These days inode64 should only control which AGs we allocate new
inodes from, while we still try to support reading all existing
inodes. To make this actually work the check ontop of xfs_iget
needs to be relaxed to allow inodes in all allocation groups instead
of just those that we allow allocating inodes from. Note that we
can't simply remove the check - it prevents us from accessing
invalid data when fed invalid inode numbers from NFS or bulkstat.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Update the per-cpu counters manually in xfs_trans_unreserve_and_mod_sb
and remove support for per-cpu counters from xfs_mod_incore_sb_batch
to simplify it. And added benefit is that we don't have to take
m_sb_lock for transactions that only modify per-cpu counters.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Export xfs_icsb_modify_counters and always use it for modifying
the per-cpu counters. Remove support for per-cpu counters from
xfs_mod_incore_sb to simplify it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Fail the mount if we can't allocate memory for the per-CPU counters.
This is consistent with how we handle everything else in the mount
path and makes the superblock counter modification a lot simpler.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
pahole reports the struct xfs_buf has quite a few holes in it, so
packing the structure better will reduce the size of it by 16 bytes.
Also, move all the fields used in cache lookups into the first
cacheline.
Before on x86_64:
/* size: 320, cachelines: 5 */
/* sum members: 298, holes: 6, sum holes: 22 */
After on x86_64:
/* size: 304, cachelines: 5 */
/* padding: 6 */
/* last cacheline: 48 bytes */
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
The buffer cache hash is showing typical hash scalability problems.
In large scale testing the number of cached items growing far larger
than the hash can efficiently handle. Hence we need to move to a
self-scaling cache indexing mechanism.
I have selected rbtrees for indexing becuse they can have O(log n)
search scalability, and insert and remove cost is not excessive,
even on large trees. Hence we should be able to cache large numbers
of buffers without incurring the excessive cache miss search
penalties that the hash is imposing on us.
To ensure we still have parallel access to the cache, we need
multiple trees. Rather than hashing the buffers by disk address to
select a tree, it seems more sensible to separate trees by typical
access patterns. Most operations use buffers from within a single AG
at a time, so rather than searching lots of different lists,
separate the buffer indexes out into per-AG rbtrees. This means that
searches during metadata operation have a much higher chance of
hitting cache resident nodes, and that updates of the tree are less
likely to disturb trees being accessed on other CPUs doing
independent operations.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Memory reclaim via shrinkers has a terrible habit of having N+M
concurrent shrinker executions (N = num CPUs, M = num kswapds) all
trying to shrink the same cache. When the cache they are all working
on is protected by a single spinlock, massive contention an
slowdowns occur.
Wrap the per-ag inode caches with a reclaim mutex to serialise
reclaim access to the AG. This will block concurrent reclaim in each
AG but still allow reclaim to scan multiple AGs concurrently. Allow
shrinkers to move on to the next AG if it can't get the lock, and if
we can't get any AG, then start blocking on locks.
To prevent reclaimers from continually scanning the same inodes in
each AG, add a cursor that tracks where the last reclaim got up to
and start from that point on the next reclaim. This should avoid
only ever scanning a small number of inodes at the satart of each AG
and not making progress. If we have a non-shrinker based reclaim
pass, ignore the cursor and reset it to zero once we are done.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Alex Elder <aelder@sgi.com>
Batch and optimise the per-ag inode lookup for reclaim to minimise
scanning overhead. This involves gang lookups on the radix trees to
get multiple inodes during each tree walk, and tighter validation of
what inodes can be reclaimed without blocking befor we take any
locks.
This is based on ideas suggested in a proof-of-concept patch
posted by Nick Piggin.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
With the reclaim code separated from the generic walking code, it is
simple to implement batched lookups for the generic walk code.
Separate out the inode validation from the execute operations and
modify the tree lookups to get a batch of inodes at a time.
Reclaim operations will be optimised separately.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
When doing read side inode cache walks, the code to validate and
grab an inode is common to all callers. Split it out of the execute
callbacks in preparation for batching lookups. Similarly, split out
the inode reference dropping from the execute callbacks into the
main lookup look to be symmetric with the grab.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
The reclaim walk requires different locking and has a slightly
different walk algorithm, so separate it out so that it can be
optimised separately.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
For RT and external log devices, we never use hashed buffers on them
now. Remove the buftarg hash tables that are set up for them.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
When we are checking we can access the last block of each device, we
do not need to use cached buffers as they will be tossed away
immediately. Use uncached buffers for size checks so that all IO
prior to full in-memory structure initialisation does not use the
buffer cache.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Filesystem level managed buffers are buffers that have their
lifecycle controlled by the filesystem layer, not the buffer cache.
