addr.c 33.7 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 14
#include "mds_client.h"
#include <linux/ceph/osd_client.h>
S
Sage Weil 已提交
15 16 17 18 19 20 21 22 23 24 25 26

/*
 * 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
27
 * count dirty pages on the inode.  In the absence of snapshots,
S
Sage Weil 已提交
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
 * 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
52 53 54 55 56
#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))

57 58 59 60 61 62
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
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

/*
 * 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;
	int undo = 0;
	struct ceph_snap_context *snapc;

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

	if (TestSetPageDirty(page)) {
		dout("%p set_page_dirty %p idx %lu -- already dirty\n",
		     mapping->host, page, page->index);
		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 */
95
	spin_lock(&ci->i_ceph_lock);
96
	if (ci->i_head_snapc == NULL)
S
Sage Weil 已提交
97 98 99
		ci->i_head_snapc = ceph_get_snap_context(snapc);
	++ci->i_wrbuffer_ref_head;
	if (ci->i_wrbuffer_ref == 0)
100
		ihold(inode);
S
Sage Weil 已提交
101 102 103 104 105 106 107
	++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);
108
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
109 110 111 112 113

	/* now adjust page */
	spin_lock_irq(&mapping->tree_lock);
	if (page->mapping) {	/* Race with truncate? */
		WARN_ON_ONCE(!PageUptodate(page));
M
Michael Rubin 已提交
114
		account_page_dirtied(page, page->mapping);
S
Sage Weil 已提交
115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
		radix_tree_tag_set(&mapping->page_tree,
				page_index(page), PAGECACHE_TAG_DIRTY);

		/*
		 * Reference snap context in page->private.  Also set
		 * PagePrivate so that we get invalidatepage callback.
		 */
		page->private = (unsigned long)snapc;
		SetPagePrivate(page);
	} else {
		dout("ANON set_page_dirty %p (raced truncate?)\n", page);
		undo = 1;
	}

	spin_unlock_irq(&mapping->tree_lock);

	if (undo)
		/* whoops, we failed to dirty the page */
		ceph_put_wrbuffer_cap_refs(ci, 1, snapc);

	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);

	BUG_ON(!PageDirty(page));
	return 1;
}

/*
 * 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.
 */
static void ceph_invalidatepage(struct page *page, unsigned long offset)
{
148
	struct inode *inode;
S
Sage Weil 已提交
149
	struct ceph_inode_info *ci;
150
	struct ceph_snap_context *snapc = page_snap_context(page);
S
Sage Weil 已提交
151 152 153 154 155

	BUG_ON(!PageLocked(page));
	BUG_ON(!PagePrivate(page));
	BUG_ON(!page->mapping);

156 157
	inode = page->mapping->host;

S
Sage Weil 已提交
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197
	/*
	 * 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);

	if (offset == 0)
		ClearPageChecked(page);

	ci = ceph_inode(inode);
	if (offset == 0) {
		dout("%p invalidatepage %p idx %lu full dirty page %lu\n",
		     inode, page, page->index, offset);
		ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
		ceph_put_snap_context(snapc);
		page->private = 0;
		ClearPagePrivate(page);
	} else {
		dout("%p invalidatepage %p idx %lu partial dirty page\n",
		     inode, page, page->index);
	}
}

/* just a sanity check */
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));
	WARN_ON(PagePrivate(page));
	return 0;
}

/*
 * read a single page, without unlocking it.
 */
static int readpage_nounlock(struct file *filp, struct page *page)
{
A
Al Viro 已提交
198
	struct inode *inode = file_inode(filp);
S
Sage Weil 已提交
199
	struct ceph_inode_info *ci = ceph_inode(inode);
200 201
	struct ceph_osd_client *osdc = 
		&ceph_inode_to_client(inode)->client->osdc;
S
Sage Weil 已提交
202 203 204 205 206 207
	int err = 0;
	u64 len = PAGE_CACHE_SIZE;

	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,
208
				  (u64) page_offset(page), &len,
S
Sage Weil 已提交
209
				  ci->i_truncate_seq, ci->i_truncate_size,
210
				  &page, 1, 0);
S
Sage Weil 已提交
211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
	if (err == -ENOENT)
		err = 0;
	if (err < 0) {
		SetPageError(page);
		goto out;
	} else if (err < PAGE_CACHE_SIZE) {
		/* zero fill remainder of page */
		zero_user_segment(page, err, PAGE_CACHE_SIZE);
	}
	SetPageUptodate(page);

