Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

Pull fuse updates from Miklos Szeredi:
 "This adds a ->writepage() implementation to fuse, improving mmaped
  writeout and paving the way for buffered writeback.

  And there's a patch to add a fix minor number for /dev/cuse, similarly
  to /dev/fuse"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: writepages: protect secondary requests from fuse file release
  fuse: writepages: update bdi writeout when deleting secondary request
  fuse: writepages: crop secondary requests
  fuse: writepages: roll back changes if request not found
  cuse: add fix minor number to /dev/cuse
  fuse: writepage: skip already in flight
  fuse: writepages: handle same page rewrites
  fuse: writepages: fix aggregation
  fuse: fix race in fuse_writepages()
  fuse: Implement writepages callback
  fuse: don't BUG on no write file
  fuse: lock page in mkwrite
  fuse: Prepare to handle multiple pages in writeback
  fuse: Getting file for writeback helper

5 files changed, 342 insertions, 27 deletions

diff --git a/Documentation/devices.txt b/Documentation/devices.txt
index 23721d3be3e..80b72419ffd 100644
--- a/Documentation/devices.txt
+++ b/Documentation/devices.txt
@@ -414,6 +414,7 @@ Your cooperation is appreciated.
 		200 = /dev/net/tun	TAP/TUN network device
 		201 = /dev/button/gulpb	Transmeta GULP-B buttons
 		202 = /dev/emd/ctl	Enhanced Metadisk RAID (EMD) control
+		203 = /dev/cuse		Cuse (character device in user-space)
 		204 = /dev/video/em8300		EM8300 DVD decoder control
 		205 = /dev/video/em8300_mv	EM8300 DVD decoder video
 		206 = /dev/video/em8300_ma	EM8300 DVD decoder audio
diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c
index adbfd66b380..24da581cb52 100644
--- a/fs/fuse/cuse.c
+++ b/fs/fuse/cuse.c
@@ -589,11 +589,14 @@ static struct attribute *cuse_class_dev_attrs[] = {
 ATTRIBUTE_GROUPS(cuse_class_dev);
 
 static struct miscdevice cuse_miscdev = {
-	.minor		= MISC_DYNAMIC_MINOR,
+	.minor		= CUSE_MINOR,
 	.name		= "cuse",
 	.fops		= &cuse_channel_fops,
 };
 
+MODULE_ALIAS_MISCDEV(CUSE_MINOR);
+MODULE_ALIAS("devname:cuse");
+
 static int __init cuse_init(void)
 {
 	int i, rc;
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 4598345ab87..7e70506297b 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -334,7 +334,8 @@ static bool fuse_page_is_writeback(struct inode *inode, pgoff_t index)
 
 		BUG_ON(req->inode != inode);
 		curr_index = req->misc.write.in.offset >> PAGE_CACHE_SHIFT;
-		if (curr_index == index) {
+		if (curr_index <= index &&
+		    index < curr_index + req->num_pages) {
 			found = true;
 			break;
 		}
@@ -1409,8 +1410,13 @@ static ssize_t fuse_direct_write(struct file *file, const char __user *buf,
 
 static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req)
 {
-	__free_page(req->pages[0]);
-	fuse_file_put(req->ff, false);
+	int i;
+
+	for (i = 0; i < req->num_pages; i++)
+		__free_page(req->pages[i]);
+
+	if (req->ff)
+		fuse_file_put(req->ff, false);
 }
 
 static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req)
@@ -1418,30 +1424,34 @@ static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req)
 	struct inode *inode = req->inode;
 	struct fuse_inode *fi = get_fuse_inode(inode);
 	struct backing_dev_info *bdi = inode->i_mapping->backing_dev_info;
+	int i;
 
 	list_del(&req->writepages_entry);
-	dec_bdi_stat(bdi, BDI_WRITEBACK);
-	dec_zone_page_state(req->pages[0], NR_WRITEBACK_TEMP);
-	bdi_writeout_inc(bdi);
+	for (i = 0; i < req->num_pages; i++) {
+		dec_bdi_stat(bdi, BDI_WRITEBACK);
+		dec_zone_page_state(req->pages[i], NR_WRITEBACK_TEMP);
+		bdi_writeout_inc(bdi);
+	}
 	wake_up(&fi->page_waitq);
 }
 
