While transaction is going to commit, it first sets its state to be
T_LOCKED and waits all outstanding handles to complete, and the
committing transaction will always be in locked state so long as it
has outstanding handles, also the whole fs will be locked and all later
fs modification operations will be stucked in wait_transaction_locked().

It's hard to tell why handles are that slow, so here we add a new staic
tracepoint to track such slow handle, and show io wait time and sched
wait time, output likes below:
  fsstress-20347 [024] ....  1570.305454: jbd2_slow_handle_stats: dev 254,17
tid 15853 type 4 line_no 3101 interval 126 sync 0 requested_blocks 24
dirtied_blocks 0 trans_wait 122 space_wait 0 sched_wait 0 io_wait 126

"trans_wait 122" means that this current committing transaction has been
locked for 122ms, due to this handle is not completed quickly.

From "io_wait 126", we can see that io is the major reason.

In this patch, we also add a per fs control file used to determine
whether a handle can be considered to be slow.
    /proc/fs/jbd2/vdb1-8/stall_thresh
default value is 100ms, users can set new threshold by echoing new value
to this file.

Later I also plan to add a proc file fs per fs to record these info.
Reviewed-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Signed-off-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>

If one process context is stucked in wait_on_buffer(), lock_buffer(),
lock_page() and wait_on_page_writeback() and wait_on_bit_io(), it's
hard to tell ture reason, for example, whether this page is under io,
or this page is just locked too long by other process context.

Normally io request has multiple bios, and every bio contains multiple
pages which will hold data to be read from or written to device, so here
we record page info or bio info in task_struct while process calls
lock_page(), lock_buffer(), wait_on_page_writeback(), wait_on_buffer()
and wait_on_bit_io(), we add a new proce interface:
[lege@localhost linux]$ cat /proc/4516/wait_res
1 ffffd0969f95d3c0 4295369599 4295381596

Above info means that thread 4516 is waitting on a page, address is
ffffd0969f95d3c0, and has waited for 11997ms.

First field denotes the page address process is waitting on.
Second field denotes the wait moment and the third denotes current moment.

In practice, if we found a process waitting on one page for too long time,
we can get page's address by reading /proc/$pid/wait_page, and search this
page address in all block devices' /sys/kernel/debug/block/${devname}/rq_hang,
if search operation hits one, we can get the request and know why this io
request hangs that long.
Reviewed-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Signed-off-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>

commit 597599268e3b91cac71faf48743f4783dec682fc upstream.

We do not unmap and clear dirty flag when forgetting a buffer without
journal or does not belongs to any transaction, so the invalid dirty
data may still be written to the disk later. It's fine if the
corresponding block is never used before the next mount, and it's also
fine that we invoke clean_bdev_aliases() related functions to unmap
the block device mapping when re-allocating such freed block as data
block. But this logic is somewhat fragile and risky that may lead to
data corruption if we forget to clean bdev aliases. So, It's better to
discard dirty data during forget time.

We have been already handled all the cases of forgetting journalled
buffer, this patch deal with the remaining two cases.

- buffer is not journalled yet,
- buffer is journalled but doesn't belongs to any transaction.

We invoke __bforget() instead of __brelese() when forgetting an
un-journalled buffer in jbd2_journal_forget(). After this patch we can
remove all clean_bdev_aliases() related calls in ext4.
Suggested-by: NJan Kara <jack@suse.cz>
Signed-off-by: Nzhangyi (F) <yi.zhang@huawei.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Reviewed-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

commit 6ba0e7dc64a5adcda2fbe65adc466891795d639e upstream.

Currently both journal_submit_inode_data_buffers() and
journal_finish_inode_data_buffers() operate on the entire address space
of each of the inodes associated with a given journal entry.  The
consequence of this is that if we have an inode where we are constantly
appending dirty pages we can end up waiting for an indefinite amount of
time in journal_finish_inode_data_buffers() while we wait for all the
pages under writeback to be written out.

The easiest way to cause this type of workload is do just dd from
/dev/zero to a file until it fills the entire filesystem.  This can
cause journal_finish_inode_data_buffers() to wait for the duration of
the entire dd operation.

We can improve this situation by scoping each of the inode dirty ranges
associated with a given transaction.  We do this via the jbd2_inode
structure so that the scoping is contained within jbd2 and so that it
follows the lifetime and locking rules for that structure.

