Everytime we write out dirty pages we search for an offset in the tree,
convert the bits in the state, and then when we wait we search for the
offset again and clear the bits.  So for every dirty range in the io tree we
are doing 4 rb searches, which is suboptimal.  With this patch we are only
doing 2 searches for every cycle (modulo weird things happening).  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

For some reason we unlock everything except the leaf we are on, set the path
blocking and then add the extent item for the extent we just finished
writing.  I can't for the life of me figure out why we would want to do
this, and the history doesn't really indicate that there was a real reason
for it, so just remove it.  This will reduce our tree lock contention on
heavy writes.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

There are a coule scenarios where farming metadata csumming off to an async
thread doesn't help.  The first is if our processor supports crc32c, in
which case the csumming will be fast and so the overhead of the async model
is not worth the cost.  The other case is for our tree log.  We will be
making that stuff dirty and writing it out and waiting for it immediately.
Even with software crc32c this gives me a ~15% increase in speed with O_SYNC
workloads.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

commit 7ca4be45 limited csum items to
PAGE_CACHE_SIZE.  It used min() with incompatible types in 32bit which
generates warnings:

fs/btrfs/file-item.c: In function ‘btrfs_csum_file_blocks’:
fs/btrfs/file-item.c:717: warning: comparison of distinct pointer types lacks a cast

This uses min_t(u32,) to fix the warnings.  u32 seemed reasonable
because btrfs_root->leafsize is u32 and PAGE_CACHE_SIZE is unsigned
long.
Signed-off-by: NZach Brown <zab@zabbo.net>

Running delayed refs is faster than running delalloc, so lets do that first
to try and reclaim space.  This makes my fs_mark test about 20% faster.
Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

With the following debug patch:

 static int btrfs_freeze(struct super_block *sb)
 {
+ 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
+	struct btrfs_transaction *trans;
+
+	spin_lock(&fs_info->trans_lock);
+	trans = fs_info->running_transaction;
+	if (trans) {
+		printk("Transid %llu, use_count %d, num_writer %d\n",
+			trans->transid, atomic_read(&trans->use_count),
+			atomic_read(&trans->num_writers));
+	}
+	spin_unlock(&fs_info->trans_lock);
 	return 0;
 }

I found there was a orphan transaction after the freeze operation was done.

It is because the transaction may not be committed when the transaction handle
end even though it is the last handle of the current transaction. This design
avoid committing the transaction frequently, but also introduce the above
problem.

So I add btrfs_attach_transaction() which can catch the current transaction
and commit it. If there is no transaction, it will return ENOENT, and do not
anything.

This function also can be used to instead of btrfs_join_transaction_freeze()
because it don't increase the writer counter and don't start a new transaction,
so it also can fix the deadlock between sync and freeze.

Besides that, it is used to instead of btrfs_join_transaction() in
transaction_kthread(), because if there is no transaction, the transaction
kthread needn't anything.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

This patch add a type field into the transaction handle structure,
in this way, we needn't implement various end-transaction functions
and can make the code more simple and readable.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

This patch fixes memory leak of the transaction handle which happened
when starting transaction failed on a freezed fs.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

The iterate_irefs in backref.c is used to build path components from inode
refs. This patch adds code to iterate extended refs as well.

I had modify the callback function signature to abstract out some of the
differences between ref structures. iref_to_path() also needed similar
changes.
Signed-off-by: NMark Fasheh <mfasheh@suse.de>

This patch adds basic support for extended inode refs. This includes support
for link and unlink of the refs, which basically gets us support for rename
as well.

Inode creation does not need changing - extended refs are only added after
the ref array is full.
Signed-off-by: NMark Fasheh <mfasheh@suse.de>

Moved part of the code into a sub function and replaced most of the gotos
by ifs, hoping that it will be easier to read now.
Signed-off-by: NJan Schmidt <list.btrfs@jan-o-sch.net>
Signed-off-by: NMark Fasheh <mfasheh@suse.de>

I started hitting warnings when running xfstest 68 in a loop because there
were EM's that were not lined up properly with the physical extents.  This
is ok, if we do something like punch a hole or write to a preallocated space
or something like that we can have an EM that doesn't cover the entire
physical extent.  So fix the tree logging stuff to cope with this case so we
don't just commit the transaction.  With this patch I no longer see the
warnings from the tree logging code.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

Call btrfs_abort_transaction as early as possible when an error
condition is detected, that way the line number reported is useful
and we're not clueless anymore which error path led to the abort.
Signed-off-by: NDavid Sterba <dsterba@suse.cz>

The macro btrfs_abort_transaction() can get the line number of the code
where the problem happens, so we should invoke it in the place that the
error occurs, or we will lose the line number.
Reported-by: NDavid Sterba <dave@jikos.cz>
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

Btrfs uses inclusive range end for lock_extent(), unlock_extent() and
related functions, so we made off-by-one errors in file clone.