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
234
 * Finish an async read(ahead) op.
S
Sage Weil 已提交
235
 */
S
Sage Weil 已提交
236
static void finish_read(struct ceph_osd_request *req, struct ceph_msg *msg)
S
Sage Weil 已提交
237
{
S
Sage Weil 已提交
238
	struct inode *inode = req->r_inode;
239
	struct ceph_osd_data *osd_data;
240 241
	int rc = req->r_result;
	int bytes = le32_to_cpu(msg->hdr.data_len);
242
	int num_pages;
S
Sage Weil 已提交
243
	int i;
S
Sage Weil 已提交
244

S
Sage Weil 已提交
245 246 247
	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 */
248
	osd_data = osd_req_op_extent_osd_data(req, 0);
249 250 251
	BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES);
	num_pages = calc_pages_for((u64)osd_data->alignment,
					(u64)osd_data->length);
252
	for (i = 0; i < num_pages; i++) {
253
		struct page *page = osd_data->pages[i];
S
Sage Weil 已提交
254 255 256 257 258

		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
259
		}
S
Sage Weil 已提交
260 261 262 263 264 265
 		dout("finish_read %p uptodate %p idx %lu\n", inode, page,
		     page->index);
		flush_dcache_page(page);
		SetPageUptodate(page);
		unlock_page(page);
		page_cache_release(page);
266
		bytes -= PAGE_CACHE_SIZE;
S
Sage Weil 已提交
267
	}
268
	kfree(osd_data->pages);
S
Sage Weil 已提交
269 270
}

271 272 273 274 275 276 277 278
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
279
/*
S
Sage Weil 已提交
280 281
 * 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
282
 */
283
static int start_read(struct inode *inode, struct list_head *page_list, int max)
S
Sage Weil 已提交
284
{
285 286
	struct ceph_osd_client *osdc =
		&ceph_inode_to_client(inode)->client->osdc;
S
Sage Weil 已提交
287 288
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct page *page = list_entry(page_list->prev, struct page, lru);
289
	struct ceph_vino vino;
S
Sage Weil 已提交
290 291
	struct ceph_osd_request *req;
	u64 off;
S
Sage Weil 已提交
292
	u64 len;
S
Sage Weil 已提交
293 294 295 296 297
	int i;
	struct page **pages;
	pgoff_t next_index;
	int nr_pages = 0;
	int ret;
S
Sage Weil 已提交
298

299
	off = (u64) page_offset(page);
S
Sage Weil 已提交
300

S
Sage Weil 已提交
301 302 303 304 305 306 307
	/* 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++;
308 309
		if (max && nr_pages == max)
			break;
S
Sage Weil 已提交
310
	}
S
Sage Weil 已提交
311
	len = nr_pages << PAGE_CACHE_SHIFT;
S
Sage Weil 已提交
312 313
	dout("start_read %p nr_pages %d is %lld~%lld\n", inode, nr_pages,
	     off, len);
314 315
	vino = ceph_vino(inode);
	req = ceph_osdc_new_request(osdc, &ci->i_layout, vino, off, &len,
316
				    1, CEPH_OSD_OP_READ,
317
				    CEPH_OSD_FLAG_READ, NULL,
S
Sage Weil 已提交
318
				    ci->i_truncate_seq, ci->i_truncate_size,
319
				    false);
320 321
	if (IS_ERR(req))
		return PTR_ERR(req);
S
Sage Weil 已提交
322

S
Sage Weil 已提交
323
	/* build page vector */
324
	nr_pages = calc_pages_for(0, len);
S
Sage Weil 已提交
325 326 327 328 329 330 331
	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
332
		list_del(&page->lru);
S
Sage Weil 已提交
333 334 335 336
		
