We need to obtain the i_mutex, i_iolock and i_ilock during the read
and write paths. Add a set of wrapper functions to neatly
encapsulate the lock ordering and shared/exclusive semantics to make
the locking easier to follow and get right.

Note that this changes some of the exclusive locking serialisation in
that serialisation will occur against the i_mutex instead of the
XFS_IOLOCK_EXCL. This does not change any behaviour, and it is
arguably more efficient to use the mutex for such serialisation than
the rw_sem.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

xfs_file_aio_write() only returns the error from synchronous
flushing of the data and inode if error == 0. At the point where
error is being checked, it is guaranteed to be > 0. Therefore any
errors returned by the data or fsync flush will never be returned.
Fix the checks so we overwrite the current error once and only if an
error really occurred.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

The log grant ticket wait queues are currently protected by the log
grant lock.  However, the queues are functionally independent from
each other, and operations on them only require serialisation
against other queue operations now that all of the other log
variables they use are atomic values.

Hence, we can make them independent of the grant lock by introducing
new locks just to protect the lists operations. because the lists
are independent, we can use a lock per list and ensure that reserve
and write head queuing do not contend.

To ensure forced shutdowns work correctly in conjunction with the
new fast paths, ensure that we check whether the log has been shut
down in the grant functions once we hold the relevant spin locks but
before we go to sleep. This is needed to co-ordinate correctly with
the wakeups that are issued on the ticket queues so we don't leave
any processes sleeping on the queues during a shutdown.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

log->l_tail_lsn is currently protected by the log grant lock. The
lock is only needed for serialising readers against writers, so we
don't really need the lock if we make the l_tail_lsn variable an
atomic. Converting the l_tail_lsn variable to an atomic64_t means we
can start to peel back the grant lock from various operations.

Also, provide functions to safely crack an atomic LSN variable into
it's component pieces and to recombined the components into an
atomic variable. Use them where appropriate.

This also removes the need for explicitly holding a spinlock to read
the l_tail_lsn on 32 bit platforms.
Signed-off-by: NDave Chinner <dchinner@redhat.com>

The log grant queues are one of the few places left using sv_t
constructs for waiting. Given we are touching this code, we should
convert them to plain wait queues. While there, convert all the
other sv_t users in the log code as well.

Seeing as this removes the last users of the sv_t type, remove the
header file defining the wrapper and the fragments that still
reference it.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Prepare for switching the grant heads to atomic variables by
combining the two 32 bit values that make up the grant head into a
single 64 bit variable.  Provide wrapper functions to combine and
split the grant heads appropriately for calculations and use them as
necessary.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

The grant write and reserve queues use a roll-your-own double linked
list, so convert it to a standard list_head structure and convert
all the list traversals to use list_for_each_entry(). We can also
get rid of the XLOG_TIC_IN_Q flag as we can use the list_empty()
check to tell if the ticket is in a list or not.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

The xfaild often tries to rest to wait for congestion to pass of for
IO to complete, but is regularly woken in tail-pushing situations.
In severe cases, the xfsaild is getting woken tens of thousands of
times a second. Reduce the number needless wakeups by only waking
the xfsaild if the new target is larger than the old one. Further
make short sleeps uninterruptible as they occur when the xfsaild has
decided it needs to back off to allow some IO to complete and being
woken early is counter-productive.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Now that the buffer reclaim infrastructure can handle different reclaim
priorities for different types of buffers, reconnect the hooks in the
XFS code that has been sitting dormant since it was ported to Linux. This
should finally give use reclaim prioritisation that is on a par with the
functionality that Irix provided XFS 15 years ago.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Introduce a per-buftarg LRU for memory reclaim to operate on. This
is the last piece we need to put in place so that we can fully
control the buffer lifecycle. This allows XFS to be responsibile for
maintaining the working set of buffers under memory pressure instead
of relying on the VM reclaim not to take pages we need out from
underneath us.

