1263
	if ((got & (CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO)) ||
	    ci->i_inline_version == CEPH_INLINE_NONE)
		ret = filemap_fault(vma, vmf);
	else
		ret = -EAGAIN;
1264 1265

	dout("filemap_fault %p %llu~%zd dropping cap refs on %s ret %d\n",
1266
	     inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got), ret);
1267 1268
	if (pinned_page)
		page_cache_release(pinned_page);
1269 1270
	ceph_put_cap_refs(ci, got);

Y
Yan, Zheng 已提交
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	if (ret != -EAGAIN)
		return ret;

	/* read inline data */
	if (off >= PAGE_CACHE_SIZE) {
		/* does not support inline data > PAGE_SIZE */
		ret = VM_FAULT_SIGBUS;
	} else {
		int ret1;
		struct address_space *mapping = inode->i_mapping;
		struct page *page = find_or_create_page(mapping, 0,
						mapping_gfp_mask(mapping) &
						~__GFP_FS);
		if (!page) {
			ret = VM_FAULT_OOM;
			goto out;
		}
		ret1 = __ceph_do_getattr(inode, page,
					 CEPH_STAT_CAP_INLINE_DATA, true);
		if (ret1 < 0 || off >= i_size_read(inode)) {
			unlock_page(page);
			page_cache_release(page);
			ret = VM_FAULT_SIGBUS;
			goto out;
		}
		if (ret1 < PAGE_CACHE_SIZE)
			zero_user_segment(page, ret1, PAGE_CACHE_SIZE);
		else
			flush_dcache_page(page);
		SetPageUptodate(page);
		vmf->page = page;
		ret = VM_FAULT_MAJOR | VM_FAULT_LOCKED;
	}
out:
	dout("filemap_fault %p %llu~%zd read inline data ret %d\n",
	     inode, off, (size_t)PAGE_CACHE_SIZE, ret);
1307 1308
	return ret;
}
S
Sage Weil 已提交
1309 1310 1311 1312 1313 1314

/*
 * Reuse write_begin here for simplicity.
 */
static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
A
Al Viro 已提交
1315
	struct inode *inode = file_inode(vma->vm_file);
1316 1317
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_file_info *fi = vma->vm_file->private_data;
1318
	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
1319
	struct page *page = vmf->page;
1320
	loff_t off = page_offset(page);
1321 1322 1323
	loff_t size = i_size_read(inode);
	size_t len;
	int want, got, ret;
1324

1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337
	if (ci->i_inline_version != CEPH_INLINE_NONE) {
		struct page *locked_page = NULL;
		if (off == 0) {
			lock_page(page);
			locked_page = page;
		}
		ret = ceph_uninline_data(vma->vm_file, locked_page);
		if (locked_page)
			unlock_page(locked_page);
		if (ret < 0)
			return VM_FAULT_SIGBUS;
	}

S
Sage Weil 已提交
1338 1339 1340 1341 1342
	if (off + PAGE_CACHE_SIZE <= size)
		len = PAGE_CACHE_SIZE;
	else
		len = size & ~PAGE_CACHE_MASK;

1343 1344 1345 1346 1347 1348 1349 1350
	dout("page_mkwrite %p %llx.%llx %llu~%zd getting caps i_size %llu\n",
	     inode, ceph_vinop(inode), off, len, size);
	if (fi->fmode & CEPH_FILE_MODE_LAZY)
		want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO;
	else
		want = CEPH_CAP_FILE_BUFFER;
	while (1) {
		got = 0;
1351 1352
		ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, off + len,
				    &got, NULL);
1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
		if (ret == 0)
			break;
		if (ret != -ERESTARTSYS) {
			WARN_ON(1);
			return VM_FAULT_SIGBUS;
		}
	}
	dout("page_mkwrite %p %llu~%zd got cap refs on %s\n",
	     inode, off, len, ceph_cap_string(got));

	/* Update time before taking page lock */
	file_update_time(vma->vm_file);
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375

	lock_page(page);

	ret = VM_FAULT_NOPAGE;
	if ((off > size) ||
	    (page->mapping != inode->i_mapping))
		goto out;

	ret = ceph_update_writeable_page(vma->vm_file, off, len, page);
	if (ret == 0) {
		/* success.  we'll keep the page locked. */
S
Sage Weil 已提交
1376 1377 1378 1379
		set_page_dirty(page);
		up_read(&mdsc->snap_rwsem);
		ret = VM_FAULT_LOCKED;
	} else {
1380 1381 1382 1383
		if (ret == -ENOMEM)
			ret = VM_FAULT_OOM;
		else
			ret = VM_FAULT_SIGBUS;
S
Sage Weil 已提交
1384
	}
1385
out:
1386
	if (ret != VM_FAULT_LOCKED)
1387
		unlock_page(page);
1388 1389
	if (ret == VM_FAULT_LOCKED ||
	    ci->i_inline_version != CEPH_INLINE_NONE) {
1390 1391
		int dirty;
		spin_lock(&ci->i_ceph_lock);
1392
		ci->i_inline_version = CEPH_INLINE_NONE;
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402
		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
		spin_unlock(&ci->i_ceph_lock);
		if (dirty)
			__mark_inode_dirty(inode, dirty);
	}

	dout("page_mkwrite %p %llu~%zd dropping cap refs on %s ret %d\n",
	     inode, off, len, ceph_cap_string(got), ret);
	ceph_put_cap_refs(ci, got);

