The i_mutex lock and flush_completed_IO() added by commit 2581fdc810
in ext4_evict_inode() causes lockdep complaining about potential
deadlock in several places. In most/all of these LOCKDEP complaints
it looks like it's a false positive, since many of the potential
circular locking cases can't take place by the time the
ext4_evict_inode() is called; but since at the very least it may mask
real problems, we need to address this.
This change removes the flush_completed_IO() and i_mutex lock in
ext4_evict_inode(). Instead, we take a different approach to resolve
the software lockup that commit 2581fdc810 intends to fix. Rather
than having ext4-dio-unwritten thread wait for grabing the i_mutex
lock of an inode, we use mutex_trylock() instead, and simply requeue
the work item if we fail to grab the inode's i_mutex lock.
This should speed up work queue processing in general and also
prevents the following deadlock scenario: During page fault,
shrink_icache_memory is called that in turn evicts another inode B.
Inode B has some pending io_end work so it calls ext4_ioend_wait()
that waits for inode B's i_ioend_count to become zero. However, inode
B's ioend work was queued behind some of inode A's ioend work on the
same cpu's ext4-dio-unwritten workqueue. As the ext4-dio-unwritten
thread on that cpu is processing inode A's ioend work, it tries to
grab inode A's i_mutex lock. Since the i_mutex lock of inode A is
still hold before the page fault happened, we enter a deadlock.
Signed-off-by: Jiaying Zhang <jiayingz@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: flush any pending end_io requests before DIO reads w/dioread_nolock
ext4: fix nomblk_io_submit option so it correctly converts uninit blocks
ext4: Resolve the hang of direct i/o read in handling EXT4_IO_END_UNWRITTEN.
ext4: call ext4_ioend_wait and ext4_flush_completed_IO in ext4_evict_inode
ext4: Fix ext4_should_writeback_data() for no-journal mode
There is a race between ext4 buffer write and direct_IO read with
dioread_nolock mount option enabled. The problem is that we clear
PageWriteback flag during end_io time but will do
uninitialized-to-initialized extent conversion later with dioread_nolock.
If an O_direct read request comes in during this period, ext4 will return
zero instead of the recently written data.
This patch checks whether there are any pending uninitialized-to-initialized
extent conversion requests before doing O_direct read to close the race.
Note that this is just a bandaid fix. The fundamental issue is that we
clear PageWriteback flag before we really complete an IO, which is
problem-prone. To fix the fundamental issue, we may need to implement an
extent tree cache that we can use to look up pending to-be-converted extents.
Signed-off-by: Jiaying Zhang <jiayingz@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
Bug discovered by Jan Kara:
Finally, commit 1449032be1 returned back
the old IO submission code but apparently it forgot to return the old
handling of uninitialized buffers so we unconditionnaly call
block_write_full_page() without specifying end_io function. So AFAICS
we never convert unwritten extents to written in some cases. For
example when I mount the fs as: mount -t ext4 -o
nomblk_io_submit,dioread_nolock /dev/ubdb /mnt and do
int fd = open(argv[1], O_RDWR | O_CREAT | O_TRUNC, 0600);
char buf[1024];
memset(buf, 'a', sizeof(buf));
fallocate(fd, 0, 0, 16384);
write(fd, buf, sizeof(buf));
I get a file full of zeros (after remounting the filesystem so that
pagecache is dropped) instead of seeing the first KB contain 'a's.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
EXT4_IO_END_UNWRITTEN flag set and the increase of i_aiodio_unwritten
should be done simultaneously since ext4_end_io_nolock always clear
the flag and decrease the counter in the same time.
We don't increase i_aiodio_unwritten when setting
EXT4_IO_END_UNWRITTEN so it will go nagative and causes some process
to wait forever.
Part of the patch came from Eric in his e-mail, but it doesn't fix the
problem met by Michael actually.
http://marc.info/?l=linux-ext4&m=131316851417460&w=2
Reported-and-Tested-by: Michael Tokarev<mjt@tls.msk.ru>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
Flush inode's i_completed_io_list before calling ext4_io_wait to
prevent the following deadlock scenario: A page fault happens while
some process is writing inode A. During page fault,
shrink_icache_memory is called that in turn evicts another inode
B. Inode B has some pending io_end work so it calls ext4_ioend_wait()
that waits for inode B's i_ioend_count to become zero. However, inode
B's ioend work was queued behind some of inode A's ioend work on the
same cpu's ext4-dio-unwritten workqueue. As the ext4-dio-unwritten
thread on that cpu is processing inode A's ioend work, it tries to
grab inode A's i_mutex lock. Since the i_mutex lock of inode A is
still hold before the page fault happened, we enter a deadlock.
