The comment is not coincident with the code. Fix it.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

div_factor{_fine} has been implemented for two times, cleanup it.
And I move them into a independent file named math.h because they are
common math functions.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@fusionio.com>

The direct-IO write path already had the i_size checks in mm/filemap.c,
but it turns out the read path did not, and removing the block size
checks in fs/block_dev.c (commit bbec0270: "blkdev_max_block: make
private to fs/buffer.c") removed the magic "shrink IO to past the end of
the device" code there.

Fix it by truncating the IO to the size of the block device, like the
write path already does.

NOTE! I suspect the write path would be *much* better off doing it this
way in fs/block_dev.c, rather than hidden deep in mm/filemap.c.  The
mm/filemap.c code is extremely hard to follow, and has various
conditionals on the target being a block device (ie the flag passed in
to 'generic_write_checks()', along with a conditional update of the
inode timestamp etc).

It is also quite possible that we should treat this whole block device
size as a "s_maxbytes" issue, and try to make the logic even more
generic.  However, in the meantime this is the fairly minimal targeted
fix.

Noted by Milan Broz thanks to a regression test for the cryptsetup
reencrypt tool.
Reported-and-tested-by: NMilan Broz <mbroz@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

READ is zero so the "rw & READ" test is always false.  The intended test
was "((rw & RW_MASK) == READ)".
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The block device access simplification that avoided accessing the (racy)
block size information (commit bbec0270: "blkdev_max_block: make
private to fs/buffer.c") no longer checks the maximum block size in the
block mapping path.

That was _almost_ as simple as just removing the code entirely, because
the readers and writers all check the size of the device anyway, so
under normal circumstances it "just worked".

However, the block size may be such that the end of the device may
straddle one single buffer_head.  At which point we may still want to
access the end of the device, but the buffer we use to access it
partially extends past the end.

The 'bd_set_size()' function intentionally sets the block size to avoid
this, but mounting the device - or setting the block size by hand to
some other value - can modify that block size.

So instead, teach 'submit_bh()' about the special case of the buffer
head straddling the end of the device, and turning such an access into a
smaller IO access, avoiding the problem.

This, btw, also means that unlike before, we can now access the whole
device regardless of device block size setting.  So now, even if the
device size is only 512-byte aligned, we can read and write even the
last sector even when having a much bigger block size for accessing the
rest of the device.

So with this, we could now get rid of the 'bd_set_size()' block size
code entirely - resulting in faster IO for the common case - but that
would be a separate patch.
Reported-and-tested-by: NRomain Francoise <romain@orebokech.com>
Reporeted-and-tested-by: NMeelis Roos <mroos@linux.ee>
Reported-by: NTony Luck <tony.luck@intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Noticed by Pavel Roskin; the thing in his patch I disagree with
was compensating for that shite in callbacks instead of fixing
it once in the iterator itself.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

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

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

We are leaking fattr and fhandle if we decide that dentry is not to
be invalidated, after all (e.g. happens to be a mountpoint).  Just
free both before that...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

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

We really don't want to look at the block size for the raw block device
accesses in fs/block-dev.c, because it may be changing from under us.
So get rid of the max_block logic entirely, since the caller should
already have done it anyway.

That leaves the only user of this function in fs/buffer.c, so move the
whole function there and make it static.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since directio can work on a raw block device, and the block size of the
device can change under it, we need to do the same thing that
fs/buffer.c now does: read the block size a single time, using
ACCESS_ONCE().

Reading it multiple times can get different results, which will then
confuse the code because it actually encodes the i_blksize in
relationship to the underlying logical blocksize.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This reverts the block-device direct access code to the previous
unlocked code, now that fs/buffer.c no longer needs external locking.

With this, fs/block_dev.c is back to the original version, apart from a
whitespace cleanup that I didn't want to revert.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This makes the buffer size handling be a per-page thing, which allows us
to not have to worry about locking too much when changing the buffer
size.  If a page doesn't have buffers, we still need to read the block
size from the inode, but we can do that with ACCESS_ONCE(), so that even
if the size is changing, we get a consistent value.

