For the DEBUGS!
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

In doing an investigation into AIL push stalls, I was looking at the
log force code to see if an async CIL push could be done instead.
This lead me to xfs_log_force_lsn() and looking at how it works.

xfs_log_force_lsn() is only called from inode synchronisation
contexts such as fsync(), and it takes the ip->i_itemp->ili_last_lsn
value as the LSN to sync the log to. This gets passed to
xlog_cil_force_lsn() via xfs_log_force_lsn() to flush the CIL to the
journal, and then used by xfs_log_force_lsn() to flush the iclogs to
the journal.

The problem is that ip->i_itemp->ili_last_lsn does not store a
log sequence number. What it stores is passed to it from the
->iop_committing method, which is called by xfs_log_commit_cil().
The value this passes to the iop_committing method is the CIL
context sequence number that the item was committed to.

As it turns out, xlog_cil_force_lsn() converts the sequence to an
actual commit LSN for the related context and returns that to
xfs_log_force_lsn(). xfs_log_force_lsn() overwrites it's "lsn"
variable that contained a sequence with an actual LSN and then uses
that to sync the iclogs.

This caused me some confusion for a while, even though I originally
wrote all this code a decade ago. ->iop_committing is only used by
a couple of log item types, and only inode items use the sequence
number it is passed.

Let's clean up the API, CIL structures and inode log item to call it
a sequence number, and make it clear that the high level code is
using CIL sequence numbers and not on-disk LSNs for integrity
synchronisation purposes.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

Currently every journal IO is issued as REQ_PREFLUSH | REQ_FUA to
guarantee the ordering requirements the journal has w.r.t. metadata
writeback. THe two ordering constraints are:

1. we cannot overwrite metadata in the journal until we guarantee
that the dirty metadata has been written back in place and is
stable.

2. we cannot write back dirty metadata until it has been written to
the journal and guaranteed to be stable (and hence recoverable) in
the journal.

The ordering guarantees of #1 are provided by REQ_PREFLUSH. This
causes the journal IO to issue a cache flush and wait for it to
complete before issuing the write IO to the journal. Hence all
completed metadata IO is guaranteed to be stable before the journal
overwrites the old metadata.

The ordering guarantees of #2 are provided by the REQ_FUA, which
ensures the journal writes do not complete until they are on stable
storage. Hence by the time the last journal IO in a checkpoint
completes, we know that the entire checkpoint is on stable storage
and we can unpin the dirty metadata and allow it to be written back.

This is the mechanism by which ordering was first implemented in XFS
way back in 2002 by commit 95d97c36e5155075ba2eb22b17562cfcc53fcf96
("Add support for drive write cache flushing") in the xfs-archive
tree.

A lot has changed since then, most notably we now use delayed
logging to checkpoint the filesystem to the journal rather than
write each individual transaction to the journal. Cache flushes on
journal IO are necessary when individual transactions are wholly
contained within a single iclog. However, CIL checkpoints are single
transactions that typically span hundreds to thousands of individual
journal writes, and so the requirements for device cache flushing
have changed.

That is, the ordering rules I state above apply to ordering of
atomic transactions recorded in the journal, not to the journal IO
itself. Hence we need to ensure metadata is stable before we start
writing a new transaction to the journal (guarantee #1), and we need
to ensure the entire transaction is stable in the journal before we
start metadata writeback (guarantee #2).

Hence we only need a REQ_PREFLUSH on the journal IO that starts a
new journal transaction to provide #1, and it is not on any other
journal IO done within the context of that journal transaction.

The CIL checkpoint already issues a cache flush before it starts
writing to the log, so we no longer need the iclog IO to issue a
REQ_REFLUSH for us. Hence if XLOG_START_TRANS is passed
to xlog_write(), we no longer need to mark the first iclog in
the log write with REQ_PREFLUSH for this case. As an added bonus,
this ordering mechanism works for both internal and external logs,
meaning we can remove the explicit data device cache flushes from
the iclog write code when using external logs.

Given the new ordering semantics of commit records for the CIL, we
need iclogs containing commit records to issue a REQ_PREFLUSH. We
also require unmount records to do this. Hence for both
XLOG_COMMIT_TRANS and XLOG_UNMOUNT_TRANS xlog_write() calls we need
to mark the first iclog being written with REQ_PREFLUSH.

For both commit records and unmount records, we also want them
immediately on stable storage, so we want to also mark the iclogs
that contain these records to be marked REQ_FUA. That means if a
record is split across multiple iclogs, they are all marked REQ_FUA
and not just the last one so that when the transaction is completed
all the parts of the record are on stable storage.

And for external logs, unmount records need a pre-write data device
cache flush similar to the CIL checkpoint cache pre-flush as the
internal iclog write code does not do this implicitly anymore.

As an optimisation, when the commit record lands in the same iclog
as the journal transaction starts, we don't need to wait for
anything and can simply use REQ_FUA to provide guarantee #2.  This
means that for fsync() heavy workloads, the cache flush behaviour is
completely unchanged and there is no degradation in performance as a
result of optimise the multi-IO transaction case.