 		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
337
					  GFP_NOFS)) {
S
Sage Weil 已提交
338
			page_cache_release(page);
S
Sage Weil 已提交
339
			dout("start_read %p add_to_page_cache failed %p\n",
S
Sage Weil 已提交
340
			     inode, page);
S
Sage Weil 已提交
341 342
			nr_pages = i;
			goto out_pages;
S
Sage Weil 已提交
343
		}
S
Sage Weil 已提交
344
		pages[i] = page;
S
Sage Weil 已提交
345
	}
346
	osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, false, false);
S
Sage Weil 已提交
347 348 349
	req->r_callback = finish_read;
	req->r_inode = inode;

350
	ceph_osdc_build_request(req, off, NULL, vino.snap, NULL);
351

S
Sage Weil 已提交
352 353 354 355 356 357 358 359
	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:
360
	ceph_unlock_page_vector(pages, nr_pages);
S
Sage Weil 已提交
361 362 363 364 365
	ceph_release_page_vector(pages, nr_pages);
out:
	ceph_osdc_put_request(req);
	return ret;
}
S
Sage Weil 已提交
366

S
Sage Weil 已提交
367 368 369 370 371 372 373 374

/*
 * 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
375
	struct inode *inode = file_inode(file);
376
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
S
Sage Weil 已提交
377
	int rc = 0;
378 379 380 381 382
	int max = 0;

	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
383

384 385
	dout("readpages %p file %p nr_pages %d max %d\n", inode,
		file, nr_pages,
386
	     max);
S
Sage Weil 已提交
387
	while (!list_empty(page_list)) {
388
		rc = start_read(inode, page_list, max);
S
Sage Weil 已提交
389 390 391 392
		if (rc < 0)
			goto out;
		BUG_ON(rc == 0);
	}
S
Sage Weil 已提交
393
out:
S
Sage Weil 已提交
394
	dout("readpages %p file %p ret %d\n", inode, file, rc);
S
Sage Weil 已提交
395 396 397 398 399 400 401
	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.
 */
402 403
static struct ceph_snap_context *get_oldest_context(struct inode *inode,
						    u64 *snap_size)
S
Sage Weil 已提交
404 405 406 407 408
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_snap_context *snapc = NULL;
	struct ceph_cap_snap *capsnap = NULL;

409
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
410 411 412 413 414 415 416 417 418 419
	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;
		}
	}
420
	if (!snapc && ci->i_wrbuffer_ref_head) {
421
		snapc = ceph_get_snap_context(ci->i_head_snapc);
S
Sage Weil 已提交
422 423 424
		dout(" head snapc %p has %d dirty pages\n",
		     snapc, ci->i_wrbuffer_ref_head);
	}
425
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
426 427 428 429 430 431 432 433 434 435 436 437 438
	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;
439
	struct ceph_fs_client *fsc;
S
Sage Weil 已提交
440
	struct ceph_osd_client *osdc;
441
	loff_t page_off = page_offset(page);
S
Sage Weil 已提交
442 443 444
	int len = PAGE_CACHE_SIZE;
	loff_t i_size;
	int err = 0;
445
	struct ceph_snap_context *snapc, *oldest;
S
Sage Weil 已提交
446
	u64 snap_size = 0;
Y
Yehuda Sadeh 已提交
447
	long writeback_stat;
S
Sage Weil 已提交
448 449 450 451 452 453 454 455 456

	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);
457 458
	fsc = ceph_inode_to_client(inode);
	osdc = &fsc->client->osdc;
S
Sage Weil 已提交
459 460

	/* verify this is a writeable snap context */
461
	snapc = page_snap_context(page);
S
Sage Weil 已提交
462 463 464 465
	if (snapc == NULL) {
		dout("writepage %p page %p not dirty?\n", inode, page);
		goto out;
	}
466 467
	oldest = get_oldest_context(inode, &snap_size);
	if (snapc->seq > oldest->seq) {
S
Sage Weil 已提交
468
		dout("writepage %p page %p snapc %p not writeable - noop\n",
469
		     inode, page, snapc);
S
Sage Weil 已提交
470 471
		/* we should only noop if called by kswapd */
		WARN_ON((current->flags & PF_MEMALLOC) == 0);
472
		ceph_put_snap_context(oldest);
S
Sage Weil 已提交
473 474
		goto out;
	}
475
	ceph_put_snap_context(oldest);
S
Sage Weil 已提交
476 477 478 479 480 481 482 483 484

