For some reason we have been waiting for lazyinit thread to start in the
ext4_run_lazyinit_thread() but it is not needed since it was jus
unnecessary complexity, so get rid of it. We can also remove li_task and
li_wait_task since it is not used anymore.
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: NEric Sandeen <sandeen@redhat.com>

In order to make lazyinit eat approx. 10% of io bandwidth at max, we
are sleeping between zeroing each single inode table. For that purpose
we are using timer which wakes up thread when it expires. It is set
via add_timer() and this may cause troubles in the case that thread
has been woken up earlier and in next iteration we call add_timer() on
still running timer hence hitting BUG_ON in add_timer(). We could fix
that by using mod_timer() instead however we can use
schedule_timeout_interruptible() for waiting and hence simplifying
things a lot.

This commit exchange the old "waiting mechanism" with simple
schedule_timeout_interruptible(), setting the time to sleep. Hence we
do not longer need li_wait_daemon waiting queue and others, so get rid
of it.

Addresses-Red-Hat-Bugzilla: #699708
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: NEric Sandeen <sandeen@redhat.com>

In order to stabilize pages during disk writes, ext4_page_mkwrite must
wait for writeback operations to complete before making a page
writable.  Furthermore, the function must return locked pages, and
recheck the writeback status if the page lock is ever dropped.  The
"someone could wander in" part of this patch was suggested by Chris
Mason.
Signed-off-by: NDarrick J. Wong <djwong@us.ibm.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

wait_on_page_writeback already checks the writeback bit, so callers of it
needn't do that test.
Signed-off-by: NDarrick J. Wong <djwong@us.ibm.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Currently, if we mkfs a new ext4 volume with s_max_mnt_count set to
zero, and mount it for the first time, we will get the warning:

	maximal mount count reached, running e2fsck is recommended

It is really misleading. So change the check so that it won't warn in
that case.
Signed-off-by: NTao Ma <boyu.mt@taobao.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

This patch addresses bugs found while testing punch hole 
with the fsx test.  The patch corrects the number of blocks
that are zeroed out while splitting an extent, and also corrects
the return value to return the number of blocks split out, instead
of the number of blocks zeroed out.

This patch has been tested in addition to the following patches: 
[Ext4 punch hole v7]
[XFS Tests Punch Hole 1/1 v2] Add Punch Hole Testing to FSX

The test ran successfully for 24 hours.
Signed-off-by: NAllison Henderson <achender@us.ibm.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

If quota is not enabled when ext4_quota_off() is called, we must not
dereference quota file inode since it is NULL.  Check properly for
this.

This fixes a bug in commit 21f97697 (ext4: remove unnecessary
[cm]time update of quota file), which was merged for 2.6.39-rc3.
Reported-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Fix for a null pointer bug found while running punch hole tests
Signed-off-by: NAllison Henderson <achender@us.ibm.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

After taking care of all group init races, all that remains is to
remove alloc_semp from ext4_allocation_context and ext4_buddy structs.
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

After online resize which adds new groups, some of the groups
in a buddy page may be initialized and uptodate, while other
(new ones) may be uninitialized.

The indication for init of new block groups is when ext4_mb_init_cache()
is called with an uptodate buddy page. In this case, initialized groups
on that buddy page must be skipped when initializing the buddy cache.
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

The old routines ext4_mb_[get|put]_buddy_cache_lock(), which used
to take grp->alloc_sem for all groups on the buddy page have been
replaced with the routines ext4_mb_[get|put]_buddy_page_lock().

The new routines take both buddy and bitmap page locks to protect
against concurrent init of groups on the same buddy page.

The GROUP_NEED_INIT flag is tested again under page lock to check
if the group was initialized by another caller.
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

The old imlementation used to take grp->alloc_sem and set the
GROUP_NEED_INIT flag, so that the buddy cache would be reloaded.