The implementation introduces a b_lru_ref counter into the buffer.
This is currently set to 1 whenever the buffer is referenced and so is used to
determine if the buffer should be added to the LRU or not when freed.
Effectively it allows lazy LRU initialisation of the buffer so we do not need
to touch the LRU list and locks in xfs_buf_find().

Instead, when the buffer is being released and we drop the last
reference to it, we check the b_lru_ref count and if it is none zero
we re-add the buffer reference and add the inode to the LRU. The
b_lru_ref counter is decremented by the shrinker, and whenever the
shrinker comes across a buffer with a zero b_lru_ref counter, if
released the LRU reference on the buffer. In the absence of a lookup
race, this will result in the buffer being freed.

This counting mechanism is used instead of a reference flag so that
it is simple to re-introduce buffer-type specific reclaim reference
counts to prioritise reclaim more effectively. We still have all
those hooks in the XFS code, so this will provide the infrastructure
to re-implement that functionality.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Before we introduce per-buftarg LRU lists, split the shrinker
implementation into per-buftarg shrinker callbacks. At the moment
we wake all the xfsbufds to run the delayed write queues to free
the dirty buffers and make their pages available for reclaim.
However, with an LRU, we want to be able to free clean, unused
buffers as well, so we need to separate the xfsbufd from the
shrinker callbacks.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>

now that we are using RCU protection for the inode cache lookups,
the lock is only needed on the modification side. Hence it is not
necessary for the lock to be a rwlock as there are no read side
holders anymore. Convert it to a spin lock to reflect it's exclusive
nature.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

With delayed logging greatly increasing the sustained parallelism of inode
operations, the inode cache locking is showing significant read vs write
contention when inode reclaim runs at the same time as lookups. There is
also a lot more write lock acquistions than there are read locks (4:1 ratio)
so the read locking is not really buying us much in the way of parallelism.

To avoid the read vs write contention, change the cache to use RCU locking on
the read side. To avoid needing to RCU free every single inode, use the built
in slab RCU freeing mechanism. This requires us to be able to detect lookups of
freed inodes, so enѕure that ever freed inode has an inode number of zero and
the XFS_IRECLAIM flag set. We already check the XFS_IRECLAIM flag in cache hit
lookup path, but also add a check for a zero inode number as well.

We canthen convert all the read locking lockups to use RCU read side locking
and hence remove all read side locking.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NAlex Elder <aelder@sgi.com>

The XFS iolock needs to be re-initialised to a new lock class before
it enters reclaim to prevent lockdep false positives. Unfortunately,
this is not sufficient protection as inodes in the XFS_IRECLAIMABLE
state can be recycled and not re-initialised before being reused.

We need to re-initialise the lock state when transfering out of
XFS_IRECLAIMABLE state to XFS_INEW, but we need to keep the same
class as if the inode was just allocated. Hence we need a specific
lockdep class variable for the iolock so that both initialisations
use the same class.

While there, add a specific class for inodes in the reclaim state so
that it is easy to tell from lockdep reports what state the inode
was in that generated the report.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Use a goto label to consolidate all block not found cases, and add a
tracepoint for them.  Also clean up a few whitespace issues.

Based on an earlier patch from Dave Chinner.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Move the buffer locking into the callers as they need to do it
wether they call xfs_map_at_offset or not.  Remove the b_bdev
assignment, which is already done by get_blocks.  Remove the
duplicate extent type asserts in xfs_convert_page just before
calling xfs_map_at_offset.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

After the last patches the code for overwrites is the same as for
delayed and unwritten extents except that it doesn't need to call
xfs_map_at_offset.  Take care of that fact to simplify
xfs_vm_writepage.