This fixes it and also fixes some style problems.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>

Hi,

the patch si simple, but it has user visible impact and I'm not quite sure how
to resolve it.

In short, $subj says it, chattr -C supports it and we want to use it.

The conditions that acutally allow to change the NOCOW flag are clear. What if
I try to set the flag on a file that is not empty? Options:

1) whole ioctl will fail, EINVAL
2.1) ioctl will succeed, the NOCOW flag will be silently removed, but the file
     will stay COW-ed and checksummed
2.2) ioctl will succeed, flag will not be removed and a syslog message will
     warn that the COW flag has not been changed
2.2.1) dtto, no syslog message

Man page of chattr states that

 "If it is set on a file which already has data blocks, it is undefined when
 the blocks assigned to the file will be fully stable."

Yes, it's undefined and with current implementation it'll never happen. So from
this end, the user cannot expect anything. I'm trying to find a reasonable
behaviour, so that a command like 'chattr -R -aijS +C' to tweak a broad set of
flags in a deep directory does not fail unnecessarily and does not pollute the
log.

My personal preference is 2.2.1, but my dev's oppinion is skewed, not counting
the fact that I know the code and otherwise would look there before consulting
the documentation.

The patch implements 2.2.1.

david

-------------8<-------------------
From: David Sterba <dsterba@suse.cz>

It's safe to turn off checksums for a zero sized file.

http://thread.gmane.org/gmane.comp.file-systems.btrfs/18030

"We cannot switch on NODATASUM for a file that already has extents that
are checksummed. The invariant here is that either all the extents or
none are checksummed.

Theoretically it's possible to add/remove all checksums from a given
file, but it's a potentially longtime operation, the file has to be in
some intermediate state where the checksums partially exist but have to
be ignored (for the csum->nocsum) until the file is fully converted,
this brings more special cases to extent handling, it has to survive
power failure and remain consistent, and probably needs to be restarted
after next mount."
Signed-off-by: NDavid Sterba <dsterba@suse.cz>

I saw the warning in btrfs_drop_extent_cache where our end is less than our
start while running xfstests 68 in a loop.  This is because we
unconditionally do drop_end = min(end, extent_end) in
__btrfs_drop_extents(), even though we may not have found an extent in the
range we were looking to drop.  So keep track of wether or not we found
something, and if we didn't just use our end.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

We do not need to do anything special to freeze or unfreeze, it's all taken
care of by the generic work, and what we currently have is wrong anyway
since we shouldn't be returnning to userspace with mutexes held anyway.
Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

The btree inode has it's own write cache pages so we can remove this write
cache pages hook as it's not used.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

We can race when checking wether PagePrivate is set on a page and we
actually have an eb saved in the pages private pointer.  We could have
easily written out this page and released it in the time that we did the
pagevec lookup and actually got around to looking at this page.  So use
mapping->private_lock to ensure we get a consistent view of the
page->private pointer.  This is inline with the alloc and releasepage paths
which use private_lock when manipulating page->private.  Thanks,
Reported-by: NDavid Sterba <dave@jikos.cz>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

Dave Sterba pointed out a sleeping while atomic bug while doing fsync.  This
is because I'm an idiot and didn't realize that rwlock's were spin locks, so
we've been holding this thing while doing allocations and such which is not
good.  This patch fixes this by dropping the write lock before we do
anything heavy and re-acquire it when it is done.  We also need to take a
ref on the em's in case their corresponding pages are evicted and mark them
as being logged so that releasepage does not remove them and doesn't remove
them from our local list.  Thanks,
Reported-by: NDave Sterba <dave@jikos.cz>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