This allows us to limit the writeback & wait in
journal_submit_inode_data_buffers() and
journal_finish_inode_data_buffers() respectively to the dirty range for
a given struct jdb2_inode, keeping us from waiting forever if the inode
in question is still being appended to.
Signed-off-by: NRoss Zwisler <zwisler@google.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 0d52154bb0a700abb459a2cbce0a30fc2549b67e upstream.

When failing from creating cache jbd2_inode_cache, we will destroy the
previously created cache jbd2_handle_cache twice.  This patch fixes
this by moving each cache initialization/destruction to its own
separate, individual function.
Signed-off-by: NChengguang Xu <cgxu519@gmail.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 01215d3edb0f384ddeaa5e4a22c1ae5ff634149f upstream.

The jh pointer may be used uninitialized in the two cases below and the
compiler complain about it when enabling JBUFFER_TRACE macro, fix them.

In file included from fs/jbd2/transaction.c:19:0:
fs/jbd2/transaction.c: In function ‘jbd2_journal_get_undo_access’:
./include/linux/jbd2.h:1637:38: warning: ‘jh’ is used uninitialized in this function [-Wuninitialized]
 #define JBUFFER_TRACE(jh, info) do { printk("%s: %d\n", __func__, jh->b_jcount);} while (0)
                                      ^
fs/jbd2/transaction.c:1219:23: note: ‘jh’ was declared here
  struct journal_head *jh;
                       ^
In file included from fs/jbd2/transaction.c:19:0:
fs/jbd2/transaction.c: In function ‘jbd2_journal_dirty_metadata’:
./include/linux/jbd2.h:1637:38: warning: ‘jh’ may be used uninitialized in this function [-Wmaybe-uninitialized]
 #define JBUFFER_TRACE(jh, info) do { printk("%s: %d\n", __func__, jh->b_jcount);} while (0)
                                      ^
fs/jbd2/transaction.c:1332:23: note: ‘jh’ was declared here
  struct journal_head *jh;
                       ^
Signed-off-by: Nzhangyi (F) <yi.zhang@huawei.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 904cdbd41d749a476863a0ca41f6f396774f26e4 upstream.

Now, we capture a data corruption problem on ext4 while we're truncating
an extent index block. Imaging that if we are revoking a buffer which
has been journaled by the committing transaction, the buffer's jbddirty
flag will not be cleared in jbd2_journal_forget(), so the commit code
will set the buffer dirty flag again after refile the buffer.

fsx                               kjournald2
                                  jbd2_journal_commit_transaction
jbd2_journal_revoke                commit phase 1~5...
 jbd2_journal_forget
   belongs to older transaction    commit phase 6
   jbddirty not clear               __jbd2_journal_refile_buffer
                                     __jbd2_journal_unfile_buffer
                                      test_clear_buffer_jbddirty
                                       mark_buffer_dirty

Finally, if the freed extent index block was allocated again as data
block by some other files, it may corrupt the file data after writing
cached pages later, such as during unmount time. (In general,
clean_bdev_aliases() related helpers should be invoked after
re-allocation to prevent the above corruption, but unfortunately we
missed it when zeroout the head of extra extent blocks in
ext4_ext_handle_unwritten_extents()).

This patch mark buffer as freed and set j_next_transaction to the new
transaction when it already belongs to the committing transaction in
jbd2_journal_forget(), so that commit code knows it should clear dirty
bits when it is done with the buffer.

This problem can be reproduced by xfstests generic/455 easily with
seeds (3246 3247 3248 3249).
Signed-off-by: Nzhangyi (F) <yi.zhang@huawei.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Do not set the b_modified flag in block's journal head should not
until after we're sure that jbd2_journal_dirty_metadat() will not
abort with an error due to there not being enough space reserved in
the jbd2 handle.

Otherwise, future attempts to modify the buffer may lead a large
number of spurious errors and warnings.

This addresses CVE-2018-10883.

https://bugzilla.kernel.org/show_bug.cgi?id=200071Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org

The kmem_cache_destroy() function already checks for null pointers, so
we can remove the check at the call site.

This patch also sets jbd2_handle_cache and jbd2_inode_cache to be NULL
after freeing them in jbd2_journal_destroy_handle_cache().
Signed-off-by: NWang Long <wanglong19@meituan.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>

If ext4 tries to start a reserved handle via
jbd2_journal_start_reserved(), and the journal has been aborted, this
can result in a NULL pointer dereference.  This is because the fields
h_journal and h_transaction in the handle structure share the same
memory, via a union, so jbd2_journal_start_reserved() will clear
h_journal before calling start_this_handle().  If this function fails
due to an aborted handle, h_journal will still be NULL, and the call
to jbd2_journal_free_reserved() will pass a NULL journal to
sub_reserve_credits().

