Btrfs: Back port to 2.6.18-el kernels

Signed-off-by: Chris Mason <chris.mason@oracle.com>

fs/btrfs/ctree.h

@@ -19,6 +19,7 @@
 #ifndef __BTRFS__
 #define __BTRFS__
 
+#include <linux/version.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
 #include <linux/fs.h>
@@ -334,7 +335,11 @@ struct btrfs_fs_info {
 	struct list_head trans_list;
 	struct list_head hashers;
 	struct list_head dead_roots;
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	struct work_struct trans_work;
+#else
 	struct delayed_work trans_work;
+#endif
 	struct kobject super_kobj;
 	struct completion kobj_unregister;
 	int do_barriers;
@@ -916,6 +921,14 @@ static inline u32 btrfs_level_size(struct btrfs_root *root, int level) {
 	((unsigned long)(btrfs_leaf_data(leaf) + \
 	btrfs_item_offset_nr(leaf, slot)))
 
+static inline struct dentry *fdentry(struct file *file) {
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	return file->f_dentry;
+#else
+	return file->f_path.dentry;
+#endif
+}
+
 /* extent-tree.c */
 u32 btrfs_count_snapshots_in_path(struct btrfs_root *root,
 				  struct btrfs_path *count_path,

fs/btrfs/disk-io.c

@@ -593,8 +593,11 @@ struct btrfs_root *open_ctree(struct super_block *sb)
 	fs_info->do_barriers = 1;
 	fs_info->closing = 0;
 	fs_info->total_pinned = 0;
-
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	INIT_WORK(&fs_info->trans_work, btrfs_transaction_cleaner, fs_info);
+#else
 	INIT_DELAYED_WORK(&fs_info->trans_work, btrfs_transaction_cleaner);
+#endif
 	BTRFS_I(fs_info->btree_inode)->root = tree_root;
 	memset(&BTRFS_I(fs_info->btree_inode)->location, 0,
 	       sizeof(struct btrfs_key));

fs/btrfs/extent_map.c

@@ -1585,8 +1585,18 @@ extent_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
 
 static int submit_one_bio(int rw, struct bio *bio)
 {
+	u64 maxsector;
 	int ret = 0;
+
 	bio_get(bio);
+
+        maxsector = bio->bi_bdev->bd_inode->i_size >> 9;
+	if (maxsector < bio->bi_sector) {
+		printk("sector too large max %Lu got %llu\n", maxsector,
+			(unsigned long long)bio->bi_sector);
+		WARN_ON(1);
+	}
+
 	submit_bio(rw, bio);
 	if (bio_flagged(bio, BIO_EOPNOTSUPP))
 		ret = -EOPNOTSUPP;
@@ -1678,8 +1688,12 @@ static int __extent_read_full_page(struct extent_map_tree *tree,
 
 	while (cur <= end) {
 		if (cur >= last_byte) {
+			char *userpage;
 			iosize = PAGE_CACHE_SIZE - page_offset;
-			zero_user_page(page, page_offset, iosize, KM_USER0);
+			userpage = kmap_atomic(page, KM_USER0);
+			memset(userpage + page_offset, 0, iosize);
+			flush_dcache_page(page);
+			kunmap_atomic(userpage, KM_USER0);
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    GFP_NOFS);
 			unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS);
@@ -1707,7 +1721,12 @@ static int __extent_read_full_page(struct extent_map_tree *tree,
 
 		/* we've found a hole, just zero and go on */
 		if (block_start == EXTENT_MAP_HOLE) {
-			zero_user_page(page, page_offset, iosize, KM_USER0);
+			char *userpage;
+			userpage = kmap_atomic(page, KM_USER0);
+			memset(userpage + page_offset, 0, iosize);
+			flush_dcache_page(page);
+			kunmap_atomic(userpage, KM_USER0);
+
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    GFP_NOFS);
 			unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS);
@@ -1804,9 +1823,14 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
 	}
 
 	if (page->index == end_index) {
+		char *userpage;
+
 		size_t offset = i_size & (PAGE_CACHE_SIZE - 1);
-		zero_user_page(page, offset,
-			       PAGE_CACHE_SIZE - offset, KM_USER0);
+
+		userpage = kmap_atomic(page, KM_USER0);
+		memset(userpage + offset, 0, PAGE_CACHE_SIZE - offset);
+		flush_dcache_page(page);
+		kunmap_atomic(userpage, KM_USER0);
 	}
 