We currently cache these buffers, which makes cleanup and cache
walking somewhat troublesome. Convert the fs managed buffers to
uncached buffers obtained by via xfs_buf_get_uncached(), and remove
the XBF_FS_MANAGED special cases from the buffer cache.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Each buffer contains both a buftarg pointer and a mount pointer. If
we add a mount pointer into the buftarg, we can avoid needing the
b_mount field in every buffer and grab it from the buftarg when
needed instead. This shrinks the xfs_buf by 8 bytes.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
To avoid the need to use cached buffers for single-shot or buffers
cached at the filesystem level, introduce a new buffer read
primitive that bypasses the cache an reads directly from disk.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
xfs_buf_get_nodaddr() is really used to allocate a buffer that is
uncached. While it is not directly assigned a disk address, the fact
that they are not cached is a more important distinction. With the
upcoming uncached buffer read primitive, we should be consistent
with this disctinction.
While there, make page allocation in xfs_buf_get_nodaddr() safe
against memory reclaim re-entrancy into the filesystem by allowing
a flags parameter to be passed.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Under heavy multi-way parallel create workloads, the VFS struggles
to write back all the inodes that have been changed in age order.
The bdi flusher thread becomes CPU bound, spending 85% of it's time
in the VFS code, mostly traversing the superblock dirty inode list
to separate dirty inodes old enough to flush.
We already keep an index of all metadata changes in age order - in
the AIL - and continued log pressure will do age ordered writeback
without any extra overhead at all. If there is no pressure on the
log, the xfssyncd will periodically write back metadata in ascending
disk address offset order so will be very efficient.
Hence we can stop marking VFS inodes dirty during transaction commit
or when changing timestamps during transactions. This will keep the
inodes in the superblock dirty list to those containing data or
unlogged metadata changes.
However, the timstamp changes are slightly more complex than this -
there are a couple of places that do unlogged updates of the
timestamps, and the VFS need to be informed of these. Hence add a
new function xfs_trans_ichgtime() for transactional changes,
and leave xfs_ichgtime() for the non-transactional changes.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Alex Elder <aelder@sgi.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
When we start taking a reference to the per-ag for every cached
buffer in the system, kernel lockstat profiling on an 8-way create
workload shows the mp->m_perag_lock has higher acquisition rates
than the inode lock and has significantly more contention. That is,
it becomes the highest contended lock in the system.
The perag lookup is trivial to convert to lock-less RCU lookups
because perag structures never go away. Hence the only thing we need
to protect against is tree structure changes during a grow. This can
be done simply by replacing the locking in xfs_perag_get() with RCU
read locking. This removes the mp->m_perag_lock completely from this
path.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
When we start taking references per cached buffer to the the perag
it is cached on, it will blow the current debug maximum reference
count assert out of the water. The assert has never caught a bug,
and we have tracing to track changes if there ever is a problem,
so just remove it.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
When commiting a transaction, we do a lock CIL state lock round trip
on every single log vector we insert into the CIL. This is resulting
in the lock being as hot as the inode and dcache locks on 8-way
create workloads. Rework the insertion loops to bring the number
of lock round trips to one per transaction for log vectors, and one
more do the busy extents.
Also change the allocation of the log vector buffer not to zero it
as we copy over the entire allocated buffer anyway.
This patch also includes a structural cleanup to the CIL item
insertion provided by Christoph Hellwig.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Ionut Gabriel Popescu <poyo_vl@yahoo.com> submitted a simple change
to eliminate some "may be used uninitialized" warnings when building
XFS. The reported condition seems to be something that GCC did not
used to recognize or report. The warnings were produced by:
gcc version 4.5.0 20100604
[gcc-4_5-branch revision 160292] (SUSE Linux)
Signed-off-by: Ionut Gabriel Popescu <poyo_vl@yahoo.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
The implementation os ->kill_root only differ by either simply
zeroing out the now unused buffer in the btree cursor in the inode
allocation btree or using xfs_btree_setbuf in the allocation btree.
Initially both of them used xfs_btree_setbuf, but the use in the
ialloc btree was removed early on because it interacted badly with
xfs_trans_binval.
In addition to zeroing out the buffer in the cursor xfs_btree_setbuf
updates the bc_ra array in the btree cursor, and calls
xfs_trans_brelse on the buffer previous occupying the slot.