This can be reproduced by running "kvm-xfstests -c dioread_nolock
generic/475".

Cc: stable@kernel.org # 3.11
Fixes: 8f7d89f3 ("jbd2: transaction reservation support")
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NAndreas Dilger <adilger@dilger.ca>
Reviewed-by: NJan Kara <jack@suse.cz>

Sphinx emits various (26) warnings when building make target 'htmldocs'.
Currently struct definitions contain duplicate documentation, some as
kernel-docs and some as standard c89 comments.  We can reduce
duplication while cleaning up the kernel docs.

Move all kernel-docs to right above each struct member.  Use the set of
all existing comments (kernel-doc and c89).  Add documentation for
missing struct members and function arguments.
Signed-off-by: NTobin C. Harding <me@tobin.cc>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org

A number of ext4 source files were skipped due because their copyright
permission statements didn't match the expected text used by the
automated conversion utilities.  I've added SPDX tags for the rest.

While looking at some of these files, I've noticed that we have quite
a bit of variation on the licenses that were used --- in particular
some of the Red Hat licenses on the jbd2 files use a GPL2+ license,
and we have some files that have a LGPL-2.1 license (which was quite
surprising).

I've not attempted to do any license changes.  Even if it is perfectly
legal to relicense to GPL 2.0-only for consistency's sake, that should
be done with ext4 developer community discussion.
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

When a transaction starts, start_this_handle() saves current
PF_MEMALLOC_NOFS value so that it can be restored at journal stop time.
Journal restart is a special case that calls start_this_handle() without
stopping the transaction. start_this_handle() isn't aware that the
original value is already stored so it overwrites it with current value.

For instance, a call sequence like below leaves PF_MEMALLOC_NOFS flag set
at the end:

  jbd2_journal_start()
  jbd2__journal_restart()
  jbd2_journal_stop()

Make jbd2__journal_restart() restore the original value before calling
start_this_handle().

Fixes: 81378da6 ("jbd2: mark the transaction context with the scope GFP_NOFS context")
Signed-off-by: NTahsin Erdogan <tahsin@google.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>

kernel-doc will try to interpret a foo() string, except if
properly escaped.
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

kernel-doc script expects that a function documentation to
be just before the function, otherwise it will be ignored.

So, move the kernel-doc markup to the right place.
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

now that we have memalloc_nofs_{save,restore} api we can mark the whole
transaction context as implicitly GFP_NOFS.  All allocations will
automatically inherit GFP_NOFS this way.  This means that we do not have
to mark any of those requests with GFP_NOFS and moreover all the
ext4_kv[mz]alloc(GFP_NOFS) are also safe now because even the hardcoded
GFP_KERNEL allocations deep inside the vmalloc will be NOFS now.

[akpm@linux-foundation.org: tweak comments]
Link: http://lkml.kernel.org/r/20170306131408.9828-7-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Chris Mason <clm@fb.com>
Cc: David Sterba <dsterba@suse.cz>
Cc: Brian Foster <bfoster@redhat.com>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: Nikolay Borisov <nborisov@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If the journal has been aborted, we shouldn't mark the underlying
buffer head as dirty, since that will cause the metadata block to get
modified.  And if the journal has been aborted, we shouldn't allow
this since it will almost certainly lead to a corrupted file system.
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org

When 'jh->b_transaction == transaction' (asserted by below)

  J_ASSERT_JH(jh, (jh->b_transaction == transaction || ...

'journal->j_list_lock' will be incorrectly unlocked, since
the the lock is aquired only at the end of if / else-if
statements (missing the else case).
Signed-off-by: NTaesoo Kim <tsgatesv@gmail.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NAndreas Dilger <adilger@dilger.ca>
Fixes: 6e4862a5
Cc: stable@vger.kernel.org # 3.14+

Thomas has reported a lockdep splat hitting in
add_transaction_credits(). The problem is that that function calls
jbd2_might_wait_for_commit() while holding j_state_lock which is wrong
(we do not really wait for transaction commit while holding that lock).

Fix the problem by moving jbd2_might_wait_for_commit() into places where
we are ready to wait for transaction commit and thus j_state_lock is
unlocked.