 	set_page_extent_mapped(page);
@@ -1921,6 +1945,129 @@ done:
 	return 0;
 }
 
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+
+/* Taken directly from 2.6.23 for 2.6.18 back port */
+typedef int (*writepage_t)(struct page *page, struct writeback_control *wbc,
+                                void *data);
+
+/**
+ * write_cache_pages - walk the list of dirty pages of the given address space
+ * and write all of them.
+ * @mapping: address space structure to write
+ * @wbc: subtract the number of written pages from *@wbc->nr_to_write
+ * @writepage: function called for each page
+ * @data: data passed to writepage function
+ *
+ * If a page is already under I/O, write_cache_pages() skips it, even
+ * if it's dirty.  This is desirable behaviour for memory-cleaning writeback,
+ * but it is INCORRECT for data-integrity system calls such as fsync().  fsync()
+ * and msync() need to guarantee that all the data which was dirty at the time
+ * the call was made get new I/O started against them.  If wbc->sync_mode is
+ * WB_SYNC_ALL then we were called for data integrity and we must wait for
+ * existing IO to complete.
+ */
+static int write_cache_pages(struct address_space *mapping,
+		      struct writeback_control *wbc, writepage_t writepage,
+		      void *data)
+{
+	struct backing_dev_info *bdi = mapping->backing_dev_info;
+	int ret = 0;
+	int done = 0;
+	struct pagevec pvec;
+	int nr_pages;
+	pgoff_t index;
+	pgoff_t end;		/* Inclusive */
+	int scanned = 0;
+	int range_whole = 0;
+
+	if (wbc->nonblocking && bdi_write_congested(bdi)) {
+		wbc->encountered_congestion = 1;
+		return 0;
+	}
+
+	pagevec_init(&pvec, 0);
+	if (wbc->range_cyclic) {
+		index = mapping->writeback_index; /* Start from prev offset */
+		end = -1;
+	} else {
+		index = 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;
+		scanned = 1;
+	}
+retry:
+	while (!done && (index <= end) &&
+	       (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
+					      PAGECACHE_TAG_DIRTY,
+					      min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1))) {
+		unsigned i;
+
+		scanned = 1;
+		for (i = 0; i < nr_pages; i++) {
+			struct page *page = pvec.pages[i];
+
+			/*
+			 * At this point we hold neither mapping->tree_lock nor
+			 * lock on the page itself: the page may be truncated or
+			 * invalidated (changing page->mapping to NULL), or even
+			 * swizzled back from swapper_space to tmpfs file
+			 * mapping
+			 */
+			lock_page(page);
+
+			if (unlikely(page->mapping != mapping)) {
+				unlock_page(page);
+				continue;
+			}
+
+			if (!wbc->range_cyclic && page->index > end) {
+				done = 1;
+				unlock_page(page);
+				continue;
+			}
+
+			if (wbc->sync_mode != WB_SYNC_NONE)
+				wait_on_page_writeback(page);
+
+			if (PageWriteback(page) ||
+			    !clear_page_dirty_for_io(page)) {
+				unlock_page(page);
+				continue;
+			}
+
+			ret = (*writepage)(page, wbc, data);
+
+			if (unlikely(ret == AOP_WRITEPAGE_ACTIVATE)) {
+				unlock_page(page);
+				ret = 0;
+			}
+			if (ret || (--(wbc->nr_to_write) <= 0))
+				done = 1;
+			if (wbc->nonblocking && bdi_write_congested(bdi)) {
+				wbc->encountered_congestion = 1;
+				done = 1;
+			}
+		}
+		pagevec_release(&pvec);
+		cond_resched();
+	}
+	if (!scanned && !done) {
+		/*
+		 * We hit the last page and there is more work to be done: wrap
+		 * back to the start of the file
+		 */
+		scanned = 1;
+		index = 0;
+		goto retry;
+	}
+	if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
+		mapping->writeback_index = index;
+	return ret;
+}
+#endif
+
 int extent_write_full_page(struct extent_map_tree *tree, struct page *page,
 			  get_extent_t *get_extent,
 			  struct writeback_control *wbc)
@@ -1945,18 +2092,20 @@ int extent_write_full_page(struct extent_map_tree *tree, struct page *page,
 	ret = __extent_writepage(page, wbc, &epd);
 
 	write_cache_pages(mapping, &wbc_writepages, __extent_writepage, &epd);
-	if (epd.bio)
+	if (epd.bio) {
 		submit_one_bio(WRITE, epd.bio);
+	}
 	return ret;
 }
 EXPORT_SYMBOL(extent_write_full_page);
 
