From 784b4e29a26617589edd290dd2919735e190c06e Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Sun, 21 Nov 2010 22:20:49 -0500
Subject: Btrfs: add migrate page for metadata inode

Migrate page will directly call the btrfs btree writepage function,
which isn't actually allowed.

Our writepage assumes that you have locked the extent_buffer and
flagged the block as written.  Without doing these steps, we can
corrupt metadata blocks.

A later commit will remove the btree writepage function since
it is really only safely used internally by btrfs.  We
use writepages for everything else.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/disk-io.c | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index b40dfe48017..a67b98d58c2 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -28,6 +28,7 @@
 #include <linux/freezer.h>
 #include <linux/crc32c.h>
 #include <linux/slab.h>
+#include <linux/migrate.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -355,6 +356,8 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
 	ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE,
 					     btrfs_header_generation(eb));
 	BUG_ON(ret);
+	WARN_ON(!btrfs_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN));
+
 	found_start = btrfs_header_bytenr(eb);
 	if (found_start != start) {
 		WARN_ON(1);
@@ -693,6 +696,26 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 				   __btree_submit_bio_done);
 }
 
+static int btree_migratepage(struct address_space *mapping,
+			struct page *newpage, struct page *page)
+{
+	/*
+	 * we can't safely write a btree page from here,
+	 * we haven't done the locking hook
+	 */
+	if (PageDirty(page))
+		return -EAGAIN;
+	/*
+	 * Buffers may be managed in a filesystem specific way.
+	 * We must have no buffers or drop them.
+	 */
+	if (page_has_private(page) &&
+	    !try_to_release_page(page, GFP_KERNEL))
+		return -EAGAIN;
+
+	return migrate_page(mapping, newpage, page);
+}
+
 static int btree_writepage(struct page *page, struct writeback_control *wbc)
 {
 	struct extent_io_tree *tree;
@@ -707,8 +730,7 @@ static int btree_writepage(struct page *page, struct writeback_control *wbc)
 	}
 
 	redirty_page_for_writepage(wbc, page);
-	eb = btrfs_find_tree_block(root, page_offset(page),
-				      PAGE_CACHE_SIZE);
+	eb = btrfs_find_tree_block(root, page_offset(page), PAGE_CACHE_SIZE);
 	WARN_ON(!eb);
 
 	was_dirty = test_and_set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags);
@@ -799,6 +821,7 @@ static const struct address_space_operations btree_aops = {
 	.releasepage	= btree_releasepage,
 	.invalidatepage = btree_invalidatepage,
 	.sync_page	= block_sync_page,
+	.migratepage	= btree_migratepage,
 };
 
 int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
-- 
cgit v1.2.2


From 0c56fa9662927354255f2f64617d1de61fc03db9 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Mon, 22 Nov 2010 03:01:39 +0000
Subject: btrfs: fix free dip and dip->csums twice

bio_endio() will free dip and dip->csums, so dip and dip->csums twice will
be freed twice. Fix it.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5132c9af888..8c027aa0020 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5731,7 +5731,7 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 
 	ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0);
 	if (ret)
-		goto out_err;
+		goto free_ordered;
 
 	if (write && !skip_sum) {
 		ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
@@ -5740,7 +5740,7 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 				   __btrfs_submit_bio_start_direct_io,
 				   __btrfs_submit_bio_done);
 		if (ret)
-			goto out_err;
+			goto free_ordered;
 		return;
 	} else if (!skip_sum)
 		btrfs_lookup_bio_sums_dio(root, inode, bio,
@@ -5748,11 +5748,8 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 
 	ret = btrfs_map_bio(root, rw, bio, 0, 1);
 	if (ret)
-		goto out_err;
+		goto free_ordered;
 	return;
-out_err:
-	kfree(dip->csums);
-	kfree(dip);
 free_ordered:
 	/*
 	 * If this is a write, we need to clean up the reserved space and kill
-- 
cgit v1.2.2


From 88f794ede7fadd4b63135b94d1561c1f2d5eb5f5 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Mon, 22 Nov 2010 03:02:55 +0000
Subject: btrfs: cleanup duplicate bio allocating functions

extent_bio_alloc() and compressed_bio_alloc() are similar, cleanup
similar source code.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/compression.c | 15 +--------------
 fs/btrfs/extent_io.c   |  8 ++++----
 fs/btrfs/extent_io.h   |  3 +++
 3 files changed, 8 insertions(+), 18 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 7845d1f7d1d..b50bc4bd5c5 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -91,23 +91,10 @@ static inline int compressed_bio_size(struct btrfs_root *root,
 static struct bio *compressed_bio_alloc(struct block_device *bdev,
 					u64 first_byte, gfp_t gfp_flags)
 {
-	struct bio *bio;
 	int nr_vecs;
 
 	nr_vecs = bio_get_nr_vecs(bdev);
-	bio = bio_alloc(gfp_flags, nr_vecs);
-
-	if (bio == NULL && (current->flags & PF_MEMALLOC)) {
-		while (!bio && (nr_vecs /= 2))
-			bio = bio_alloc(gfp_flags, nr_vecs);
-	}
-
-	if (bio) {
-		bio->bi_size = 0;
-		bio->bi_bdev = bdev;
-		bio->bi_sector = first_byte >> 9;
-	}
-	return bio;
+	return btrfs_bio_alloc(bdev, first_byte >> 9, nr_vecs, gfp_flags);
 }
 
 static int check_compressed_csum(struct inode *inode,
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 3b7eaee0f91..f60aa3c35c2 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -1828,9 +1828,9 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err)
 	bio_put(bio);
 }
 
-static struct bio *
-extent_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
-		 gfp_t gfp_flags)
+struct bio *
+btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
+		gfp_t gfp_flags)
 {
 	struct bio *bio;
 
@@ -1919,7 +1919,7 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree,
 	else
 		nr = bio_get_nr_vecs(bdev);
 
-	bio = extent_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH);
+	bio = btrfs_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH);
 
 	bio_add_page(bio, page, page_size, offset);
 	bio->bi_end_io = end_io_func;
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 1c6d4f342ef..4183c8178f0 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -310,4 +310,7 @@ int extent_clear_unlock_delalloc(struct inode *inode,
 				struct extent_io_tree *tree,
 				u64 start, u64 end, struct page *locked_page,
 				unsigned long op);
+struct bio *
+btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
+		gfp_t gfp_flags);
 #endif
-- 
cgit v1.2.2


From e65e1535542931e51189832264cd282e5899e4b9 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Mon, 22 Nov 2010 03:04:43 +0000
Subject: btrfs: fix panic caused by direct IO

btrfs paniced when we write >64KB data by direct IO at one time.

Reproduce steps:
 # mkfs.btrfs /dev/sda5 /dev/sda6
 # mount /dev/sda5 /mnt
 # dd if=/dev/zero of=/mnt/tmpfile bs=100K count=1 oflag=direct

Then btrfs paniced:
mapping failed logical 1103155200 bio len 69632 len 12288
------------[ cut here ]------------
kernel BUG at fs/btrfs/volumes.c:3010!
[SNIP]
Pid: 1992, comm: btrfs-worker-0 Not tainted 2.6.37-rc1 #1 D2399/PRIMERGY
RIP: 0010:[<ffffffffa03d1462>]  [<ffffffffa03d1462>] btrfs_map_bio+0x202/0x210 [btrfs]
[SNIP]
Call Trace:
 [<ffffffffa03ab3eb>] __btrfs_submit_bio_done+0x1b/0x20 [btrfs]
 [<ffffffffa03a35ff>] run_one_async_done+0x9f/0xb0 [btrfs]
 [<ffffffffa03d3d20>] run_ordered_completions+0x80/0xc0 [btrfs]
 [<ffffffffa03d45a4>] worker_loop+0x154/0x5f0 [btrfs]
 [<ffffffffa03d4450>] ? worker_loop+0x0/0x5f0 [btrfs]
 [<ffffffffa03d4450>] ? worker_loop+0x0/0x5f0 [btrfs]
 [<ffffffff81083216>] kthread+0x96/0xa0
 [<ffffffff8100cec4>] kernel_thread_helper+0x4/0x10
 [<ffffffff81083180>] ? kthread+0x0/0xa0
 [<ffffffff8100cec0>] ? kernel_thread_helper+0x0/0x10

We fix this problem by splitting bios when we submit bios.

Reported-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Tested-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 205 +++++++++++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 184 insertions(+), 21 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8c027aa0020..a47e4faa8c4 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5535,13 +5535,21 @@ struct btrfs_dio_private {
 	u64 bytes;
 	u32 *csums;
 	void *private;
+
+	/* number of bios pending for this dio */
+	atomic_t pending_bios;
+
+	/* IO errors */
+	int errors;
+
+	struct bio *orig_bio;
 };
 
 static void btrfs_endio_direct_read(struct bio *bio, int err)
 {
+	struct btrfs_dio_private *dip = bio->bi_private;
 	struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1;
 	struct bio_vec *bvec = bio->bi_io_vec;
-	struct btrfs_dio_private *dip = bio->bi_private;
 	struct inode *inode = dip->inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	u64 start;
@@ -5684,6 +5692,176 @@ static int __btrfs_submit_bio_start_direct_io(struct inode *inode, int rw,
 	return 0;
 }
 