This doesn't convert all functions - many of the buffer functions are
used purely by filesystems, which in turn results in the buffer size
being fixed at mount-time.  So they don't have the same consistency
issues that the raw device access can have.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 6bdf6dbd caused a regression
in setattr codepath that leads to files with wrong attributes.
Signed-off-by: NPavel Shilovsky <piastry@etersoft.ru>
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <smfrench@gmail.com>

Commit eddb079d created a regression in the writepages codepath.
Previously, whenever it needed to check the size of the file, it did so
by consulting the inode->i_size field directly. With that patch, the
i_size was fetched once on entry into the writepages code and that value
was used henceforth.

If the file is changing size though (for instance, if someone is writing
to it or has truncated it), then that value is likely to be wrong. This
can lead to data corruption. Pages past the EOF at the time that the
writepages call was issued may be silently dropped and ignored because
cifs_writepages wrongly assumes that the file must have been truncated
in the interim.

Fix cifs_writepages to properly fetch the size from the inode->i_size
field instead to properly account for this possibility.

Original bug report is here:

    https://bugzilla.kernel.org/show_bug.cgi?id=50991Reported-and-Tested-by: NMaxim Britov <ungifted01@gmail.com>
Reviewed-by: NSuresh Jayaraman <sjayaraman@suse.com>
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <smfrench@gmail.com>

Commit 169ebd90 ("writeback: Avoid iput() from flusher thread")
removed iget-iput pair from inode writeback.  As a side effect, inodes
that are dirty during iput_final() call won't be ever added to inode LRU
(iput_final() doesn't add dirty inodes to LRU and later when the inode
is cleaned there's noone to add the inode there).  Thus inodes are
effectively unreclaimable until someone looks them up again.

The practical effect of this bug is limited by the fact that inodes are
pinned by a dentry for long enough that the inode gets cleaned.  But
still the bug can have nasty consequences leading up to OOM conditions
under certain circumstances.  Following can easily reproduce the
problem:

  for (( i = 0; i < 1000; i++ )); do
    mkdir $i
    for (( j = 0; j < 1000; j++ )); do
      touch $i/$j
      echo 2 > /proc/sys/vm/drop_caches
    done
  done

then one needs to run 'sync; ls -lR' to make inodes reclaimable again.

We fix the issue by inserting unused clean inodes into the LRU after
writeback finishes in inode_sync_complete().
Signed-off-by: NJan Kara <jack@suse.cz>
Reported-by: NOGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: <stable@vger.kernel.org>		[3.5+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 7b540d06 ("proc_map_files_readdir(): don't bother with
grabbing files") switched proc_map_files_readdir() to use @f_mode
directly instead of grabbing @file reference, but same time the test for
@vm_file presence was lost leading to nil dereference.  The patch brings
the test back.

The all proc_map_files feature is CONFIG_CHECKPOINT_RESTORE wrapped
(which is set to 'n' by default) so the bug doesn't affect regular
kernels.

The regression is 3.7-rc1 only as far as I can tell.

[gorcunov@openvz.org: provided changelog]
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Acked-by: NCyrill Gorcunov <gorcunov@openvz.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 09e05d48 introduced a wait for transaction commit into
journal_unmap_buffer() in the case we are truncating a buffer undergoing commit
in the page stradding i_size on a filesystem with blocksize < pagesize. Sadly
we forgot to drop buffer lock before waiting for transaction commit and thus
deadlock is possible when kjournald wants to lock the buffer.

Fix the problem by dropping the buffer lock before waiting for transaction
commit. Since we are still holding page lock (and that is OK), buffer cannot
disappear under us.

CC: stable@vger.kernel.org # Wherever commit 09e05d48 was taken
Signed-off-by: NJan Kara <jack@suse.cz>

Currently if len argument in ext3_trim_fs() is smaller than one block,
the 'end' variable underflow. Avoid that by returning EINVAL if len is
smaller than file system block.

Also remove useless unlikely().
Signed-off-by: NLukas Czerner <lczerner@redhat.com>
Signed-off-by: NJan Kara <jack@suse.cz>

Calls into highlevel quota code cannot happen under the write lock. These
calls take dqio_mutex which ranks above write lock. So drop write lock
before calling back into quota code.

CC: stable@vger.kernel.org # >= 3.0
Signed-off-by: NJan Kara <jack@suse.cz>

Calls into reiserfs journalling code and reiserfs_get_block() need to
be protected with write lock. We remove write lock around calls to high
level quota code in the next patch so these paths would suddently become
unprotected.