 /* Called under fc->lock, may release and reacquire it */
-static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req)
+static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req,
+				loff_t size)
 __releases(fc->lock)
 __acquires(fc->lock)
 {
 	struct fuse_inode *fi = get_fuse_inode(req->inode);
-	loff_t size = i_size_read(req->inode);
 	struct fuse_write_in *inarg = &req->misc.write.in;
+	__u64 data_size = req->num_pages * PAGE_CACHE_SIZE;
 
 	if (!fc->connected)
 		goto out_free;
 
-	if (inarg->offset + PAGE_CACHE_SIZE <= size) {
-		inarg->size = PAGE_CACHE_SIZE;
+	if (inarg->offset + data_size <= size) {
+		inarg->size = data_size;
 	} else if (inarg->offset < size) {
-		inarg->size = size & (PAGE_CACHE_SIZE - 1);
+		inarg->size = size - inarg->offset;
 	} else {
 		/* Got truncated off completely */
 		goto out_free;
@@ -1472,12 +1482,13 @@ __acquires(fc->lock)
 {
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_inode *fi = get_fuse_inode(inode);
+	size_t crop = i_size_read(inode);
 	struct fuse_req *req;
 
 	while (fi->writectr >= 0 && !list_empty(&fi->queued_writes)) {
 		req = list_entry(fi->queued_writes.next, struct fuse_req, list);
 		list_del_init(&req->list);
-		fuse_send_writepage(fc, req);
+		fuse_send_writepage(fc, req, crop);
 	}
 }
 
@@ -1488,12 +1499,62 @@ static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_req *req)
 
 	mapping_set_error(inode->i_mapping, req->out.h.error);
 	spin_lock(&fc->lock);
+	while (req->misc.write.next) {
+		struct fuse_conn *fc = get_fuse_conn(inode);
+		struct fuse_write_in *inarg = &req->misc.write.in;
+		struct fuse_req *next = req->misc.write.next;
+		req->misc.write.next = next->misc.write.next;
+		next->misc.write.next = NULL;
+		next->ff = fuse_file_get(req->ff);
+		list_add(&next->writepages_entry, &fi->writepages);
+
+		/*
+		 * Skip fuse_flush_writepages() to make it easy to crop requests
+		 * based on primary request size.
+		 *
+		 * 1st case (trivial): there are no concurrent activities using
+		 * fuse_set/release_nowrite.  Then we're on safe side because
+		 * fuse_flush_writepages() would call fuse_send_writepage()
+		 * anyway.
+		 *
+		 * 2nd case: someone called fuse_set_nowrite and it is waiting
+		 * now for completion of all in-flight requests.  This happens
+		 * rarely and no more than once per page, so this should be
+		 * okay.
+		 *
+		 * 3rd case: someone (e.g. fuse_do_setattr()) is in the middle
+		 * of fuse_set_nowrite..fuse_release_nowrite section.  The fact
+		 * that fuse_set_nowrite returned implies that all in-flight
+		 * requests were completed along with all of their secondary
+		 * requests.  Further primary requests are blocked by negative
+		 * writectr.  Hence there cannot be any in-flight requests and
+		 * no invocations of fuse_writepage_end() while we're in
+		 * fuse_set_nowrite..fuse_release_nowrite section.
+		 */
+		fuse_send_writepage(fc, next, inarg->offset + inarg->size);
+	}
 	fi->writectr--;
 	fuse_writepage_finish(fc, req);
 	spin_unlock(&fc->lock);
 	fuse_writepage_free(fc, req);
 }
 
+static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc,
+					     struct fuse_inode *fi)
+{
+	struct fuse_file *ff = NULL;
+
+	spin_lock(&fc->lock);
+	if (!WARN_ON(list_empty(&fi->write_files))) {
+		ff = list_entry(fi->write_files.next, struct fuse_file,
+				write_entry);
+		fuse_file_get(ff);
+	}
+	spin_unlock(&fc->lock);
+
+	return ff;
+}
+
 static int fuse_writepage_locked(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
@@ -1501,8 +1562,8 @@ static int fuse_writepage_locked(struct page *page)
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_inode *fi = get_fuse_inode(inode);
 	struct fuse_req *req;
-	struct fuse_file *ff;
 	struct page *tmp_page;
+	int error = -ENOMEM;
 
 	set_page_writeback(page);
 