+
 int extent_writepages(struct extent_map_tree *tree,
 		      struct address_space *mapping,
 		      get_extent_t *get_extent,
 		      struct writeback_control *wbc)
 {
-	int ret;
+	int ret = 0;
 	struct extent_page_data epd = {
 		.bio = NULL,
 		.tree = tree,
@@ -1964,8 +2113,9 @@ int extent_writepages(struct extent_map_tree *tree,
 	};
 
 	ret = write_cache_pages(mapping, wbc, __extent_writepage, &epd);
-	if (epd.bio)
+	if (epd.bio) {
 		submit_one_bio(WRITE, epd.bio);
+	}
 	return ret;
 }
 EXPORT_SYMBOL(extent_writepages);
@@ -2106,7 +2256,9 @@ int extent_prepare_write(struct extent_map_tree *tree,
 			flush_dcache_page(page);
 			kunmap_atomic(kaddr, KM_USER0);
 		}
-		if (!isnew && !PageUptodate(page) &&
+		if ((em->block_start != EXTENT_MAP_HOLE &&
+		     em->block_start != EXTENT_MAP_INLINE) && 
+		    !isnew && !PageUptodate(page) &&
 		    (block_off_end > to || block_off_start < from) &&
 		    !test_range_bit(tree, block_start, cur_end,
 				    EXTENT_UPTODATE, 1)) {

fs/btrfs/file.c

@@ -231,7 +231,7 @@ static int dirty_and_release_pages(struct btrfs_trans_handle *trans,
 {
 	int err = 0;
 	int i;
-	struct inode *inode = file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(file)->d_inode;
 	struct extent_map *em;
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	u64 hint_byte;
@@ -652,7 +652,7 @@ static int prepare_pages(struct btrfs_root *root,
 {
 	int i;
 	unsigned long index = pos >> PAGE_CACHE_SHIFT;
-	struct inode *inode = file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(file)->d_inode;
 	int err = 0;
 	u64 start_pos;
 
@@ -666,7 +666,11 @@ static int prepare_pages(struct btrfs_root *root,
 			err = -ENOMEM;
 			BUG_ON(1);
 		}
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+		ClearPageDirty(pages[i]);
+#else
 		cancel_dirty_page(pages[i], PAGE_CACHE_SIZE);
+#endif
 		wait_on_page_writeback(pages[i]);
 		set_page_extent_mapped(pages[i]);
 		WARN_ON(!PageLocked(pages[i]));
@@ -682,7 +686,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
 	ssize_t num_written = 0;
 	ssize_t err = 0;
 	int ret = 0;
-	struct inode *inode = file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct page **pages = NULL;
 	int nrptrs;
@@ -707,7 +711,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
 		goto out;
 	if (count == 0)
 		goto out;
-	err = remove_suid(file->f_path.dentry);
+	err = remove_suid(fdentry(file));
 	if (err)
 		goto out;
 	file_update_time(file);
@@ -862,6 +866,9 @@ struct file_operations btrfs_file_operations = {
 	.read		= do_sync_read,
 	.aio_read       = generic_file_aio_read,
 	.splice_read	= generic_file_splice_read,
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	.sendfile	= generic_file_sendfile,
+#endif
 	.write		= btrfs_file_write,
 	.mmap		= btrfs_file_mmap,
 	.open		= generic_file_open,

fs/btrfs/inode.c

@@ -532,7 +532,11 @@ err:
 		dir->i_size -= name_len * 2;
 		dir->i_mtime = dir->i_ctime = CURRENT_TIME;
 		btrfs_update_inode(trans, root, dir);
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+		dentry->d_inode->i_nlink--;
+#else
 		drop_nlink(dentry->d_inode);
+#endif
 		ret = btrfs_update_inode(trans, root, dentry->d_inode);
 		dir->i_sb->s_dirt = 1;
 	}
@@ -1139,7 +1143,7 @@ static unsigned char btrfs_filetype_table[] = {
 
 static int btrfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 {
-	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct inode *inode = filp->f_dentry->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_item *item;
 	struct btrfs_dir_item *di;
@@ -1554,7 +1558,11 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	if (inode->i_nlink == 0)
 		return -ENOENT;
 
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	inode->i_nlink++;
+#else
 	inc_nlink(inode);
+#endif
 	mutex_lock(&root->fs_info->fs_mutex);
 	trans = btrfs_start_transaction(root, 1);
 
