When doing large parallel file creates on a 16p machines, large amounts of
time is being spent in _xfs_buf_find(). A system wide profile with perf top
shows this:

          1134740.00 19.3% _xfs_buf_find
           733142.00 12.5% __ticket_spin_lock

The problem is that the hash contains 45,000 buffers, and the hash table width
is only 256 buffers. That means we've got around 200 buffers per chain, and
searching it is quite expensive. The hash table size needs to increase.

Secondly, every time we do a lookup, we promote the buffer we find to the head
of the hash chain. This is causing cachelines to be dirtied and causes
invalidation of cachelines across all CPUs that may have walked the hash chain
recently. hence every walk of the hash chain is effectively a cold cache walk.
Remove the promotion to avoid this invalidation.

The results are:

          1045043.00 21.2% __ticket_spin_lock
           326184.00  6.6% _xfs_buf_find

A 70% drop in the CPU usage when looking up buffers. Unfortunately that does
not result in an increase in performance underthis workload as contention on
the inode_lock soaks up most of the reduction in CPU usage.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

If xfs_map_blocks returns EAGAIN because of lock contention we must redirty the
page and not disard the pagecache content and return an error from writepage.
We used to do this correctly, but the logic got lost during the recent
reshuffle of the writepage code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reported-by: NMike Gao <ygao.linux@gmail.com>
Tested-by: NMike Gao <ygao.linux@gmail.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <dchinner@redhat.com>

When we  need to cover the log, we issue dummy transactions to ensure
the current log tail is on disk. Unfortunately we currently use the
root inode in the dummy transaction, and the act of committing the
transaction dirties the inode at the VFS level.

As a result, the VFS writeback of the dirty inode will prevent the
filesystem from idling long enough for the log covering state
machine to complete. The state machine gets stuck in a loop issuing
new dummy transactions to cover the log and never makes progress.

To avoid this problem, the dummy transactions should not cause
externally visible state changes. To ensure this occurs, make sure
that dummy transactions log an unchanging field in the superblock as
it's state is never propagated outside the filesystem. This allows
the log covering state machine to complete successfully and the
filesystem now correctly enters a fully idle state about 90s after
the last modification was made.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Because of delayed updates to sb_icount field in the super block, it
is possible to allocate over maxicount number of inodes.  This
causes the arithmetic to calculate a negative number of free inodes
in user commands like df or stat -f.

Since maxicount is a somewhat arbitrary number, a slight over
allocation is not critical but user commands should be displayed as
0 or greater and never go negative.  To do this the value in the
stats buffer f_ffree is capped to never go negative.

[ Modified to use max_t as per Christoph's comment. ]
Signed-off-by: NStu Brodsky <sbrodsky@sgi.com>
Signed-off-by: NDave Chinner <dchinner@redhat.com>

During data integrity (WB_SYNC_ALL) writeback, wbc->nr_to_write will
go negative on inodes with more than 1024 dirty pages due to
implementation details of write_cache_pages(). Currently XFS will
abort page clustering in writeback once nr_to_write drops below
zero, and so for data integrity writeback we will do very
inefficient page at a time allocation and IO submission for inodes
with large numbers of dirty pages.

Fix this by only aborting the page clustering code when
wbc->nr_to_write is negative and the sync mode is WB_SYNC_NONE.

Cc: <stable@kernel.org>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

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

add I_CLEAR instead of replacing I_FREEING with it.  I_CLEAR is
equivalent to I_FREEING for almost all code looking at either;
it's there to keep track of having called clear_inode() exactly
once per inode lifetime, at some point after having set I_FREEING.
I_CLEAR and I_FREEING never get set at the same time with the
current code, so we can switch to setting i_flags to I_FREEING | I_CLEAR
instead of I_CLEAR without loss of information.  As the result of
such change, checks become simpler and the amount of code that needs
to know about I_CLEAR shrinks a lot.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Convert XFS to the new truncate sequence.  We still can have errors after
updating the file size in xfs_setattr, but these are real I/O errors and lead
to a transaction abort and filesystem shutdown, so they are not an issue.

Errors from ->write_begin and write_end can now be handled correctly because
we can actually get rid of the delalloc extents while previous the buffer
state was stipped in block_invalidatepage.

There is still no error handling for ->direct_IO, because doing so will need
some major restructuring given that we only have the iolock shared and do not
hold i_mutex at all.  Fortunately leaving the normally allocated blocks behind
there is not a major issue and this will get cleaned up by xfs_free_eofblock
later.

Note: the patch is against Al's vfs.git tree as that contains the nessecary
preparations.  I'd prefer to get it applied there so that we can get some
testing in linux-next.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Move the call to vmtruncate to get rid of accessive blocks to the callers
in preparation of the new truncate sequence and rename the non-truncating
version to block_write_begin.