So we start our freeze, somebody comes in and does an fsync() on a file
where we have to commit a transaction for whatever reason, and we will
deadlock because the freeze is waiting on FS_FREEZE people to stop writing
to the file system, but the transaction is waiting for its free space inodes
to be written out, which are in turn waiting on sb_start_intwrite while
trying to write the file extents.  To fix this we'll just skip the
sb_start_intwrite() if we TRANS_JOIN_NOLOCK since we're being waited on by a
transaction commit so we're safe wrt to freeze and this will keep us from
deadlocking.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

nocow_only is now an obsolete argument.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>

fs_info->hashers is now an obsolete one.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>

There is already an 'add free space' phrase in front of this one, we
needn't to redo it.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>

I screwed this up, there is a race between checking if there is a running
transaction and actually starting a transaction in sync where we could race
with a freezer and get ourselves into trouble.  To fix this we need to make
a new join type to only do the try lock on the freeze stuff.  If it fails
we'll return EPERM and just return from sync.  This fixes a hang Liu Bo
reported when running xfstest 68 in a loop.  Thanks,
Reported-by: NLiu Bo <bo.li.liu@oracle.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

A subvolume cannot be deleted via rmdir, but the error code ENOTEMPTY
is confusing. Return EPERM instead, as this is not permitted.
Signed-off-by: NDavid Sterba <dsterba@suse.cz>

Using for_each_set_bit_from() to simplify the code.

spatch with a semantic match is used to found this.
(http://coccinelle.lip6.fr/)
Signed-off-by: NWei Yongjun <yongjun_wei@trendmicro.com.cn>

Developing service cmds needs it.
Signed-off-by: NAnand Jain <anand.jain@oracle.com>

Unnecessary lookup_extent_mapping() is removed because an error is
returned to the caller.
This patch was made based on the advice from Stefan Behrens, thanks.
Signed-off-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>

This patch simplifies a little complex error processing in
btree_get_extent().
Signed-off-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>

This reverts commit 0885ef5b

After applying the above patch, the performance slowed down because the dirty
page flush can only be done by one task, so revert it.

The following is the test result of sysbench:
	Before		After
	24MB/s		39MB/s
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

Everybody is just making stuff up, and it's just used to see if we really do
need to alloc a chunk, and since we do this when we already know we really
do it's just a waste of space.  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

So we have lots of places where we try to preallocate chunks in order to
make sure we have enough space as we make our allocations.  This has
historically meant that we're constantly tweaking when we should allocate a
new chunk, and historically we have gotten this horribly wrong so we way
over allocate either metadata or data.  To try and keep this from happening
we are going to make it so that the block group item insertion is done out
of band at the end of a transaction.  This will allow us to create chunks
even if we are trying to make an allocation for the extent tree.  With this
patch my enospc tests run faster (didn't expect this) and more efficiently
use the disk space (this is what I wanted).  Thanks,
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

For immutable bio vecs, I've been auditing and removing bi_idx
references. These were harmless, but removing them will make auditing
easier.

scrub_bio_end_io_worker() was open coding a bio_reset() - but this
doesn't appear to have been needed for anything as right after it does a
bio_put(), and perusing the code it doesn't appear anything else was
holding a reference to the bio.

The other use end_bio_extent_readpage() was just for a pr_debug() -
changed it to something that might be a bit more useful.
Signed-off-by: NKent Overstreet <koverstreet@google.com>
CC: Chris Mason <chris.mason@oracle.com>
CC: Stefan Behrens <sbehrens@giantdisaster.de>

When we wrote some data by compress mode into a btrfs filesystem which was full
of the fragments, the kernel will report:
	BTRFS warning (device xxx): Aborting unused transaction.

The reason is:
We can not find a long enough free space to store the compressed data because
of the fragmentary free space, and the compressed data can not be splited,
so the kernel outputed the above message.

