Hello everyone,

Btrfs v0.10 is now available for download from:

http://oss.oracle.com/projects/btrfs/

Btrfs is still in an early alpha state, and the disk format is not finalized.
v0.10 introduces a new disk format, and is not compatible with v0.9.

The core of this release is explicit back references for all metadata blocks,
data extents, and directory items.  These are a crucial building block for
future features such as online fsck and migration between devices.  The back
references are verified during deletes, and the extent back references are
checked by the existing offline fsck tool.

For all of the details of how the back references are maintained, please
see the design document:

http://oss.oracle.com/projects/btrfs/dist/documentation/btrfs-design.html

Other new features (described in detail below):

* Online resizing (including shrinking)
* In place conversion from Ext3 to Btrfs
* data=ordered support
* Mount options to disable data COW and checksumming 
* Barrier support for sata and IDE drives

[ Resizing ]

In order to demonstrate and test the back references, I've added an online
resizer, which can both grow and shrink the filesystem:

mount -t btrfs /dev/xxx /mnt

# add 2GB to the FS
btrfsctl -r +2g /mnt

# shrink the FS by 4GB
btrfsctl -r -4g /mnt

# Explicitly set the FS size
btrfsctl -r 20g /mnt

# Use 'max' to grow the FS to the limit of the device
btrfsctl -r max /mnt

[ Conversion from Ext3 ]

This is an offline, in place, conversion program written by Yan Zheng.  It
has been through basic testing, but should not be trusted with critical data.

To build the conversion program, run 'make convert' in the btrfs-progs
tree. It depends on libe2fs and acl development libraries.

The conversion program uses the copy on write nature of Btrfs to preserve the
original Ext3 FS, sharing the data blocks between Btrfs and Ext3 metadata.
Btrfs metadata is created inside the free space of the Ext3 filesystem, and it
is possible to either make th...
[ message continues ]

Well, it turns out this release had a few small problems:

* data=ordered deadlock on older kernels (including 2.6.23)
* Compile problems when ACLs were not enabled in the kernel

So, I've put v0.11 out there.  It fixes those two problems and will also 
compile on older (2.6.18) enterprise kernels.

v0.11 does not have any disk format changes.

-chris


-

Ok, back to the suspend problem I mentioned:

Jan 18 00:04:40 revo WARNING: at fs/btrfs/tree-defrag.c:74 defrag_walk_down()
Jan 18 00:04:40 revo Pid: 258, comm: btrfs/0 Not tainted 2.6.24-rc8 #1
Jan 18 00:04:40 revo [<b01ddef3>] btrfs_defrag_leaves+0x273/0x8c0
Jan 18 00:04:40 revo [<b01d5b57>] btrfs_defrag_root+0x67/0xe0
Jan 18 00:04:40 revo [<b01d5c43>] btrfs_defrag_dirty_roots+0x73/0x80
Jan 18 00:04:40 revo [<b01d5c50>] btrfs_transaction_cleaner+0x0/0xd0
Jan 18 00:04:40 revo [<b01d5cf2>] btrfs_transaction_cleaner+0xa2/0xd0
Jan 18 00:04:40 revo [<b013911d>] run_workqueue+0xad/0x130
Jan 18 00:04:40 revo [<b0139b80>] worker_thread+0x0/0xf0
Jan 18 00:04:40 revo [<b0139c0c>] worker_thread+0x8c/0xf0
Jan 18 00:04:40 revo [<b013d100>] autoremove_wake_function+0x0/0x40
Jan 18 00:04:40 revo [<b0139b80>] worker_thread+0x0/0xf0
Jan 18 00:04:40 revo [<b013ce42>] kthread+0x42/0x70
Jan 18 00:04:40 revo [<b013ce00>] kthread+0x0/0x70
Jan 18 00:04:40 revo [<b0104ea7>] kernel_thread_helper+0x7/0x10
Jan 18 00:04:40 revo =======================
Jan 18 00:04:40 revo BUG: unable to handle kernel NULL pointer dereference at 
virtual address 0000001c
Jan 18 00:04:40 revo printing eip: b01d3c58 *pde = 00000000
Jan 18 00:04:40 revo Oops: 0000 [#1] SMP
Jan 18 00:04:40 revo Modules linked in: iwl3945
Jan 18 00:04:40 revo
Jan 18 00:04:40 revo Pid: 258, comm: btrfs/0 Not tainted (2.6.24-rc8 #1)
Jan 18 00:04:40 revo EIP: 0060:[<b01d3c58>] EFLAGS: 00010292 CPU: 0
Jan 18 00:04:40 revo EIP is at btrfs_clear_buffer_defrag+0x18/0x80
Jan 18 00:04:40 revo EAX: 00000000 EBX: 00000000 ECX: eec6d528 EDX: eec6d528
Jan 18 00:04:40 revo ESI: ee544200 EDI: ee40c000 EBP: 00000040 ESP: ee40de8c
Jan 18 00:04:40 revo DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Jan 18 00:04:40 revo Process btrfs/0 (pid: 258, ti=ee40c000 task=ef0ab570 
task.ti=ee40c000)
Jan 18 00:04:40 revo Stack: 00100fff 00000000 00000040 00000000 d343bdbc 
eec37634 ee544200 ...
[ message continues ]

Looks like metadata corruption.  How are you suspending?