Also moves ext4_flush_completed_IO and ext4_ioend_wait from
ext4_destroy_inode() to ext4_evict_inode(). During inode deleteion,
ext4_evict_inode() is called before ext4_destroy_inode() and in
ext4_evict_inode(), we may call ext4_truncate() without holding
i_mutex lock. As a result, there is a race between flush_completed_IO
that is called from ext4_ext_truncate() and ext4_end_io_work, which
may cause corruption on an io_end structure. This change moves
ext4_flush_completed_IO and ext4_ioend_wait from ext4_destroy_inode()
to ext4_evict_inode() to resolve the race between ext4_truncate() and
ext4_end_io_work during inode deletion.
Signed-off-by: Jiaying Zhang <jiayingz@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
ext4_should_writeback_data() had an incorrect sequence of
tests to determine if it should return 0 or 1: in
particular, even in no-journal mode, 0 was being returned
for a non-regular-file inode.
This meant that, in non-journal mode, we would use
ext4_journalled_aops for directories, symlinks, and other
non-regular files. However, calling journalled aop
callbacks when there is no valid handle, can cause problems.
This would cause a kernel crash with Jan Kara's commit
2d859db3e4 ("ext4: fix data corruption in inodes with
journalled data"), because we now dereference 'handle' in
ext4_journalled_write_end().
I also added BUG_ONs to check for a valid handle in the
obviously journal-only aops callbacks.
I tested this running xfstests with a scratch device in
these modes:
- no-journal
- data=ordered
- data=writeback
- data=journal
All work fine; the data=journal run has many failures and a
crash in xfstests 074, but this is no different from a
vanilla kernel.
Signed-off-by: Curt Wohlgemuth <curtw@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
Commit df5e622340 ("ext4: fix deadlock in ext4_symlink() in ENOSPC
conditions") recalculated the number of credits needed for a long
symlink, in the process of splitting it into two transactions. However,
the first credit calculation under-counted because if selinux is
enabled, credits are needed to create the selinux xattr as well.
Overrunning the reservation will result in an OOPS in
jbd2_journal_dirty_metadata() due to this assert:
J_ASSERT_JH(jh, handle->h_buffer_credits > 0);
Fix this by increasing the reservation size.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 9933fc0i (ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and
ext4_kvfree()) intruduced wrappers around k*alloc/vmalloc but introduced
a typo for ext4_kzalloc() by not using kzalloc() but kmalloc().
Signed-off-by: Mathias Krause <minipli@googlemail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (60 commits)
ext4: prevent memory leaks from ext4_mb_init_backend() on error path
ext4: use EXT4_BAD_INO for buddy cache to avoid colliding with valid inode #
ext4: use ext4_msg() instead of printk in mballoc
ext4: use ext4_kvzalloc()/ext4_kvmalloc() for s_group_desc and s_group_info
ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and ext4_kvfree()
ext4: use the correct error exit path in ext4_init_inode_table()
ext4: add missing kfree() on error return path in add_new_gdb()
ext4: change umode_t in tracepoint headers to be an explicit __u16
ext4: fix races in ext4_sync_parent()
ext4: Fix overflow caused by missing cast in ext4_fallocate()
ext4: add action of moving index in ext4_ext_rm_idx for Punch Hole
ext4: simplify parameters of reserve_backup_gdb()
ext4: simplify parameters of add_new_gdb()
ext4: remove lock_buffer in bclean() and setup_new_group_blocks()
ext4: simplify journal handling in setup_new_group_blocks()
ext4: let setup_new_group_blocks() set multiple bits at a time
ext4: fix a typo in ext4_group_extend()
ext4: let ext4_group_add_blocks() handle 0 blocks quickly
ext4: let ext4_group_add_blocks() return an error code
ext4: rename ext4_add_groupblocks() to ext4_group_add_blocks()
...
Fix up conflict in fs/ext4/inode.c: commit aacfc19c62 ("fs: simplify
the blockdev_direct_IO prototype") had changed the ext4_ind_direct_IO()
function for the new simplified calling convention, while commit
dae1e52cb1 ("ext4: move ext4_ind_* functions from inode.c to
indirect.c") moved the function to another file.
In ext4_mb_init(), if the s_locality_group allocation fails it will
currently cause the allocations made in ext4_mb_init_backend() to
be leaked. Moving the ext4_mb_init_backend() allocation after the
s_locality_group allocation avoids that problem.