	/* is this a partial page at end of file? */
	if (snap_size)
		i_size = snap_size;
	else
		i_size = i_size_read(inode);
	if (i_size < page_off + len)
		len = i_size - page_off;

S
Sage Weil 已提交
485 486
	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
487

488
	writeback_stat = atomic_long_inc_return(&fsc->writeback_count);
Y
Yehuda Sadeh 已提交
489
	if (writeback_stat >
490 491
	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
		set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
Y
Yehuda Sadeh 已提交
492

S
Sage Weil 已提交
493 494 495 496 497
	set_page_writeback(page);
	err = ceph_osdc_writepages(osdc, ceph_vino(inode),
				   &ci->i_layout, snapc,
				   page_off, len,
				   ci->i_truncate_seq, ci->i_truncate_size,
498
				   &inode->i_mtime, &page, 1);
S
Sage Weil 已提交
499 500 501 502 503 504 505 506 507 508 509 510 511 512
	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);
513
	ceph_put_snap_context(snapc);  /* page's reference */
S
Sage Weil 已提交
514 515 516 517 518 519
out:
	return err;
}

static int ceph_writepage(struct page *page, struct writeback_control *wbc)
{
520 521 522
	int err;
	struct inode *inode = page->mapping->host;
	BUG_ON(!inode);
523
	ihold(inode);
524
	err = writepage_nounlock(page, wbc);
S
Sage Weil 已提交
525
	unlock_page(page);
526
	iput(inode);
S
Sage Weil 已提交
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
	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);
560
	struct ceph_osd_data *osd_data;
S
Sage Weil 已提交
561 562
	unsigned wrote;
	struct page *page;
563
	int num_pages;
S
Sage Weil 已提交
564 565 566
	int i;
	struct ceph_snap_context *snapc = req->r_snapc;
	struct address_space *mapping = inode->i_mapping;
567
	int rc = req->r_result;
568
	u64 bytes = req->r_ops[0].extent.length;
569
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
Y
Yehuda Sadeh 已提交
570
	long writeback_stat;
S
Sage Weil 已提交
571
	unsigned issued = ceph_caps_issued(ci);
S
Sage Weil 已提交
572

573
	osd_data = osd_req_op_extent_osd_data(req, 0);
574 575 576
	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
577
	if (rc >= 0) {
578 579 580 581 582 583
		/*
		 * 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.
		 */
584
		wrote = num_pages;
S
Sage Weil 已提交
585 586 587 588 589 590 591 592
	} 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 */
593
	for (i = 0; i < num_pages; i++) {
594
		page = osd_data->pages[i];
S
Sage Weil 已提交
595 596 597
		BUG_ON(!page);
		WARN_ON(!PageUptodate(page));

Y
Yehuda Sadeh 已提交
598
		writeback_stat =
599
			atomic_long_dec_return(&fsc->writeback_count);
Y
Yehuda Sadeh 已提交
600
		if (writeback_stat <
601 602
		    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
			clear_bdi_congested(&fsc->backing_dev_info,
Y
Yehuda Sadeh 已提交
603 604
					    BLK_RW_ASYNC);

605
		ceph_put_snap_context(page_snap_context(page));
S
Sage Weil 已提交
606 607 608 609
		page->private = 0;
		ClearPagePrivate(page);
		dout("unlocking %d %p\n", i, page);
		end_page_writeback(page);
610 611 612 613 614 615 616

		/*
		 * 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
		 */
617
		if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0)
618 619
			generic_error_remove_page(inode->i_mapping, page);

S
Sage Weil 已提交
620 621 622
		unlock_page(page);
	}
	dout("%p wrote+cleaned %d pages\n", inode, wrote);
623
	ceph_put_wrbuffer_cap_refs(ci, num_pages, snapc);
S
Sage Weil 已提交
624

625 626 627
	ceph_release_pages(osd_data->pages, num_pages);
	if (osd_data->pages_from_pool)
		mempool_free(osd_data->pages,
628
			     ceph_sb_to_client(inode->i_sb)->wb_pagevec_pool);
S
Sage Weil 已提交
629
	else
630
		kfree(osd_data->pages);
S
Sage Weil 已提交
631 632 633
	ceph_osdc_put_request(req);
}