In fact, btrfs can deal with this problem very well: it fall back to
uncompressed IO, split the uncompressed data into small ones, and then
store them into to the fragmentary free space. So we shouldn't output the
above warning message.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>

Wade Cline reported a problem where he was getting garbage and warnings when
writing to a preallocated range via O_DIRECT.  This is because we weren't
creating our normal pinned extent_map for the range we were writing to,
which was causing all sorts of issues.  This patch fixes the problem and
makes his testcase much happier.  Thanks,
Reported-by: NWade Cline <clinew@linux.vnet.ibm.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

Sage reported the following lockdep backtrace

=====================================
[ BUG: bad unlock balance detected! ]
3.6.0-rc2-ceph-00171-gc7ed62d #1 Not tainted
-------------------------------------
btrfs-cleaner/7607 is trying to release lock (sb_internal) at:
[<ffffffffa00422ae>] btrfs_commit_transaction+0xa6e/0xb20 [btrfs]
but there are no more locks to release!

other info that might help us debug this:
1 lock held by btrfs-cleaner/7607:
 #0:  (&fs_info->cleaner_mutex){+.+...}, at: [<ffffffffa003b405>] cleaner_kthread+0x95/0x120 [btrfs]

stack backtrace:
Pid: 7607, comm: btrfs-cleaner Not tainted 3.6.0-rc2-ceph-00171-gc7ed62d #1
Call Trace:
 [<ffffffffa00422ae>] ? btrfs_commit_transaction+0xa6e/0xb20 [btrfs]
 [<ffffffff810afa9e>] print_unlock_inbalance_bug+0xfe/0x110
 [<ffffffff810b289e>] lock_release_non_nested+0x1ee/0x310
 [<ffffffff81172f9b>] ? kmem_cache_free+0x7b/0x160
 [<ffffffffa004106c>] ? put_transaction+0x8c/0x130 [btrfs]
 [<ffffffffa00422ae>] ? btrfs_commit_transaction+0xa6e/0xb20 [btrfs]
 [<ffffffff810b2a95>] lock_release+0xd5/0x220
 [<ffffffff81173071>] ? kmem_cache_free+0x151/0x160
 [<ffffffff8117d9ed>] __sb_end_write+0x7d/0x90
 [<ffffffffa00422ae>] btrfs_commit_transaction+0xa6e/0xb20 [btrfs]
 [<ffffffff81079850>] ? __init_waitqueue_head+0x60/0x60
 [<ffffffff81634c6b>] ? _raw_spin_unlock+0x2b/0x40
 [<ffffffffa0042758>] __btrfs_end_transaction+0x368/0x3c0 [btrfs]
 [<ffffffffa0042808>] btrfs_end_transaction_throttle+0x18/0x20 [btrfs]
 [<ffffffffa00318f0>] btrfs_drop_snapshot+0x410/0x600 [btrfs]
 [<ffffffff8132babd>] ? do_raw_spin_unlock+0x5d/0xb0
 [<ffffffffa00430ef>] btrfs_clean_old_snapshots+0xaf/0x150 [btrfs]
 [<ffffffffa003b405>] ? cleaner_kthread+0x95/0x120 [btrfs]
 [<ffffffffa003b419>] cleaner_kthread+0xa9/0x120 [btrfs]
 [<ffffffffa003b370>] ? btrfs_destroy_delayed_refs.isra.102+0x220/0x220 [btrfs]
 [<ffffffff810791ee>] kthread+0xae/0xc0
 [<ffffffff810b379d>] ? trace_hardirqs_on+0xd/0x10
 [<ffffffff8163e744>] kernel_thread_helper+0x4/0x10
 [<ffffffff81635430>] ? retint_restore_args+0x13/0x13
 [<ffffffff81079140>] ? flush_kthread_work+0x1a0/0x1a0
 [<ffffffff8163e740>] ? gs_change+0x13/0x13

This is because the throttle stuff can commit the transaction, which expects to
be the one stopping the intwrite stuff, but we've already done it in the
__btrfs_end_transaction.  Moving the sb_end_intewrite after this logic makes the
lockdep go away.  Thanks,
Tested-by: NSage Weil <sage@inktank.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

This is the change of the kernel side.

Translation of logical to inode used to have an upper limit 4k on
inode container's size, but the limit is not large enough for a data
with a great many of refs, so when resolving logical address,
we can end up with
"ioctl ret=0, bytes_left=0, bytes_missing=19944, cnt=510, missed=2493"

This changes to regard 64k as the upper limit and use vmalloc instead of
kmalloc to get memory more easily.
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>

We already have a helper, iterate_inodes_from_logical(), for logical resolve,
so just use it.
Signed-off-by: NLiu Bo <bo.li.liu@oracle.com>