+static void btrfs_end_dio_bio(struct bio *bio, int err)
+{
+	struct btrfs_dio_private *dip = bio->bi_private;
+
+	if (err) {
+		printk(KERN_ERR "btrfs direct IO failed ino %lu rw %lu "
+		      "disk_bytenr %lu len %u err no %d\n",
+		      dip->inode->i_ino, bio->bi_rw, bio->bi_sector,
+		      bio->bi_size, err);
+		dip->errors = 1;
+
+		/*
+		 * before atomic variable goto zero, we must make sure
+		 * dip->errors is perceived to be set.
+		 */
+		smp_mb__before_atomic_dec();
+	}
+
+	/* if there are more bios still pending for this dio, just exit */
+	if (!atomic_dec_and_test(&dip->pending_bios))
+		goto out;
+
+	if (dip->errors)
+		bio_io_error(dip->orig_bio);
+	else {
+		set_bit(BIO_UPTODATE, &dip->orig_bio->bi_flags);
+		bio_endio(dip->orig_bio, 0);
+	}
+out:
+	bio_put(bio);
+}
+
+static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev,
+				       u64 first_sector, gfp_t gfp_flags)
+{
+	int nr_vecs = bio_get_nr_vecs(bdev);
+	return btrfs_bio_alloc(bdev, first_sector, nr_vecs, gfp_flags);
+}
+
+static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
+					 int rw, u64 file_offset, int skip_sum,
+					 u32 *csums)
+{
+	int write = rw & REQ_WRITE;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret;
+
+	bio_get(bio);
+	ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0);
+	if (ret)
+		goto err;
+
+	if (write && !skip_sum) {
+		ret = btrfs_wq_submit_bio(root->fs_info,
+				   inode, rw, bio, 0, 0,
+				   file_offset,
+				   __btrfs_submit_bio_start_direct_io,
+				   __btrfs_submit_bio_done);
+		goto err;
+	} else if (!skip_sum)
+		btrfs_lookup_bio_sums_dio(root, inode, bio,
+					  file_offset, csums);
+
+	ret = btrfs_map_bio(root, rw, bio, 0, 1);
+err:
+	bio_put(bio);
+	return ret;
+}
+
+static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
+				    int skip_sum)
+{
+	struct inode *inode = dip->inode;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
+	struct bio *bio;
+	struct bio *orig_bio = dip->orig_bio;
+	struct bio_vec *bvec = orig_bio->bi_io_vec;
+	u64 start_sector = orig_bio->bi_sector;
+	u64 file_offset = dip->logical_offset;
+	u64 submit_len = 0;
+	u64 map_length;
+	int nr_pages = 0;
+	u32 *csums = dip->csums;
+	int ret = 0;
+
+	bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
+	if (!bio)
+		return -ENOMEM;
+	bio->bi_private = dip;
+	bio->bi_end_io = btrfs_end_dio_bio;
+	atomic_inc(&dip->pending_bios);
+
+	map_length = orig_bio->bi_size;
+	ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+			      &map_length, NULL, 0);
+	if (ret) {
+		bio_put(bio);
+		return -EIO;
+	}
+
+	while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) {
+		if (unlikely(map_length < submit_len + bvec->bv_len ||
+		    bio_add_page(bio, bvec->bv_page, bvec->bv_len,
+				 bvec->bv_offset) < bvec->bv_len)) {
+			/*
+			 * inc the count before we submit the bio so
+			 * we know the end IO handler won't happen before
+			 * we inc the count. Otherwise, the dip might get freed
+			 * before we're done setting it up
+			 */
+			atomic_inc(&dip->pending_bios);
+			ret = __btrfs_submit_dio_bio(bio, inode, rw,
+						     file_offset, skip_sum,
+						     csums);
+			if (ret) {
+				bio_put(bio);
+				atomic_dec(&dip->pending_bios);
+				goto out_err;
+			}
+
+			if (!skip_sum)
+				csums = csums + nr_pages;
+			start_sector += submit_len >> 9;
+			file_offset += submit_len;
+
+			submit_len = 0;
+			nr_pages = 0;
+
+			bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev,
+						  start_sector, GFP_NOFS);
+			if (!bio)
+				goto out_err;
+			bio->bi_private = dip;
+			bio->bi_end_io = btrfs_end_dio_bio;
+
+			map_length = orig_bio->bi_size;
+			ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+					      &map_length, NULL, 0);
+			if (ret) {
+				bio_put(bio);
+				goto out_err;
+			}
+		} else {
+			submit_len += bvec->bv_len;
+			nr_pages ++;
+			bvec++;
+		}
+	}
+
+	ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum,
+				     csums);
+	if (!ret)
+		return 0;
+
+	bio_put(bio);
+out_err:
+	dip->errors = 1;
+	/*
+	 * before atomic variable goto zero, we must
+	 * make sure dip->errors is perceived to be set.
+	 */
+	smp_mb__before_atomic_dec();
+	if (atomic_dec_and_test(&dip->pending_bios))
+		bio_io_error(dip->orig_bio);
+
+	/* bio_end_io() will handle error, so we needn't return it */
+	return 0;
+}
+
 static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 				loff_t file_offset)
 {
@@ -5723,33 +5901,18 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 
 	dip->disk_bytenr = (u64)bio->bi_sector << 9;
 	bio->bi_private = dip;
+	dip->errors = 0;
+	dip->orig_bio = bio;
+	atomic_set(&dip->pending_bios, 0);
 
 	if (write)
 		bio->bi_end_io = btrfs_endio_direct_write;
 	else
 		bio->bi_end_io = btrfs_endio_direct_read;
 
-	ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0);
-	if (ret)
-		goto free_ordered;
-
-	if (write && !skip_sum) {
-		ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
-				   inode, rw, bio, 0, 0,
-				   dip->logical_offset,
-				   __btrfs_submit_bio_start_direct_io,
-				   __btrfs_submit_bio_done);
-		if (ret)
-			goto free_ordered;
+	ret = btrfs_submit_direct_hook(rw, dip, skip_sum);
+	if (!ret)
 		return;
-	} else if (!skip_sum)
-		btrfs_lookup_bio_sums_dio(root, inode, bio,
-					  dip->logical_offset, dip->csums);
-
-	ret = btrfs_map_bio(root, rw, bio, 0, 1);
-	if (ret)
-		goto free_ordered;
-	return;
 free_ordered:
 	/*
 	 * If this is a write, we need to clean up the reserved space and kill
-- 
cgit v1.2.2


From 6f33434850ed87dc5e56b60ebbad3d3cf405f296 Mon Sep 17 00:00:00 2001
From: Arne Jansen <sensille@gmx.net>
Date: Fri, 12 Nov 2010 23:17:56 +0000
Subject: btrfs: Fix early enospc because 'unused' calculated with wrong sign.

'unused' calculated with wrong sign in reserve_metadata_bytes().
This might have lead to unwanted over-reservations.

Signed-off-by: Arne Jansen <sensille@gmx.net>
Reviewed-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent-tree.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index a541bc87f04..ddaf6340fe7 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3413,7 +3413,7 @@ again:
 	 * our reservation.
 	 */
 	if (unused <= space_info->total_bytes) {
-		unused -= space_info->total_bytes;
+		unused = space_info->total_bytes - unused;
 		if (unused >= num_bytes) {
 			if (!reserved)
 				space_info->bytes_reserved += orig_bytes;
-- 
cgit v1.2.2


From 0de90876c6cb774d4a424dafc1fc9ec50071b81b Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 13:40:41 +0000
Subject: Btrfs: handle the space_cache option properly

When I added the clear_cache option I screwed up and took the break out of
the space_cache case statement, so whenever you mount with space_cache you also
get clear_cache, which does you no good if you say set space_cache in fstab so
it always gets set.  This patch adds the break back in properly.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/super.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 718b10de204..66e4612a791 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -244,6 +244,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 		case Opt_space_cache:
 			printk(KERN_INFO "btrfs: enabling disk space caching\n");
 			btrfs_set_opt(info->mount_opt, SPACE_CACHE);
+			break;
 		case Opt_clear_cache:
 			printk(KERN_INFO "btrfs: force clearing of disk cache\n");
 			btrfs_set_opt(info->mount_opt, CLEAR_CACHE);
-- 
cgit v1.2.2


From 2a6b8daedaf3682bed3fc1d4e2390491f6e19c49 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 19 Nov 2010 01:36:10 +0000
Subject: btrfs: Check if dest_offset is block-size aligned before cloning file

We've done the check for src_offset and src_length, and We should
also check dest_offset, otherwise we'll corrupt the destination
file:

  (After cloning file1 to file2 with unaligned dest_offset)
  # cat /mnt/file2
  cat: /mnt/file2: Input/output error

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ioctl.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 463d91b4dd3..81b47bd8a55 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1669,12 +1669,11 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 		olen = len = src->i_size - off;
 	/* if we extend to eof, continue to block boundary */
 	if (off + len == src->i_size)
-		len = ((src->i_size + bs-1) & ~(bs-1))
-			- off;
+		len = ALIGN(src->i_size, bs) - off;
 
 	/* verify the end result is block aligned */
-	if ((off & (bs-1)) ||
-	    ((off + len) & (bs-1)))
+	if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) ||
+	    !IS_ALIGNED(destoff, bs))
 		goto out_unlock;
 
 	/* do any pending delalloc/csum calc on src, one way or
-- 
cgit v1.2.2


From 5f3888ff6f0b9dce60705765752b788a92557644 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 19 Nov 2010 01:36:34 +0000
Subject: btrfs: Set file size correctly in file clone

Set src_offset = 0, src_length = 20K, dest_offset = 20K. And the
original filesize of the dest file 'file2' is 30K:

  # ls -l /mnt/file2
  -rw-r--r-- 1 root root 30720 Nov 18 16:42 /mnt/file2

Now clone file1 to file2, the dest file should be 40K, but it
still shows 30K:

  # ls -l /mnt/file2
  -rw-r--r-- 1 root root 30720 Nov 18 16:42 /mnt/file2

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ioctl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 81b47bd8a55..6b4bfa72bf8 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1873,8 +1873,8 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 			 * but shouldn't round up the file size
 			 */
 			endoff = new_key.offset + datal;
-			if (endoff > off+olen)
-				endoff = off+olen;
+			if (endoff > destoff+olen)
+				endoff = destoff+olen;
 			if (endoff > inode->i_size)
 				btrfs_i_size_write(inode, endoff);
 
-- 
cgit v1.2.2


From f209561ad83c5ffd561dc4bc3a3c90b704fe9231 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 19 Nov 2010 02:05:24 +0000
Subject: btrfs: Show device attr correctly for symlinks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Symlinks and files of other types show different device numbers, though
they are on the same partition:

 $ touch tmp; ln -s tmp tmp2; stat tmp tmp2
   File: `tmp'
   Size: 0         	Blocks: 0          IO Block: 4096   regular empty file
 Device: 15h/21d	Inode: 984027      Links: 1
 --- snip ---
   File: `tmp2' -> `tmp'
   Size: 3         	Blocks: 0          IO Block: 4096   symbolic link
 Device: 13h/19d	Inode: 984028      Links: 1

Reported-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index a47e4faa8c4..eed357ff6c9 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7299,6 +7299,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.follow_link	= page_follow_link_light,
 	.put_link	= page_put_link,
+	.getattr	= btrfs_getattr,
 	.permission	= btrfs_permission,
 	.setxattr	= btrfs_setxattr,
 	.getxattr	= btrfs_getxattr,
-- 
cgit v1.2.2


From 0410c94aff109c02b6774a0ed00114987cda7ce5 Mon Sep 17 00:00:00 2001
From: Mariusz Kozlowski <mk@lab.zgora.pl>
Date: Sat, 20 Nov 2010 12:03:07 +0000
Subject: btrfs: make 1-bit signed fileds unsigned

Fixes these sparse warnings:
fs/btrfs/ctree.h:811:17: error: dubious one-bit signed bitfield
fs/btrfs/ctree.h:812:20: error: dubious one-bit signed bitfield
fs/btrfs/ctree.h:813:19: error: dubious one-bit signed bitfield

Signed-off-by: Mariusz Kozlowski <mk@lab.zgora.pl>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ctree.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 8db9234f6b4..af52f6d7a4d 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -808,9 +808,9 @@ struct btrfs_block_group_cache {
 	int extents_thresh;
 	int free_extents;
 	int total_bitmaps;
-	int ro:1;
-	int dirty:1;
-	int iref:1;
+	unsigned int ro:1;
+	unsigned int dirty:1;
+	unsigned int iref:1;
 
 	int disk_cache_state;
 
-- 
cgit v1.2.2


From 2ede0daf01549cecf4bb0962c46dc47382047523 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Wed, 17 Nov 2010 18:54:54 +0000
Subject: Btrfs: handle NFS lookups properly

People kept reporting NFS issues, specifically getting ESTALE alot.  I figured
out how to reproduce the problem

SERVER
mkfs.btrfs /dev/sda1
mount /dev/sda1 /mnt/btrfs-test
<add /mnt/btrfs-test to /etc/exports>
btrfs subvol create /mnt/btrfs-test/foo
service nfs start

CLIENT
mount server:/mnt/btrfs /mnt/test
cd /mnt/test/foo
ls

SERVER
echo 3 > /proc/sys/vm/drop_caches

CLIENT
ls			<-- get an ESTALE here

This is because the standard way to lookup a name in nfsd is to use readdir, and
what it does is do a readdir on the parent directory looking for the inode of
the child.  So in this case the parent being / and the child being foo.  Well
subvols all have the same inode number, so doing a readdir of / looking for
inode 256 will return '.', which obviously doesn't match foo.  So instead we
need to have our own .get_name so that we can find the right name.