634 635
static struct ceph_osd_request *
ceph_writepages_osd_request(struct inode *inode, u64 offset, u64 *len,
636
				struct ceph_snap_context *snapc, int num_ops)
637 638 639 640 641 642 643 644 645 646 647
{
	struct ceph_fs_client *fsc;
	struct ceph_inode_info *ci;
	struct ceph_vino vino;

	fsc = ceph_inode_to_client(inode);
	ci = ceph_inode(inode);
	vino = ceph_vino(inode);
	/* BUG_ON(vino.snap != CEPH_NOSNAP); */

	return ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
648
			vino, offset, len, num_ops, CEPH_OSD_OP_WRITE,
649 650 651 652
			CEPH_OSD_FLAG_WRITE|CEPH_OSD_FLAG_ONDISK,
			snapc, ci->i_truncate_seq, ci->i_truncate_size, true);
}

S
Sage Weil 已提交
653 654 655 656 657 658 659 660
/*
 * 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);
661
	struct ceph_fs_client *fsc;
S
Sage Weil 已提交
662 663 664 665
	pgoff_t index, start, end;
	int range_whole = 0;
	int should_loop = 1;
	pgoff_t max_pages = 0, max_pages_ever = 0;
666
	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
S
Sage Weil 已提交
667 668 669 670 671 672
	struct pagevec pvec;
	int done = 0;
	int rc = 0;
	unsigned wsize = 1 << inode->i_blkbits;
	struct ceph_osd_request *req = NULL;
	int do_sync;
673
	u64 snap_size;
S
Sage Weil 已提交
674 675 676 677 678 679 680 681 682 683 684 685 686 687

	/*
	 * 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.
	 */
	do_sync = wbc->sync_mode == WB_SYNC_ALL;
	if (ceph_caps_revoking(ci, CEPH_CAP_FILE_BUFFER))
		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"));

688 689
	fsc = ceph_inode_to_client(inode);
	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
S
Sage Weil 已提交
690 691 692
		pr_warning("writepage_start %p on forced umount\n", inode);
		return -EIO; /* we're in a forced umount, don't write! */
	}
693 694
	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
		wsize = fsc->mount_options->wsize;
S
Sage Weil 已提交
695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718
	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);
719
	snap_size = 0;
S
Sage Weil 已提交
720 721 722 723 724 725 726
	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;
	}
727 728
	if (snap_size == 0)
		snap_size = i_size_read(inode);
S
Sage Weil 已提交
729 730 731 732 733 734 735 736 737 738 739 740
	dout(" oldest snapc is %p seq %lld (%d snaps)\n",
	     snapc, snapc->seq, snapc->num_snaps);
	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) {
741 742
		int num_ops = do_sync ? 2 : 1;
		struct ceph_vino vino;
S
Sage Weil 已提交
743 744 745 746
		unsigned i;
		int first;
		pgoff_t next;
		int pvec_pages, locked_pages;
747 748
		struct page **pages = NULL;
		mempool_t *pool = NULL;	/* Becomes non-null if mempool used */
S
Sage Weil 已提交
749 750 751
		struct page *page;
		int want;
		u64 offset, len;
Y
Yehuda Sadeh 已提交
752
		long writeback_stat;
S
Sage Weil 已提交
753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799

		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);
			}
800 801
			if (page_offset(page) >= snap_size) {
				dout("%p page eof %llu\n", page, snap_size);
S
Sage Weil 已提交
802 803 804 805 806 807 808 809 810 811 812
				done = 1;
				unlock_page(page);
				break;
			}
			if (PageWriteback(page)) {
				dout("%p under writeback\n", page);
				unlock_page(page);
				break;
			}

			/* only if matching snap context */
813
			pgsnapc = page_snap_context(page);
814 815 816
			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
817 818 819 820 821 822 823 824 825 826 827 828
				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;
			}

829 830 831 832 833 834
			/*
			 * 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
835
			if (locked_pages == 0) {
A
Alex Elder 已提交
836
				size_t size;
837 838