CC: stable@vger.kernel.org # >= 3.0
Signed-off-by: NJan Kara <jack@suse.cz>

In reiserfs_quota_on() we do quite some work - for example unpacking
tail of a quota file. Thus we have to hold write lock until a moment
we call back into the quota code.

CC: stable@vger.kernel.org # >= 3.0
Signed-off-by: NJan Kara <jack@suse.cz>

When remounting reiserfs dquot_suspend() or dquot_resume() can be called.
These functions take dqonoff_mutex which ranks above write lock so we have
to drop it before calling into quota code.

CC: stable@vger.kernel.org # >= 3.0
Signed-off-by: NJan Kara <jack@suse.cz>

If the FAN_Q_OVERFLOW bit set in event->mask, the fanotify event
metadata will not contain a valid file descriptor, but
copy_event_to_user() didn't check for that, and unconditionally does a
fd_install() on the file descriptor.

Which in turn will cause a BUG_ON() in __fd_install().

Introduced by commit 352e3b24 ("fanotify: sanitize failure exits in
copy_event_to_user()")

Mea culpa - missed that path ;-/
Reported-by: NAlex Shi <lkml.alex@gmail.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Error handling in xfs_buf_ioapply_map() does not handle IO reference
counts correctly. We increment the b_io_remaining count before
building the bio, but then fail to decrement it in the failure case.
This leads to the buffer never running IO completion and releasing
the reference that the IO holds, so at unmount we can leak the
buffer. This leak is captured by this assert failure during unmount:

XFS: Assertion failed: atomic_read(&pag->pag_ref) == 0, file: fs/xfs/xfs_mount.c, line: 273

This is not a new bug - the b_io_remaining accounting has had this
problem for a long, long time - it's just very hard to get a
zero length bio being built by this code...

Further, the buffer IO error can be overwritten on a multi-segment
buffer by subsequent bio completions for partial sections of the
buffer. Hence we should only set the buffer error status if the
buffer is not already carrying an error status. This ensures that a
partial IO error on a multi-segment buffer will not be lost. This
part of the problem is a regression, however.

cc: <stable@vger.kernel.org>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

When we shut down the filesystem, it might first be detected in
writeback when we are allocating a inode size transaction. This
happens after we have moved all the pages into the writeback state
and unlocked them. Unfortunately, if we fail to set up the
transaction we then abort writeback and try to invalidate the
current page. This then triggers are BUG() in block_invalidatepage()
because we are trying to invalidate an unlocked page.

Fixing this is a bit of a chicken and egg problem - we can't
allocate the transaction until we've clustered all the pages into
the IO and we know the size of it (i.e. whether the last block of
the IO is beyond the current EOF or not). However, we don't want to
hold pages locked for long periods of time, especially while we lock
other pages to cluster them into the write.

To fix this, we need to make a clear delineation in writeback where
errors can only be handled by IO completion processing. That is,
once we have marked a page for writeback and unlocked it, we have to
report errors via IO completion because we've already started the
IO. We may not have submitted any IO, but we've changed the page
state to indicate that it is under IO so we must now use the IO
completion path to report errors.

To do this, add an error field to xfs_submit_ioend() to pass it the
error that occurred during the building on the ioend chain. When
this is non-zero, mark each ioend with the error and call
xfs_finish_ioend() directly rather than building bios. This will
immediately push the ioends through completion processing with the
error that has occurred.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

In certain circumstances, a double split of an attribute tree is
needed to insert or replace an attribute. In rare situations, this
can go wrong, leaving the attribute tree corrupted. In this case,
the attr being replaced is the last attr in a leaf node, and the
replacement is larger so doesn't fit in the same leaf node.
When we have the initial condition of a node format attribute
btree with two leaves at index 1 and 2. Call them L1 and L2.  The
leaf L1 is completely full, there is not a single byte of free space
in it. L2 is mostly empty.  The attribute being replaced - call it X
- is the last attribute in L1.

The way an attribute replace is executed is that the replacement
attribute - call it Y - is first inserted into the tree, but has an
INCOMPLETE flag set on it so that list traversals ignore it. Once
this transaction is committed, a second transaction it run to
atomically mark Y as COMPLETE and X as INCOMPLETE, so that a
traversal will now find Y and skip X. Once that transaction is
committed, attribute X is then removed.