S
Sage Weil 已提交
1403 1404 1405
	return ret;
}

Y
Yan, Zheng 已提交
1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
void ceph_fill_inline_data(struct inode *inode, struct page *locked_page,
			   char	*data, size_t len)
{
	struct address_space *mapping = inode->i_mapping;
	struct page *page;

	if (locked_page) {
		page = locked_page;
	} else {
		if (i_size_read(inode) == 0)
			return;
		page = find_or_create_page(mapping, 0,
					   mapping_gfp_mask(mapping) & ~__GFP_FS);
		if (!page)
			return;
		if (PageUptodate(page)) {
			unlock_page(page);
			page_cache_release(page);
			return;
		}
	}

1428
	dout("fill_inline_data %p %llx.%llx len %zu locked_page %p\n",
Y
Yan, Zheng 已提交
1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448
	     inode, ceph_vinop(inode), len, locked_page);

	if (len > 0) {
		void *kaddr = kmap_atomic(page);
		memcpy(kaddr, data, len);
		kunmap_atomic(kaddr);
	}

	if (page != locked_page) {
		if (len < PAGE_CACHE_SIZE)
			zero_user_segment(page, len, PAGE_CACHE_SIZE);
		else
			flush_dcache_page(page);

		SetPageUptodate(page);
		unlock_page(page);
		page_cache_release(page);
	}
}

1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512
int ceph_uninline_data(struct file *filp, struct page *locked_page)
{
	struct inode *inode = file_inode(filp);
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
	struct ceph_osd_request *req;
	struct page *page = NULL;
	u64 len, inline_version;
	int err = 0;
	bool from_pagecache = false;

	spin_lock(&ci->i_ceph_lock);
	inline_version = ci->i_inline_version;
	spin_unlock(&ci->i_ceph_lock);

	dout("uninline_data %p %llx.%llx inline_version %llu\n",
	     inode, ceph_vinop(inode), inline_version);

	if (inline_version == 1 || /* initial version, no data */
	    inline_version == CEPH_INLINE_NONE)
		goto out;

	if (locked_page) {
		page = locked_page;
		WARN_ON(!PageUptodate(page));
	} else if (ceph_caps_issued(ci) &
		   (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) {
		page = find_get_page(inode->i_mapping, 0);
		if (page) {
			if (PageUptodate(page)) {
				from_pagecache = true;
				lock_page(page);
			} else {
				page_cache_release(page);
				page = NULL;
			}
		}
	}

	if (page) {
		len = i_size_read(inode);
		if (len > PAGE_CACHE_SIZE)
			len = PAGE_CACHE_SIZE;
	} else {
		page = __page_cache_alloc(GFP_NOFS);
		if (!page) {
			err = -ENOMEM;
			goto out;
		}
		err = __ceph_do_getattr(inode, page,
					CEPH_STAT_CAP_INLINE_DATA, true);
		if (err < 0) {
			/* no inline data */
			if (err == -ENODATA)
				err = 0;
			goto out;
		}
		len = err;
	}

	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
				    ceph_vino(inode), 0, &len, 0, 1,
				    CEPH_OSD_OP_CREATE,
				    CEPH_OSD_FLAG_ONDISK | CEPH_OSD_FLAG_WRITE,
1513
				    ceph_empty_snapc, 0, 0, false);
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530
	if (IS_ERR(req)) {
		err = PTR_ERR(req);
		goto out;
	}

	ceph_osdc_build_request(req, 0, NULL, CEPH_NOSNAP, &inode->i_mtime);
	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
	if (!err)
		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
	ceph_osdc_put_request(req);
	if (err < 0)
		goto out;

	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
				    ceph_vino(inode), 0, &len, 1, 3,
				    CEPH_OSD_OP_WRITE,
				    CEPH_OSD_FLAG_ONDISK | CEPH_OSD_FLAG_WRITE,
1531
				    ceph_empty_snapc,
1532 1533 1534 1535 1536 1537 1538 1539 1540
				    ci->i_truncate_seq, ci->i_truncate_size,
				    false);
	if (IS_ERR(req)) {
		err = PTR_ERR(req);
		goto out;
	}

	osd_req_op_extent_osd_data_pages(req, 1, &page, len, 0, false, false);