The buffer loop now first checks the type of buffer and checks/sets
the ioend type, or continues to the next buffer if it's not
interesting to us.  Only after that we validate the iomap and
perform the block mapping if needed, all in common code for the
cases where we have to do work.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

The all_bh flag is always set when entering the page clustering
machinery with a regular written extent, which means the check for
it is superflous.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

xfs_map_blocks always calls xfs_bmapi with the XFS_BMAPI_ENTIRE
entire flag, which tells it to not cap the extent at the passed in
size, but just treat the size as an minimum to map.  This means
xfs_probe_cluster is entirely useless as we'll always get the whole
extent back anyway.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

No need to lock the extent map exclusive when performing an
overwrite, we know the extent map must already have been loaded by
get_blocks.  Apply the non-blocking inode semantics to all mapping
types instead of just delayed allocations.  Remove the handling of
not yet allocated blocks for the IO_UNWRITTEN case - if an extent is
marked as unwritten allocated in the buffer it must already have an
extent on disk.

Add asserts to verify all the assumptions above in debug builds.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Opencode the xfs_iomap code in it's two callers.  The overlap of
passed flags already was minimal and will be further reduced in the
next patch.

As a side effect the BMAPI_* flags for xfs_bmapi and the IO_* flags
for I/O end processing are merged into a single set of flags, which
should be a bit more descriptive of the operation we perform.

Also improve the tracing by giving each caller it's own type set of
tracepoints.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Don't trylock the buffer.  We are the only one ever locking it for a
regular file address space, and trylock was only copied from the
generic code which did it due to the old buffer based writeout in
jbd.  Also make sure to only write out the buffer if the iomap
actually is valid, because we wouldn't have a proper mapping
otherwise.  In practice we will never get an invalid mapping here as
the page lock guarantees truncate doesn't race with us, but better
be safe than sorry.  Also make sure we allocate a new ioend when
crossing boundaries between mappings, just like we do for delalloc
and unwritten extents.  Again this currently doesn't matter as the
I/O end handler only cares for the boundaries for unwritten extents,
but this makes the code fully correct and the same as for
delalloc/unwritten extents.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

We'll never have BIO_EOPNOTSUPP set after calling submit_bio as this
can only happen for discards, and used to happen for barriers, none
of which is every submitted by xfs_submit_ioend_bio.  Also remove
the loop around bio_alloc as it will never fail due to it's mempool
backing.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Currently we only refuse a "read-only" mapping for writing out
unwritten and delayed buffers, and refuse any other for overwrites.
Improve the checks to require delalloc mappings for delayed buffers,
and unwritten extent mappings for unwritten extents.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

We now support mounting and using filesystems with 64-bit inodes
even when not mounted with the inode64 option (which now only
controls if we allocate new inodes in that space or not).  Make sure
we always use large NFS file handles when exporting a filesystem
that may contain 64-bit inodes.  Note that this only affects newly
generated file handles, any outstanding 32-bit file handle is still
accepted.

[hch: the comment and commit log are mine, the rest is from a patch
 snipplet from Samuel]
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

As reported by Nick Piggin, XFS is suffering from long pauses under
highly concurrent workloads when hosted on ramdisks. The problem is
that an inode buffer is stuck in the pinned state in memory and as a
result either the inode buffer or one of the inodes within the
buffer is stopping the tail of the log from being moved forward.

The system remains in this state until a periodic log force issued
by xfssyncd causes the buffer to be unpinned. The main problem is
that these are stale buffers, and are hence held locked until the
transaction/checkpoint that marked them state has been committed to
disk. When the filesystem gets into this state, only the xfssyncd
can cause the async transactions to be committed to disk and hence
unpin the inode buffer.

This problem was encountered when scaling the busy extent list, but
only the blocking lock interface was fixed to solve the problem.
Extend the same fix to the buffer trylock operations - if we fail to
lock a pinned, stale buffer, then force the log immediately so that
when the next attempt to lock it comes around, it will have been
unpinned.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Since the move to the new truncate sequence we call xfs_setattr to
truncate down excessively instanciated blocks.  As shown by the testcase
in kernel.org BZ #22452 that doesn't work too well.  Due to the confusion
of the internal inode size, and the VFS inode i_size it zeroes data that
it shouldn't.

But full blown truncate seems like overkill here.  We only instanciate
delayed allocations in the write path, and given that we never released
the iolock we can't have converted them to real allocations yet either.