The bc_ra update should be done for the alloc btree updated too,
although the lack of it does not cause serious problems. The
xfs_trans_brelse call on the other hand is effectively a no-op in
the end - it keeps decrementing the bli_recur refcount until it hits
zero, and then just skips out because the buffer will always be
dirty at this point. So removing it for the allocation btree is
just fine.
So unify the code and move it to xfs_btree.c. While we're at it
also replace the call to xfs_btree_setbuf with a NULL bp argument in
xfs_btree_del_cursor with a direct call to xfs_trans_brelse given
that the cursor is beeing freed just after this and the state
updates are superflous. After this xfs_btree_setbuf is only used
with a non-NULL bp argument and can thus be simplified.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
In xfs_qm_dqflush we know that q_blkno must be initialized already from a
previous xfs_qm_dqread. So instead of calling xfs_qm_dqtobp we can
simply read the quota buffer directly. This also saves us from a duplicate
xfs_qm_dqcheck call check and allows xfs_qm_dqtobp to be simplified now
that it is always called for a newly initialized inode. In addition to
that properly unwind all locks in xfs_qm_dqflush when xfs_qm_dqcheck
fails.
This mirrors a similar cleanup in the inode lookup done earlier.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
There is no need to have the users and group/project quota locked at the
same time. Get rid of xfs_qm_dqget_noattach and just do a xfs_qm_dqget
inside xfs_qm_quotacheck_dqadjust for the quota we are operating on
right now. The new version of xfs_qm_quotacheck_dqadjust holds the
inode lock over it's operations, which is not a problem as it simply
increments counters and there is no concern about log contention
during mount time.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
XFS_IOC_ZERO_RANGE is the equivalent of an atomic XFS_IOC_UNRESVSP/
XFS_IOC_RESVSP call pair. It enabled ranges of written data to be
turned into zeroes without requiring IO or having to free and
reallocate the extents in the range given as would occur if we had
to punch and then preallocate them separately. This enables
applications to zero parts of files very quickly without changing
the layout of the files in any way.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
While XFS passes ranges to operate on from the core code, the
functions being called ignore the either the entire range or the end
of the range. This is historical because when the function were
written linux didn't have the necessary range operations. Update the
functions to use the correct operations.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
When marking an inode reclaimable, a per-AG counter is increased, the
inode is tagged reclaimable in its per-AG tree, and, when this is the
first reclaimable inode in the AG, the AG entry in the per-mount tree
is also tagged.
When an inode is finally reclaimed, however, it is only deleted from
the per-AG tree. Neither the counter is decreased, nor is the parent
tree's AG entry untagged properly.
Since the tags in the per-mount tree are not cleared, the inode
shrinker iterates over all AGs that have had reclaimable inodes at one
point in time.
The counters on the other hand signal an increasing amount of slab
objects to reclaim. Since "70e60ce xfs: convert inode shrinker to
per-filesystem context" this is not a real issue anymore because the
shrinker bails out after one iteration.
But the problem was observable on a machine running v2.6.34, where the
reclaimable work increased and each process going into direct reclaim
eventually got stuck on the xfs inode shrinking path, trying to scan
several million objects.
Fix this by properly unwinding the reclaimable-state tracking of an
inode when it is reclaimed.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: stable@kernel.org
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
I have been seeing occasional pauses in transaction throughput up to
30s long under heavy parallel workloads. The only notable thing was
that the xfsaild was trying to be active during the pauses, but
making no progress. It was running exactly 20 times a second (on the
50ms no-progress backoff), and the number of pushbuf events was
constant across this time as well. IOWs, the xfsaild appeared to be
stuck on buffers that it could not push out.
Further investigation indicated that it was trying to push out inode
buffers that were pinned and/or locked. The xfsbufd was also getting
woken at the same frequency (by the xfsaild, no doubt) to push out
delayed write buffers. The xfsbufd was not making any progress
because all the buffers in the delwri queue were pinned. This scan-
and-make-no-progress dance went one in the trace for some seconds,
before the xfssyncd came along an issued a log force, and then
things started going again.
However, I noticed something strange about the log force - there
were way too many IO's issued. 516 log buffers were written, to be
exact. That added up to 129MB of log IO, which got me very
interested because it's almost exactly 25% of the size of the log.
He delayed logging code is suppose to aggregate the minimum of 25%
of the log or 8MB worth of changes before flushing. That's what
really puzzled me - why did a log force write 129MB instead of only
8MB?
Essentially what has happened is that no CIL pushes had occurred
since the previous tail push which cleared out 25% of the log space.
That caused all the new transactions to block because there wasn't
log space for them, but they kick the xfsaild to push the tail.
