1 files changed, 123 insertions, 164 deletions

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 9fee3af83a73..e7af802dcfa6 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -376,8 +376,6 @@ static void cifsFileInfo_put_final(struct cifsFileInfo *cifs_file)
 	struct cifsLockInfo *li, *tmp;
 	struct super_block *sb = inode->i_sb;
 
-	cifs_fscache_release_inode_cookie(inode);
-
 	/*
 	 * Delete any outstanding lock records. We'll lose them when the file
 	 * is closed anyway.
@@ -570,7 +568,7 @@ int cifs_open(struct inode *inode, struct file *file)
 			spin_lock(&CIFS_I(inode)->deferred_lock);
 			cifs_del_deferred_close(cfile);
 			spin_unlock(&CIFS_I(inode)->deferred_lock);
-			goto out;
+			goto use_cache;
 		} else {
 			_cifsFileInfo_put(cfile, true, false);
 		}
@@ -632,8 +630,6 @@ int cifs_open(struct inode *inode, struct file *file)
 		goto out;
 	}
 
-	cifs_fscache_set_inode_cookie(inode, file);
-
 	if ((oplock & CIFS_CREATE_ACTION) && !posix_open_ok && tcon->unix_ext) {
 		/*
 		 * Time to set mode which we can not set earlier due to
@@ -652,6 +648,15 @@ int cifs_open(struct inode *inode, struct file *file)
 				       cfile->pid);
 	}
 
+use_cache:
+	fscache_use_cookie(cifs_inode_cookie(file_inode(file)),
+			   file->f_mode & FMODE_WRITE);
+	if (file->f_flags & O_DIRECT &&
+	    (!((file->f_flags & O_ACCMODE) != O_RDONLY) ||
+	     file->f_flags & O_APPEND))
+		cifs_invalidate_cache(file_inode(file),
+				      FSCACHE_INVAL_DIO_WRITE);
+
 out:
 	free_dentry_path(page);
 	free_xid(xid);
@@ -876,6 +881,8 @@ int cifs_close(struct inode *inode, struct file *file)
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	struct cifs_deferred_close *dclose;
 
+	cifs_fscache_unuse_inode_cookie(inode, file->f_mode & FMODE_WRITE);
+
 	if (file->private_data != NULL) {
 		cfile = file->private_data;
 		file->private_data = NULL;
@@ -886,7 +893,6 @@ int cifs_close(struct inode *inode, struct file *file)
 		    dclose) {
 			if (test_and_clear_bit(CIFS_INO_MODIFIED_ATTR, &cinode->flags)) {
 				inode->i_ctime = inode->i_mtime = current_time(inode);
-				cifs_fscache_update_inode_cookie(inode);
 			}
 			spin_lock(&cinode->deferred_lock);
 			cifs_add_deferred_close(cfile, dclose);
@@ -4198,10 +4204,12 @@ static vm_fault_t
 cifs_page_mkwrite(struct vm_fault *vmf)
 {
 	struct page *page = vmf->page;
-	struct file *file = vmf->vma->vm_file;
-	struct inode *inode = file_inode(file);
 
-	cifs_fscache_wait_on_page_write(inode, page);
+#ifdef CONFIG_CIFS_FSCACHE
+	if (PageFsCache(page) &&
+	    wait_on_page_fscache_killable(page) < 0)
+		return VM_FAULT_RETRY;
+#endif
 
 	lock_page(page);
 	return VM_FAULT_LOCKED;
@@ -4261,8 +4269,6 @@ cifs_readv_complete(struct work_struct *work)
 	for (i = 0; i < rdata->nr_pages; i++) {
 		struct page *page = rdata->pages[i];
 
-		lru_cache_add(page);
-
 		if (rdata->result == 0 ||
 		    (rdata->result == -EAGAIN && got_bytes)) {
 			flush_dcache_page(page);
@@ -4270,13 +4276,11 @@ cifs_readv_complete(struct work_struct *work)
 		} else
 			SetPageError(page);
 
-		unlock_page(page);
-
 		if (rdata->result == 0 ||
 		    (rdata->result == -EAGAIN && got_bytes))
 			cifs_readpage_to_fscache(rdata->mapping->host, page);
-		else
-			cifs_fscache_uncache_page(rdata->mapping->host, page);
+
+		unlock_page(page);
 
 		got_bytes -= min_t(unsigned int, PAGE_SIZE, got_bytes);
 