The only nasty case is pre-existing preallocation which we need to skip.
We already do this for page discard during writeback, so make the delayed
allocation block punching a generic function and call it from the failed
write path as well as xfs_aops_discard_page. The callers are
responsible for ensuring that partial blocks are not truncated away,
and that they hold the ilock.

Based on a fix originally from Christoph Hellwig. This version used
filesystem blocks as the range unit.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

In commit 20cb52eb, titled
"xfs: simplify xfs_vm_writepage" I added an assert that any !mapped and
uptodate buffers are not dirty.  That asserts turns out to trigger a lot
when running fsx on filesystems with small block sizes.  The reason for
that is that the assert is simply incorrect.  !mapped and uptodate
just mean this buffer covers a hole, and whenever we do a set_page_dirty
we mark all blocks in the page dirty, no matter if they have data or
not.  So remove the assert, and update the comment above the condition
to match reality.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

XFS does not need it's inodes to actuall be hashed in the VFS inode
cache, but we require the inode to be marked hashed for the
writeback code to work.

Insted of using insert_inode_hash, which requires a second
inode_lock roundtrip after the partial merge of the inode
scalability patches in 2.6.37-rc simply use the new hlist_add_fake
helper to mark it hashed without requiring a lock or touching a
global cache line.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

The delayed write buffer split trace currently issues a trace for
every buffer it scans. These buffers are not necessarily queued for
delayed write. Indeed, when buffers are pinned, there can be
thousands of traces of buffers that aren't actually queued for
delayed write and the ones that are are lost in the noise. Move the
trace point to record only buffers that are split out for IO to be
issued on.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

The walk fails to decrement the per-ag reference count when the
non-blocking walk fails to obtain the per-ag reclaim lock, leading
to an assert failure on debug kernels when unmounting a filesystem.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

al_hreq is copied from userland.  If al_hreq.buflen is not properly aligned
then xfs_attr_list will ignore the last bytes of kbuf.  These bytes are
unitialized.  It leads to leaking of contents of kernel stack memory.
Signed-off-by: NVasiliy Kulikov <segooon@gmail.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

We promised to do this for 2.6.37, and the code looks stable enough to
keep that promise.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

... and switch of the obvious get_sb_bdev() users to ->mount()
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This removes more dead code that was somehow missed by commit 0d99519e
(writeback: remove unused nonblocking and congestion checks).  There are
no behavior change except for the removal of two entries from one of the
ext4 tracing interface.

The nonblocking checks in ->writepages are no longer used because the
flusher now prefer to block on get_request_wait() than to skip inodes on
IO congestion.  The latter will lead to more seeky IO.

The nonblocking checks in ->writepage are no longer used because it's
redundant with the WB_SYNC_NONE check.

We no long set ->nonblocking in VM page out and page migration, because
a) it's effectively redundant with WB_SYNC_NONE in current code
b) it's old semantic of "Don't get stuck on request queues" is mis-behavior:
   that would skip some dirty inodes on congestion and page out others, which
   is unfair in terms of LRU age.

Inspired by Christoph Hellwig. Thanks!
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: David Howells <dhowells@redhat.com>
Cc: Sage Weil <sage@newdream.net>
Cc: Steve French <sfrench@samba.org>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Instead of always assigning an increasing inode number in new_inode
move the call to assign it into those callers that actually need it.
For now callers that need it is estimated conservatively, that is
the call is added to all filesystems that do not assign an i_ino
by themselves.  For a few more filesystems we can avoid assigning
any inode number given that they aren't user visible, and for others
it could be done lazily when an inode number is actually needed,
but that's left for later patches.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Clones an existing reference to inode; caller must already hold one.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Split up inode_add_to_list/__inode_add_to_list.  Locking for the two
lists will be split soon so these helpers really don't buy us much
anymore.

The __ prefixes for the sb list helpers will go away soon, but until
inode_lock is gone we'll need them to distinguish between the locked
and unlocked variants.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>