Y
Yan, Zheng 已提交
1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561
	{
		__le64 xattr_buf = cpu_to_le64(inline_version);
		err = osd_req_op_xattr_init(req, 0, CEPH_OSD_OP_CMPXATTR,
					    "inline_version", &xattr_buf,
					    sizeof(xattr_buf),
					    CEPH_OSD_CMPXATTR_OP_GT,
					    CEPH_OSD_CMPXATTR_MODE_U64);
		if (err)
			goto out_put;
	}

	{
		char xattr_buf[32];
		int xattr_len = snprintf(xattr_buf, sizeof(xattr_buf),
					 "%llu", inline_version);
		err = osd_req_op_xattr_init(req, 2, CEPH_OSD_OP_SETXATTR,
					    "inline_version",
					    xattr_buf, xattr_len, 0, 0);
		if (err)
			goto out_put;
	}
1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584

	ceph_osdc_build_request(req, 0, NULL, CEPH_NOSNAP, &inode->i_mtime);
	err = ceph_osdc_start_request(&fsc->client->osdc, req, false);
	if (!err)
		err = ceph_osdc_wait_request(&fsc->client->osdc, req);
out_put:
	ceph_osdc_put_request(req);
	if (err == -ECANCELED)
		err = 0;
out:
	if (page && page != locked_page) {
		if (from_pagecache) {
			unlock_page(page);
			page_cache_release(page);
		} else
			__free_pages(page, 0);
	}

	dout("uninline_data %p %llx.%llx inline_version %llu = %d\n",
	     inode, ceph_vinop(inode), inline_version, err);
	return err;
}

S
Sage Weil 已提交
1585
static struct vm_operations_struct ceph_vmops = {
1586
	.fault		= ceph_filemap_fault,
S
Sage Weil 已提交
1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599
	.page_mkwrite	= ceph_page_mkwrite,
};

int ceph_mmap(struct file *file, struct vm_area_struct *vma)
{
	struct address_space *mapping = file->f_mapping;

	if (!mapping->a_ops->readpage)
		return -ENOEXEC;
	file_accessed(file);
	vma->vm_ops = &ceph_vmops;
	return 0;
}
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654

enum {
	POOL_READ	= 1,
	POOL_WRITE	= 2,
};

static int __ceph_pool_perm_get(struct ceph_inode_info *ci, u32 pool)
{
	struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->vfs_inode);
	struct ceph_mds_client *mdsc = fsc->mdsc;
	struct ceph_osd_request *rd_req = NULL, *wr_req = NULL;
	struct rb_node **p, *parent;
	struct ceph_pool_perm *perm;
	struct page **pages;
	int err = 0, err2 = 0, have = 0;

	down_read(&mdsc->pool_perm_rwsem);
	p = &mdsc->pool_perm_tree.rb_node;
	while (*p) {
		perm = rb_entry(*p, struct ceph_pool_perm, node);
		if (pool < perm->pool)
			p = &(*p)->rb_left;
		else if (pool > perm->pool)
			p = &(*p)->rb_right;
		else {
			have = perm->perm;
			break;
		}
	}
	up_read(&mdsc->pool_perm_rwsem);
	if (*p)
		goto out;

	dout("__ceph_pool_perm_get pool %u no perm cached\n", pool);

	down_write(&mdsc->pool_perm_rwsem);
	parent = NULL;
	while (*p) {
		parent = *p;
		perm = rb_entry(parent, struct ceph_pool_perm, node);
		if (pool < perm->pool)
			p = &(*p)->rb_left;
		else if (pool > perm->pool)
			p = &(*p)->rb_right;
		else {
			have = perm->perm;
			break;
		}
	}
	if (*p) {
		up_write(&mdsc->pool_perm_rwsem);
		goto out;
	}

	rd_req = ceph_osdc_alloc_request(&fsc->client->osdc,
1655
					 ceph_empty_snapc,
1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669
					 1, false, GFP_NOFS);
	if (!rd_req) {
		err = -ENOMEM;
		goto out_unlock;
	}

	rd_req->r_flags = CEPH_OSD_FLAG_READ;
	osd_req_op_init(rd_req, 0, CEPH_OSD_OP_STAT, 0);
	rd_req->r_base_oloc.pool = pool;
	snprintf(rd_req->r_base_oid.name, sizeof(rd_req->r_base_oid.name),
		 "%llx.00000000", ci->i_vino.ino);
	rd_req->r_base_oid.name_len = strlen(rd_req->r_base_oid.name);

	wr_req = ceph_osdc_alloc_request(&fsc->client->osdc,
1670
					 ceph_empty_snapc,
1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
					 1, false, GFP_NOFS);
	if (!wr_req) {
		err = -ENOMEM;
		goto out_unlock;
	}

	wr_req->r_flags = CEPH_OSD_FLAG_WRITE |
			  CEPH_OSD_FLAG_ACK | CEPH_OSD_FLAG_ONDISK;
	osd_req_op_init(wr_req, 0, CEPH_OSD_OP_CREATE, CEPH_OSD_OP_FLAG_EXCL);
	wr_req->r_base_oloc.pool = pool;
	wr_req->r_base_oid = rd_req->r_base_oid;