Our .get_name will either lookup the inode backref or the root backref,
whichever we're looking for, and return the name we find.  Running the above
reproducer with this patch results in everything acting the way its supposed to.
Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/export.c | 76 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 76 insertions(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 951ef09b82f..6f044447359 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -232,9 +232,85 @@ fail:
 	return ERR_PTR(ret);
 }
 
+static int btrfs_get_name(struct dentry *parent, char *name,
+			  struct dentry *child)
+{
+	struct inode *inode = child->d_inode;
+	struct inode *dir = parent->d_inode;
+	struct btrfs_path *path;
+	struct btrfs_root *root = BTRFS_I(dir)->root;
+	struct btrfs_inode_ref *iref;
+	struct btrfs_root_ref *rref;
+	struct extent_buffer *leaf;
+	unsigned long name_ptr;
+	struct btrfs_key key;
+	int name_len;
+	int ret;
+
+	if (!dir || !inode)
+		return -EINVAL;
+
+	if (!S_ISDIR(dir->i_mode))
+		return -EINVAL;
+
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+	path->leave_spinning = 1;
+
+	if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+		key.objectid = BTRFS_I(inode)->root->root_key.objectid;
+		key.type = BTRFS_ROOT_BACKREF_KEY;
+		key.offset = (u64)-1;
+		root = root->fs_info->tree_root;
+	} else {
+		key.objectid = inode->i_ino;
+		key.offset = dir->i_ino;
+		key.type = BTRFS_INODE_REF_KEY;
+	}
+
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0) {
+		btrfs_free_path(path);
+		return ret;
+	} else if (ret > 0) {
+		if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+			path->slots[0]--;
+		} else {
+			btrfs_free_path(path);
+			return -ENOENT;
+		}
+	}
+	leaf = path->nodes[0];
+
+	if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+	       rref = btrfs_item_ptr(leaf, path->slots[0],
+				     struct btrfs_root_ref);
+	       name_ptr = (unsigned long)(rref + 1);
+	       name_len = btrfs_root_ref_name_len(leaf, rref);
+	} else {
+		iref = btrfs_item_ptr(leaf, path->slots[0],
+				      struct btrfs_inode_ref);
+		name_ptr = (unsigned long)(iref + 1);
+		name_len = btrfs_inode_ref_name_len(leaf, iref);
+	}
+
+	read_extent_buffer(leaf, name, name_ptr, name_len);
+	btrfs_free_path(path);
+
+	/*
+	 * have to add the null termination to make sure that reconnect_path
+	 * gets the right len for strlen
+	 */
+	name[name_len] = '\0';
+
+	return 0;
+}
+
 const struct export_operations btrfs_export_ops = {
 	.encode_fh	= btrfs_encode_fh,
 	.fh_to_dentry	= btrfs_fh_to_dentry,
 	.fh_to_parent	= btrfs_fh_to_parent,
 	.get_parent	= btrfs_get_parent,
+	.get_name	= btrfs_get_name,
 };
-- 
cgit v1.2.2


From 76195853903ca613ba722203db9b747d70478fc7 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 02:18:02 +0000
Subject: Btrfs: fix more ESTALE problems with NFS

When creating new inodes we don't setup inode->i_generation.  So if we generate
an fh with a newly created inode we save the generation of 0, but if we flush
the inode to disk and have to read it back when getting the inode on the server
we'll have the right i_generation, so gens wont match and we get ESTALE.  This
patch properly sets inode->i_generation when we create the new inode and now I'm
no longer getting ESTALE.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index eed357ff6c9..fc22f556aa2 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4501,6 +4501,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	BTRFS_I(inode)->index_cnt = 2;
 	BTRFS_I(inode)->root = root;
 	BTRFS_I(inode)->generation = trans->transid;
+	inode->i_generation = BTRFS_I(inode)->generation;
 	btrfs_set_inode_space_info(root, inode);
 
 	if (mode & S_IFDIR)
-- 
cgit v1.2.2


From 6a912213046ecb6511fdf35531a0c7de3de963c9 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Sat, 20 Nov 2010 09:48:00 +0000
Subject: Btrfs: use dget_parent where we can UPDATED

There are lots of places where we do dentry->d_parent->d_inode without holding
the dentry->d_lock.  This could cause problems with rename.  So instead we need
to use dget_parent() and hold the reference to the parent as long as we are
going to use it's inode and then dput it at the end.

Signed-off-by: Josef Bacik <josef@redhat.com>
Cc: raven@themaw.net
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c       |  9 ++++++---
 fs/btrfs/ioctl.c       | 20 ++++++++++++++++----
 fs/btrfs/transaction.c |  5 ++++-
 fs/btrfs/tree-log.c    | 21 +++++++++++++++++----
 4 files changed, 43 insertions(+), 12 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index fc22f556aa2..c0faf47d0cd 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4811,10 +4811,12 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	if (err) {
 		drop_inode = 1;
 	} else {
+		struct dentry *parent = dget_parent(dentry);
 		btrfs_update_inode_block_group(trans, dir);
 		err = btrfs_update_inode(trans, root, inode);
 		BUG_ON(err);
-		btrfs_log_new_name(trans, inode, NULL, dentry->d_parent);
+		btrfs_log_new_name(trans, inode, NULL, parent);
+		dput(parent);
 	}
 
 	nr = trans->blocks_used;
@@ -6768,8 +6770,9 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	BUG_ON(ret);
 
 	if (old_inode->i_ino != BTRFS_FIRST_FREE_OBJECTID) {
-		btrfs_log_new_name(trans, old_inode, old_dir,
-				   new_dentry->d_parent);
+		struct dentry *parent = dget_parent(new_dentry);
+		btrfs_log_new_name(trans, old_inode, old_dir, parent);
+		dput(parent);
 		btrfs_end_log_trans(root);
 	}
 out_fail:
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 6b4bfa72bf8..f1c9bb4079e 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -233,7 +233,8 @@ static noinline int create_subvol(struct btrfs_root *root,
 	struct btrfs_inode_item *inode_item;
 	struct extent_buffer *leaf;
 	struct btrfs_root *new_root;
-	struct inode *dir = dentry->d_parent->d_inode;
+	struct dentry *parent = dget_parent(dentry);
+	struct inode *dir;
 	int ret;
 	int err;
 	u64 objectid;
@@ -242,8 +243,13 @@ static noinline int create_subvol(struct btrfs_root *root,
 
 	ret = btrfs_find_free_objectid(NULL, root->fs_info->tree_root,
 				       0, &objectid);
-	if (ret)
+	if (ret) {
+		dput(parent);
 		return ret;
+	}
+
+	dir = parent->d_inode;
+
 	/*
 	 * 1 - inode item
 	 * 2 - refs
@@ -251,8 +257,10 @@ static noinline int create_subvol(struct btrfs_root *root,
 	 * 2 - dir items
 	 */
 	trans = btrfs_start_transaction(root, 6);
-	if (IS_ERR(trans))
+	if (IS_ERR(trans)) {
+		dput(parent);
 		return PTR_ERR(trans);
+	}
 
 	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
 				      0, objectid, NULL, 0, 0, 0);
@@ -339,6 +347,7 @@ static noinline int create_subvol(struct btrfs_root *root,
 
 	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
 fail:
+	dput(parent);
 	if (async_transid) {
 		*async_transid = trans->transid;
 		err = btrfs_commit_transaction_async(trans, root, 1);
@@ -354,6 +363,7 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 			   char *name, int namelen, u64 *async_transid)
 {
 	struct inode *inode;
+	struct dentry *parent;
 	struct btrfs_pending_snapshot *pending_snapshot;
 	struct btrfs_trans_handle *trans;
 	int ret;
@@ -396,7 +406,9 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 
 	btrfs_orphan_cleanup(pending_snapshot->snap);
 
-	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
+	parent = dget_parent(dentry);
+	inode = btrfs_lookup_dentry(parent->d_inode, dentry);
+	dput(parent);
 	if (IS_ERR(inode)) {
 		ret = PTR_ERR(inode);
 		goto fail;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 1fffbc017bd..f50e931fc21 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -902,6 +902,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	struct btrfs_root *root = pending->root;
 	struct btrfs_root *parent_root;
 	struct inode *parent_inode;
+	struct dentry *parent;
 	struct dentry *dentry;
 	struct extent_buffer *tmp;
 	struct extent_buffer *old;
@@ -941,7 +942,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	trans->block_rsv = &pending->block_rsv;
 
 	dentry = pending->dentry;
-	parent_inode = dentry->d_parent->d_inode;
+	parent = dget_parent(dentry);
+	parent_inode = parent->d_inode;
 	parent_root = BTRFS_I(parent_inode)->root;
 	record_root_in_trans(trans, parent_root);
 
@@ -989,6 +991,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 				 parent_inode->i_ino, index,
 				 dentry->d_name.name, dentry->d_name.len);
 	BUG_ON(ret);
+	dput(parent);
 
 	key.offset = (u64)-1;
 	pending->snap = btrfs_read_fs_root_no_name(root->fs_info, &key);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index a29f19384a2..054744ac571 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -2869,6 +2869,7 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
 {
 	int ret = 0;
 	struct btrfs_root *root;
+	struct dentry *old_parent = NULL;
 
 	/*
 	 * for regular files, if its inode is already on disk, we don't
@@ -2910,10 +2911,13 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
 		if (IS_ROOT(parent))
 			break;
 
-		parent = parent->d_parent;
+		parent = dget_parent(parent);
+		dput(old_parent);
+		old_parent = parent;
 		inode = parent->d_inode;
 
 	}
+	dput(old_parent);
 out:
 	return ret;
 }
@@ -2945,6 +2949,7 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 {
 	int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
 	struct super_block *sb;
+	struct dentry *old_parent = NULL;
 	int ret = 0;
 	u64 last_committed = root->fs_info->last_trans_committed;
 
@@ -3016,10 +3021,13 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
 		if (IS_ROOT(parent))
 			break;
 
-		parent = parent->d_parent;
+		parent = dget_parent(parent);
+		dput(old_parent);
+		old_parent = parent;
 	}
 	ret = 0;
 end_trans:
+	dput(old_parent);
 	if (ret < 0) {
 		BUG_ON(ret != -ENOSPC);
 		root->fs_info->last_trans_log_full_commit = trans->transid;
@@ -3039,8 +3047,13 @@ end_no_trans:
 int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root, struct dentry *dentry)
 {
-	return btrfs_log_inode_parent(trans, root, dentry->d_inode,
-				      dentry->d_parent, 0);
+	struct dentry *parent = dget_parent(dentry);
+	int ret;
+
+	ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent, 0);
+	dput(parent);
+
+	return ret;
 }
 
 /*
-- 
cgit v1.2.2


From 495e86779f4f319828bc10dfc0c9ac2161868077 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 20:36:10 +0000
Subject: Btrfs: hold i_mutex when calling btrfs_log_dentry_safe

Since we walk up the path logging all of the parts of the inode's path, we need
to hold i_mutex to make sure that the inode is not renamed while we're logging
everything.  btrfs_log_dentry_safe does dget_parent and all of that jazz, but we
may get unexpected results if the rename changes the inode's location while
we're higher up the path logging those dentries, so do this for safety reasons.
Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/file.c | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index e354c33df08..c1faded5fca 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1047,8 +1047,14 @@ out:
 
 		if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
 			trans = btrfs_start_transaction(root, 0);
+			if (IS_ERR(trans)) {
+				num_written = PTR_ERR(trans);
+				goto done;
+			}
+			mutex_lock(&inode->i_mutex);
 			ret = btrfs_log_dentry_safe(trans, root,
 						    file->f_dentry);
+			mutex_unlock(&inode->i_mutex);
 			if (ret == 0) {
 				ret = btrfs_sync_log(trans, root);
 				if (ret == 0)
@@ -1067,6 +1073,7 @@ out:
 			     (start_pos + num_written - 1) >> PAGE_CACHE_SHIFT);
 		}
 	}
+done:
 	current->backing_dev_info = NULL;
 	return num_written ? num_written : err;
 }
-- 
cgit v1.2.2