Signed-off-by: Yu Jian <yujian@whamcloud.com>
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Introduce new helper functions which try kmalloc, and then fall back
to vmalloc if necessary, and use them for allocating and deallocating
s_flex_groups.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch lets ext4_init_inode_table() handle errors right.
ext4_init_inode_table() should down_write() alloc_sem which
has been up_write()ed and stop the started journal handle.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
We added some more error handling in b40971426a "ext4: add error
checking to calls to ext4_handle_dirty_metadata()". But we need to
call kfree() as well to avoid a memory leak.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix problems if fsync() races against a rename of a parent directory
as pointed out by Al Viro in his own inimitable way:
>While we are at it, could somebody please explain what the hell is ext4
>doing in
>static int ext4_sync_parent(struct inode *inode)
>{
> struct writeback_control wbc;
> struct dentry *dentry = NULL;
> int ret = 0;
>
> while (inode && ext4_test_inode_state(inode, EXT4_STATE_NEWENTRY)) {
> ext4_clear_inode_state(inode, EXT4_STATE_NEWENTRY);
> dentry = list_entry(inode->i_dentry.next,
> struct dentry, d_alias);
> if (!dentry || !dentry->d_parent || !dentry->d_parent->d_inode)
> break;
> inode = dentry->d_parent->d_inode;
> ret = sync_mapping_buffers(inode->i_mapping);
> ...
>Note that dentry obviously can't be NULL there. dentry->d_parent is never
>NULL. And dentry->d_parent would better not be negative, for crying out
>loud! What's worse, there's no guarantees that dentry->d_parent will
>remain our parent over that sync_mapping_buffers() *and* that inode won't
>just be freed under us (after rename() and memory pressure leading to
>eviction of what used to be our dentry->d_parent)......
Reported-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The logical block number in map.l_blk is a __u32, and so before we
shift it left, by the block size, we neeed cast it to a 64-bit size.
Otherwise i_size can be corrupted on an ENOSPC.
# df -T /mnt/mp1
Filesystem Type 1K-blocks Used Available Use% Mounted on
/dev/sda6 ext4 9843276 153056 9190200 2% /mnt/mp1
# fallocate -o 0 -l 2199023251456 /mnt/mp1/testfile
fallocate: /mnt/mp1/testfile: fallocate failed: No space left on device
# stat /mnt/mp1/testfile
File: `/mnt/mp1/testfile'
Size: 4293656576 Blocks: 19380440 IO Block: 4096 regular file
Device: 806h/2054d Inode: 12 Links: 1
Access: (0644/-rw-r--r--) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2011-07-25 13:01:31.414490496 +0900
Modify: 2011-07-25 13:01:31.414490496 +0900
Change: 2011-07-25 13:01:31.454490495 +0900
Signed-off-by: Utako Kusaka <u-kusaka@wm.jp.nec.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
--
fs/ext4/extents.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
The old function ext4_ext_rm_idx is used only for truncate case
because it just remove last index in extent-index-block. When punching
hole, it usually needed to remove "middle" index, therefore we must
move indexes which after it forward.
(I create a file with 1 depth extent tree and punch hole in the middle
of it, the last index in index-block strangly gone, so I find out this
bug)
Signed-off-by: Robin Dong <sanbai@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The reserve_backup_gdb() function only needs the block group number;
there's no need to pass a pointer to struct ext4_new_group_data to it.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
add_new_gdb() only needs the block group number; there is no need to
pass a pointer to struct ext4_new_group_data to add_new_gdb().
Instead of filling in a pointer the struct buffer_head in
add_new_gdb(), it's simpler to have the caller fetch it from the
s_group_desc[] array.
[Fixed error path to handle the case where struct buffer_head *primary
hasn't been set yet. -- Ted]
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
There is no need to lock the buffers since no one else should be
touching these buffers besides the file system.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch simplifies journal handling in setup_new_group_blocks().
In previous code, block bitmap is modified everywhere in
setup_new_group_blocks(), ext4_get_write_access() in
extend_or_restart_transaction() is used to guarantee that the block
bitmap stays in the new handle, this makes things complicated.