The new implementation updates the buddy cache by freeing the added
blocks and making them available for use, so there is no need to
reload the buddy cache and there is no need to take grp->alloc_sem.
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

The block allocation code used to use jbd2_journal_get_undo_access as
a way to make changes that wouldn't show up until the commit took
place.  The new multi-block allocation code has a its own way of
preventing newly freed blocks from getting reused until the commit
takes place (it avoids updating the buddy bitmaps until the commit is
done), so we don't need to use jbd2_journal_get_undo_access(), which
has extra overhead compared to jbd2_journal_get_write_access().

There was one last vestigal use of ext4_journal_get_undo_access() in
ext4_add_groupblocks(); change it to use ext4_journal_get_write_access()
and then remove the ext4_journal_get_undo_access() support.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

In preparation for the next patch, the function ext4_add_groupblocks()
is moved to mballoc.c, where it could use some static functions.
Signed-off-by: NAmir Goldstein <amir73il@users.sf.net>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

There is already an #ifdef CONFIG_QUOTA some lines above,
so this one is totally useless.
Signed-off-by: NWANG Cong <amwang@redhat.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

We have checked first_not_zeroed == ngroups already above, so remove
this redundant check.

sbi->s_li_request = NULL above is also removed since it is NULL
already.

Cc: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: NTao Ma <boyu.mt@taobao.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

In __ext4_get_inode_loc, we calculate inodes_per_block every time by
EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb).  AFAICS, this function is a
hot path for ext4, so we'd better use s_inodes_per_block directly
instead of calculating every time.
Signed-off-by: NTao Ma <boyu.mt@taobao.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

We have EXT4FS_DEBUG for some old debug and CONFIG_EXT4_DEBUG
for the new mballoc debug, but there isn't any EXT4_DEBUG.

As CONFIG_EXT4_DEBUG seems to be only used in mballoc, use
EXT4FS_DEBUG in fsync.c.

[ It doesn't really matter; although I'm including this commit for
  consistency's sake.  The whole point of the #ifdef's is to disable
  the debugging code.  In general you're not going to want to enable
  all of the code protected by EXT4FS_DEBUG at the same time.  -- Ted ]
Signed-off-by: NTao Ma <boyu.mt@taobao.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

If we somehow wrap, we don't want to keep printing the warning message
over and over again.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

In do_get_write_access() we wait on BH_Unshadow bit for buffer to get
from shadow state. The waking code in journal_commit_transaction() has
a bug because it does not issue a memory barrier after the buffer is
moved from the shadow state and before wake_up_bit() is called. Thus a
waitqueue check can happen before the buffer is actually moved from
the shadow state and waiting process may never be woken. Fix the
problem by issuing proper barrier.
Reported-by: NTao Ma <boyu.mt@taobao.com>
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Reimplement ext4_ext_convert_to_initialized() and
ext4_split_unwritten_extents() using ext4_split_extent()
Signed-off-by: NYongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Tested-by: NAllison Henderson <achender@linux.vnet.ibm.com>

Add two functions: ext4_split_extent_at(), which splits an extent into
two extents at given logical block, and ext4_split_extent() which
splits an extent into three extents.
Signed-off-by: NYongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Tested-by: NAllison Henderson <achender@linux.vnet.ibm.com>

1) Rename ext4_ext_try_to_merge() to ext4_ext_try_to_merge_right().

2) Add a new function ext4_ext_try_to_merge() which tries to merge
   an extent both left and right.

3) Use the new function in ext4_ext_convert_unwritten_endio() and
   ext4_ext_insert_extent().
Signed-off-by: NYongqiang Yang <xiaoqiangnk@gmail.com>
Tested-by: NAllison Henderson <achender@linux.vnet.ibm.com>

ext4_symlink() cannot call __page_symlink() with transaction open.
__page_symlink() calls ext4_write_begin() which can wait for
transaction commit if we are running out of space thus causing a
deadlock. Also error recovery in ext4_truncate_failed_write() does not
count with the transaction being already started (although I'm not
aware of any particular deadlock here).