				BUG_ON(pages);
839

S
Sage Weil 已提交
840
				/* prepare async write request */
841
				offset = (u64)page_offset(page);
S
Sage Weil 已提交
842
				len = wsize;
843 844
				req = ceph_writepages_osd_request(inode,
							offset, &len, snapc,
845
							num_ops);
846

847 848
				if (IS_ERR(req)) {
					rc = PTR_ERR(req);
849 850 851 852
					unlock_page(page);
					break;
				}

A
Alex Elder 已提交
853 854 855 856 857 858 859 860 861
				req->r_callback = writepages_finish;
				req->r_inode = inode;

				max_pages = calc_pages_for(0, (u64)len);
				size = max_pages * sizeof (*pages);
				pages = kmalloc(size, GFP_NOFS);
				if (!pages) {
					pool = fsc->wb_pagevec_pool;
					pages = mempool_alloc(pool, GFP_NOFS);
862
					BUG_ON(!pages);
A
Alex Elder 已提交
863
				}
S
Sage Weil 已提交
864 865 866 867 868 869 870
			}

			/* 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
871

S
Sage Weil 已提交
872
			writeback_stat =
873
			       atomic_long_inc_return(&fsc->writeback_count);
S
Sage Weil 已提交
874
			if (writeback_stat > CONGESTION_ON_THRESH(
875 876
				    fsc->mount_options->congestion_kb)) {
				set_bdi_congested(&fsc->backing_dev_info,
S
Sage Weil 已提交
877
						  BLK_RW_ASYNC);
Y
Yehuda Sadeh 已提交
878 879
			}

S
Sage Weil 已提交
880
			set_page_writeback(page);
881
			pages[locked_pages] = page;
S
Sage Weil 已提交
882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909
			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;
		}

910 911 912
		/* Format the osd request message and submit the write */

		offset = page_offset(pages[0]);
913
		len = min(snap_size - offset,
S
Sage Weil 已提交
914 915 916 917
			  (u64)locked_pages << PAGE_CACHE_SHIFT);
		dout("writepages got %d pages at %llu~%llu\n",
		     locked_pages, offset, len);

918
		osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0,
919
							!!pool, false);
920 921 922 923 924 925

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

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

926
		osd_req_op_extent_update(req, 0, len);
927 928

		vino = ceph_vino(inode);
929 930
		ceph_osdc_build_request(req, offset, snapc, vino.snap,
					&inode->i_mtime);
S
Sage Weil 已提交
931

932 933
		rc = ceph_osdc_start_request(&fsc->client->osdc, req, true);
		BUG_ON(rc);
S
Sage Weil 已提交
934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978
		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);
979 980 981 982
	int ret = !oldest || snapc->seq <= oldest->seq;

	ceph_put_snap_context(oldest);
	return ret;
S
Sage Weil 已提交
983 984 985 986 987
}

/*
 * We are only allowed to write into/dirty the page if the page is
 * clean, or already dirty within the same snap context.
988 989 990 991
 *
 * called with page locked.
 * return success with page locked,
 * or any failure (incl -EAGAIN) with page unlocked.
S
Sage Weil 已提交
992
 */
993 994 995
static int ceph_update_writeable_page(struct file *file,
			    loff_t pos, unsigned len,
			    struct page *page)
S
Sage Weil 已提交
996
{
A
Al Viro 已提交
997
	struct inode *inode = file_inode(file);
S
Sage Weil 已提交
998
	struct ceph_inode_info *ci = ceph_inode(inode);
999
	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
S
Sage Weil 已提交
1000 1001 1002 1003 1004
	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;
1005
	struct ceph_snap_context *snapc, *oldest;
S
Sage Weil 已提交
1006 1007 1008 1009 1010 1011 1012 1013 1014

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);
1015
	snapc = page_snap_context(page);
1016
	if (snapc && snapc != ci->i_head_snapc) {
S
Sage Weil 已提交
1017 1018 1019 1020
		/*
		 * this page is already dirty in another (older) snap
		 * context!  is it writeable now?
		 */
1021
		oldest = get_oldest_context(inode, NULL);
S
Sage Weil 已提交
1022 1023
		up_read(&mdsc->snap_rwsem);