Cc: stable@vger.kernel.org
Fixes: 1eaa566dReported-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

So far we were tracking only dependency on transaction commit due to
starting a new handle (which may require commit to start a new
transaction). Now add tracking also for other cases where we wait for
transaction commit. This way lockdep can catch deadlocks e. g. because we
call jbd2_journal_stop() for a synchronous handle with some locks held
which rank below transaction start.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

Currently lockdep map is tracked in each journal handle. To be able to
expand lockdep support to cover also other cases where we depend on
transaction commit and where handle is not available, move lockdep map
into struct journal_s. Since this makes the lockdep map shared for all
handles, we have to use rwsem_acquire_read() for acquisitions now.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

The transaction the handle references is free to commit once we've
decremented t_updates counter. Move the lockdep instrumentation to that
place. Currently it was a bit later which did not really matter but
subsequent improvements to lockdep instrumentation would cause false
positives with it.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

Currently when filesystem needs to make sure data is on permanent
storage before committing a transaction it adds inode to transaction's
inode list. During transaction commit, jbd2 writes back all dirty
buffers that have allocated underlying blocks and waits for the IO to
finish. However when doing writeback for delayed allocated data, we
allocate blocks and immediately submit the data. Thus asking jbd2 to
write dirty pages just unnecessarily adds more work to jbd2 possibly
writing back other redirtied blocks.

Add support to jbd2 to allow filesystem to ask jbd2 to only wait for
outstanding data writes before committing a transaction and thus avoid
unnecessary writes.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

This patch fix spelling typos found in DocBook/filesystem.xml.
It is because the file was generated from comments in code,
I have to fix the comments in codes, instead of xml file.
Signed-off-by: NMasanari Iida <standby24x7@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.

This promise never materialized.  And unlikely will.

We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE.  And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.

Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.

Let's stop pretending that pages in page cache are special.  They are
not.

The changes are pretty straight-forward:

 - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;

 - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;

 - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};

 - page_cache_get() -> get_page();

 - page_cache_release() -> put_page();

This patch contains automated changes generated with coccinelle using
script below.  For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.

The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.

There are few places in the code where coccinelle didn't reach.  I'll
fix them manually in a separate patch.  Comments and documentation also
will be addressed with the separate patch.

virtual patch

@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E

@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E

@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT

@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE

@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK

@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)

@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)

@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Journal transaction might fail prematurely because the frozen_buffer
is allocated by GFP_NOFS request:
[   72.440013] do_get_write_access: OOM for frozen_buffer
[   72.440014] EXT4-fs: ext4_reserve_inode_write:4729: aborting transaction: Out of memory in __ext4_journal_get_write_access
[   72.440015] EXT4-fs error (device sda1) in ext4_reserve_inode_write:4735: Out of memory
(...snipped....)
[   72.495559] do_get_write_access: OOM for frozen_buffer
[   72.495560] EXT4-fs: ext4_reserve_inode_write:4729: aborting transaction: Out of memory in __ext4_journal_get_write_access
[   72.496839] do_get_write_access: OOM for frozen_buffer
[   72.496841] EXT4-fs: ext4_reserve_inode_write:4729: aborting transaction: Out of memory in __ext4_journal_get_write_access
[   72.505766] Aborting journal on device sda1-8.
[   72.505851] EXT4-fs (sda1): Remounting filesystem read-only

This wasn't a problem until "mm: page_alloc: do not lock up GFP_NOFS
allocations upon OOM" because small GPF_NOFS allocations never failed.
This allocation seems essential for the journal and GFP_NOFS is too
restrictive to the memory allocator so let's use __GFP_NOFAIL here to
emulate the previous behavior.
Signed-off-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

Signed-off-by: NDmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

introduced jbd2_write_access_granted() to improve write|undo_access
speed, but missed to check the status of b_committed_data which caused
a kernel panic on ocfs2.