Fix the problem by stopping a transaction before calling
__page_symlink() (we have to be careful and put inode to orphan list
so that it gets deleted in case of crash) and starting another one
after __page_symlink() returns for addition of symlink into a
directory.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

When make_indexed_dir() fails (e.g. because of ENOSPC) after it has
allocated block for index tree root, we did not properly mark all
changed buffers dirty.  This lead to only some of these buffers being
written out and thus effectively corrupting the directory.

Fix the issue by marking all changed data dirty even in the error
failure case.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Fix a typo that was introduced in commit 07a03824 (in 2.6.36) which
caused the extents flag not to be set at the conclusion of converting
an inode to use extents.
Reported-by: NPeter Uchno <peter.uchno@gmail.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

If an application program does not make any changes to the indirect
blocks or extent tree, i_datasync_tid will not get updated.  If there
are enough commits (i.e., 2**31) such that tid_geq()'s calculations
wrap, and there isn't a currently active transaction at the time of
the fdatasync() call, this can end up triggering a BUG_ON in
fs/jbd2/commit.c:

	J_ASSERT(journal->j_running_transaction != NULL);

It's pretty rare that this can happen, since it requires the use of
fdatasync() plus *very* frequent and excessive use of fsync().  But
with the right workload, it can.

We fix this by replacing the use of tid_geq() with an equality test,
since there's only one valid transaction id that we is valid for us to
wait until it is commited: namely, the currently running transaction
(if it exists).
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

percpu_counter_sum_positive() never returns a negative value.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

This is an effective revert of commit a30eec2a: "ext4: stop issuing
discards if not supported by device".  The problem is that there are
some devices that may return errors in response to a discard request
some times but not others.  (One example would be a hybrid dm device
which concatenates an SSD and an HDD device).

By this logic, I also removed the error checking from ext4's FITRIM
code; so that an error from a discard will not stop the FITRIM from
trying to trim the rest of the file system.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

In the bio completion routine, we should not be setting
PageUptodate at all -- it's set at sys_write() time, and is
unaffected by success/failure of the write to disk.

This can cause a page corruption bug when the file system's
block size is less than the architecture's VM page size.

if we have only written a single block -- we might end up
setting the page's PageUptodate flag, indicating that page
is completely read into memory, which may not be true.
This could cause subsequent reads to get bad data.

This commit also takes the opportunity to clean up error
handling in ext4_end_bio(), and remove some extraneous code:

   - fixes ext4_end_bio() to set AS_EIO in the
     page->mapping->flags on error, which was left out by
     mistake.  This is needed so that fsync() will
     return an error if there was an I/O error.
   - remove the clear_buffer_dirty() call on unmapped
     buffers for each page.
   - consolidate page/buffer error handling in a single
     section.
Signed-off-by: NCurt Wohlgemuth <curtw@google.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Reported-by: NJim Meyering <jim@meyering.net>
Reported-by: NHugh Dickins <hughd@google.com>
Cc: Mingming Cao <cmm@us.ibm.com>

Provide better emulation for ext[23] mode by enforcing that the file
system does not have any unsupported file system features as defined
by ext[23] when emulating the ext[23] file system driver when
CONFIG_EXT4_USE_FOR_EXT23 is defined.

This causes the file system type information in /proc/mounts to be
correct for the automatically mounted root file system.  This also
means that "mount -t ext2 /dev/sda /mnt" will fail if /dev/sda
contains an ext3 or ext4 file system, just as one would expect if the
original ext2 file system driver were in use.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Add missing page_cache_release in the error path of ext4_mb_load_buddy
Signed-off-by: NYang Ruirui <ruirui.r.yang@tieto.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org

ia64 throws away .exit sections for the built-in CONFIG case, so routines
that are used in other circumstances should not be tagged as __exit.
Signed-off-by: NTony Luck <tony.luck@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Revert commit 6de9843d, since it
caused a data corruption regression with BitTorrent downloads.  Thanks
to Damien for discovering and bisecting to find the problem commit.

https://bugzilla.kernel.org/show_bug.cgi?id=32972Reported-by: NDamien Grassart <damien@grassart.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

We can create 4402345721856 byte file with indirect block mapping.
However, if we grow an indirect-block file to the size with ftruncate(),
we can see an ext4 warning. The following patch fixes this problem.