While we're at it also remove several unused arguments to block_write_begin.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Move the call to vmtruncate to get rid of accessive blocks to the callers
in prepearation of the new truncate calling sequence.  This was only done
for DIO_LOCKING filesystems, so the __blockdev_direct_IO_newtrunc variant
was not needed anyway.  Get rid of blockdev_direct_IO_no_locking and
its _newtrunc variant while at it as just opencoding the two additional
paramters is shorted than the name suffix.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Filesystems with unwritten extent support must not complete an AIO request
until the transaction to convert the extent has been commited.  That means
the aio_complete calls needs to be moved into the ->end_io callback so
that the filesystem can control when to call it exactly.

This makes a bit of a mess out of dio_complete and the ->end_io callback
prototype even more complicated. 
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz> 
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

Our current handling of direct I/O completions is rather suboptimal,
because we defer it to a workqueue more often than needed, and we
perform a much to aggressive flush of the workqueue in case unwritten
extent conversions happen.

This patch changes the direct I/O reads to not even use a completion
handler, as we don't bother to use it at all, and to perform the unwritten
extent conversions in caller context for synchronous direct I/O.

For a small I/O size direct I/O workload on a consumer grade SSD, such as
the untar of a kernel tree inside qemu this patch gives speedups of
about 5%.  Getting us much closer to the speed of a native block device,
or a fully allocated XFS file.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

If we write into an unwritten extent using AIO we need to complete the AIO
request after the extent conversion has finished.  Without that a read could
race to see see the extent still unwritten and return zeros.   For synchronous
I/O we already take care of that by flushing the xfsconvertd workqueue (which
might be a bit of overkill).

To do that add iocb and result fields to struct xfs_ioend, so that we can
call aio_complete from xfs_end_io after the extent conversion has happened.
Note that we need a new result field as io_error is used for positive errno
values, while the AIO code can return negative error values and positive
transfer sizes.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Filesystems with unwritten extent support must not complete an AIO request
until the transaction to convert the extent has been commited.  That means
the aio_complete calls needs to be moved into the ->end_io callback so
that the filesystem can control when to call it exactly.

This makes a bit of a mess out of dio_complete and the ->end_io callback
prototype even more complicated.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NAlex Elder <aelder@sgi.com>

The b_strat callback is used by xfs_buf_iostrategy to perform additional
checks before submitting a buffer.  It is used in xfs_bwrite and when
writing out delayed buffers.  In xfs_bwrite it we can de-virtualize the
call easily as b_strat is set a few lines above the call to
xfs_buf_iostrategy.  For the delayed buffers the rationale is a bit
more complicated:

 - there are three callers of xfs_buf_delwri_queue, which places buffers
   on the delwri list:
    (1) xfs_bdwrite - this sets up b_strat, so it's fine
    (2) xfs_buf_iorequest.  None of the callers can have XBF_DELWRI set:
	- xlog_bdstrat is only used for log buffers, which are never delwri
	- _xfs_buf_read explicitly clears the delwri flag
	- xfs_buf_iodone_work retries log buffers only
	- xfsbdstrat - only used for reads, superblock writes without the
	  delwri flag, log I/O and file zeroing with explicitly allocated
	  buffers.
	- xfs_buf_iostrategy - only calls xfs_buf_iorequest if b_strat is
	  not set
    (3) xfs_buf_unlock
	- only puts the buffer on the delwri list if the DELWRI flag is
	  already set.  The DELWRI flag is only ever set in xfs_bwrite,
	  xfs_buf_iodone_callbacks, or xfs_trans_log_buf.  For
	  xfs_buf_iodone_callbacks and xfs_trans_log_buf we require
	  an initialized buf item, which means b_strat was set to
	  xfs_bdstrat_cb in xfs_buf_item_init.

Conclusion: we can just get rid of the callback and replace it with
explicit calls to xfs_bdstrat_cb.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Since Linux 2.6.33 the kernel has support for real O_SYNC, which made
the osyncisosync option a no-op.  Warn the users about this and remove
the mount flag for it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

When CONFIG_XFS_POSIX_ACL is not set "xfs_check_acl" is #defined
to NULL - which breaks the code attempting to add a tracepoint
on this function.

Only define the tracepoint when the function exists.
Signed-off-by: NTony Luck <tony.luck@intel.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Avoid a lockdep warning by preventing page cache allocation from
recursing back into the filesystem during memory reclaim.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

xfs_ireclaim has to get and put te pag structure because it is only
called with the inode to reclaim. The one caller of this function
already has a reference on the pag and a pointer to is, so move the
radix tree delete to the caller and remove xfs_ireclaim completely.
This avoids a xfs_perag_get/put on every inode being reclaimed.