@@ -1515,16 +1576,16 @@ static int fuse_writepage_locked(struct page *page)
 	if (!tmp_page)
 		goto err_free;
 
-	spin_lock(&fc->lock);
-	BUG_ON(list_empty(&fi->write_files));
-	ff = list_entry(fi->write_files.next, struct fuse_file, write_entry);
-	req->ff = fuse_file_get(ff);
-	spin_unlock(&fc->lock);
+	error = -EIO;
+	req->ff = fuse_write_file_get(fc, fi);
+	if (!req->ff)
+		goto err_free;
 
-	fuse_write_fill(req, ff, page_offset(page), 0);
+	fuse_write_fill(req, req->ff, page_offset(page), 0);
 
 	copy_highpage(tmp_page, page);
 	req->misc.write.in.write_flags |= FUSE_WRITE_CACHE;
+	req->misc.write.next = NULL;
 	req->in.argpages = 1;
 	req->num_pages = 1;
 	req->pages[0] = tmp_page;
@@ -1550,19 +1611,263 @@ err_free:
 	fuse_request_free(req);
 err:
 	end_page_writeback(page);
-	return -ENOMEM;
+	return error;
 }
 
 static int fuse_writepage(struct page *page, struct writeback_control *wbc)
 {
 	int err;
 
+	if (fuse_page_is_writeback(page->mapping->host, page->index)) {
+		/*
+		 * ->writepages() should be called for sync() and friends.  We
+		 * should only get here on direct reclaim and then we are
+		 * allowed to skip a page which is already in flight
+		 */
+		WARN_ON(wbc->sync_mode == WB_SYNC_ALL);
+
+		redirty_page_for_writepage(wbc, page);
+		return 0;
+	}
+
 	err = fuse_writepage_locked(page);
 	unlock_page(page);
 