However, the xfsaild was not making progress because there were
buffers it could not lock and flush, and the xfsbufd could not flush
them because they were pinned. As a result, both the xfsaild and the
xfsbufd could not move the tail of the log forward without the CIL
first committing.
The cause of the problem was that the background CIL push, which
should happen when 8MB of aggregated changes have been committed, is
being held off by the concurrent transaction commit load. The
background push does a down_write_trylock() which will fail if there
is a concurrent transaction commit holding the push lock in read
mode. With 8 CPUs all doing transactions as fast as they can, there
was enough concurrent transaction commits to hold off the background
push until tail-pushing could no longer free log space, and the halt
would occur.
It should be noted that there is no reason why it would halt at 25%
of log space used by a single CIL checkpoint. This bug could
definitely violate the "no transaction should be larger than half
the log" requirement and hence result in corruption if the system
crashed under heavy load. This sort of bug is exactly the reason why
delayed logging was tagged as experimental....
The fix is to start blocking background pushes once the threshold
has been exceeded. Rework the threshold calculations to keep the
amount of log space a CIL checkpoint can use to below that of the
AIL push threshold to avoid the problem completely.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Alex Elder <aelder@sgi.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The workqueue implementation in 2.6.36-rcX has changed, resulting
in the workqueues no longer having dedicated threads for work
processing. This has caused severe livelocks under heavy parallel
create workloads because the log IO completions have been getting
held up behind metadata IO completions. Hence log commits would
stall, memory allocation would stall because pages could not be
cleaned, and lock contention on the AIL during inode IO completion
processing was being seen to slow everything down even further.
By making the log Io completion workqueue a high priority workqueue,
they are queued ahead of all data/metadata IO completions and
processed before the data/metadata completions. Hence the log never
gets stalled, and operations needed to clean memory can continue as
quickly as possible. This avoids the livelock conditions and allos
the system to keep running under heavy load as per normal.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
The XFS_IOC_FSGETXATTR ioctl allows unprivileged users to read 12
bytes of uninitialized stack memory, because the fsxattr struct
declared on the stack in xfs_ioc_fsgetxattr() does not alter (or zero)
the 12-byte fsx_pad member before copying it back to the user. This
patch takes care of it.
Signed-off-by: Dan Rosenberg <dan.j.rosenberg@gmail.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
In xfs_vn_fiemap, we set bvm_count to fi_extent_max + 1 and want
to return fi_extent_max extents, but actually it won't work for
a sparse file. The reason is that in xfs_getbmap we will
calculate holes and set it in 'out', while out is malloced by
bmv_count(fi_extent_max+1) which didn't consider holes. So in the
worst case, if 'out' vector looks like
[hole, extent, hole, extent, hole, ... hole, extent, hole],
we will only return half of fi_extent_max extents.
This patch add a new parameter BMV_IF_NO_HOLES for bvm_iflags.
So with this flags, we don't use our 'out' in xfs_getbmap for
a hole. The solution is a bit ugly by just don't increasing
index of 'out' vector. I felt that it is not easy to skip it
at the very beginning since we have the complicated check and
some function like xfs_getbmapx_fix_eof_hole to adjust 'out'.
Cc: Dave Chinner <david@fromorbit.com>
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
If we attempt to preallocate more than 2^32 blocks of space in a
single syscall, the transaction block reservation will overflow
leading to a hangs in the superblock block accounting code. This
is trivially reproduced with xfs_io. Fix the problem by capping the
allocation reservation to the maximum number of blocks a single
xfs_bmapi() call can allocate (2^21 blocks).
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Currently on-disk structure is able to keep only 16bit project quota
id, so disallow 32bit ones. This fixes a problem where parts of
kernel structures holding project quota id are 32bit while parts
(on-disk) are 16bit variables which causes project quota member
files to be inaccessible for some operations (like mv/rm).
Signed-off-by: Arkadiusz Mi?kiewicz <arekm@maven.pl>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
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: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph 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: Christoph Hellwig <hch@lst.de>
Reported-by: Mike Gao <ygao.linux@gmail.com>
Tested-by: Mike Gao <ygao.linux@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Formatting items requires memory allocation when using delayed
logging. Currently that memory allocation is done while holding the
CIL context lock in read mode. This means that if memory allocation
takes some time (e.g. enters reclaim), we cannot push on the CIL
until the allocation(s) required by formatting complete. This can
stall CIL pushes for some time, and once a push is stalled so are
all new transaction commits.