The overhead was noticed in a bug report at:

https://bugzilla.kernel.org/show_bug.cgi?id=16348Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

xfs_buf_read() fails to detect dispatch errors before attempting to
wait on sychronous IO. If there was an error, it will get stuck
forever, waiting for an I/O that was never started. Make sure the
error is detected correctly.

Further, such a failure can leave locked pages in the page cache
which will cause a later operation to hang on the page. Ensure that
we correctly process pages in the buffers when we get a dispatch
error.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDave Chinner <david@fromorbit.com>

I missed Dave Chinner's second revision of this change, and pushed
his first version out to the repository instead.

	commit a476c59ebb279d738718edc0e3fb76aab3687114
	Author: Dave Chinner <dchinner@redhat.com>

This commit compensates for that by moving a block of code up a bit
further, with a result that matches the the effect of Dave's second
version.

Dave's first version was:
Reviewed-by: NEric Sandeen <sandeen@redhat.com>
Dave's second version was:
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>
Reviewed-by: NEric Sandeen <sandeen@redhat.com>

The open_exec file operation is only added by the external dmapi
patch.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAlex Elder <aelder@sgi.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

These days we always have buffers thanks to ->page_mkwrite.  And we
already have an assert a few lines above tripping in case that was
not true due to a bug.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

We only need disable I/O from direct or memcg reclaim.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Currently we don't remove the XFS mount from the shrinker list until
late in the unmount path. By this time, we have already torn down
the internals of the filesystem (e.g. the per-ag structures), and
hence if the shrinker is executed between the teardown and the
unregistering, the shrinker will get NULL per-ag structure pointers
and panic trying to dereference them.

Fix this by removing the xfs mount from the shrinker list before
tearing down it's internal structures.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NEric Sandeen <sandeen@redhat.com>
Signed-off-by: NAlex Elder <aelder@sgi.com>

Replace the xfs_itrace_entry catchall with specific trace points.  For
most simple callers we now use the simple inode class, which used to
be the iget class, but add more details tracing for namespace events,
which now includes the name of the directory entries manipulated.

Remove the xfs_inactive trace point, which is a duplicate of the clear_inode
one, and the xfs_change_file_space trace point, which is immediately
followed by the more specific alloc/free space trace points.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

We never get an i_mode of 0 or a locked VFS inode until we pass in the
XFS_IGET_CREATE flag to xfs_iget, which makes xfs_iput_new equivalent to
xfs_iput for the only caller.  In addition to that xfs_nfs_get_inode
does not even need to lock the inode given that the generation never changes
for a life inode, so just pass a 0 lock_flags to xfs_iget and release
the inode using IRELE in the error path.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

The xfs_iget_alloc/found tracepoints are a bit misnamed and misplaced.
Rename them to xfs_iget_hit/xfs_iget_miss and move them to the beggining
of the xfs_iget_cache_hit/miss functions.  Add a new xfs_iget_reclaim_fail
tracepoint for the case where we fail to re-initialize a VFS inode,
and add a second instance of the xfs_iget_skip tracepoint for the case
of a failed igrab() call.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

The tracing code can't print flags defined as enums.  Most flags that
we want to print are defines as macros already, but move the few remaining
ones over to make the trace output more useful.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

On the final put of a superblock the VFS already calls sync_filesystem
for us to write out all data and wait for it.  No need to start another
asynchronous writeback inside ->put_super.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Remove the flags argument to  __xfs_get_blocks as we can easily derive
it from the direct argument, and remove the unused BMAPI_MMAP flag.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

We already rely on the fact that the sync code will cause a synchronous
log force later on (currently via xfs_fs_sync_fs -> xfs_quiesce_data ->
xfs_sync_data), so no need to do this here.  This allows us to avoid
a lot of synchronous log forces during sync, which pays of especially
with delayed logging enabled.   Some compilebench numbers that show
this:

xfs (delayed logging, 256k logbufs)
===================================

intial create		  25.94 MB/s	  25.75 MB/s	  25.64 MB/s
create			   8.54 MB/s	   9.12 MB/s	   9.15 MB/s
patch			   2.47 MB/s	   2.47 MB/s	   3.17 MB/s
compile			  29.65 MB/s	  30.51 MB/s	  27.33 MB/s
clean			  90.92 MB/s	  98.83 MB/s	 128.87 MB/s
read tree		  11.90 MB/s	  11.84 MB/s	   8.56 MB/s
read compiled		  28.75 MB/s	  29.96 MB/s	  24.25 MB/s
delete tree		8.39 seconds	8.12 seconds	8.46 seconds
delete compiled		8.35 seconds	8.44 seconds	5.11 seconds
stat tree		6.03 seconds	5.59 seconds	5.19 seconds
stat compiled tree	9.00 seconds	9.52 seconds	8.49 seconds