 	return err;
 }
 
+struct fuse_fill_wb_data {
+	struct fuse_req *req;
+	struct fuse_file *ff;
+	struct inode *inode;
+	struct page **orig_pages;
+};
+
+static void fuse_writepages_send(struct fuse_fill_wb_data *data)
+{
+	struct fuse_req *req = data->req;
+	struct inode *inode = data->inode;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_inode *fi = get_fuse_inode(inode);
+	int num_pages = req->num_pages;
+	int i;
+
+	req->ff = fuse_file_get(data->ff);
+	spin_lock(&fc->lock);
+	list_add_tail(&req->list, &fi->queued_writes);
+	fuse_flush_writepages(inode);
+	spin_unlock(&fc->lock);
+
+	for (i = 0; i < num_pages; i++)
+		end_page_writeback(data->orig_pages[i]);
+}
+
+static bool fuse_writepage_in_flight(struct fuse_req *new_req,
+				     struct page *page)
+{
+	struct fuse_conn *fc = get_fuse_conn(new_req->inode);
+	struct fuse_inode *fi = get_fuse_inode(new_req->inode);
+	struct fuse_req *tmp;
+	struct fuse_req *old_req;
+	bool found = false;
+	pgoff_t curr_index;
+
+	BUG_ON(new_req->num_pages != 0);
+
+	spin_lock(&fc->lock);
+	list_del(&new_req->writepages_entry);
+	list_for_each_entry(old_req, &fi->writepages, writepages_entry) {
+		BUG_ON(old_req->inode != new_req->inode);
+		curr_index = old_req->misc.write.in.offset >> PAGE_CACHE_SHIFT;
+		if (curr_index <= page->index &&
+		    page->index < curr_index + old_req->num_pages) {
+			found = true;
+			break;
+		}
+	}
+	if (!found) {
+		list_add(&new_req->writepages_entry, &fi->writepages);
+		goto out_unlock;
+	}
+
+	new_req->num_pages = 1;
+	for (tmp = old_req; tmp != NULL; tmp = tmp->misc.write.next) {
+		BUG_ON(tmp->inode != new_req->inode);
+		curr_index = tmp->misc.write.in.offset >> PAGE_CACHE_SHIFT;
+		if (tmp->num_pages == 1 &&
+		    curr_index == page->index) {
+			old_req = tmp;
+		}
+	}
+
+	if (old_req->num_pages == 1 && (old_req->state == FUSE_REQ_INIT ||
+					old_req->state == FUSE_REQ_PENDING)) {
+		struct backing_dev_info *bdi = page->mapping->backing_dev_info;
+
+		copy_highpage(old_req->pages[0], page);
+		spin_unlock(&fc->lock);
+
+		dec_bdi_stat(bdi, BDI_WRITEBACK);
+		dec_zone_page_state(page, NR_WRITEBACK_TEMP);
+		bdi_writeout_inc(bdi);
+		fuse_writepage_free(fc, new_req);
+		fuse_request_free(new_req);
+		goto out;
+	} else {
+		new_req->misc.write.next = old_req->misc.write.next;
+		old_req->misc.write.next = new_req;
+	}
+out_unlock:
+	spin_unlock(&fc->lock);
+out:
+	return found;
+}
+
+static int fuse_writepages_fill(struct page *page,
+		struct writeback_control *wbc, void *_data)
+{
+	struct fuse_fill_wb_data *data = _data;
+	struct fuse_req *req = data->req;
+	struct inode *inode = data->inode;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct page *tmp_page;
+	bool is_writeback;
+	int err;
+
+	if (!data->ff) {
+		err = -EIO;
+		data->ff = fuse_write_file_get(fc, get_fuse_inode(inode));
+		if (!data->ff)
+			goto out_unlock;
+	}
+
+	/*
+	 * Being under writeback is unlikely but possible.  For example direct
+	 * read to an mmaped fuse file will set the page dirty twice; once when
+	 * the pages are faulted with get_user_pages(), and then after the read
+	 * completed.
+	 */
+	is_writeback = fuse_page_is_writeback(inode, page->index);
+
+	if (req && req->num_pages &&
+	    (is_writeback || req->num_pages == FUSE_MAX_PAGES_PER_REQ ||
+	     (req->num_pages + 1) * PAGE_CACHE_SIZE > fc->max_write ||
+	     data->orig_pages[req->num_pages - 1]->index + 1 != page->index)) {
+		fuse_writepages_send(data);
+		data->req = NULL;
+	}
+	err = -ENOMEM;
+	tmp_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+	if (!tmp_page)
+		goto out_unlock;
+
+	/*
+	 * The page must not be redirtied until the writeout is completed
+	 * (i.e. userspace has sent a reply to the write request).  Otherwise
+	 * there could be more than one temporary page instance for each real
+	 * page.
+	 *
+	 * This is ensured by holding the page lock in page_mkwrite() while
+	 * checking fuse_page_is_writeback().  We already hold the page lock
+	 * since clear_page_dirty_for_io() and keep it held until we add the
+	 * request to the fi->writepages list and increment req->num_pages.
+	 * After this fuse_page_is_writeback() will indicate that the page is
+	 * under writeback, so we can release the page lock.
+	 */
+	if (data->req == NULL) {
+		struct fuse_inode *fi = get_fuse_inode(inode);
+
+		err = -ENOMEM;
+		req = fuse_request_alloc_nofs(FUSE_MAX_PAGES_PER_REQ);
+		if (!req) {
+			__free_page(tmp_page);
+			goto out_unlock;
+		}
+
+		fuse_write_fill(req, data->ff, page_offset(page), 0);
+		req->misc.write.in.write_flags |= FUSE_WRITE_CACHE;
+		req->misc.write.next = NULL;
+		req->in.argpages = 1;
+		req->background = 1;
+		req->num_pages = 0;
+		req->end = fuse_writepage_end;
+		req->inode = inode;
+
+		spin_lock(&fc->lock);
+		list_add(&req->writepages_entry, &fi->writepages);
+		spin_unlock(&fc->lock);
+
+		data->req = req;
+	}
+	set_page_writeback(page);
+
+	copy_highpage(tmp_page, page);
+	req->pages[req->num_pages] = tmp_page;
+	req->page_descs[req->num_pages].offset = 0;
+	req->page_descs[req->num_pages].length = PAGE_SIZE;
+
+	inc_bdi_stat(page->mapping->backing_dev_info, BDI_WRITEBACK);
+	inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP);
+
+	err = 0;
+	if (is_writeback && fuse_writepage_in_flight(req, page)) {
+		end_page_writeback(page);
+		data->req = NULL;
+		goto out_unlock;
+	}
+	data->orig_pages[req->num_pages] = page;
+
+	/*
+	 * Protected by fc->lock against concurrent access by
+	 * fuse_page_is_writeback().
+	 */
+	spin_lock(&fc->lock);
+	req->num_pages++;
+	spin_unlock(&fc->lock);
+
+out_unlock:
+	unlock_page(page);
+
+	return err;
+}
+
+static int fuse_writepages(struct address_space *mapping,
+			   struct writeback_control *wbc)
+{
+	struct inode *inode = mapping->host;
+	struct fuse_fill_wb_data data;
+	int err;
+
+	err = -EIO;
+	if (is_bad_inode(inode))
+		goto out;
+
+	data.inode = inode;
+	data.req = NULL;
+	data.ff = NULL;
+
+	err = -ENOMEM;
+	data.orig_pages = kzalloc(sizeof(struct page *) *
+				  FUSE_MAX_PAGES_PER_REQ,
+				  GFP_NOFS);
+	if (!data.orig_pages)
+		goto out;
+
+	err = write_cache_pages(mapping, wbc, fuse_writepages_fill, &data);
+	if (data.req) {
+		/* Ignore errors if we can write at least one page */
+		BUG_ON(!data.req->num_pages);
+		fuse_writepages_send(&data);
+		err = 0;
+	}
+	if (data.ff)
+		fuse_file_put(data.ff, false);
+
+	kfree(data.orig_pages);
+out:
+	return err;
+}
+
 static int fuse_launder_page(struct page *page)
 {
 	int err = 0;
@@ -1602,14 +1907,17 @@ static void fuse_vma_close(struct vm_area_struct *vma)
 static int fuse_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct page *page = vmf->page;
-	/*
-	 * Don't use page->mapping as it may become NULL from a
-	 * concurrent truncate.
-	 */
-	struct inode *inode = vma->vm_file->f_mapping->host;
+	struct inode *inode = file_inode(vma->vm_file);
+
+	file_update_time(vma->vm_file);
+	lock_page(page);
+	if (page->mapping != inode->i_mapping) {
+		unlock_page(page);
+		return VM_FAULT_NOPAGE;
+	}
 