From a1b075d28da563c5e2325577f282c042494254ba Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 20:36:11 +0000
Subject: Btrfs: make btrfs_add_nondir take parent inode as an argument

Everybody who calls btrfs_add_nondir just passes in the dentry of the new file
and then dereference dentry->d_parent->d_inode, but everybody who calls
btrfs_add_nondir() are already passed the parent's inode.  So instead of
dereferencing dentry->d_parent, just make btrfs_add_nondir take the dir inode as
an argument and pass that along so we don't have to worry about d_parent.
Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 38 ++++++++++++++++----------------------
 1 file changed, 16 insertions(+), 22 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index c0faf47d0cd..37cc1776a5d 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4623,12 +4623,12 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
 }
 
 static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
-			    struct dentry *dentry, struct inode *inode,
-			    int backref, u64 index)
+			    struct inode *dir, struct dentry *dentry,
+			    struct inode *inode, int backref, u64 index)
 {
-	int err = btrfs_add_link(trans, dentry->d_parent->d_inode,
-				 inode, dentry->d_name.name,
-				 dentry->d_name.len, backref, index);
+	int err = btrfs_add_link(trans, dir, inode,
+				 dentry->d_name.name, dentry->d_name.len,
+				 backref, index);
 	if (!err) {
 		d_instantiate(dentry, inode);
 		return 0;
@@ -4669,8 +4669,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 	btrfs_set_trans_block_group(trans, dir);
 
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
-				dentry->d_name.len,
-				dentry->d_parent->d_inode->i_ino, objectid,
+				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, mode, &index);
 	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
@@ -4683,7 +4682,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 	}
 
 	btrfs_set_trans_block_group(trans, inode);
-	err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
@@ -4731,10 +4730,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	btrfs_set_trans_block_group(trans, dir);
 
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
-				dentry->d_name.len,
-				dentry->d_parent->d_inode->i_ino,
-				objectid, BTRFS_I(dir)->block_group, mode,
-				&index);
+				dentry->d_name.len, dir->i_ino, objectid,
+				BTRFS_I(dir)->block_group, mode, &index);
 	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
 		goto out_unlock;
@@ -4746,7 +4743,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	}
 
 	btrfs_set_trans_block_group(trans, inode);
-	err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
@@ -4806,7 +4803,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	btrfs_set_trans_block_group(trans, dir);
 	atomic_inc(&inode->i_count);
 
-	err = btrfs_add_nondir(trans, dentry, inode, 1, index);
+	err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index);
 
 	if (err) {
 		drop_inode = 1;
@@ -4856,8 +4853,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	btrfs_set_trans_block_group(trans, dir);
 
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
-				dentry->d_name.len,
-				dentry->d_parent->d_inode->i_ino, objectid,
+				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, S_IFDIR | mode,
 				&index);
 	if (IS_ERR(inode)) {
@@ -4880,9 +4876,8 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (err)
 		goto out_fail;
 
-	err = btrfs_add_link(trans, dentry->d_parent->d_inode,
-				 inode, dentry->d_name.name,
-				 dentry->d_name.len, 0, index);
+	err = btrfs_add_link(trans, dir, inode, dentry->d_name.name,
+			     dentry->d_name.len, 0, index);
 	if (err)
 		goto out_fail;
 
@@ -6922,8 +6917,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	btrfs_set_trans_block_group(trans, dir);
 
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
-				dentry->d_name.len,
-				dentry->d_parent->d_inode->i_ino, objectid,
+				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO,
 				&index);
 	err = PTR_ERR(inode);
@@ -6937,7 +6931,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	}
 
 	btrfs_set_trans_block_group(trans, inode);
-	err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
-- 
cgit v1.2.2


From 45f49bce99d008d6864a20324548f35936ba46fb Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Sun, 21 Nov 2010 22:27:44 -0500
Subject: Btrfs: avoid NULL pointer deref in try_release_extent_buffer

If we fail to find a pointer in the radix tree, don't try
to deref the NULL one we do have.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent_io.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index f60aa3c35c2..143d3f541d6 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -3837,8 +3837,10 @@ int try_release_extent_buffer(struct extent_io_tree *tree, struct page *page)
 
 	spin_lock(&tree->buffer_lock);
 	eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT);
-	if (!eb)
-		goto out;
+	if (!eb) {
+		spin_unlock(&tree->buffer_lock);
+		return ret;
+	}
 
 	if (test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) {
 		ret = 0;
-- 
cgit v1.2.2


From 55a61d1d06a3dc443d0db8aaa613365dcb83b98a Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Mon, 22 Nov 2010 18:50:32 +0000
Subject: Btrfs: fix typo in fallocate to make it honor actual size

There is a typo in __btrfs_prealloc_file_range() where we set the i_size to
actual_len/cur_offset, and then just set it to cur_offset again, and do the same
with btrfs_ordered_update_i_size().  This fixes it back to keeping i_size in a
local variable and then updating i_size properly.  Tested this with

xfs_io -F -f -c "falloc 0 1" -c "pwrite 0 1" foo

stat'ing foo gives us a size of 1 instead of 4096 like it was.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 37cc1776a5d..0058fb3c256 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7002,6 +7002,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
 	u64 cur_offset = start;
+	u64 i_size;
 	int ret = 0;
 	bool own_trans = true;
 
@@ -7043,11 +7044,11 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		    (actual_len > inode->i_size) &&
 		    (cur_offset > inode->i_size)) {
 			if (cur_offset > actual_len)
-				i_size_write(inode, actual_len);
+				i_size = actual_len;
 			else
-				i_size_write(inode, cur_offset);
-			i_size_write(inode, cur_offset);
-			btrfs_ordered_update_i_size(inode, cur_offset, NULL);
+				i_size = cur_offset;
+			i_size_write(inode, i_size);
+			btrfs_ordered_update_i_size(inode, i_size, NULL);
 		}
 
 		ret = btrfs_update_inode(trans, root, inode);
-- 
cgit v1.2.2


From 0ed42a63f3edb144b091d9528401fce95c3c4d8d Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Mon, 22 Nov 2010 18:55:39 +0000
Subject: Btrfs: make sure new inode size is ok in fallocate

We have been failing xfstest 228 forever, because we don't check to make sure
the new inode size is acceptable as far as RLIMIT is concerned.  Just check to
make sure it's ok to create a inode with this new size and error out if not.
With this patch we now pass 228.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 0058fb3c256..0eeacd93e8e 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7102,6 +7102,10 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 	btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start);
 
 	mutex_lock(&inode->i_mutex);
+	ret = inode_newsize_ok(inode, alloc_end);
+	if (ret)
+		goto out;
+
 	if (alloc_start > inode->i_size) {
 		ret = btrfs_cont_expand(inode, alloc_start);
 		if (ret)
-- 
cgit v1.2.2


From bc1cbf1f86aa2501efa9ca637c736fce6bcc4b1d Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Tue, 23 Nov 2010 19:50:59 +0000
Subject: Btrfs: update inode ctime when using links

Currently we fail xfstest 236 because we're not updating the inode ctime on
link.  This is a simple fix, and makes it so we pass 236 now.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 0eeacd93e8e..6df921f218f 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4785,6 +4785,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		return -EPERM;
 
 	btrfs_inc_nlink(inode);
+	inode->i_ctime = CURRENT_TIME;
 
 	err = btrfs_set_inode_index(dir, &index);
 	if (err)
-- 
cgit v1.2.2


From 619c8c763928841b1112e1d417f88bc1d44daecb Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Mon, 22 Nov 2010 02:21:38 +0000
Subject: Btrfs - fix race between btrfs_get_sb() and umount

When mounting a btrfs file system btrfs_test_super() may attempt to
use sb->s_fs_info, the btrfs root, of a super block that is going away
and that has had the btrfs root set to NULL in its ->put_super(). But
if the super block is going away it cannot be an existing super block
so we can return false in this case.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/super.c | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 66e4612a791..141fb317d3b 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -566,6 +566,12 @@ static int btrfs_test_super(struct super_block *s, void *data)
 	struct btrfs_fs_devices *test_fs_devices = data;
 	struct btrfs_root *root = btrfs_sb(s);
 
+	/*
+	 * If this super block is going away, return false as it
+	 * can't match as an existing super block.
+	 */
+	if (!atomic_read(&s->s_active))
+		return 0;
 	return root->fs_info->fs_devices == test_fs_devices;
 }
 
-- 
cgit v1.2.2


From 975f84fee2e8a77ee5f41bfe7c5682bf29366b10 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Tue, 23 Nov 2010 19:36:57 +0000
Subject: Btrfs: fix fiemap

There are two big problems currently with FIEMAP

1) We return extents for holes.  This isn't supposed to happen, we just don't
return extents for holes and then userspace interprets the lack of an extent as
a hole.

2) We sometimes don't set FIEMAP_EXTENT_LAST properly.  This is because we wait
to see a EXTENT_FLAG_VACANCY flag on the em, but this won't happen if say we ask
fiemap to map up to the last extent in a file, and there is nothing but holes up
to the i_size.  To fix this we need to lookup the last extent in this file and
save the logical offset, so if we happen to try and map that extent we can be
sure to set FIEMAP_EXTENT_LAST.

With this patch we now pass xfstest 225, which we never have before.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent_io.c | 63 ++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 54 insertions(+), 9 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 143d3f541d6..5e7a94d7da8 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2901,21 +2901,53 @@ out:
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len, get_extent_t *get_extent)
 {
-	int ret;
+	int ret = 0;
 	u64 off = start;
 	u64 max = start + len;
 	u32 flags = 0;
+	u32 found_type;
+	u64 last;
 	u64 disko = 0;
+	struct btrfs_key found_key;
 	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
+	struct btrfs_path *path;
+	struct btrfs_file_extent_item *item;
 	int end = 0;
 	u64 em_start = 0, em_len = 0;
 	unsigned long emflags;
-	ret = 0;
+	int hole = 0;
 
 	if (len == 0)
 		return -EINVAL;
 
+	path = btrfs_alloc_path();
+	if (!path)
+		return -ENOMEM;
+	path->leave_spinning = 1;
+
+	ret = btrfs_lookup_file_extent(NULL, BTRFS_I(inode)->root,
+				       path, inode->i_ino, -1, 0);
+	if (ret < 0) {
+		btrfs_free_path(path);
+		return ret;
+	}
+	WARN_ON(!ret);
+	path->slots[0]--;
+	item = btrfs_item_ptr(path->nodes[0], path->slots[0],
+			      struct btrfs_file_extent_item);
+	btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]);
+	found_type = btrfs_key_type(&found_key);
+
+	/* No extents, just return */
+	if (found_key.objectid != inode->i_ino ||
+	    found_type != BTRFS_EXTENT_DATA_KEY) {
+		btrfs_free_path(path);
+		return 0;
+	}
+	last = found_key.offset;
+	btrfs_free_path(path);
+
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0,
 			 &cached_state, GFP_NOFS);
 	em = get_extent(inode, NULL, 0, off, max - off, 0);
@@ -2925,11 +2957,18 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		ret = PTR_ERR(em);
 		goto out;
 	}
+
 	while (!end) {
+		hole = 0;
 		off = em->start + em->len;
 		if (off >= max)
 			end = 1;
 
+		if (em->block_start == EXTENT_MAP_HOLE) {
+			hole = 1;
+			goto next;
+		}
+
 		em_start = em->start;
 		em_len = em->len;
 