@@ -4334,7 +4338,6 @@ readpages_fill_pages(struct TCP_Server_Info *server,
 			 * fill them until the writes are flushed.
 			 */
 			zero_user(page, 0, PAGE_SIZE);
-			lru_cache_add(page);
 			flush_dcache_page(page);
 			SetPageUptodate(page);
 			unlock_page(page);
@@ -4344,7 +4347,6 @@ readpages_fill_pages(struct TCP_Server_Info *server,
 			continue;
 		} else {
 			/* no need to hold page hostage */
-			lru_cache_add(page);
 			unlock_page(page);
 			put_page(page);
 			rdata->pages[i] = NULL;
@@ -4387,92 +4389,20 @@ cifs_readpages_copy_into_pages(struct TCP_Server_Info *server,
 	return readpages_fill_pages(server, rdata, iter, iter->count);
 }
 
-static int
-readpages_get_pages(struct address_space *mapping, struct list_head *page_list,
-		    unsigned int rsize, struct list_head *tmplist,
-		    unsigned int *nr_pages, loff_t *offset, unsigned int *bytes)
+static void cifs_readahead(struct readahead_control *ractl)
 {
-	struct page *page, *tpage;
-	unsigned int expected_index;
 	int rc;
-	gfp_t gfp = readahead_gfp_mask(mapping);
-
-	INIT_LIST_HEAD(tmplist);
-
-	page = lru_to_page(page_list);
-
-	/*
-	 * Lock the page and put it in the cache. Since no one else
-	 * should have access to this page, we're safe to simply set
-	 * PG_locked without checking it first.
-	 */
-	__SetPageLocked(page);
-	rc = add_to_page_cache_locked(page, mapping,
-				      page->index, gfp);
-
-	/* give up if we can't stick it in the cache */
-	if (rc) {
-		__ClearPageLocked(page);
-		return rc;
-	}
-
-	/* move first page to the tmplist */
-	*offset = (loff_t)page->index << PAGE_SHIFT;
-	*bytes = PAGE_SIZE;
-	*nr_pages = 1;
-	list_move_tail(&page->lru, tmplist);
-
-	/* now try and add more pages onto the request */
-	expected_index = page->index + 1;
-	list_for_each_entry_safe_reverse(page, tpage, page_list, lru) {
-		/* discontinuity ? */
-		if (page->index != expected_index)
-			break;
-
-		/* would this page push the read over the rsize? */
-		if (*bytes + PAGE_SIZE > rsize)
-			break;
-
-		__SetPageLocked(page);
-		rc = add_to_page_cache_locked(page, mapping, page->index, gfp);
-		if (rc) {
-			__ClearPageLocked(page);
-			break;
-		}
-		list_move_tail(&page->lru, tmplist);
-		(*bytes) += PAGE_SIZE;
-		expected_index++;
-		(*nr_pages)++;
-	}
-	return rc;
-}
-
-static int cifs_readpages(struct file *file, struct address_space *mapping,
-	struct list_head *page_list, unsigned num_pages)
-{
-	int rc;
-	int err = 0;
-	struct list_head tmplist;
-	struct cifsFileInfo *open_file = file->private_data;
-	struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file);
+	struct cifsFileInfo *open_file = ractl->file->private_data;
+	struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(ractl->file);
 	struct TCP_Server_Info *server;
 	pid_t pid;
-	unsigned int xid;
+	unsigned int xid, nr_pages, last_batch_size = 0, cache_nr_pages = 0;
+	pgoff_t next_cached = ULONG_MAX;
+	bool caching = fscache_cookie_enabled(cifs_inode_cookie(ractl->mapping->host)) &&
+		cifs_inode_cookie(ractl->mapping->host)->cache_priv;
+	bool check_cache = caching;
 
 	xid = get_xid();
-	/*
-	 * Reads as many pages as possible from fscache. Returns -ENOBUFS
-	 * immediately if the cookie is negative
-	 *
-	 * After this point, every page in the list might have PG_fscache set,
-	 * so we will need to clean that up off of every page we don't use.
-	 */
-	rc = cifs_readpages_from_fscache(mapping->host, mapping, page_list,
-					 &num_pages);
-	if (rc == 0) {
-		free_xid(xid);
-		return rc;
-	}
 
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
 		pid = open_file->pid;
@@ -4483,39 +4413,73 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 	server = cifs_pick_channel(tlink_tcon(open_file->tlink)->ses);
 
 	cifs_dbg(FYI, "%s: file=%p mapping=%p num_pages=%u\n",
-		 __func__, file, mapping, num_pages);
+		 __func__, ractl->file, ractl->mapping, readahead_count(ractl));
 