 	fuse_wait_on_page_writeback(inode, page->index);
-	return 0;
+	return VM_FAULT_LOCKED;
 }
 
 static const struct vm_operations_struct fuse_file_vm_ops = {
@@ -2581,6 +2889,7 @@ static const struct file_operations fuse_direct_io_file_operations = {
 static const struct address_space_operations fuse_file_aops  = {
 	.readpage	= fuse_readpage,
 	.writepage	= fuse_writepage,
+	.writepages	= fuse_writepages,
 	.launder_page	= fuse_launder_page,
 	.readpages	= fuse_readpages,
 	.set_page_dirty	= __set_page_dirty_nobuffers,
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 5b9e6f3b6ae..643274852c8 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -321,6 +321,7 @@ struct fuse_req {
 		struct {
 			struct fuse_write_in in;
 			struct fuse_write_out out;
+			struct fuse_req *next;
 		} write;
 		struct fuse_notify_retrieve_in retrieve_in;
 		struct fuse_lk_in lk_in;
diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h
index cb358355ef4..f7eaf2d6008 100644
--- a/include/linux/miscdevice.h
+++ b/include/linux/miscdevice.h
@@ -31,6 +31,7 @@
 #define I2O_MINOR		166
 #define MICROCODE_MINOR		184
 #define TUN_MINOR		200
+#define CUSE_MINOR		203
 #define MWAVE_MINOR		219	/* ACP/Mwave Modem */
 #define MPT_MINOR		220
 #define MPT2SAS_MINOR		221