@@ -2939,8 +2978,6 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		if (em->block_start == EXTENT_MAP_LAST_BYTE) {
 			end = 1;
 			flags |= FIEMAP_EXTENT_LAST;
-		} else if (em->block_start == EXTENT_MAP_HOLE) {
-			flags |= FIEMAP_EXTENT_UNWRITTEN;
 		} else if (em->block_start == EXTENT_MAP_INLINE) {
 			flags |= (FIEMAP_EXTENT_DATA_INLINE |
 				  FIEMAP_EXTENT_NOT_ALIGNED);
@@ -2953,10 +2990,10 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
 			flags |= FIEMAP_EXTENT_ENCODED;
 
+next:
 		emflags = em->flags;
 		free_extent_map(em);
 		em = NULL;
-
 		if (!end) {
 			em = get_extent(inode, NULL, 0, off, max - off, 0);
 			if (!em)
@@ -2967,15 +3004,23 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			}
 			emflags = em->flags;
 		}
+
 		if (test_bit(EXTENT_FLAG_VACANCY, &emflags)) {
 			flags |= FIEMAP_EXTENT_LAST;
 			end = 1;
 		}
 
-		ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
-					em_len, flags);
-		if (ret)
-			goto out_free;
+		if (em_start == last) {
+			flags |= FIEMAP_EXTENT_LAST;
+			end = 1;
+		}
+
+		if (!hole) {
+			ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
+						em_len, flags);
+			if (ret)
+				goto out_free;
+		}
 	}
 out_free:
 	free_extent_map(em);
-- 
cgit v1.2.2


From 450ba0ea06b6ed3612d27f2b7127a9de4160f285 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 14:59:15 -0500
Subject: Btrfs: setup blank root and fs_info for mount time

There is a problem with how we use sget, it searches through the list of supers
attached to the fs_type looking for a super with the same fs_devices as what
we're trying to mount.  This depends on sb->s_fs_info being filled, but we don't
fill that in until we get to btrfs_fill_super, so we could hit supers on the
fs_type super list that have a null s_fs_info.  In order to fix that we need to
go ahead and setup a blank root with a blank fs_info to hold fs_devices, that
way our test will work out right and then we can set s_fs_info in
btrfs_set_super, and then open_ctree will simply use our pre-allocated root and
fs_info when setting everything up.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/disk-io.c |  6 ++----
 fs/btrfs/super.c   | 34 +++++++++++++++++++++++++++++++---
 2 files changed, 33 insertions(+), 7 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index a67b98d58c2..57c9d8eeb7d 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1561,10 +1561,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 						 GFP_NOFS);
 	struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
 						 GFP_NOFS);
-	struct btrfs_root *tree_root = kzalloc(sizeof(struct btrfs_root),
-					       GFP_NOFS);
-	struct btrfs_fs_info *fs_info = kzalloc(sizeof(*fs_info),
-						GFP_NOFS);
+	struct btrfs_root *tree_root = btrfs_sb(sb);
+	struct btrfs_fs_info *fs_info = tree_root->fs_info;
 	struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
 						GFP_NOFS);
 	struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 141fb317d3b..47bf67cbe6b 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -563,7 +563,7 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 
 static int btrfs_test_super(struct super_block *s, void *data)
 {
-	struct btrfs_fs_devices *test_fs_devices = data;
+	struct btrfs_root *test_root = data;
 	struct btrfs_root *root = btrfs_sb(s);
 
 	/*
@@ -572,9 +572,17 @@ static int btrfs_test_super(struct super_block *s, void *data)
 	 */
 	if (!atomic_read(&s->s_active))
 		return 0;
-	return root->fs_info->fs_devices == test_fs_devices;
+	return root->fs_info->fs_devices == test_root->fs_info->fs_devices;
 }
 
+static int btrfs_set_super(struct super_block *s, void *data)
+{
+	s->s_fs_info = data;
+
+	return set_anon_super(s, data);
+}
+
+
 /*
  * Find a superblock for the given device / mount point.
  *
@@ -588,6 +596,8 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 	struct super_block *s;
 	struct dentry *root;
 	struct btrfs_fs_devices *fs_devices = NULL;
+	struct btrfs_root *tree_root = NULL;
+	struct btrfs_fs_info *fs_info = NULL;
 	fmode_t mode = FMODE_READ;
 	char *subvol_name = NULL;
 	u64 subvol_objectid = 0;
@@ -615,8 +625,24 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 		goto error_close_devices;
 	}
 
+	/*
+	 * Setup a dummy root and fs_info for test/set super.  This is because
+	 * we don't actually fill this stuff out until open_ctree, but we need
+	 * it for searching for existing supers, so this lets us do that and
+	 * then open_ctree will properly initialize everything later.
+	 */
+	fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS);
+	tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+	if (!fs_info || !tree_root) {
+		error = -ENOMEM;
+		goto error_close_devices;
+	}
+	fs_info->tree_root = tree_root;
+	fs_info->fs_devices = fs_devices;
+	tree_root->fs_info = fs_info;
+
 	bdev = fs_devices->latest_bdev;
-	s = sget(fs_type, btrfs_test_super, set_anon_super, fs_devices);
+	s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root);
 	if (IS_ERR(s))
 		goto error_s;
 
@@ -685,6 +711,8 @@ error_s:
 	error = PTR_ERR(s);
 error_close_devices:
 	btrfs_close_devices(fs_devices);
+	kfree(fs_info);
+	kfree(tree_root);
 error_free_subvol_name:
 	kfree(subvol_name);
 	return error;
-- 
cgit v1.2.2


From 163cf09c2a0ee5cac6285f9347975bd1e97725da Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Sun, 28 Nov 2010 19:56:33 -0500
Subject: Btrfs: deal with DIO bios that span more than one ordered extent

The new DIO bio splitting code has problems when the bio
spans more than one ordered extent.  This will happen as the
generic DIO code merges our get_blocks calls together into
a bigger single bio.

This fixes things by walking forward in the ordered extent
code finding all the overlapping ordered extents and completing them
all at once.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c        | 23 ++++++++++++++---
 fs/btrfs/ordered-data.c | 67 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/ordered-data.h |  3 +++
 3 files changed, 89 insertions(+), 4 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 6df921f218f..0f34cae0a63 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5602,15 +5602,18 @@ static void btrfs_endio_direct_write(struct bio *bio, int err)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_ordered_extent *ordered = NULL;
 	struct extent_state *cached_state = NULL;
+	u64 ordered_offset = dip->logical_offset;
+	u64 ordered_bytes = dip->bytes;
 	int ret;
 
 	if (err)
 		goto out_done;
-
-	ret = btrfs_dec_test_ordered_pending(inode, &ordered,
-					     dip->logical_offset, dip->bytes);
+again:
+	ret = btrfs_dec_test_first_ordered_pending(inode, &ordered,
+						   &ordered_offset,
+						   ordered_bytes);
 	if (!ret)
-		goto out_done;
+		goto out_test;
 
 	BUG_ON(!ordered);
 
@@ -5670,8 +5673,20 @@ out_unlock:
 out:
 	btrfs_delalloc_release_metadata(inode, ordered->len);
 	btrfs_end_transaction(trans, root);
+	ordered_offset = ordered->file_offset + ordered->len;
 	btrfs_put_ordered_extent(ordered);
 	btrfs_put_ordered_extent(ordered);
+
+out_test:
+	/*
+	 * our bio might span multiple ordered extents.  If we haven't
+	 * completed the accounting for the whole dio, go back and try again
+	 */
+	if (ordered_offset < dip->logical_offset + dip->bytes) {
+		ordered_bytes = dip->logical_offset + dip->bytes -
+			ordered_offset;
+		goto again;
+	}
 out_done:
 	bio->bi_private = dip->private;
 
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index f4621f6deca..ae7737e352c 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -248,6 +248,73 @@ int btrfs_add_ordered_sum(struct inode *inode,
 	return 0;
 }
 
+/*
+ * this is used to account for finished IO across a given range
+ * of the file.  The IO may span ordered extents.  If
+ * a given ordered_extent is completely done, 1 is returned, otherwise
+ * 0.
+ *
+ * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used
+ * to make sure this function only returns 1 once for a given ordered extent.
+ *
+ * file_offset is updated to one byte past the range that is recorded as
+ * complete.  This allows you to walk forward in the file.
+ */
+int btrfs_dec_test_first_ordered_pending(struct inode *inode,
+				   struct btrfs_ordered_extent **cached,
+				   u64 *file_offset, u64 io_size)
+{
+	struct btrfs_ordered_inode_tree *tree;
+	struct rb_node *node;
+	struct btrfs_ordered_extent *entry = NULL;
+	int ret;
+	u64 dec_end;
+	u64 dec_start;
+	u64 to_dec;
+
+	tree = &BTRFS_I(inode)->ordered_tree;
+	spin_lock(&tree->lock);
+	node = tree_search(tree, *file_offset);
+	if (!node) {
+		ret = 1;
+		goto out;
+	}
+
+	entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
+	if (!offset_in_entry(entry, *file_offset)) {
+		ret = 1;
+		goto out;
+	}
+
+	dec_start = max(*file_offset, entry->file_offset);
+	dec_end = min(*file_offset + io_size, entry->file_offset +
+		      entry->len);
+	*file_offset = dec_end;
+	if (dec_start > dec_end) {
+		printk(KERN_CRIT "bad ordering dec_start %llu end %llu\n",
+		       (unsigned long long)dec_start,
+		       (unsigned long long)dec_end);
+	}
+	to_dec = dec_end - dec_start;
+	if (to_dec > entry->bytes_left) {
+		printk(KERN_CRIT "bad ordered accounting left %llu size %llu\n",
+		       (unsigned long long)entry->bytes_left,
+		       (unsigned long long)to_dec);
+	}
+	entry->bytes_left -= to_dec;
+	if (entry->bytes_left == 0)
+		ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
+	else
+		ret = 1;
+out:
+	if (!ret && cached && entry) {
+		*cached = entry;
+		atomic_inc(&entry->refs);
+	}
+	spin_unlock(&tree->lock);
+	return ret == 0;
+}
+
 /*
  * this is used to account for finished IO across a given range
  * of the file.  The IO should not span ordered extents.  If
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 8ac365492a3..61dca83119d 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -141,6 +141,9 @@ int btrfs_remove_ordered_extent(struct inode *inode,
 int btrfs_dec_test_ordered_pending(struct inode *inode,
 				   struct btrfs_ordered_extent **cached,
 				   u64 file_offset, u64 io_size);
+int btrfs_dec_test_first_ordered_pending(struct inode *inode,
+				   struct btrfs_ordered_extent **cached,
+				   u64 *file_offset, u64 io_size);
 int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 			     u64 start, u64 len, u64 disk_len, int type);
 int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
-- 
cgit v1.2.2


From 5a92bc88cef279261d3f138e25850c122df67045 Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Mon, 29 Nov 2010 09:49:11 -0500
Subject: Btrfs: don't use migrate page without CONFIG_MIGRATION

Fixes compile error

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/disk-io.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 57c9d8eeb7d..33b6d459494 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -712,8 +712,11 @@ static int btree_migratepage(struct address_space *mapping,
 	if (page_has_private(page) &&
 	    !try_to_release_page(page, GFP_KERNEL))
 		return -EAGAIN;
-
+#ifdef CONFIG_MIGRATION
 	return migrate_page(mapping, newpage, page);
+#else
+	return -ENOSYS;
+#endif
 }
 
 static int btree_writepage(struct page *page, struct writeback_control *wbc)
@@ -821,7 +824,9 @@ static const struct address_space_operations btree_aops = {
 	.releasepage	= btree_releasepage,
 	.invalidatepage = btree_invalidatepage,
 	.sync_page	= block_sync_page,
+#ifdef CONFIG_MIGRATION
 	.migratepage	= btree_migratepage,
+#endif
 };
 
 int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
-- 
cgit v1.2.2


From 955256f2c3e25c94ad373c43fbc38d2ac8af2a71 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 19 Nov 2010 09:41:10 -0500
Subject: Btrfs: fix use after free in O_DIRECT