@@ -1825,6 +1833,9 @@ insert:
 	if (ret == -EEXIST) {
 		free_extent_map(em);
 		em = NULL;
+		if (0 && failed_insert == 1) {
+			btrfs_drop_extent_cache(inode, start, end);
+		}
 		failed_insert++;
 		if (failed_insert > 5) {
 			printk("failing to insert %Lu %Lu\n", start, end);
@@ -1942,7 +1953,7 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
  */
 int btrfs_page_mkwrite(struct vm_area_struct *vma, struct page *page)
 {
-	struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(vma->vm_file)->d_inode;
 	unsigned long end;
 	loff_t size;
 	int ret = -EINVAL;
@@ -2243,7 +2254,7 @@ static unsigned long force_ra(struct address_space *mapping,
 }
 
 int btrfs_defrag_file(struct file *file) {
-	struct inode *inode = file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(file)->d_inode;
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	struct page *page;
 	unsigned long last_index;
@@ -2329,7 +2340,7 @@ static int btrfs_ioctl_snap_create(struct btrfs_root *root, void __user *arg)
 
 static int btrfs_ioctl_defrag(struct file *file)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
+	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 
 	switch (inode->i_mode & S_IFMT) {
@@ -2350,7 +2361,7 @@ static int btrfs_ioctl_defrag(struct file *file)
 long btrfs_ioctl(struct file *file, unsigned int
 		cmd, unsigned long arg)
 {
-	struct btrfs_root *root = BTRFS_I(file->f_path.dentry->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
 
 	switch (cmd) {
 	case BTRFS_IOC_SNAP_CREATE:

fs/btrfs/super.c

@@ -35,6 +35,7 @@
 #include <linux/compat.h>
 #include <linux/parser.h>
 #include <linux/ctype.h>
+#include <linux/namei.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"

fs/btrfs/sysfs.c

@@ -189,12 +189,29 @@ static struct kset btrfs_kset;
 int btrfs_sysfs_add_super(struct btrfs_fs_info *fs)
 {
 	int error;
+	char *name;
+	char c;
+	int len = strlen(fs->sb->s_id) + 1;
+	int i;
+
+	name = kmalloc(len, GFP_NOFS);
+	if (!name) {
+		error = -ENOMEM;
+		goto fail;
+	}
+
+	for (i = 0; i < len; i++) {
+		c = fs->sb->s_id[i];
+		if (c == '/' || c == '\\')
+			c = '!';
+		name[i] = c;
+	}
+	name[len] = '\0';
 
 	fs->super_kobj.kset = &btrfs_kset;
 	fs->super_kobj.ktype = &btrfs_super_ktype;
 
-	error = kobject_set_name(&fs->super_kobj, "%s",
-				 fs->sb->s_id);
+	error = kobject_set_name(&fs->super_kobj, "%s", name);
 	if (error)
 		goto fail;
 
@@ -202,9 +219,11 @@ int btrfs_sysfs_add_super(struct btrfs_fs_info *fs)
 	if (error)
 		goto fail;
 
+	kfree(name);
 	return 0;
 
 fail:
+	kfree(name);
 	printk(KERN_ERR "btrfs: sysfs creation for super failed\n");
 	return error;
 }

fs/btrfs/transaction.c

@@ -614,12 +614,20 @@ int btrfs_clean_old_snapshots(struct btrfs_root *root)
 	}
 	return 0;
 }
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+void btrfs_transaction_cleaner(void *p)
+#else
 void btrfs_transaction_cleaner(struct work_struct *work)
+#endif
 {
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+	struct btrfs_fs_info *fs_info = p;
+#else
 	struct btrfs_fs_info *fs_info = container_of(work,
 						     struct btrfs_fs_info,
 						     trans_work.work);
 
+#endif
 	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_transaction *cur;
 	struct btrfs_trans_handle *trans;

fs/btrfs/transaction.h

@@ -73,7 +73,12 @@ int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
 int btrfs_commit_tree_roots(struct btrfs_trans_handle *trans,
 			    struct btrfs_root *root);
 
+#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,18)
+void btrfs_transaction_cleaner(void *p);
+#else
 void btrfs_transaction_cleaner(struct work_struct *work);
+#endif
+
 void btrfs_transaction_flush_work(struct btrfs_root *root);
 void btrfs_transaction_queue_work(struct btrfs_root *root, int delay);
 void btrfs_init_transaction_sys(void);