 	/*
-	 * Start with the page at end of list and move it to private
-	 * list. Do the same with any following pages until we hit
-	 * the rsize limit, hit an index discontinuity, or run out of
-	 * pages. Issue the async read and then start the loop again
-	 * until the list is empty.
-	 *
-	 * Note that list order is important. The page_list is in
-	 * the order of declining indexes. When we put the pages in
-	 * the rdata->pages, then we want them in increasing order.
+	 * Chop the readahead request up into rsize-sized read requests.
 	 */
-	while (!list_empty(page_list) && !err) {
-		unsigned int i, nr_pages, bytes, rsize;
-		loff_t offset;
-		struct page *page, *tpage;
+	while ((nr_pages = readahead_count(ractl) - last_batch_size)) {
+		unsigned int i, got, rsize;
+		struct page *page;
 		struct cifs_readdata *rdata;
 		struct cifs_credits credits_on_stack;
 		struct cifs_credits *credits = &credits_on_stack;
+		pgoff_t index = readahead_index(ractl) + last_batch_size;
+
+		/*
+		 * Find out if we have anything cached in the range of
+		 * interest, and if so, where the next chunk of cached data is.
+		 */
+		if (caching) {
+			if (check_cache) {
+				rc = cifs_fscache_query_occupancy(
+					ractl->mapping->host, index, nr_pages,
+					&next_cached, &cache_nr_pages);
+				if (rc < 0)
+					caching = false;
+				check_cache = false;
+			}
+
+			if (index == next_cached) {
+				/*
+				 * TODO: Send a whole batch of pages to be read
+				 * by the cache.
+				 */
+				page = readahead_page(ractl);
+				last_batch_size = 1 << thp_order(page);
+				if (cifs_readpage_from_fscache(ractl->mapping->host,
+							       page) < 0) {
+					/*
+					 * TODO: Deal with cache read failure
+					 * here, but for the moment, delegate
+					 * that to readpage.
+					 */
+					caching = false;
+				}
+				unlock_page(page);
+				next_cached++;
+				cache_nr_pages--;
+				if (cache_nr_pages == 0)
+					check_cache = true;
+				continue;
+			}
+		}
 
 		if (open_file->invalidHandle) {
 			rc = cifs_reopen_file(open_file, true);
-			if (rc == -EAGAIN)
-				continue;
-			else if (rc)
+			if (rc) {
+				if (rc == -EAGAIN)
+					continue;
 				break;
+			}
 		}
 
 		rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize,
 						   &rsize, credits);
 		if (rc)
 			break;
+		nr_pages = min_t(size_t, rsize / PAGE_SIZE, readahead_count(ractl));
+		nr_pages = min_t(size_t, nr_pages, next_cached - index);
 
 		/*
 		 * Give up immediately if rsize is too small to read an entire
@@ -4523,16 +4487,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 		 * reach this point however since we set ra_pages to 0 when the
 		 * rsize is smaller than a cache page.
 		 */
-		if (unlikely(rsize < PAGE_SIZE)) {
-			add_credits_and_wake_if(server, credits, 0);
-			free_xid(xid);
-			return 0;
-		}
-
-		nr_pages = 0;
-		err = readpages_get_pages(mapping, page_list, rsize, &tmplist,
-					 &nr_pages, &offset, &bytes);
-		if (!nr_pages) {
+		if (unlikely(!nr_pages)) {
 			add_credits_and_wake_if(server, credits, 0);
 			break;
 		}
@@ -4540,36 +4495,31 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 		rdata = cifs_readdata_alloc(nr_pages, cifs_readv_complete);
 		if (!rdata) {
 			/* best to give up if we're out of mem */
-			list_for_each_entry_safe(page, tpage, &tmplist, lru) {
-				list_del(&page->lru);
-				lru_cache_add(page);
-				unlock_page(page);
-				put_page(page);
-			}
-			rc = -ENOMEM;
 			add_credits_and_wake_if(server, credits, 0);
 			break;
 		}
 
-		rdata->cfile = cifsFileInfo_get(open_file);
-		rdata->server = server;
-		rdata->mapping = mapping;
-		rdata->offset = offset;
-		rdata->bytes = bytes;
-		rdata->pid = pid;
-		rdata->pagesz = PAGE_SIZE;
-		rdata->tailsz = PAGE_SIZE;
+		got = __readahead_batch(ractl, rdata->pages, nr_pages);
+		if (got != nr_pages) {
+			pr_warn("__readahead_batch() returned %u/%u\n",
+				got, nr_pages);
+			nr_pages = got;
+		}
+
+		rdata->nr_pages = nr_pages;
+		rdata->bytes	= readahead_batch_length(ractl);
+		rdata->cfile	= cifsFileInfo_get(open_file);
+		rdata->server	= server;
+		rdata->mapping	= ractl->mapping;
+		rdata->offset	= readahead_pos(ractl);
+		rdata->pid	= pid;
+		rdata->pagesz	= PAGE_SIZE;
+		rdata->tailsz	= PAGE_SIZE;
 		rdata->read_into_pages = cifs_readpages_read_into_pages;
 		rdata->copy_into_pages = cifs_readpages_copy_into_pages;
-		rdata->credits = credits_on_stack;
-
-		list_for_each_entry_safe(page, tpage, &tmplist, lru) {
-			list_del(&page->lru);
-			rdata->pages[rdata->nr_pages++] = page;
-		}
+		rdata->credits	= credits_on_stack;
 