This fixes a bug where we use dip after we have freed it.  Instead just use the
file_offset that was passed to the function.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/inode.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 0f34cae0a63..ae6c0d190bc 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5934,8 +5934,7 @@ free_ordered:
 	 */
 	if (write) {
 		struct btrfs_ordered_extent *ordered;
-		ordered = btrfs_lookup_ordered_extent(inode,
-						      dip->logical_offset);
+		ordered = btrfs_lookup_ordered_extent(inode, file_offset);
 		if (!test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags) &&
 		    !test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags))
 			btrfs_free_reserved_extent(root, ordered->start,
-- 
cgit v1.2.2


From 2b20982e3154266106573beac2a4d4ba57a2789a Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 3 Dec 2010 13:17:53 -0500
Subject: Btrfs: deal with space cache errors better

Currently if the space cache inode generation number doesn't match the
generation number in the space cache header we will just fail to load the space
cache, but we won't mark the space cache as an error, so we'll keep getting that
error each time somebody tries to cache that block group until we actually clear
the thing.  Fix this by marking the space cache as having an error so we only
get the message once.  This patch also makes it so that we don't try and setup
space cache for a block group that isn't cached, since we won't be able to write
it out anyway.  None of these problems are actual problems, they are just
annoying and sub-optimal.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/extent-tree.c      | 10 ++++++----
 fs/btrfs/free-space-cache.c | 12 +++++++-----
 2 files changed, 13 insertions(+), 9 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index ddaf6340fe7..8c56f5b3894 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2742,6 +2742,7 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group,
 	struct btrfs_root *root = block_group->fs_info->tree_root;
 	struct inode *inode = NULL;
 	u64 alloc_hint = 0;
+	int dcs = BTRFS_DC_ERROR;
 	int num_pages = 0;
 	int retries = 0;
 	int ret = 0;
@@ -2796,6 +2797,8 @@ again:
 
 	spin_lock(&block_group->lock);
 	if (block_group->cached != BTRFS_CACHE_FINISHED) {
+		/* We're not cached, don't bother trying to write stuff out */
+		dcs = BTRFS_DC_WRITTEN;
 		spin_unlock(&block_group->lock);
 		goto out_put;
 	}
@@ -2822,6 +2825,8 @@ again:
 	ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, num_pages,
 					      num_pages, num_pages,
 					      &alloc_hint);
+	if (!ret)
+		dcs = BTRFS_DC_SETUP;
 	btrfs_free_reserved_data_space(inode, num_pages);
 out_put:
 	iput(inode);
@@ -2829,10 +2834,7 @@ out_free:
 	btrfs_release_path(root, path);
 out:
 	spin_lock(&block_group->lock);
-	if (ret)
-		block_group->disk_cache_state = BTRFS_DC_ERROR;
-	else
-		block_group->disk_cache_state = BTRFS_DC_SETUP;
+	block_group->disk_cache_state = dcs;
 	spin_unlock(&block_group->lock);
 
 	return ret;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 22ee0dc2e6b..60d68426695 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -290,7 +290,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
 		       (unsigned long long)BTRFS_I(inode)->generation,
 		       (unsigned long long)generation,
 		       (unsigned long long)block_group->key.objectid);
-		goto out;
+		goto free_cache;
 	}
 
 	if (!num_entries)
@@ -524,6 +524,12 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 		return 0;
 	}
 
+	node = rb_first(&block_group->free_space_offset);
+	if (!node) {
+		iput(inode);
+		return 0;
+	}
+
 	last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
 	filemap_write_and_wait(inode->i_mapping);
 	btrfs_wait_ordered_range(inode, inode->i_size &
@@ -543,10 +549,6 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	 */
 	first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64);
 
-	node = rb_first(&block_group->free_space_offset);
-	if (!node)
-		goto out_free;
-
 	/*
 	 * Lock all pages first so we can lock the extent safely.
 	 *
-- 
cgit v1.2.2


From b8399dee478db7939cd0d6fda8ecacddf2facd03 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Wed, 8 Dec 2010 09:15:11 -0500
Subject: Btrfs: do not do fast caching if we are allocating blocks for
 tree_root

Since the fast caching uses normal tree locking, we can possibly deadlock if we
get to the caching via a btrfs_search_slot() on the tree_root.  So just check to
see if the root we are on is the tree root, and just don't do the fast caching.

Reported-by: Sage Weil <sage@newdream.net>
Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/extent-tree.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 8c56f5b3894..cec05e10014 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -429,6 +429,7 @@ err:
 
 static int cache_block_group(struct btrfs_block_group_cache *cache,
 			     struct btrfs_trans_handle *trans,
+			     struct btrfs_root *root,
 			     int load_cache_only)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
@@ -442,9 +443,12 @@ static int cache_block_group(struct btrfs_block_group_cache *cache,
 
 	/*
 	 * We can't do the read from on-disk cache during a commit since we need
-	 * to have the normal tree locking.
+	 * to have the normal tree locking.  Also if we are currently trying to
+	 * allocate blocks for the tree root we can't do the fast caching since
+	 * we likely hold important locks.
 	 */
-	if (!trans->transaction->in_commit) {
+	if (!trans->transaction->in_commit &&
+	    (root && root != root->fs_info->tree_root)) {
 		spin_lock(&cache->lock);
 		if (cache->cached != BTRFS_CACHE_NO) {
 			spin_unlock(&cache->lock);
@@ -4083,7 +4087,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 		 * space back to the block group, otherwise we will leak space.
 		 */
 		if (!alloc && cache->cached == BTRFS_CACHE_NO)
-			cache_block_group(cache, trans, 1);
+			cache_block_group(cache, trans, NULL, 1);
 
 		byte_in_group = bytenr - cache->key.objectid;
 		WARN_ON(byte_in_group > cache->key.offset);
@@ -4937,7 +4941,8 @@ have_block_group:
 		if (unlikely(block_group->cached == BTRFS_CACHE_NO)) {
 			u64 free_percent;
 
-			ret = cache_block_group(block_group, trans, 1);
+			ret = cache_block_group(block_group, trans,
+						orig_root, 1);
 			if (block_group->cached == BTRFS_CACHE_FINISHED)
 				goto have_block_group;
 
@@ -4961,7 +4966,8 @@ have_block_group:
 			if (loop > LOOP_CACHING_NOWAIT ||
 			    (loop > LOOP_FIND_IDEAL &&
 			     atomic_read(&space_info->caching_threads) < 2)) {
-				ret = cache_block_group(block_group, trans, 0);
+				ret = cache_block_group(block_group, trans,
+							orig_root, 0);
 				BUG_ON(ret);
 			}
 			found_uncached_bg = true;
@@ -5518,7 +5524,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 	u64 num_bytes = ins->offset;
 
 	block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid);
-	cache_block_group(block_group, trans, 0);
+	cache_block_group(block_group, trans, NULL, 0);
 	caching_ctl = get_caching_control(block_group);
 
 	if (!caching_ctl) {
-- 
cgit v1.2.2


From 7e1fea731da8c1b5fcf5d8e157befd389b030760 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Wed, 8 Dec 2010 12:22:34 -0500
Subject: Btrfs: fixup return code for btrfs_del_orphan_item

If the orphan item doesn't exist, we return 1, which doesn't make any sense to
the callers.  Instead return -ENOENT if we didn't find the item.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/orphan.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/orphan.c b/fs/btrfs/orphan.c
index 79cba5fbc28..f8be250963a 100644
--- a/fs/btrfs/orphan.c
+++ b/fs/btrfs/orphan.c
@@ -56,8 +56,12 @@ int btrfs_del_orphan_item(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret)
+	if (ret < 0)
 		goto out;
+	if (ret) {
+		ret = -ENOENT;
+		goto out;
+	}
 
 	ret = btrfs_del_item(trans, root, path);
 
-- 
cgit v1.2.2


From 84cd948cb11041f205242de457e680b9bb872a36 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Wed, 8 Dec 2010 12:24:01 -0500
Subject: Btrfs: do not BUG if we fail to remove the orphan item for dead
 snapshots

Not being able to delete an orphan item isn't a horrible thing.  The worst that
happens is the next time around we try and do the orphan cleanup and we can't
find the referenced object and just delete the item and move on.

Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/extent-tree.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index cec05e10014..41133b064d7 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -6309,9 +6309,13 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 					   NULL, NULL);
 		BUG_ON(ret < 0);
 		if (ret > 0) {
-			ret = btrfs_del_orphan_item(trans, tree_root,
-						    root->root_key.objectid);
-			BUG_ON(ret);
+			/* if we fail to delete the orphan item this time
+			 * around, it'll get picked up the next time.
+			 *
+			 * The most common failure here is just -ENOENT.
+			 */
+			btrfs_del_orphan_item(trans, tree_root,
+					      root->root_key.objectid);
 		}
 	}
 
-- 
cgit v1.2.2


From 24ae63656a165c870c0d69fcc8aac1dc35e25e34 Mon Sep 17 00:00:00 2001
From: "Yan, Zheng" <zheng.z.yan@linux.intel.com>
Date: Mon, 6 Dec 2010 07:02:36 +0000
Subject: Btrfs: Fix page leak in compressed writeback path

"start + num_bytes >= actual_end" can happen when compressed page writeback races
with file truncation. In that case we need unlock and release pages past the end
of file.

Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index ae6c0d190bc..4875d69871b 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -495,7 +495,7 @@ again:
 		add_async_extent(async_cow, start, num_bytes,
 				 total_compressed, pages, nr_pages_ret);
 
-		if (start + num_bytes < end && start + num_bytes < actual_end) {
+		if (start + num_bytes < end) {
 			start += num_bytes;
 			pages = NULL;
 			cond_resched();
-- 
cgit v1.2.2


From 75eaa0e22c055e38982df267d0f84cc510ba38bf Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Fri, 10 Dec 2010 00:36:28 +0000
Subject: Btrfs: fix sync subvol/snapshot creation

We were incorrectly taking the async path even for the sync ioctls by
passing in &transid unconditionally.

There's ample room for further cleanup here, but this keeps the fix simple.

Signed-off-by: Sage Weil <sage@newdream.net>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ioctl.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index f1c9bb4079e..7cc2e8e075b 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -964,6 +964,15 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 		name = async_vol_args->name;
 		fd = async_vol_args->fd;
 		async_vol_args->name[BTRFS_SNAPSHOT_NAME_MAX] = '\0';
+
+		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
+						      subvol, &transid);
+
+		if (ret == 0 &&
+		    copy_to_user(arg +
+				 offsetof(struct btrfs_ioctl_async_vol_args,
+					  transid), &transid, sizeof(transid)))
+			ret = -EFAULT;
 	} else {
 		vol_args = memdup_user(arg, sizeof(*vol_args));
 		if (IS_ERR(vol_args))
@@ -971,16 +980,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 		name = vol_args->name;
 		fd = vol_args->fd;
 		vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
-	}
 
-	ret = btrfs_ioctl_snap_create_transid(file, name, fd,
-					      subvol, &transid);
-
-	if (!ret && async) {
-		if (copy_to_user(arg +
-				offsetof(struct btrfs_ioctl_async_vol_args,
-				transid), &transid, sizeof(transid)))
-			return -EFAULT;
+		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
+						      subvol, NULL);
 	}
 
 	kfree(vol_args);
-- 
cgit v1.2.2


From f106e82caaa0d943e47cacc184f5b40d538e0044 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Tue, 7 Dec 2010 01:51:26 +0000
Subject: Btrfs: Fix a crash when mounting a subvolume

We should drop dentry before deactivating the superblock, otherwise
we can hit this bug:

BUG: Dentry f349a690{i=100,n=/} still in use (1) [unmount of btrfs loop1]
...