The previous commit changed things so that the modifications on the
block bitmap are batched and done by ext4_set_bits() at the end of the
for loop. This allows us to simplify things.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Rename mb_set_bits() to ext4_set_bits() and make it a global function
so that setup_new_group_blocks() can use it.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
If ext4_group_add_blocks() is called with 0 block, make it return 0
without doing any extra work.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch lets ext4_group_add_blocks() return an error code if it
fails, so that upper functions can handle error correctly.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
A filesystem with errors is not allowed to being resized, otherwise,
it is easy to destroy the filesystem.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Before this patch, parallel resizers are allowed and protected by a
mutex lock, actually, there is no need to support parallel resizer, so
this patch prevents parallel resizers by atmoic bit ops, like
lock_page() and unlock_page() do.
To do this, the patch removed the mutex lock s_resize_lock from struct
ext4_sb_info and added a unsigned long field named s_resize_flags
which inidicates if there is a resizer.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
When journalling data for an inode (either because it is a symlink or
because the filesystem is mounted in data=journal mode), ext4_evict_inode()
can discard unwritten data by calling truncate_inode_pages(). This is
because we don't mark the buffer / page dirty when journalling data but only
add the buffer to the running transaction and thus mm does not know there
are still unwritten data.
Fix the problem by carefully tracking transaction containing inode's data,
committing this transaction, and writing uncheckpointed buffers when inode
should be reaped.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Replace the ->check_acl method with a ->get_acl method that simply reads an
ACL from disk after having a cache miss. This means we can replace the ACL
checking boilerplate code with a single implementation in namei.c.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
new helper: posix_acl_create(&acl, gfp, mode_p). Replaces acl with
modified clone, on failure releases acl and replaces with NULL.
Returns 0 or -ve on error. All callers of posix_acl_create_masq()
switched.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
new helper: posix_acl_chmod(&acl, gfp, mode). Replaces acl with modified
clone or with NULL if that has failed; returns 0 or -ve on error. All
callers of posix_acl_chmod_masq() switched to that - they'd been doing
exactly the same thing.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This moves logic for checking the cached ACL values from low-level
filesystems into generic code. The end result is a streamlined ACL
check that doesn't need to load the inode->i_op->check_acl pointer at
all for the common cached case.
The filesystems also don't need to check for a non-blocking RCU walk
case in their acl_check() functions, because that is all handled at a
VFS layer.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The debug message in ext4_ext_insert_extent before moving extent
is incorrect (the "from xx to xx").
Signed-off-by: Robin Dong <sanbai@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The argument "inode" in function ext4_ext_next_allocated_block looks useless,
so clean it.
Signed-off-by: Robin Dong <sanbai@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
ac_repeats isn't referenced in the mballoc code. So remove it.
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext4_mb_release, we use s_mb_buddies_generated++. Although
the output is OK, but I don't think we need this extra ++.
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
ext4_mb_load_buddy() calls ext4_get_group_info() for setting both
"grp" and "e4b->bd_info", but it could do "e4b->bd_info = grp".
Reported-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Btrfs needs to be able to control how filemap_write_and_wait_range() is called
in fsync to make it less of a painful operation, so push down taking i_mutex and
the calling of filemap_write_and_wait() down into the ->fsync() handlers. Some
file systems can drop taking the i_mutex altogether it seems, like ext3 and
ocfs2. For correctness sake I just pushed everything down in all cases to make
sure that we keep the current behavior the same for everybody, and then each
individual fs maintainer can make up their mind about what to do from there.
Thanks,
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Since Ext4 has its own lseek we need to make sure it handles
SEEK_HOLE/SEEK_DATA. For now just do the same thing that is done in the generic
case, somebody else can come along and make it do fancy things later. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
For filesystems that delay their end_io processing we should keep our
i_dio_count until the the processing is done. Enable this by moving
the inode_dio_done call to the end_io handler if one exist. Note that
the actual move to the workqueue for ext4 and XFS is not done in
this patch yet, but left to the filesystem maintainers. At least
for XFS it's not needed yet either as XFS has an internal equivalent
to i_dio_count.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Simple filesystems always pass inode->i_sb_bdev as the block device
argument, and never need a end_io handler. Let's simply things for
them and for my grepping activity by dropping these arguments. The
only thing not falling into that scheme is ext4, which passes and
end_io handler without needing special flags (yet), but given how
messy the direct I/O code there is use of __blockdev_direct_IO
in one instead of two out of three cases isn't going to make a large
difference anyway.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Let filesystems handle waiting for direct I/O requests themselves instead
of doing it beforehand. This means filesystem-specific locks to prevent
new dio referenes from appearing can be held. This is important to allow
generalizing i_dio_count to non-DIO_LOCKING filesystems.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