So, the initial condition is:

     +--------+     +--------+
     |   L1   |     |   L2   |
     | fwd: 2 |---->| fwd: 0 |
     | bwd: 0 |<----| bwd: 1 |
     | fsp: 0 |     | fsp: N |
     |--------|     |--------|
     | attr A |     | attr 1 |
     |--------|     |--------|
     | attr B |     | attr 2 |
     |--------|     |--------|
     ..........     ..........
     |--------|     |--------|
     | attr X |     | attr n |
     +--------+     +--------+

So now we go to replace X, and see that L1:fsp = 0 - it is full so
we can't insert Y in the same leaf. So we record the the location of
attribute X so we can track it for later use, then we split L1 into
L1 and L3 and reblance across the two leafs. We end with:

     +--------+     +--------+     +--------+
     |   L1   |     |   L3   |     |   L2   |
     | fwd: 3 |---->| fwd: 2 |---->| fwd: 0 |
     | bwd: 0 |<----| bwd: 1 |<----| bwd: 3 |
     | fsp: M |     | fsp: J |     | fsp: N |
     |--------|     |--------|     |--------|
     | attr A |     | attr X |     | attr 1 |
     |--------|     +--------+     |--------|
     | attr B |                    | attr 2 |
     |--------|                    |--------|
     ..........                    ..........
     |--------|                    |--------|
     | attr W |                    | attr n |
     +--------+                    +--------+

And we track that the original attribute is now at L3:0.

We then try to insert Y into L1 again, and find that there isn't
enough room because the new attribute is larger than the old one.
Hence we have to split again to make room for Y. We end up with
this:

     +--------+     +--------+     +--------+     +--------+
     |   L1   |     |   L4   |     |   L3   |     |   L2   |
     | fwd: 4 |---->| fwd: 3 |---->| fwd: 2 |---->| fwd: 0 |
     | bwd: 0 |<----| bwd: 1 |<----| bwd: 4 |<----| bwd: 3 |
     | fsp: M |     | fsp: J |     | fsp: J |     | fsp: N |
     |--------|     |--------|     |--------|     |--------|
     | attr A |     | attr Y |     | attr X |     | attr 1 |
     |--------|     + INCOMP +     +--------+     |--------|
     | attr B |     +--------+                    | attr 2 |
     |--------|                                   |--------|
     ..........                                   ..........
     |--------|                                   |--------|
     | attr W |                                   | attr n |
     +--------+                                   +--------+

And now we have the new (incomplete) attribute @ L4:0, and the
original attribute at L3:0. At this point, the first transaction is
committed, and we move to the flipping of the flags.

This is where we are supposed to end up with this:

     +--------+     +--------+     +--------+     +--------+
     |   L1   |     |   L4   |     |   L3   |     |   L2   |
     | fwd: 4 |---->| fwd: 3 |---->| fwd: 2 |---->| fwd: 0 |
     | bwd: 0 |<----| bwd: 1 |<----| bwd: 4 |<----| bwd: 3 |
     | fsp: M |     | fsp: J |     | fsp: J |     | fsp: N |
     |--------|     |--------|     |--------|     |--------|
     | attr A |     | attr Y |     | attr X |     | attr 1 |
     |--------|     +--------+     + INCOMP +     |--------|
     | attr B |                    +--------+     | attr 2 |
     |--------|                                   |--------|
     ..........                                   ..........
     |--------|                                   |--------|
     | attr W |                                   | attr n |
     +--------+                                   +--------+

But that doesn't happen properly - the attribute tracking indexes
are not pointing to the right locations. What we end up with is both
the old attribute to be removed pointing at L4:0 and the new
attribute at L4:1.  On a debug kernel, this assert fails like so:

XFS: Assertion failed: args->index2 < be16_to_cpu(leaf2->hdr.count), file: fs/xfs/xfs_attr_leaf.c, line: 2725

because the new attribute location does not exist. On a production
kernel, this goes unnoticed and the code proceeds ahead merrily and
removes L4 because it thinks that is the block that is no longer
needed. This leaves the hash index node pointing to entries
L1, L4 and L2, but only blocks L1, L3 and L2 to exist. Further, the
leaf level sibling list is L1 <-> L4 <-> L2, but L4 is now free
space, and so everything is busted. This corruption is caused by the
removal of the old attribute triggering a join - it joins everything
correctly but then frees the wrong block.

xfs_repair will report something like:

bad sibling back pointer for block 4 in attribute fork for inode 131
problem with attribute contents in inode 131
would clear attr fork
bad nblocks 8 for inode 131, would reset to 3
bad anextents 4 for inode 131, would reset to 0

The problem lies in the assignment of the old/new blocks for
tracking purposes when the double leaf split occurs. The first split
tries to place the new attribute inside the current leaf (i.e.
"inleaf == true") and moves the old attribute (X) to the new block.
This sets up the old block/index to L1:X, and newly allocated
block to L3:0. It then moves attr X to the new block and tries to
insert attr Y at the old index. That fails, so it splits again.