Steps to reproduce the bug:

  # mount /dev/loop1 /mnt
  # mkdir save
  # btrfs subvolume snapshot /mnt save/snap1
  # umount /mnt
  # mount -o subvol=save/snap1 /dev/loop1 /mnt
  (crash)

Reported-by: Michael Niederle <mniederle@gmx.at>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 47bf67cbe6b..61bd79abb80 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -685,9 +685,9 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
 		mutex_unlock(&root->d_inode->i_mutex);
 
 		if (IS_ERR(new_root)) {
+			dput(root);
 			deactivate_locked_super(s);
 			error = PTR_ERR(new_root);
-			dput(root);
 			goto error_free_subvol_name;
 		}
 		if (!new_root->d_inode) {
-- 
cgit v1.2.2


From 914ee295af418e936ec20a08c1663eaabe4cd07a Mon Sep 17 00:00:00 2001
From: Xin Zhong <xin.zhong@intel.com>
Date: Thu, 9 Dec 2010 09:30:14 +0000
Subject: Btrfs: pwrite blocked when writing from the mmaped buffer of the same
 page

This problem is found in meego testing:
http://bugs.meego.com/show_bug.cgi?id=6672
A file in btrfs is mmaped and the mmaped buffer is passed to pwrite to write to the same page
of the same file. In btrfs_file_aio_write(), the pages is locked by prepare_pages(). So when
btrfs_copy_from_user() is called, page fault happens and the same page needs to be locked again
in filemap_fault(). The fix is to move iov_iter_fault_in_readable() before prepage_pages() to make page
fault happen before pages are locked. And also disable page fault in critical region in
btrfs_copy_from_user().

Reviewed-by: Yan, Zheng<zheng.z.yan@intel.com>
Signed-off-by: Zhong, Xin <xin.zhong@intel.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/file.c | 92 +++++++++++++++++++++++++++++++++++++--------------------
 1 file changed, 60 insertions(+), 32 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index c1faded5fca..66836d85763 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -48,30 +48,34 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
 					 struct page **prepared_pages,
 					 struct iov_iter *i)
 {
-	size_t copied;
+	size_t copied = 0;
 	int pg = 0;
 	int offset = pos & (PAGE_CACHE_SIZE - 1);
+	int total_copied = 0;
 
 	while (write_bytes > 0) {
 		size_t count = min_t(size_t,
 				     PAGE_CACHE_SIZE - offset, write_bytes);
 		struct page *page = prepared_pages[pg];
-again:
-		if (unlikely(iov_iter_fault_in_readable(i, count)))
-			return -EFAULT;
-
-		/* Copy data from userspace to the current page */
-		copied = iov_iter_copy_from_user(page, i, offset, count);
+		/*
+		 * Copy data from userspace to the current page
+		 *
+		 * Disable pagefault to avoid recursive lock since
+		 * the pages are already locked
+		 */
+		pagefault_disable();
+		copied = iov_iter_copy_from_user_atomic(page, i, offset, count);
+		pagefault_enable();
 
 		/* Flush processor's dcache for this page */
 		flush_dcache_page(page);
 		iov_iter_advance(i, copied);
 		write_bytes -= copied;
+		total_copied += copied;
 
+		/* Return to btrfs_file_aio_write to fault page */
 		if (unlikely(copied == 0)) {
-			count = min_t(size_t, PAGE_CACHE_SIZE - offset,
-				      iov_iter_single_seg_count(i));
-			goto again;
+			break;
 		}
 
 		if (unlikely(copied < PAGE_CACHE_SIZE - offset)) {
@@ -81,7 +85,7 @@ again:
 			offset = 0;
 		}
 	}
-	return 0;
+	return total_copied;
 }
 
 /*
@@ -854,6 +858,8 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	unsigned long last_index;
 	int will_write;
 	int buffered = 0;
+	int copied = 0;
+	int dirty_pages = 0;
 
 	will_write = ((file->f_flags & O_DSYNC) || IS_SYNC(inode) ||
 		      (file->f_flags & O_DIRECT));
@@ -970,7 +976,17 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 		WARN_ON(num_pages > nrptrs);
 		memset(pages, 0, sizeof(struct page *) * nrptrs);
 
-		ret = btrfs_delalloc_reserve_space(inode, write_bytes);
+		/*
+		 * Fault pages before locking them in prepare_pages
+		 * to avoid recursive lock
+		 */
+		if (unlikely(iov_iter_fault_in_readable(&i, write_bytes))) {
+			ret = -EFAULT;
+			goto out;
+		}
+
+		ret = btrfs_delalloc_reserve_space(inode,
+					num_pages << PAGE_CACHE_SHIFT);
 		if (ret)
 			goto out;
 
@@ -978,37 +994,49 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 				    pos, first_index, last_index,
 				    write_bytes);
 		if (ret) {
-			btrfs_delalloc_release_space(inode, write_bytes);
+			btrfs_delalloc_release_space(inode,
+					num_pages << PAGE_CACHE_SHIFT);
 			goto out;
 		}
 
-		ret = btrfs_copy_from_user(pos, num_pages,
+		copied = btrfs_copy_from_user(pos, num_pages,
 					   write_bytes, pages, &i);
-		if (ret == 0) {
+		dirty_pages = (copied + PAGE_CACHE_SIZE - 1) >>
+					PAGE_CACHE_SHIFT;
+
+		if (num_pages > dirty_pages) {
+			if (copied > 0)
+				atomic_inc(
+					&BTRFS_I(inode)->outstanding_extents);
+			btrfs_delalloc_release_space(inode,
+					(num_pages - dirty_pages) <<
+					PAGE_CACHE_SHIFT);
+		}
+
+		if (copied > 0) {
 			dirty_and_release_pages(NULL, root, file, pages,
-						num_pages, pos, write_bytes);
+						dirty_pages, pos, copied);
 		}
 
 		btrfs_drop_pages(pages, num_pages);
-		if (ret) {
-			btrfs_delalloc_release_space(inode, write_bytes);
-			goto out;
-		}
 
-		if (will_write) {
-			filemap_fdatawrite_range(inode->i_mapping, pos,
-						 pos + write_bytes - 1);
-		} else {
-			balance_dirty_pages_ratelimited_nr(inode->i_mapping,
-							   num_pages);
-			if (num_pages <
-			    (root->leafsize >> PAGE_CACHE_SHIFT) + 1)
-				btrfs_btree_balance_dirty(root, 1);
-			btrfs_throttle(root);
+		if (copied > 0) {
+			if (will_write) {
+				filemap_fdatawrite_range(inode->i_mapping, pos,
+							 pos + copied - 1);
+			} else {
+				balance_dirty_pages_ratelimited_nr(
+							inode->i_mapping,
+							dirty_pages);
+				if (dirty_pages <
+				(root->leafsize >> PAGE_CACHE_SHIFT) + 1)
+					btrfs_btree_balance_dirty(root, 1);
+				btrfs_throttle(root);
+			}
 		}
 
-		pos += write_bytes;
-		num_written += write_bytes;
+		pos += copied;
+		num_written += copied;
 
 		cond_resched();
 	}
-- 
cgit v1.2.2


From fdfb1e4f6c61477a61890b64974d65cdc3a98702 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 10 Dec 2010 06:41:56 +0000
Subject: Btrfs: Make async snapshot ioctl more generic

If we had reserved some bytes in struct btrfs_ioctl_vol_args, we
wouldn't have to create a new structure for async snapshot creation.

Here we convert async snapshot ioctl to use a more generic ABI, as
we'll add more ioctls for snapshots/subvolumes in the future, readonly
snapshots for example.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ioctl.c | 44 +++++++++++++++++++++++++++-----------------
 fs/btrfs/ioctl.h | 14 +++++++++-----
 2 files changed, 36 insertions(+), 22 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 7cc2e8e075b..f87552a1d7e 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -947,31 +947,41 @@ out:
 
 static noinline int btrfs_ioctl_snap_create(struct file *file,
 					    void __user *arg, int subvol,
-					    int async)
+					    int v2)
 {
 	struct btrfs_ioctl_vol_args *vol_args = NULL;
-	struct btrfs_ioctl_async_vol_args *async_vol_args = NULL;
+	struct btrfs_ioctl_vol_args_v2 *vol_args_v2 = NULL;
 	char *name;
 	u64 fd;
-	u64 transid = 0;
 	int ret;
 
-	if (async) {
-		async_vol_args = memdup_user(arg, sizeof(*async_vol_args));
-		if (IS_ERR(async_vol_args))
-			return PTR_ERR(async_vol_args);
+	if (v2) {
+		u64 transid = 0;
+		u64 *ptr = NULL;
 
-		name = async_vol_args->name;
-		fd = async_vol_args->fd;
-		async_vol_args->name[BTRFS_SNAPSHOT_NAME_MAX] = '\0';
+		vol_args_v2 = memdup_user(arg, sizeof(*vol_args_v2));
+		if (IS_ERR(vol_args_v2))
+			return PTR_ERR(vol_args_v2);
+
+		if (vol_args_v2->flags & ~BTRFS_SUBVOL_CREATE_ASYNC) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		name = vol_args_v2->name;
+		fd = vol_args_v2->fd;
+		vol_args_v2->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
+
+		if (vol_args_v2->flags & BTRFS_SUBVOL_CREATE_ASYNC)
+			ptr = &transid;
 
 		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
-						      subvol, &transid);
+						      subvol, ptr);
 
-		if (ret == 0 &&
+		if (ret == 0 && ptr &&
 		    copy_to_user(arg +
-				 offsetof(struct btrfs_ioctl_async_vol_args,
-					  transid), &transid, sizeof(transid)))
+				 offsetof(struct btrfs_ioctl_vol_args_v2,
+					  transid), ptr, sizeof(*ptr)))
 			ret = -EFAULT;
 	} else {
 		vol_args = memdup_user(arg, sizeof(*vol_args));
@@ -984,9 +994,9 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 		ret = btrfs_ioctl_snap_create_transid(file, name, fd,
 						      subvol, NULL);
 	}
-
+out:
 	kfree(vol_args);
-	kfree(async_vol_args);
+	kfree(vol_args_v2);
 
 	return ret;
 }
@@ -2248,7 +2258,7 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_getversion(file, argp);
 	case BTRFS_IOC_SNAP_CREATE:
 		return btrfs_ioctl_snap_create(file, argp, 0, 0);
-	case BTRFS_IOC_SNAP_CREATE_ASYNC:
+	case BTRFS_IOC_SNAP_CREATE_V2:
 		return btrfs_ioctl_snap_create(file, argp, 0, 1);
 	case BTRFS_IOC_SUBVOL_CREATE:
 		return btrfs_ioctl_snap_create(file, argp, 1, 0);
diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h
index 17c99ebdf96..c344d12c646 100644
--- a/fs/btrfs/ioctl.h
+++ b/fs/btrfs/ioctl.h
@@ -30,11 +30,15 @@ struct btrfs_ioctl_vol_args {
 	char name[BTRFS_PATH_NAME_MAX + 1];
 };
 