xfs + write_inode log_force removal
===================================
intial create		  25.87 MB/s	  25.76 MB/s	  25.87 MB/s
create			  15.18 MB/s	  14.80 MB/s	  14.94 MB/s
patch			   3.13 MB/s	   3.14 MB/s	   3.11 MB/s
compile			  36.74 MB/s	  37.17 MB/s	  36.84 MB/s
clean			 226.02 MB/s	 222.58 MB/s	 217.94 MB/s
read tree		  15.14 MB/s	  15.02 MB/s	  15.14 MB/s
read compiled tree	  29.30 MB/s	  29.31 MB/s	  29.32 MB/s
delete tree		6.22 seconds	6.14 seconds	6.15 seconds
delete compiled tree	5.75 seconds	5.92 seconds	5.81 seconds
stat tree		4.60 seconds	4.51 seconds	4.56 seconds
stat compiled tree	4.07 seconds	3.87 seconds	3.96 seconds

In addition to that also remove the delwri inode flush that is unessecary
now that bulkstat is always coherent.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

The writepage implementation in XFS still tries to deal with dirty but
unmapped buffers which used to caused by writes through shared mmaps.  Since
the introduction of ->page_mkwrite these can't happen anymore, so remove the
code dealing with them.

Note that the all_bh variable which causes us to start I/O on all buffers on
the pages was controlled by the count of unmapped buffers, which also
included those not actually dirty.  It's now unconditionally initialized to
0 but set to 1 for the case of small file size extensions.  It probably can
be removed entirely, but that's left for another patch.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Currently the xfs releasepage implementation has code to deal with converting
delayed allocated and unwritten space.  But we never get called for those as
we always convert delayed and unwritten space when cleaning a page, or drop
the state from the buffers in block_invalidatepage.  We still keep a WARN_ON
on those cases for now, but remove all the case dealing with it, which allows
to fold xfs_page_state_convert into xfs_vm_writepage and remove the !startio
case from the whole writeback path.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

xfstests 194 first truncats a file back and then extends it again by
truncating it to a larger size.  This causes discard_buffer to drop
the mapped, but not the uptodate bit and thus creates something that
xfs_page_state_convert takes for unmapped space created by mmap because
it doesn't check for the dirty bit, which also gets cleared by
discard_buffer and checked by other ->writepage implementations like
block_write_full_page.  Handle this kind of buffers early, and unlike
Eric's first version of the patch simply ASSERT that the buffers is
dirty, given that the mmap write case can't happen anymore since the
introduction of ->page_mkwrite.  The now dead code dealing with that
will be deleted in a follow on patch.
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

This code was introduced four years ago in commit
3e57ecf6 without any review and has
been unused since.  Remove it just as the rest of the code introduced
in that commit to reduce that stack usage and complexity in this central
piece of code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Currently we need to either call IHOLD or xfs_trans_ihold on an inode when
joining it to a transaction via xfs_trans_ijoin.

This patches instead makes xfs_trans_ijoin usable on it's own by doing
an implicity xfs_trans_ihold, which also allows us to drop the third
argument.  For the case where we want to hold a reference on the inode
a xfs_trans_ijoin_ref wrapper is added which does the IHOLD and marks
the inode for needing an xfs_iput.  In addition to the cleaner interface
to the caller this also simplifies the implementation.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Get rid of the xfs_buf_pin/xfs_buf_unpin/xfs_buf_ispin helpers and opencode
them in their only callers, just like we did for the inode pinning a while
ago.  Also remove duplicate trace points - the bufitem tracepoints cover
all the information that is present in a buffer tracepoint.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Currently we track log item descriptor belonging to a transaction using a
complex opencoded chunk allocator.  This code has been there since day one
and seems to work around the lack of an efficient slab allocator.

This patch replaces it with dynamically allocated log item descriptors
from a dedicated slab pool, linked to the transaction by a linked list.

This allows to greatly simplify the log item descriptor tracking to the
point where it's just a couple hundred lines in xfs_trans.c instead of
a separate file.  The external API has also been simplified while we're
at it - the xfs_trans_add_item and xfs_trans_del_item functions to add/
delete items from a transaction have been simplified to the bare minium,
and the xfs_trans_find_item function is replaced with a direct dereference
of the li_desc field.  All debug code walking the list of log items in
a transaction is down to a simple list_for_each_entry.

Note that we could easily use a singly linked list here instead of the
double linked list from list.h as the fastpath only does deletion from
sequential traversal.  But given that we don't have one available as
a library function yet I use the list.h functions for simplicity.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>