1024
		if (snapc->seq > oldest->seq) {
1025
			ceph_put_snap_context(oldest);
S
Sage Weil 已提交
1026
			dout(" page %p snapc %p not current or oldest\n",
1027
			     page, snapc);
S
Sage Weil 已提交
1028 1029 1030 1031
			/*
			 * queue for writeback, and wait for snapc to
			 * be writeable or written
			 */
1032
			snapc = ceph_get_snap_context(snapc);
S
Sage Weil 已提交
1033
			unlock_page(page);
1034
			ceph_queue_writeback(inode);
1035
			r = wait_event_interruptible(ci->i_cap_wq,
S
Sage Weil 已提交
1036 1037
			       context_is_writeable_or_written(inode, snapc));
			ceph_put_snap_context(snapc);
1038 1039
			if (r == -ERESTARTSYS)
				return r;
1040
			return -EAGAIN;
S
Sage Weil 已提交
1041
		}
1042
		ceph_put_snap_context(oldest);
S
Sage Weil 已提交
1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097

		/* 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 (i_size + len > inode->i_sb->s_maxbytes) {
		/* file is too big */
		r = -EINVAL;
		goto fail;
	}

	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:
	up_read(&mdsc->snap_rwsem);
fail_nosnap:
	unlock_page(page);
	return r;
}

1098 1099 1100 1101 1102 1103 1104 1105
/*
 * 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
1106
	struct inode *inode = file_inode(file);
1107 1108
	struct page *page;
	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
S
Sage Weil 已提交
1109
	int r;
1110 1111

	do {
1112
		/* get a page */
1113
		page = grab_cache_page_write_begin(mapping, index, 0);
S
Sage Weil 已提交
1114 1115 1116
		if (!page)
			return -ENOMEM;
		*pagep = page;
1117 1118

		dout("write_begin file %p inode %p page %p %d~%d\n", file,
S
Sage Weil 已提交
1119
		     inode, page, (int)pos, (int)len);
1120 1121 1122 1123 1124 1125 1126

		r = ceph_update_writeable_page(file, pos, len, page);
	} while (r == -EAGAIN);

	return r;
}

S
Sage Weil 已提交
1127 1128 1129 1130 1131 1132 1133 1134 1135
/*
 * 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
1136
	struct inode *inode = file_inode(file);
1137 1138
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 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
	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.
 */
static ssize_t ceph_direct_io(int rw, struct kiocb *iocb,
			      const struct iovec *iov,
			      loff_t pos, unsigned long nr_segs)
{
	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
 */

/*
 * 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
1205
	struct inode *inode = file_inode(vma->vm_file);
S
Sage Weil 已提交
1206
	struct page *page = vmf->page;
1207
	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
1208
	loff_t off = page_offset(page);
S
Sage Weil 已提交
1209 1210 1211
	loff_t size, len;
	int ret;

1212 1213 1214
	/* Update time before taking page lock */
	file_update_time(vma->vm_file);

S
Sage Weil 已提交
1215 1216 1217 1218 1219 1220 1221 1222
	size = i_size_read(inode);
	if (off + PAGE_CACHE_SIZE <= size)
		len = PAGE_CACHE_SIZE;
	else
		len = size & ~PAGE_CACHE_MASK;

	dout("page_mkwrite %p %llu~%llu page %p idx %lu\n", inode,
	     off, len, page, page->index);
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233

	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
1234 1235 1236 1237
		set_page_dirty(page);
		up_read(&mdsc->snap_rwsem);
		ret = VM_FAULT_LOCKED;
	} else {
1238 1239 1240 1241
		if (ret == -ENOMEM)
			ret = VM_FAULT_OOM;
		else
			ret = VM_FAULT_SIGBUS;
S
Sage Weil 已提交
1242
	}
1243
out:
S
Sage Weil 已提交
1244
	dout("page_mkwrite %p %llu~%llu = %d\n", inode, off, len, ret);
1245 1246
	if (ret != VM_FAULT_LOCKED)
		unlock_page(page);
S
Sage Weil 已提交
1247 1248 1249 1250 1251 1252
	return ret;
}

static struct vm_operations_struct ceph_vmops = {
	.fault		= filemap_fault,
	.page_mkwrite	= ceph_page_mkwrite,
1253
	.remap_pages	= generic_file_remap_pages,
S
Sage Weil 已提交
1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265
};

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;
}