-#define BTRFS_SNAPSHOT_NAME_MAX 4079
-struct btrfs_ioctl_async_vol_args {
+#define BTRFS_SUBVOL_CREATE_ASYNC	(1ULL << 0)
+
+#define BTRFS_SUBVOL_NAME_MAX 4039
+struct btrfs_ioctl_vol_args_v2 {
 	__s64 fd;
 	__u64 transid;
-	char name[BTRFS_SNAPSHOT_NAME_MAX + 1];
+	__u64 flags;
+	__u64 unused[4];
+	char name[BTRFS_SUBVOL_NAME_MAX + 1];
 };
 
 #define BTRFS_INO_LOOKUP_PATH_MAX 4080
@@ -187,6 +191,6 @@ struct btrfs_ioctl_space_args {
 				    struct btrfs_ioctl_space_args)
 #define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64)
 #define BTRFS_IOC_WAIT_SYNC  _IOW(BTRFS_IOCTL_MAGIC, 22, __u64)
-#define BTRFS_IOC_SNAP_CREATE_ASYNC _IOW(BTRFS_IOCTL_MAGIC, 23, \
-				   struct btrfs_ioctl_async_vol_args)
+#define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \
+				   struct btrfs_ioctl_vol_args_v2)
 #endif
-- 
cgit v1.2.2


From 3dd1462e82bcab7625cec129952f26dae7a8b742 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 7 Dec 2010 14:54:09 +0000
Subject: Btrfs: fix compiler warnings

... regarding an unused function when !MIGRATION, and regarding a
printk() format string vs argument mismatch.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/disk-io.c | 6 ++----
 fs/btrfs/inode.c   | 6 +++---
 2 files changed, 5 insertions(+), 7 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 33b6d459494..b803c266767 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -696,6 +696,7 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
 				   __btree_submit_bio_done);
 }
 
+#ifdef CONFIG_MIGRATION
 static int btree_migratepage(struct address_space *mapping,
 			struct page *newpage, struct page *page)
 {
@@ -712,12 +713,9 @@ static int btree_migratepage(struct address_space *mapping,
 	if (page_has_private(page) &&
 	    !try_to_release_page(page, GFP_KERNEL))
 		return -EAGAIN;
-#ifdef CONFIG_MIGRATION
 	return migrate_page(mapping, newpage, page);
-#else
-	return -ENOSYS;
-#endif
 }
+#endif
 
 static int btree_writepage(struct page *page, struct writeback_control *wbc)
 {
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 4875d69871b..5f9194438f7 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5712,9 +5712,9 @@ static void btrfs_end_dio_bio(struct bio *bio, int err)
 
 	if (err) {
 		printk(KERN_ERR "btrfs direct IO failed ino %lu rw %lu "
-		      "disk_bytenr %lu len %u err no %d\n",
-		      dip->inode->i_ino, bio->bi_rw, bio->bi_sector,
-		      bio->bi_size, err);
+		      "sector %#Lx len %u err no %d\n",
+		      dip->inode->i_ino, bio->bi_rw,
+		      (unsigned long long)bio->bi_sector, bio->bi_size, err);
 		dip->errors = 1;
 
 		/*
-- 
cgit v1.2.2


From 68433b73b104bff388aac376631d32abbbd872b0 Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Mon, 13 Dec 2010 14:47:58 -0500
Subject: Btrfs: EIO when we fail to read tree roots

If we just get a plain IO error when we read tree roots, the code
wasn't properly sending that error up the chain.  This allowed mounts to
continue when they should failed, and allowed operations
on partially setup root structs.  The end result was usually oopsen
on spinlocks that hadn't been spun up correctly.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/disk-io.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index b803c266767..a5d2249e6da 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1007,7 +1007,10 @@ static int find_and_setup_root(struct btrfs_root *tree_root,
 	blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
 	root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
 				     blocksize, generation);
-	BUG_ON(!root->node);
+	if (!root->node || !btrfs_buffer_uptodate(root->node, generation)) {
+		free_extent_buffer(root->node);
+		return -EIO;
+	}
 	root->commit_root = btrfs_root_node(root);
 	return 0;
 }
-- 
cgit v1.2.2


From cd02dca56442e1504fd6bc5b96f7f1870162b266 Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Mon, 13 Dec 2010 14:56:23 -0500
Subject: Btrfs: account for missing devices in RAID allocation profiles

When we mount in RAID degraded mode without adding a new device to
replace the failed one, we can end up using the wrong RAID flags for
allocations.

This results in strange combinations of block groups (raid1 in a raid10
filesystem) and corruptions when we try to allocate blocks from single
spindle chunks on drives that are actually missing.

The first device has two small 4MB chunks in it that mkfs creates and
these are usually unused in a raid1 or raid10 setup.  But, in -o degraded,
the allocator will fall back to these because the mask of desired raid groups
isn't correct.

The fix here is to count the missing devices as we build up the list
of devices in the system.  This count is used when picking the
raid level to make sure we continue using the same levels that were
in place before we lost a drive.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent-tree.c | 17 +++++++++++++++--
 fs/btrfs/volumes.c     | 20 +++++++++++++++++++-
 fs/btrfs/volumes.h     |  2 ++
 3 files changed, 36 insertions(+), 3 deletions(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 41133b064d7..4be231e0d2b 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3044,7 +3044,13 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
 
 u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
 {
-	u64 num_devices = root->fs_info->fs_devices->rw_devices;
+	/*
+	 * we add in the count of missing devices because we want
+	 * to make sure that any RAID levels on a degraded FS
+	 * continue to be honored.
+	 */
+	u64 num_devices = root->fs_info->fs_devices->rw_devices +
+		root->fs_info->fs_devices->missing_devices;
 
 	if (num_devices == 1)
 		flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0);
@@ -7891,7 +7897,14 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
 	u64 stripped = BTRFS_BLOCK_GROUP_RAID0 |
 		BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10;
 
-	num_devices = root->fs_info->fs_devices->rw_devices;
+	/*
+	 * we add in the count of missing devices because we want
+	 * to make sure that any RAID levels on a degraded FS
+	 * continue to be honored.
+	 */
+	num_devices = root->fs_info->fs_devices->rw_devices +
+		root->fs_info->fs_devices->missing_devices;
+
 	if (num_devices == 1) {
 		stripped |= BTRFS_BLOCK_GROUP_DUP;
 		stripped = flags & ~stripped;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 91851b555e2..177b7317959 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -413,12 +413,16 @@ static noinline int device_list_add(const char *path,
 
 		device->fs_devices = fs_devices;
 		fs_devices->num_devices++;
-	} else if (strcmp(device->name, path)) {
+	} else if (!device->name || strcmp(device->name, path)) {
 		name = kstrdup(path, GFP_NOFS);
 		if (!name)
 			return -ENOMEM;
 		kfree(device->name);
 		device->name = name;
+		if (device->missing) {
+			fs_devices->missing_devices--;
+			device->missing = 0;
+		}
 	}
 
 	if (found_transid > fs_devices->latest_trans) {
@@ -1238,6 +1242,9 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 
 	device->fs_devices->num_devices--;
 
+	if (device->missing)
+		root->fs_info->fs_devices->missing_devices--;
+
 	next_device = list_entry(root->fs_info->fs_devices->devices.next,
 				 struct btrfs_device, dev_list);
 	if (device->bdev == root->fs_info->sb->s_bdev)
@@ -3084,7 +3091,9 @@ static struct btrfs_device *add_missing_dev(struct btrfs_root *root,
 	device->devid = devid;
 	device->work.func = pending_bios_fn;
 	device->fs_devices = fs_devices;
+	device->missing = 1;
 	fs_devices->num_devices++;
+	fs_devices->missing_devices++;
 	spin_lock_init(&device->io_lock);
 	INIT_LIST_HEAD(&device->dev_alloc_list);
 	memcpy(device->uuid, dev_uuid, BTRFS_UUID_SIZE);
@@ -3282,6 +3291,15 @@ static int read_one_dev(struct btrfs_root *root,
 			device = add_missing_dev(root, devid, dev_uuid);
 			if (!device)
 				return -ENOMEM;
+		} else if (!device->missing) {
+			/*
+			 * this happens when a device that was properly setup
+			 * in the device info lists suddenly goes bad.
+			 * device->bdev is NULL, and so we have to set
+			 * device->missing to one here
+			 */
+			root->fs_info->fs_devices->missing_devices++;
+			device->missing = 1;
 		}
 	}
 
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 31b0fabdd2e..a668c011698 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -45,6 +45,7 @@ struct btrfs_device {
 	int barriers;
 	int writeable;
 	int in_fs_metadata;
+	int missing;
 
 	spinlock_t io_lock;
 
@@ -94,6 +95,7 @@ struct btrfs_fs_devices {
 	u64 num_devices;
 	u64 open_devices;
 	u64 rw_devices;
+	u64 missing_devices;
 	u64 total_rw_bytes;
 	struct block_device *latest_bdev;
 
-- 
cgit v1.2.2


From 83a50de97fe96aca82389e061862ed760ece2283 Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Mon, 13 Dec 2010 15:06:46 -0500
Subject: Btrfs: prevent RAID level downgrades when space is low

The extent allocator has code that allows us to fill
allocations from any available block group, even if it doesn't
match the raid level we've requested.

This was put in because adding a new drive to a filesystem
made with the default mkfs options actually upgrades the metadata from
single spindle dup to full RAID1.

But, the code also allows us to allocate from a raid0 chunk when we
really want a raid1 or raid10 chunk.  This can cause big trouble because
mkfs creates a small (4MB) raid0 chunk for data and metadata which then
goes unused for raid1/raid10 installs.

The allocator will happily wander in and allocate from that chunk when
things get tight, which is not correct.

The fix here is to make sure that we provide duplication when the
caller has asked for it.  It does all the dups to be any raid level,
which preserves the dup->raid1 upgrade abilities.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent-tree.c | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 4be231e0d2b..7e5162e5c41 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -4943,6 +4943,25 @@ search:
 		btrfs_get_block_group(block_group);
 		search_start = block_group->key.objectid;
 
+		/*
+		 * this can happen if we end up cycling through all the
+		 * raid types, but we want to make sure we only allocate
+		 * for the proper type.
+		 */
+		if (!block_group_bits(block_group, data)) {
+		    u64 extra = BTRFS_BLOCK_GROUP_DUP |
+				BTRFS_BLOCK_GROUP_RAID1 |
+				BTRFS_BLOCK_GROUP_RAID10;
+
+			/*
+			 * if they asked for extra copies and this block group
+			 * doesn't provide them, bail.  This does allow us to
+			 * fill raid0 from raid1.
+			 */
+			if ((data & extra) && !(block_group->flags & extra))
+				goto loop;
+		}
+
 have_block_group:
 		if (unlikely(block_group->cached == BTRFS_CACHE_NO)) {
 			u64 free_percent;
@@ -8273,7 +8292,6 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 			break;
 		if (ret != 0)
 			goto error;
-
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
 		cache = kzalloc(sizeof(*cache), GFP_NOFS);
-- 
cgit v1.2.2


From 3cb50ddf97a0a1ca4c68bc12fa1e727a6b45fbf2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ZenIV.linux.org.uk>
Date: Mon, 20 Dec 2010 15:53:18 +0000
Subject: Fix btrfs b0rkage

Buggered-in: 76dda93c6ae2 ("Btrfs: add snapshot/subvolume destroy
ioctl")

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/btrfs/export.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/btrfs')

diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 6f044447359..659f532d26a 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -166,7 +166,7 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh,
 static struct dentry *btrfs_get_parent(struct dentry *child)
 {
 	struct inode *dir = child->d_inode;
-	static struct dentry *dentry;
+	struct dentry *dentry;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
-- 
cgit v1.2.2