-chris



-

No, that's not metadata corruption. After reset I can use the image without 
problems. After resume the first access to the fs oopses.
I use kernel 2.6.24-rc8 with tuxonice 3.0-rc3 and btrfs 0.11.

Back in early december I reported the problem for btrfs 0.9. Seems like the 
lockfs call still is not implemented. Any hints what I need when I try to 
code it myself?
-- 
Regards,
Chris
-

Please try this dirty patch.  I think it can solve your problem.

Regards
YZ
---
diff -r ac53d7df4c11 super.c
--- a/super.c	Thu Jan 17 12:58:00 2008 -0500
+++ b/super.c	Mon Jan 21 17:10:00 2008 +0800
@@ -423,6 +423,18 @@ static struct file_system_type btrfs_fs_
 	.fs_flags	= FS_REQUIRES_DEV,
 };

+static void btrfs_write_super_lockfs(struct super_block *sb)
+{
+	struct btrfs_root *root = btrfs_sb(sb);
+	btrfs_transaction_flush_work(root);
+}
+
+static void btrfs_unlockfs(struct super_block *sb)
+{
+	struct btrfs_root *root = btrfs_sb(sb);
+	btrfs_transaction_queue_work(root, HZ * 30);
+}
+
 static struct super_operations btrfs_super_ops = {
 	.delete_inode	= btrfs_delete_inode,
 	.put_inode	= btrfs_put_inode,
@@ -435,6 +447,8 @@ static struct super_operations btrfs_sup
 	.alloc_inode	= btrfs_alloc_inode,
 	.destroy_inode	= btrfs_destroy_inode,
 	.statfs		= btrfs_statfs,
+	.write_super_lockfs = btrfs_write_super_lockfs,
+	.unlockfs 	= btrfs_unlockfs,
 };

 static int __init init_btrfs_fs(void)
-

Looks good... No oops so far. ;) Thanks a lot!
-- 
Regards,
Chris
-

Hi Chris,

First, massive congratulations for bringing this to fruition in such a
short time.

Now back to the regular carping: why even support older kernels?

Regards,

Daniel
-

The general answer is the backports are small and easy.  I don't test them 
heavily, and I don't go out of my way to make things work. 

But, they do make it easier for people to try out, and to figure how to use 
all these new features to solve problems.  Small changes that enable more 
testers are always welcome.

In general, the core parts of the kernel that btrfs uses haven't had many 
interface changes since 2.6.18, so this isn't a huge deal.

-chris
-

It does not even compile for me, tested with 2.6.24-rc{7,8}. I will look at 
that later.

fs/built-in.o: In function `btrfs_xattr_set_acl':
acl.c:(.text+0x68f33): undefined reference to `posix_acl_from_xattr'
acl.c:(.text+0x68f47): undefined reference to `posix_acl_valid'
make: *** [.tmp_vmlinux1] Error 1

Is this release supposed to fix the suspend problem?
-- 
Regards,
Chris
-

See this build fix for kernels without acl by Yan Zheng on the
btrfs-devel list

http://oss.oracle.com/pipermail/btrfs-devel/2008-January/000386.html


Simon Holm Thøgersen

-

Looks like fun. btrfsck fails to check if it actually received a
dev argument though, so if you don't pass a device, we get a nice
segfault.

Signed-off-by: Kyle McMartin <kmcmartin@redhat.com>

---
diff -Nur btrfs-progs-0.10/btrfsck.c btrfs-progs-0.10-kyle/btrfsck.c
--- btrfs-progs-0.10/btrfsck.c	2008-01-15 10:33:32.000000000 -0500
+++ btrfs-progs-0.10-kyle/btrfsck.c	2008-01-15 11:49:24.000000000 -0500
@@ -709,6 +709,11 @@
 	return err;
 }
 
+void print_usage(void) {
+	fprintf(stderr, "usage: btrfsck dev\n");
+	exit(1);
+}
+
 int main(int ac, char **av) {
 	struct btrfs_root *root;
 	struct cache_tree extent_cache;
@@ -727,6 +732,9 @@
 	int slot;
 	struct btrfs_root_item ri;
 
+	if (ac < 2)
+		print_usage();
+
 	radix_tree_init();
 	cache_tree_init(&extent_cache);
 	cache_tree_init(&seen);
-