With the second split, the rebalance ends up placing the new attr in
the second new block - L4:0 - and this is where the code goes wrong.
What is does is it sets both the new and old block index to the
second new block. Hence it inserts attr Y at the right place (L4:0)
but overwrites the current location of the attr to replace that is
held in the new block index (currently L3:0). It over writes it with
L4:1 - the index we later assert fail on.

Hopefully this table will show this in a foramt that is a bit easier
to understand:

Split		old attr index		new attr index
		vanilla	patched		vanilla	patched
before 1st	L1:26	L1:26		N/A	N/A
after 1st	L3:0	L3:0		L1:26	L1:26
after 2nd	L4:0	L3:0		L4:1	L4:0
                ^^^^			^^^^
		wrong			wrong

The fix is surprisingly simple, for all this analysis - just stop
the rebalance on the out-of leaf case from overwriting the new attr
index - it's already correct for the double split case.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

This is mostly a revert of 01dc52eb ("oom: remove deprecated oom_adj")
from Davidlohr Bueso.

It reintroduces /proc/pid/oom_adj for backwards compatibility with earlier
kernels.  It simply scales the value linearly when /proc/pid/oom_score_adj
is written.

The major difference is that its scheduled removal is no longer included
in Documentation/feature-removal-schedule.txt.  We do warn users with a
single printk, though, to suggest the more powerful and supported
/proc/pid/oom_score_adj interface.
Reported-by: NArtem S. Tashkinov <t.artem@lycos.com>
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Passing a NULL id causes a NULL pointer deference in writers such as
erst_writer and efi_pstore_write because they expect to update this id.
Pass a dummy id instead.

This avoids a cascade of oopses caused when the initial
pstore_console_write passes a null which in turn causes writes to the
console causing further oopses in subsequent pstore_console_write calls.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Acked-by: NKees Cook <keescook@chromium.org>
Cc: stable@vger.kernel.org
Signed-off-by: NAnton Vorontsov <anton.vorontsov@linaro.org>

It can be legitimately triggered via procfs access.  Now, at least
2 of 3 of get_files_struct() callers in procfs are useless, but
when and if we get rid of those we can always add WARN_ON() here.
BUG_ON() at that spot is simply wrong.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

jffs2_write_begin() first acquires the page lock, then f->sem. This
causes an AB-BA deadlock with jffs2_garbage_collect_live(), which first
acquires f->sem, then the page lock:

jffs2_garbage_collect_live
    mutex_lock(&f->sem)                         (A)
    jffs2_garbage_collect_dnode
        jffs2_gc_fetch_page
            read_cache_page_async
                do_read_cache_page
                    lock_page(page)             (B)

jffs2_write_begin
    grab_cache_page_write_begin
        find_lock_page
            lock_page(page)                     (B)
    mutex_lock(&f->sem)                         (A)

We fix this by restructuring jffs2_write_begin() to take f->sem before
the page lock. However, we make sure that f->sem is not held when
calling jffs2_reserve_space(), as this is not permitted by the locking
rules.

The deadlock above was observed multiple times on an SoC with a dual
ARMv7 (Cortex-A9), running the long-term 3.4.11 kernel; it occurred
when using scp to copy files from a host system to the ARM target
system. The fix was heavily tested on the same target system.