How to reproduce:
# dd if=/dev/zero of=/mnt/mp1/hoge bs=1 count=0 seek=4402345721856
0+0 records in
0+0 records out
0 bytes (0 B) copied, 0.000221428 s, 0.0 kB/s
# tail -n 1 /var/log/messages
Nov 25 15:10:27 test kernel: EXT4-fs warning (device sda8): ext4_block_to_path:345: block 1074791436 > max in inode 12
Signed-off-by: NKazuya Mio <k-mio@sx.jp.nec.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

ext4_journal_start_sb() should not prevent an active handle from being
started due to s_frozen.  Otherwise, deadlock is easy to happen, below
is a situation.

================================================
     freeze         |       truncate
================================================
                    |  ext4_ext_truncate()
    freeze_super()  |   starts a handle
    sets s_frozen   |
                    |  ext4_ext_truncate()
                    |  holds i_data_sem
  ext4_freeze()     |
  waits for updates |
                    |  ext4_free_blocks()
                    |  calls dquot_free_block()
                    |
                    |  dquot_free_blocks()
                    |  calls ext4_dirty_inode()
                    |
                    |  ext4_dirty_inode()
                    |  trys to start an active
                    |  handle
                    |
                    |  block due to s_frozen
================================================
Signed-off-by: NYongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Reported-by: NAmir Goldstein <amir73il@users.sf.net>
Reviewed-by: NJan Kara <jack@suse.cz>
Reviewed-by: NAndreas Dilger <adilger@dilger.ca>

ext4 has taken the stance that, in the absence of a journal,
when an fsync/fdatasync of an inode is done, the parent
directory should be sync'ed if this inode entry is new.
ext4_sync_parent(), which implements this, does indeed sync
the dirent pages for parent directories, but it does not
sync the directory *inode*.  This patch fixes this.

Also now return error status from ext4_sync_parent().

I tested this using a power fail test, which panics a
machine running a file server getting requests from a
client.  Without this patch, on about every other test run,
the server is missing many, many files that had been synced.
With this patch, on > 6 runs, I see zero files being lost.

Google-Bug-Id: 4179519
Signed-off-by: NCurt Wohlgemuth <curtw@google.com>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Lock stateid's can have access_bmap 0 if they were only partially
initialized (due to a failed lock request); handle that case in
free_generic_stateid.

------------[ cut here ]------------
kernel BUG at fs/nfsd/nfs4state.c:380!
invalid opcode: 0000 [#1] SMP
last sysfs file: /sys/kernel/mm/ksm/run
Modules linked in: nfs fscache md4 nls_utf8 cifs ip6table_filter ip6_tables ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat bridge stp llc nfsd lockd nfs_acl auth_rpcgss sunrpc ipv6 ppdev parport_pc parport pcnet32 mii pcspkr microcode i2c_piix4 BusLogic floppy [last unloaded: mperf]