[ 6538.405938] ------------[ cut here ]------------
[ 6538.406686] kernel BUG at fs/ocfs2/suballoc.c:2400!
[ 6538.406686] invalid opcode: 0000 [#1] SMP
[ 6538.406686] Modules linked in: ocfs2 nfsd lockd grace nfs_acl auth_rpcgss sunrpc autofs4 ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm ocfs2_nodemanager ocfs2_stackglue configfs sd_mod sg ip6t_REJECT nf_reject_ipv6 nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables be2iscsi iscsi_boot_sysfs bnx2i cnic uio cxgb4i cxgb4 cxgb3i libcxgbi cxgb3 mdio ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr ipv6 iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ppdev xen_kbdfront xen_netfront xen_fbfront parport_pc parport pcspkr i2c_piix4 acpi_cpufreq ext4 jbd2 mbcache xen_blkfront floppy pata_acpi ata_generic ata_piix cirrus ttm drm_kms_helper drm fb_sys_fops sysimgblt sysfillrect i2c_core syscopyarea dm_mirror dm_region_hash dm_log dm_mod
[ 6538.406686] CPU: 1 PID: 16265 Comm: mmap_truncate Not tainted 4.3.0 #1
[ 6538.406686] Hardware name: Xen HVM domU, BIOS 4.3.1OVM 05/14/2014
[ 6538.406686] task: ffff88007c2bab00 ti: ffff880075b78000 task.ti: ffff880075b78000
[ 6538.406686] RIP: 0010:[<ffffffffa06a286b>]  [<ffffffffa06a286b>] ocfs2_block_group_clear_bits+0x23b/0x250 [ocfs2]
[ 6538.406686] RSP: 0018:ffff880075b7b7f8  EFLAGS: 00010246
[ 6538.406686] RAX: ffff8800760c5b40 RBX: ffff88006c06a000 RCX: ffffffffa06e6df0
[ 6538.406686] RDX: 0000000000000000 RSI: ffff88007a6f6ea0 RDI: ffff88007a760430
[ 6538.406686] RBP: ffff880075b7b878 R08: 0000000000000002 R09: 0000000000000001
[ 6538.406686] R10: ffffffffa06769be R11: 0000000000000000 R12: 0000000000000001
[ 6538.406686] R13: ffffffffa06a1750 R14: 0000000000000001 R15: ffff88007a6f6ea0
[ 6538.406686] FS:  00007f17fde30720(0000) GS:ffff88007f040000(0000) knlGS:0000000000000000
[ 6538.406686] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6538.406686] CR2: 0000000000601730 CR3: 000000007aea0000 CR4: 00000000000406e0
[ 6538.406686] Stack:
[ 6538.406686]  ffff88007c2bb5b0 ffff880075b7b8e0 ffff88007a7604b0 ffff88006c640800
[ 6538.406686]  ffff88007a7604b0 ffff880075d77390 0000000075b7b878 ffffffffa06a309d
[ 6538.406686]  ffff880075d752d8 ffff880075b7b990 ffff880075b7b898 0000000000000000
[ 6538.406686] Call Trace:
[ 6538.406686]  [<ffffffffa06a309d>] ? ocfs2_read_group_descriptor+0x6d/0xa0 [ocfs2]
[ 6538.406686]  [<ffffffffa06a3654>] _ocfs2_free_suballoc_bits+0xe4/0x320 [ocfs2]
[ 6538.406686]  [<ffffffffa06a1750>] ? ocfs2_put_slot+0xf0/0xf0 [ocfs2]
[ 6538.406686]  [<ffffffffa06a397e>] _ocfs2_free_clusters+0xee/0x210 [ocfs2]
[ 6538.406686]  [<ffffffffa06a1750>] ? ocfs2_put_slot+0xf0/0xf0 [ocfs2]
[ 6538.406686]  [<ffffffffa06a1750>] ? ocfs2_put_slot+0xf0/0xf0 [ocfs2]
[ 6538.406686]  [<ffffffffa0682d50>] ? ocfs2_extend_trans+0x50/0x1a0 [ocfs2]
[ 6538.406686]  [<ffffffffa06a3ad5>] ocfs2_free_clusters+0x15/0x20 [ocfs2]
[ 6538.406686]  [<ffffffffa065072c>] ocfs2_replay_truncate_records+0xfc/0x290 [ocfs2]
[ 6538.406686]  [<ffffffffa06843ac>] ? ocfs2_start_trans+0xec/0x1d0 [ocfs2]
[ 6538.406686]  [<ffffffffa0654600>] __ocfs2_flush_truncate_log+0x140/0x2d0 [ocfs2]
[ 6538.406686]  [<ffffffffa0654394>] ? ocfs2_reserve_blocks_for_rec_trunc.clone.0+0x44/0x170 [ocfs2]
[ 6538.406686]  [<ffffffffa065acd4>] ocfs2_remove_btree_range+0x374/0x630 [ocfs2]
[ 6538.406686]  [<ffffffffa017486b>] ? jbd2_journal_stop+0x25b/0x470 [jbd2]
[ 6538.406686]  [<ffffffffa065d5b5>] ocfs2_commit_truncate+0x305/0x670 [ocfs2]
[ 6538.406686]  [<ffffffffa0683430>] ? ocfs2_journal_access_eb+0x20/0x20 [ocfs2]
[ 6538.406686]  [<ffffffffa067adb7>] ocfs2_truncate_file+0x297/0x380 [ocfs2]
[ 6538.406686]  [<ffffffffa01759e4>] ? jbd2_journal_begin_ordered_truncate+0x64/0xc0 [jbd2]
[ 6538.406686]  [<ffffffffa067c7a2>] ocfs2_setattr+0x572/0x860 [ocfs2]
[ 6538.406686]  [<ffffffff810e4a3f>] ? current_fs_time+0x3f/0x50
[ 6538.406686]  [<ffffffff812124b7>] notify_change+0x1d7/0x340
[ 6538.406686]  [<ffffffff8121abf9>] ? generic_getxattr+0x79/0x80
[ 6538.406686]  [<ffffffff811f5876>] do_truncate+0x66/0x90
[ 6538.406686]  [<ffffffff81120e30>] ? __audit_syscall_entry+0xb0/0x110
[ 6538.406686]  [<ffffffff811f5bb3>] do_sys_ftruncate.clone.0+0xf3/0x120
[ 6538.406686]  [<ffffffff811f5bee>] SyS_ftruncate+0xe/0x10
[ 6538.406686]  [<ffffffff816aa2ae>] entry_SYSCALL_64_fastpath+0x12/0x71
[ 6538.406686] Code: 28 48 81 ee b0 04 00 00 48 8b 92 50 fb ff ff 48 8b 80 b0 03 00 00 48 39 90 88 00 00 00 0f 84 30 fe ff ff 0f 0b eb fe 0f 0b eb fe <0f> 0b 0f 1f 00 eb fb 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00
[ 6538.406686] RIP  [<ffffffffa06a286b>] ocfs2_block_group_clear_bits+0x23b/0x250 [ocfs2]
[ 6538.406686]  RSP <ffff880075b7b7f8>
[ 6538.691128] ---[ end trace 31cd7011d6770d7e ]---
[ 6538.694492] Kernel panic - not syncing: Fatal exception
[ 6538.695484] Kernel Offset: disabled