Fix this splitting the item formatting into two steps. The first
step which does the allocation and memcpy() into the allocated
buffer is now done outside the CIL context lock, and only the CIL
insert is done inside the CIL context lock. This avoids the stall
issue.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Delayed logging adds some serialisation to the log force process to
ensure that it does not deference a bad commit context structure
when determining if a CIL push is necessary or not. It does this by
grabing the CIL context lock exclusively, then dropping it before
pushing the CIL if necessary. This causes serialisation of all log
forces and pushes regardless of whether a force is necessary or not.
As a result fsync heavy workloads (like dbench) can be significantly
slower with delayed logging than without.
To avoid this penalty, copy the current sequence from the context to
the CIL structure when they are swapped. This allows us to do
unlocked checks on the current sequence without having to worry
about dereferencing context structures that may have already been
freed. Hence we can remove the CIL context locking in the forcing
code and only call into the push code if the current context matches
the sequence we need to force.
By passing the sequence into the push code, we can check the
sequence again once we have the CIL lock held exclusive and abort if
the sequence has already been pushed. This avoids a lock round-trip
and unnecessary CIL pushes when we have racing push calls.
The result is that the regression in dbench performance goes away -
this change improves dbench performance on a ramdisk from ~2100MB/s
to ~2500MB/s. This compares favourably to not using delayed logging
which retuns ~2500MB/s for the same workload.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
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: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph 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: Stu Brodsky <sbrodsky@sgi.com>
Signed-off-by: Dave 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: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Commit 7124fe0a5b ("xfs: validate untrusted inode
numbers during lookup") changes the inode lookup code to do btree lookups for
untrusted inode numbers. This change made an invalid assumption about the
alignment of inodes and hence incorrectly calculated the first inode in the
cluster. As a result, some inode numbers were being incorrectly considered
invalid when they were actually valid.
The issue was not picked up by the xfstests suite because it always runs fsr
and dump (the two utilities that utilise the bulkstat interface) on cache hot
inodes and hence the lookup code in the cold cache path was not sufficiently
exercised to uncover this intermittent problem.
Fix the issue by relaxing the btree lookup criteria and then checking if the
record returned contains the inode number we are lookup for. If it we get an
incorrect record, then the inode number is invalid.
Cc: <stable@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Under heavy load parallel metadata loads (e.g. dbench), we can fail
to mark all the inodes in a cluster being freed as XFS_ISTALE as we
skip inodes we cannot get the XFS_ILOCK_EXCL or the flush lock on.
When this happens and the inode cluster buffer has already been
marked stale and freed, inode reclaim can try to write the inode out
as it is dirty and not marked stale. This can result in writing th
metadata to an freed extent, or in the case it has already
been overwritten trigger a magic number check failure and return an
EUCLEAN error such as:
Filesystem "ram0": inode 0x442ba1 background reclaim flush failed with 117
Fix this by ensuring that we hoover up all in memory inodes in the
cluster and mark them XFS_ISTALE when freeing the cluster.
Cc: <stable@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
When we commit a transaction using delayed logging, we need to
unlock the items in the transaciton before we unlock the CIL context
and allow it to be checkpointed. If we unlock them after we release
the CIl context lock, the CIL can checkpoint and complete before
we free the log items. This breaks stale buffer item unlock and
unpin processing as there is an implicit assumption that the unlock
will occur before the unpin.
Also, some log items need to store the LSN of the transaction commit
in the item (inodes and EFIs) and so can race with other transaction
completions if we don't prevent the CIL from checkpointing before
the unlock occurs.
Cc: <stable@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (96 commits)
no need for list_for_each_entry_safe()/resetting with superblock list
Fix sget() race with failing mount
vfs: don't hold s_umount over close_bdev_exclusive() call
sysv: do not mark superblock dirty on remount
sysv: do not mark superblock dirty on mount
btrfs: remove junk sb_dirt change
BFS: clean up the superblock usage
AFFS: wait for sb synchronization when needed
AFFS: clean up dirty flag usage
cifs: truncate fallout
mbcache: fix shrinker function return value
mbcache: Remove unused features
add f_flags to struct statfs(64)
pass a struct path to vfs_statfs
update VFS documentation for method changes.
All filesystems that need invalidate_inode_buffers() are doing that explicitly
convert remaining ->clear_inode() to ->evict_inode()
Make ->drop_inode() just return whether inode needs to be dropped
fs/inode.c:clear_inode() is gone
fs/inode.c:evict() doesn't care about delete vs. non-delete paths now
...
Fix up trivial conflicts in fs/nilfs2/super.c
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: Al 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: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