Pid: 1468, comm: nfsd Not tainted 2.6.38+ #120 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform
EIP: 0060:[<e24f180d>] EFLAGS: 00010297 CPU: 0
EIP is at nfs4_access_to_omode+0x1c/0x29 [nfsd]
EAX: ffffffff EBX: dd758120 ECX: 00000000 EDX: 00000004
ESI: dd758120 EDI: ddfe657c EBP: dd54dde0 ESP: dd54dde0
 DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Process nfsd (pid: 1468, ti=dd54c000 task=ddc92580 task.ti=dd54c000)
Stack:
 dd54ddf0 e24f19ca 00000000 ddfe6560 dd54de08 e24f1a5d dd758130 deee3a20
 ddfe6560 31270000 dd54df1c e24f52fd 0000000f dd758090 e2505dd0 0be304cf
 dbb51d68 0000000e ddfe657c ddcd8020 dd758130 dd758128 dd7580d8 dd54de68
Call Trace:
 [<e24f19ca>] free_generic_stateid+0x1c/0x3e [nfsd]
 [<e24f1a5d>] release_lockowner+0x71/0x8a [nfsd]
 [<e24f52fd>] nfsd4_lock+0x617/0x66c [nfsd]
 [<e24e57b6>] ? nfsd_setuser+0x199/0x1bb [nfsd]
 [<e24e056c>] ? nfsd_setuser_and_check_port+0x65/0x81 [nfsd]
 [<c07a0052>] ? _cond_resched+0x8/0x1c
 [<c04ca61f>] ? slab_pre_alloc_hook.clone.33+0x23/0x27
 [<c04cac01>] ? kmem_cache_alloc+0x1a/0xd2
 [<c04835a0>] ? __call_rcu+0xd7/0xdd
 [<e24e0dfb>] ? fh_verify+0x401/0x452 [nfsd]
 [<e24f0b61>] ? nfsd4_encode_operation+0x52/0x117 [nfsd]
 [<e24ea0d7>] ? nfsd4_putfh+0x33/0x3b [nfsd]
 [<e24f4ce6>] ? nfsd4_delegreturn+0xd4/0xd4 [nfsd]
 [<e24ea2c9>] nfsd4_proc_compound+0x1ea/0x33e [nfsd]
 [<e24de6ee>] nfsd_dispatch+0xd1/0x1a5 [nfsd]
 [<e1d6e1c7>] svc_process_common+0x282/0x46f [sunrpc]
 [<e1d6e578>] svc_process+0xdc/0xfa [sunrpc]
 [<e24de0fa>] nfsd+0xd6/0x115 [nfsd]
 [<e24de024>] ? nfsd_shutdown+0x24/0x24 [nfsd]
 [<c0454322>] kthread+0x62/0x67
 [<c04542c0>] ? kthread_worker_fn+0x114/0x114
 [<c07a6ebe>] kernel_thread_helper+0x6/0x10
Code: eb 05 b8 00 00 27 4f 8d 65 f4 5b 5e 5f 5d c3 83 e0 03 55 83 f8 02 89 e5 74 17 83 f8 03 74 05 48 75 09 eb 09 b8 02 00 00 00 eb 0b <0f> 0b 31 c0 eb 05 b8 01 00 00 00 5d c3 55 89 e5 57 56 89 d6 8d
EIP: [<e24f180d>] nfs4_access_to_omode+0x1c/0x29 [nfsd] SS:ESP 0068:dd54dde0
---[ end trace 2b0bf6c6557cb284 ]---

The trace route is:

 -> nfsd4_lock()
   -> if (lock->lk_is_new) {
     -> alloc_init_lock_stateid()

        3739: stp->st_access_bmap = 0;

   ->if (status && lock->lk_is_new && lock_sop)
     -> release_lockowner()
      -> free_generic_stateid()
       -> nfs4_access_bmap_to_omode()
          -> nfs4_access_to_omode()

        380: BUG();   *****

This problem was introduced by 0997b173.
Reported-by: NMi Jinlong <mijinlong@cn.fujitsu.com>
Tested-by: NMi Jinlong <mijinlong@cn.fujitsu.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Add proper blk_start_plug/blk_finish_plug pairs for the two places where
we issue buffer I/O, and remove the blk_flush_plug in xfs_buf_lock and
xfs_buf_iowait, given that context switches already flush the per-process
plugging lists.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

For a CONFIG_XFS_DEBUG=n build gcc complains about statements with no
effect in xfs_debug:

fs/xfs/quota/xfs_qm_syscalls.c: In function 'xfs_qm_scall_trunc_qfiles':
fs/xfs/quota/xfs_qm_syscalls.c:291:3: warning: statement with no effect

The reason for that is that the various new xfs message functions have a
return value which is never used, and in case of the non-debug build
xfs_debug the macro evaluates to a plain 0 which produces the above
warnings.  This can be fixed by turning xfs_debug into an inline function
instead of a macro, but in addition to that I've also changed all the
message helpers to return void as we never use their return values.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>