Fixes: de92c8ca("jbd2: speedup jbd2_journal_get_[write|undo]_access()")
Cc: <stable@vger.kernel.org>
Signed-off-by: NJunxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

Ted and Namjae have reported that truncated pages don't get timely
reclaimed after being truncated in data=journal mode. The following test
triggers the issue easily:

for (i = 0; i < 1000; i++) {
	pwrite(fd, buf, 1024*1024, 0);
	fsync(fd);
	fsync(fd);
	ftruncate(fd, 0);
}

The reason is that journal_unmap_buffer() finds that truncated buffers
are not journalled (jh->b_transaction == NULL), they are part of
checkpoint list of a transaction (jh->b_cp_transaction != NULL) and have
been already written out (!buffer_dirty(bh)). We clean such buffers but
we leave them in the checkpoint list. Since checkpoint transaction holds
a reference to the journal head, these buffers cannot be released until
the checkpoint transaction is cleaned up. And at that point we don't
call release_buffer_page() anymore so pages detached from mapping are
lingering in the system waiting for reclaim to find them and free them.

Fix the problem by removing buffers from transaction checkpoint lists
when journal_unmap_buffer() finds out they don't have to be there
anymore.
Reported-and-tested-by: NNamjae Jeon <namjae.jeon@samsung.com>
Fixes: de1b7941Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org

mm, page_alloc: distinguish between being unable to sleep, unwilling to sleep and avoiding waking kswapd

__GFP_WAIT has been used to identify atomic context in callers that hold
spinlocks or are in interrupts.  They are expected to be high priority and
have access one of two watermarks lower than "min" which can be referred
to as the "atomic reserve".  __GFP_HIGH users get access to the first
lower watermark and can be called the "high priority reserve".

Over time, callers had a requirement to not block when fallback options
were available.  Some have abused __GFP_WAIT leading to a situation where
an optimisitic allocation with a fallback option can access atomic
reserves.