Cc: stable@vger.kernel.org
Signed-off-by: NThomas Betker <thomas.betker@rohde-schwarz.com>
Acked-by: NJoakim Tjernlund <Joakim.Tjernlund@transmode.se>
Signed-off-by: NArtem Bityutskiy <artem.bityutskiy@linux.intel.com>

Anders Blomdell noted in 2010 that Fanotify lost events and provided a
test case.  Eric Paris confirmed it was a bug and posted a fix to the
list

  https://groups.google.com/forum/?fromgroups=#!topic/linux.kernel/RrJfTfyW2BE

but never applied it.  Repeated attempts over time to actually get him
to apply it have never had a reply from anyone who has raised it

So apply it anyway
Signed-off-by: NAlan Cox <alan@linux.intel.com>
Reported-by: NAnders Blomdell <anders.blomdell@control.lth.se>
Cc: Eric Paris <eparis@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Revert commit 03a7beb5 ("epoll: support for disabling items, and a
self-test app") pending resolution of the issues identified by Michael
Kerrisk, copied below.

We'll revisit this for 3.8.

: I've taken a look at this patch as it currently stands in 3.7-rc1, and
: done a bit of testing. (By the way, the test program
: tools/testing/selftests/epoll/test_epoll.c does not compile...)
:
: There are one or two places where the behavior seems a little strange,
: so I have a question or two at the end of this mail. But other than
: that, I want to check my understanding so that the interface can be
: correctly documented.
:
: Just to go though my understanding, the problem is the following
: scenario in a multithreaded application:
:
: 1. Multiple threads are performing epoll_wait() operations,
:    and maintaining a user-space cache that contains information
:    corresponding to each file descriptor being monitored by
:    epoll_wait().
:
: 2. At some point, a thread wants to delete (EPOLL_CTL_DEL)
:    a file descriptor from the epoll interest list, and
:    delete the corresponding record from the user-space cache.
:
: 3. The problem with (2) is that some other thread may have
:    previously done an epoll_wait() that retrieved information
:    about the fd in question, and may be in the middle of using
:    information in the cache that relates to that fd. Thus,
:    there is a potential race.
:
: 4. The race can't solved purely in user space, because doing
:    so would require applying a mutex across the epoll_wait()
:    call, which would of course blow thread concurrency.
:
: Right?
:
: Your solution is the EPOLL_CTL_DISABLE operation. I want to
: confirm my understanding about how to use this flag, since
: the description that has accompanied the patches so far
: has been a bit sparse
:
: 0. In the scenario you're concerned about, deleting a file
:    descriptor means (safely) doing the following:
:    (a) Deleting the file descriptor from the epoll interest list
:        using EPOLL_CTL_DEL
:    (b) Deleting the corresponding record in the user-space cache
:
: 1. It's only meaningful to use this EPOLL_CTL_DISABLE in
:    conjunction with EPOLLONESHOT.
:
: 2. Using EPOLL_CTL_DISABLE without using EPOLLONESHOT in
:    conjunction is a logical error.
:
: 3. The correct way to code multithreaded applications using
:    EPOLL_CTL_DISABLE and EPOLLONESHOT is as follows:
:
:    a. All EPOLL_CTL_ADD and EPOLL_CTL_MOD operations should
:       should EPOLLONESHOT.
:
:    b. When a thread wants to delete a file descriptor, it
:       should do the following:
:
:       [1] Call epoll_ctl(EPOLL_CTL_DISABLE)
:       [2] If the return status from epoll_ctl(EPOLL_CTL_DISABLE)
:           was zero, then the file descriptor can be safely
:           deleted by the thread that made this call.
:       [3] If the epoll_ctl(EPOLL_CTL_DISABLE) fails with EBUSY,
:           then the descriptor is in use. In this case, the calling
:           thread should set a flag in the user-space cache to
:           indicate that the thread that is using the descriptor
:           should perform the deletion operation.
:
: Is all of the above correct?
:
: The implementation depends on checking on whether
: (events & ~EP_PRIVATE_BITS) == 0
: This replies on the fact that EPOLL_CTL_AD and EPOLL_CTL_MOD always
: set EPOLLHUP and EPOLLERR in the 'events' mask, and EPOLLONESHOT
: causes those flags (as well as all others in ~EP_PRIVATE_BITS) to be
: cleared.
:
: A corollary to the previous paragraph is that using EPOLL_CTL_DISABLE
: is only useful in conjunction with EPOLLONESHOT. However, as things
: stand, one can use EPOLL_CTL_DISABLE on a file descriptor that does
: not have EPOLLONESHOT set in 'events' This results in the following
: (slightly surprising) behavior:
:
: (a) The first call to epoll_ctl(EPOLL_CTL_DISABLE) returns 0
:     (the indicator that the file descriptor can be safely deleted).
: (b) The next call to epoll_ctl(EPOLL_CTL_DISABLE) fails with EBUSY.
:
: This doesn't seem particularly useful, and in fact is probably an
: indication that the user made a logic error: they should only be using
: epoll_ctl(EPOLL_CTL_DISABLE) on a file descriptor for which
: EPOLLONESHOT was set in 'events'. If that is correct, then would it
: not make sense to return an error to user space for this case?

Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: "Paton J. Lewis" <palewis@adobe.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 44396476 ("xfs: reset buffer pointers before freeing them") in
3.0-rc1 introduced a regression when recovering log buffers that
wrapped around the end of log. The second part of the log buffer at
the start of the physical log was being read into the header buffer
rather than the data buffer, and hence recovery was seeing garbage
in the data buffer when it got to the region of the log buffer that
was incorrectly read.

Cc: <stable@vger.kernel.org> # 3.0.x, 3.2.x, 3.4.x 3.6.x
Reported-by: NTorsten Kaiser <just.for.lkml@googlemail.com>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

When we shut down the filesystem, we have to unpin and free all the
buffers currently active in the CIL. To do this we unpin and remove
them in one operation as a result of a failed iclogbuf write. For
buffers, we do this removal via a simultated IO completion of after
marking the buffer stale.

At the time we do this, we have two references to the buffer - the
active LRU reference and the buf log item.  The LRU reference is
removed by marking the buffer stale, and the active CIL reference is
by the xfs_buf_iodone() callback that is run by
xfs_buf_do_callbacks() during ioend processing (via the bp->b_iodone
callback).

However, ioend processing requires one more reference - that of the
IO that it is completing. We don't have this reference, so we free
the buffer prematurely and use it after it is freed. For buffers
marked with XBF_ASYNC, this leads to assert failures in
xfs_buf_rele() on debug kernels because the b_hold count is zero.

Fix this by making sure we take the necessary IO reference before
starting IO completion processing on the stale buffer, and set the
XBF_ASYNC flag to ensure that IO completion processing removes all
the active references from the buffer to ensure it is fully torn
down.

Cc: <stable@vger.kernel.org>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Inode buffers do not need to be mapped as inodes are read or written
directly from/to the pages underlying the buffer. This fixes a
regression introduced by commit 611c9946 ("xfs: make XBF_MAPPED the
default behaviour").
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

When we free a block from the alloc btree tree, we move it to the
freelist held in the AGFL and mark it busy in the busy extent tree.
This typically happens when we merge btree blocks.

Once the transaction is committed and checkpointed, the block can
remain on the free list for an indefinite amount of time.  Now, this
isn't the end of the world at this point - if the free list is
shortened, the buffer is invalidated in the transaction that moves
it back to free space. If the buffer is allocated as metadata from
the free list, then all the modifications getted logged, and we have
no issues, either. And if it gets allocated as userdata direct from
the freelist, it gets invalidated and so will never get written.

However, during the time it sits on the free list, pressure on the
log can cause the AIL to be pushed and the buffer that covers the
block gets pushed for write. IOWs, we end up writing a freed
metadata block to disk. Again, this isn't the end of the world
because we know from the above we are only writing to free space.

The problem, however, is for validation callbacks. If the block was
on old btree root block, then the level of the block is going to be
higher than the current tree root, and so will fail validation.
There may be other inconsistencies in the block as well, and
currently we don't care because the block is in free space. Shutting
down the filesystem because a freed block doesn't pass write
validation, OTOH, is rather unfriendly.

So, make sure we always invalidate buffers as they move from the
free space trees to the free list so that we guarantee they never
get written to disk while on the free list.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NPhil White <pwhite@sgi.com>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

Uninitialised variable build warning introduced by 2903ff01 ("switch
simple cases of fget_light to fdget"), gcc is not smart enough to
work out that the variable is not used uninitialised, and the commit
removed the initialisation at declaration that the old variable had.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NMark Tinguely <tinguely@sgi.com>
Signed-off-by: NBen Myers <bpm@sgi.com>

When updating new secondary superblocks in a growfs operation, the
superblock buffer is read from the newly grown region of the
underlying device. This is not guaranteed to be zero, so violates
the underlying assumption that the unused parts of superblocks are
zero filled. Get a new buffer for these secondary superblocks to
ensure that the unused regions are zero filled correctly.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NCarlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: NBen Myers <bpm@sgi.com>