 		rc = adjust_credits(server, &rdata->credits, rdata->bytes);
-
 		if (!rc) {
 			if (rdata->cfile->invalidHandle)
 				rc = -EAGAIN;
@@ -4581,7 +4531,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 			add_credits_and_wake_if(server, &rdata->credits, 0);
 			for (i = 0; i < rdata->nr_pages; i++) {
 				page = rdata->pages[i];
-				lru_cache_add(page);
 				unlock_page(page);
 				put_page(page);
 			}
@@ -4591,15 +4540,10 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 		}
 
 		kref_put(&rdata->refcount, cifs_readdata_release);
+		last_batch_size = nr_pages;
 	}
 
-	/* Any pages that have been shown to fscache but didn't get added to
-	 * the pagecache must be uncached before they get returned to the
-	 * allocator.
-	 */
-	cifs_fscache_readpages_cancel(mapping->host, page_list);
 	free_xid(xid);
-	return rc;
 }
 
 /*
@@ -4801,17 +4745,19 @@ static int cifs_release_page(struct page *page, gfp_t gfp)
 {
 	if (PagePrivate(page))
 		return 0;
-
-	return cifs_fscache_release_page(page, gfp);
+	if (PageFsCache(page)) {
+		if (current_is_kswapd() || !(gfp & __GFP_FS))
+			return false;
+		wait_on_page_fscache(page);
+	}
+	fscache_note_page_release(cifs_inode_cookie(page->mapping->host));
+	return true;
 }
 
 static void cifs_invalidate_page(struct page *page, unsigned int offset,
 				 unsigned int length)
 {
-	struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host);
-
-	if (offset == 0 && length == PAGE_SIZE)
-		cifs_fscache_invalidate_page(page, &cifsi->vfs_inode);
+	wait_on_page_fscache(page);
 }
 
 static int cifs_launder_page(struct page *page)
@@ -4831,7 +4777,7 @@ static int cifs_launder_page(struct page *page)
 	if (clear_page_dirty_for_io(page))
 		rc = cifs_writepage_locked(page, &wbc);
 
-	cifs_fscache_invalidate_page(page, page->mapping->host);
+	wait_on_page_fscache(page);
 	return rc;
 }
 
@@ -4921,7 +4867,7 @@ oplock_break_done:
  * In the non-cached mode (mount with cache=none), we shunt off direct read and write requests
  * so this method should never be called.
  *
- * Direct IO is not yet supported in the cached mode. 
+ * Direct IO is not yet supported in the cached mode.
  */
 static ssize_t
 cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter)
@@ -4988,14 +4934,27 @@ static void cifs_swap_deactivate(struct file *file)
 	/* do we need to unpin (or unlock) the file */
 }
 
+/*
+ * Mark a page as having been made dirty and thus needing writeback.  We also
+ * need to pin the cache object to write back to.
+ */
+#ifdef CONFIG_CIFS_FSCACHE
+static int cifs_set_page_dirty(struct page *page)
+{
+	return fscache_set_page_dirty(page, cifs_inode_cookie(page->mapping->host));
+}
+#else
+#define cifs_set_page_dirty __set_page_dirty_nobuffers
+#endif
+
 const struct address_space_operations cifs_addr_ops = {
 	.readpage = cifs_readpage,
-	.readpages = cifs_readpages,
+	.readahead = cifs_readahead,
 	.writepage = cifs_writepage,
 	.writepages = cifs_writepages,
 	.write_begin = cifs_write_begin,
 	.write_end = cifs_write_end,
-	.set_page_dirty = __set_page_dirty_nobuffers,
+	.set_page_dirty = cifs_set_page_dirty,
 	.releasepage = cifs_release_page,
 	.direct_IO = cifs_direct_io,
 	.invalidatepage = cifs_invalidate_page,
@@ -5020,7 +4979,7 @@ const struct address_space_operations cifs_addr_ops_smallbuf = {
 	.writepages = cifs_writepages,
 	.write_begin = cifs_write_begin,
 	.write_end = cifs_write_end,
-	.set_page_dirty = __set_page_dirty_nobuffers,
+	.set_page_dirty = cifs_set_page_dirty,
 	.releasepage = cifs_release_page,
 	.invalidatepage = cifs_invalidate_page,
 	.launder_page = cifs_launder_page,