This patch uses __GFP_ATOMIC to identify callers that are truely atomic,
cannot sleep and have no alternative.  High priority users continue to use
__GFP_HIGH.  __GFP_DIRECT_RECLAIM identifies callers that can sleep and
are willing to enter direct reclaim.  __GFP_KSWAPD_RECLAIM to identify
callers that want to wake kswapd for background reclaim.  __GFP_WAIT is
redefined as a caller that is willing to enter direct reclaim and wake
kswapd for background reclaim.

This patch then converts a number of sites

o __GFP_ATOMIC is used by callers that are high priority and have memory
  pools for those requests. GFP_ATOMIC uses this flag.

o Callers that have a limited mempool to guarantee forward progress clear
  __GFP_DIRECT_RECLAIM but keep __GFP_KSWAPD_RECLAIM. bio allocations fall
  into this category where kswapd will still be woken but atomic reserves
  are not used as there is a one-entry mempool to guarantee progress.

o Callers that are checking if they are non-blocking should use the
  helper gfpflags_allow_blocking() where possible. This is because
  checking for __GFP_WAIT as was done historically now can trigger false
  positives. Some exceptions like dm-crypt.c exist where the code intent
  is clearer if __GFP_DIRECT_RECLAIM is used instead of the helper due to
  flag manipulations.

o Callers that built their own GFP flags instead of starting with GFP_KERNEL
  and friends now also need to specify __GFP_KSWAPD_RECLAIM.

The first key hazard to watch out for is callers that removed __GFP_WAIT
and was depending on access to atomic reserves for inconspicuous reasons.
In some cases it may be appropriate for them to use __GFP_HIGH.

The second key hazard is callers that assembled their own combination of
GFP flags instead of starting with something like GFP_KERNEL.  They may
now wish to specify __GFP_KSWAPD_RECLAIM.  It's almost certainly harmless
if it's missed in most cases as other activity will wake kswapd.
Signed-off-by: NMel Gorman <mgorman@techsingularity.net>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Vitaly Wool <vitalywool@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently there is no limitation on number of reserved credits we can
ask for. If we ask for more reserved credits than 1/2 of maximum
transaction size, or if total number of credits exceeds the maximum
transaction size per operation (which is currently only possible with
the former) we will spin forever in start_this_handle().

Fix this by adding this limitation at the start of start_this_handle().

This patch also removes the credit limitation 1/2 of maximum transaction
size, since we really only want to limit the number of reserved credits.
There is not much point to limit the credits if there is still space in
the journal.

This accidentally also fixes the online resize, where due to the
limitation of the journal credits we're unable to grow file systems with
1k block size and size between 16M and 32M. It has been partially fixed
by 2c869b26, but not entirely.

Thanks Jan Kara for helping me getting the correct fix.
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

It is often the case that we mark buffer as having dirty metadata when
the buffer is already in that state (frequent for bitmaps, inode table
blocks, superblock). Thus it is unnecessary to contend on grabbing
journal head reference and bh_state lock. Avoid that by checking whether
any modification to the buffer is needed before grabbing any locks or
references.

[ Note: this is a fixed version of commit 2143c196, which was
  reverted in ebeaa8dd due to a false positive triggering of an
  assertion check. -- Ted ]
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

This reverts commit 2143c196.

This commit seems to be the cause of the following jbd2 assertion
failure:

   ------------[ cut here ]------------
   kernel BUG at fs/jbd2/transaction.c:1325!
   invalid opcode: 0000 [#1] SMP
   Modules linked in: bnep bluetooth fuse ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 nf_conntrack_ipv6 ...
   CPU: 7 PID: 5509 Comm: gcc Not tainted 4.1.0-10944-g2a298679 #1
   Hardware name:                  /DH87RL, BIOS RLH8710H.86A.0327.2014.0924.1645 09/24/2014
   task: ffff8803bf866040 ti: ffff880308528000 task.ti: ffff880308528000
   RIP: jbd2_journal_dirty_metadata+0x237/0x290
   Call Trace:
     __ext4_handle_dirty_metadata+0x43/0x1f0
     ext4_handle_dirty_dirent_node+0xde/0x160
     ? jbd2_journal_get_write_access+0x36/0x50
     ext4_delete_entry+0x112/0x160
     ? __ext4_journal_start_sb+0x52/0xb0
     ext4_unlink+0xfa/0x260
     vfs_unlink+0xec/0x190
     do_unlinkat+0x24a/0x270
     SyS_unlink+0x11/0x20
     entry_SYSCALL_64_fastpath+0x12/0x6a
   ---[ end trace ae033ebde8d080b4 ]---

which is not easily reproducible (I've seen it just once, and then Ted
was able to reproduce it once).  Revert it while Ted and Jan try to
figure out what is wrong.