	/* one page should be large enough for STAT data */
	pages = ceph_alloc_page_vector(1, GFP_KERNEL);
	if (IS_ERR(pages)) {
		err = PTR_ERR(pages);
		goto out_unlock;
	}

	osd_req_op_raw_data_in_pages(rd_req, 0, pages, PAGE_SIZE,
				     0, false, true);
	ceph_osdc_build_request(rd_req, 0, NULL, CEPH_NOSNAP,
				&ci->vfs_inode.i_mtime);
	err = ceph_osdc_start_request(&fsc->client->osdc, rd_req, false);

	ceph_osdc_build_request(wr_req, 0, NULL, CEPH_NOSNAP,
				&ci->vfs_inode.i_mtime);
	err2 = ceph_osdc_start_request(&fsc->client->osdc, wr_req, false);

	if (!err)
		err = ceph_osdc_wait_request(&fsc->client->osdc, rd_req);
	if (!err2)
		err2 = ceph_osdc_wait_request(&fsc->client->osdc, wr_req);

	if (err >= 0 || err == -ENOENT)
		have |= POOL_READ;
	else if (err != -EPERM)
		goto out_unlock;

	if (err2 == 0 || err2 == -EEXIST)
		have |= POOL_WRITE;
	else if (err2 != -EPERM) {
		err = err2;
		goto out_unlock;
	}

	perm = kmalloc(sizeof(*perm), GFP_NOFS);
	if (!perm) {
		err = -ENOMEM;
		goto out_unlock;
	}

	perm->pool = pool;
	perm->perm = have;
	rb_link_node(&perm->node, parent, p);
	rb_insert_color(&perm->node, &mdsc->pool_perm_tree);
	err = 0;
out_unlock:
	up_write(&mdsc->pool_perm_rwsem);

	if (rd_req)
		ceph_osdc_put_request(rd_req);
	if (wr_req)
		ceph_osdc_put_request(wr_req);
out:
	if (!err)
		err = have;
	dout("__ceph_pool_perm_get pool %u result = %d\n", pool, err);
	return err;
}

int ceph_pool_perm_check(struct ceph_inode_info *ci, int need)
{
	u32 pool;
	int ret, flags;

	if (ceph_test_mount_opt(ceph_inode_to_client(&ci->vfs_inode),
				NOPOOLPERM))
		return 0;

	spin_lock(&ci->i_ceph_lock);
	flags = ci->i_ceph_flags;
	pool = ceph_file_layout_pg_pool(ci->i_layout);
	spin_unlock(&ci->i_ceph_lock);
check:
	if (flags & CEPH_I_POOL_PERM) {
		if ((need & CEPH_CAP_FILE_RD) && !(flags & CEPH_I_POOL_RD)) {
			dout("ceph_pool_perm_check pool %u no read perm\n",
			     pool);
			return -EPERM;
		}
		if ((need & CEPH_CAP_FILE_WR) && !(flags & CEPH_I_POOL_WR)) {
			dout("ceph_pool_perm_check pool %u no write perm\n",
			     pool);
			return -EPERM;
		}
		return 0;
	}

	ret = __ceph_pool_perm_get(ci, pool);
	if (ret < 0)
		return ret;

	flags = CEPH_I_POOL_PERM;
	if (ret & POOL_READ)
		flags |= CEPH_I_POOL_RD;
	if (ret & POOL_WRITE)
		flags |= CEPH_I_POOL_WR;

	spin_lock(&ci->i_ceph_lock);
	if (pool == ceph_file_layout_pg_pool(ci->i_layout)) {
		ci->i_ceph_flags = flags;
        } else {
		pool = ceph_file_layout_pg_pool(ci->i_layout);
		flags = ci->i_ceph_flags;
	}
	spin_unlock(&ci->i_ceph_lock);
	goto check;
}

void ceph_pool_perm_destroy(struct ceph_mds_client *mdsc)
{
	struct ceph_pool_perm *perm;
	struct rb_node *n;

	while (!RB_EMPTY_ROOT(&mdsc->pool_perm_tree)) {
		n = rb_first(&mdsc->pool_perm_tree);
		perm = rb_entry(n, struct ceph_pool_perm, node);
		rb_erase(n, &mdsc->pool_perm_tree);
		kfree(perm);
	}
}