Cc: Jan Kara <jack@suse.cz>
Acked-by: NTheodore Ts'o <tytso@mit.edu>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It is often the case that we mark buffer as having dirty metadata when
the buffer is already in that state (frequent for bitmaps, inode table
blocks, superblock). Thus it is unnecessary to contend on grabbing
journal head reference and bh_state lock. Avoid that by checking whether
any modification to the buffer is needed before grabbing any locks or
references.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

jbd2_journal_get_write_access() and jbd2_journal_get_create_access() are
frequently called for buffers that are already part of the running
transaction - most frequently it is the case for bitmaps, inode table
blocks, and superblock. Since in such cases we have nothing to do, it is
unfortunate we still grab reference to journal head, lock the bh, lock
bh_state only to find out there's nothing to do.

Improving this is a bit subtle though since until we find out journal
head is attached to the running transaction, it can disappear from under
us because checkpointing / commit decided it's no longer needed. We deal
with this by protecting journal_head slab with RCU. We still have to be
careful about journal head being freed & reallocated within slab and
about exposing journal head in consistent state (in particular
b_modified and b_frozen_data must be in correct state before we allow
user to touch the buffer).
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

Check for the simple case of unjournaled buffer first, handle it and
bail out. This allows us to remove one if and unindent the difficult case
by one tab. The result is easier to read.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

We were acquiring bh_state_lock when allocation of buffer failed in
do_get_write_access() only to be able to jump to a label that releases
the lock and does all other checks that don't make sense for this error
path. Just jump into the right label instead.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

needs_copy is set only in one place in do_get_write_access(), just move
the frozen buffer copying into that place and factor it out to a
separate function to make do_get_write_access() slightly more readable.
Signed-off-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>

This basically reverts 47def826 (jbd2: Remove __GFP_NOFAIL from jbd2
layer). The deprecation of __GFP_NOFAIL was a bad choice because it led
to open coding the endless loop around the allocator rather than
removing the dependency on the non failing allocation. So the
deprecation was a clear failure and the reality tells us that
__GFP_NOFAIL is not even close to go away.

It is still true that __GFP_NOFAIL allocations are generally discouraged
and new uses should be evaluated and an alternative (pre-allocations or
reservations) should be considered but it doesn't make any sense to lie
the allocator about the requirements. Allocator can take steps to help
making a progress if it knows the requirements.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Acked-by: NDavid Rientjes <rientjes@google.com>

Currently when journal restart fails, we'll have the h_transaction of
the handle set to NULL to indicate that the handle has been effectively
aborted. We handle this situation quietly in the jbd2_journal_stop() and just
free the handle and exit because everything else has been done before we
attempted (and failed) to restart the journal.

Unfortunately there are a number of problems with that approach
introduced with commit

41a5b913 "jbd2: invalidate handle if jbd2_journal_restart()
fails"

First of all in ext4 jbd2_journal_stop() will be called through
__ext4_journal_stop() where we would try to get a hold of the superblock
by dereferencing h_transaction which in this case would lead to NULL
pointer dereference and crash.

In addition we're going to free the handle regardless of the refcount
which is bad as well, because others up the call chain will still
reference the handle so we might potentially reference already freed
memory.

Moreover it's expected that we'll get aborted handle as well as detached
handle in some of the journalling function as the error propagates up
the stack, so it's unnecessary to call WARN_ON every time we get
detached handle.

And finally we might leak some memory by forgetting to free reserved
handle in jbd2_journal_stop() in the case where handle was detached from
the transaction (h_transaction is NULL).

Fix the NULL pointer dereference in __ext4_journal_stop() by just
calling jbd2_journal_stop() quietly as suggested by Jan Kara. Also fix
the potential memory leak in jbd2_journal_stop() and use proper
handle refcounting before we attempt to free it to avoid use-after-free
issues.

And finally remove all WARN_ON(!transaction) from the code so that we do
not get random traces when something goes wrong because when journal
restart fails we will get to some of those functions.

Cc: stable@vger.kernel.org
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: NTheodore Ts'o <tytso@mit.edu>
Reviewed-by: NJan Kara <jack@suse.cz>