The most notable sign that there is less IO latency on my test
machine (nvme SSDs) is that the "noiclogs" rate has dropped
substantially. This metric indicates that the CIL push is blocking
in xlog_get_iclog_space() waiting for iclog IO completion to occur.
With 8 iclogs of 256kB, the rate is appoximately 1 noiclog event to
every 4 iclog writes. IOWs, every 4th call to xlog_get_iclog_space()
is blocking waiting for log IO. With the changes in this patch, this
drops to 1 noiclog event for every 100 iclog writes. Hence it is
clear that log IO is completing much faster than it was previously,
but it is also clear that for large iclog sizes, this isn't the
performance limiting factor on this hardware.

With smaller iclogs (32kB), however, there is a substantial
difference. With the cache flush modifications, the journal is now
running at over 4000 write IOPS, and the journal throughput is
largely identical to the 256kB iclogs and the noiclog event rate
stays low at about 1:50 iclog writes. The existing code tops out at
about 2500 IOPS as the number of cache flushes dominate performance
and latency. The noiclog event rate is about 1:4, and the
performance variance is quite large as the journal throughput can
fall to less than half the peak sustained rate when the cache flush
rate prevents metadata writeback from keeping up and the log runs
out of space and throttles reservations.

As a result:

	logbsize	fsmark create rate	rm -rf
before	32kb		152851+/-5.3e+04	5m28s
patched	32kb		221533+/-1.1e+04	5m24s

before	256kb		220239+/-6.2e+03	4m58s
patched	256kb		228286+/-9.2e+03	5m06s

The rm -rf times are included because I ran them, but the
differences are largely noise. This workload is largely metadata
read IO latency bound and the changes to the journal cache flushing
doesn't really make any noticable difference to behaviour apart from
a reduction in noiclog events from background CIL pushing.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

The CIL push is the only call to xlog_write that sets this variable
to true. The other callers don't need a start rec, and they tell
xlog_write what to do by passing the type of ophdr they need written
in the flags field. The need_start_rec parameter essentially tells
xlog_write to to write an extra ophdr with a XLOG_START_TRANS type,
so get rid of the variable to do this and pass XLOG_START_TRANS as
the flag value into xlog_write() from the CIL push.

$ size fs/xfs/xfs_log.o*
  text	   data	    bss	    dec	    hex	filename
 27595	    560	      8	  28163	   6e03	fs/xfs/xfs_log.o.orig
 27454	    560	      8	  28022	   6d76	fs/xfs/xfs_log.o.patched
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

It's a one line wrapper around blkdev_issue_flush(). Just replace it
with direct calls to blkdev_issue_flush().
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

To allow for iclog IO device cache flush behaviour to be optimised,
we first need to separate out the commit record iclog IO from the
rest of the checkpoint so we can wait for the checkpoint IO to
complete before we issue the commit record.

This separation is only necessary if the commit record is being
written into a different iclog to the start of the checkpoint as the
upcoming cache flushing changes requires completion ordering against
the other iclogs submitted by the checkpoint.

If the entire checkpoint and commit is in the one iclog, then they
are both covered by the one set of cache flush primitives on the
iclog and hence there is no need to separate them for ordering.

Otherwise, we need to wait for all the previous iclogs to complete
so they are ordered correctly and made stable by the REQ_PREFLUSH
that the commit record iclog IO issues. This guarantees that if a
reader sees the commit record in the journal, they will also see the
entire checkpoint that commit record closes off.

This also provides the guarantee that when the commit record IO
completes, we can safely unpin all the log items in the checkpoint
so they can be written back because the entire checkpoint is stable
in the journal.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NChandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

On 32-bit (e.g. m68k):

    ERROR: modpost: "__udivdi3" [fs/xfs/xfs.ko] undefined!

Fix this by using a uint32_t intermediate, like before.

Reported-by: noreply@ellerman.id.au
Fixes: 7660a5b48fbef958 ("xfs: log stripe roundoff is a property of the log")
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

We don't need to look at the xfs_mount and superblock every time we
need to do an iclog roundoff calculation. The property is fixed for
the life of the log, so store the roundoff in the log at mount time
and use that everywhere.

On a debug build:

$ size fs/xfs/xfs_log.o.*
   text	   data	    bss	    dec	    hex	filename
  27360	    560	      8	  27928	   6d18	fs/xfs/xfs_log.o.orig
  27219	    560	      8	  27787	   6c8b	fs/xfs/xfs_log.o.patched
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NChandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>

While running generic/050 with an external log, I observed this warning
in dmesg:

Trying to write to read-only block-device sda4 (partno 4)
WARNING: CPU: 2 PID: 215677 at block/blk-core.c:704 submit_bio_checks+0x256/0x510
Call Trace:
 submit_bio_noacct+0x2c/0x430
 _xfs_buf_ioapply+0x283/0x3c0 [xfs]
 __xfs_buf_submit+0x6a/0x210 [xfs]
 xfs_buf_delwri_submit_buffers+0xf8/0x270 [xfs]
 xfsaild+0x2db/0xc50 [xfs]
 kthread+0x14b/0x170

I think this happened because we tried to cover the log after a readonly
mount, and the AIL tried to write the primary superblock to the data
device.  The test marks the data device readonly, but it doesn't do the
same to the external log device.  Therefore, XFS thinks that the log is
writable, even though AIL writes whine to dmesg because the data device
is read only.

Fix this by amending xfs_log_writable to prevent writes when the AIL
can't possible write anything into the filesystem.

Note: As for the external log or the rt devices being readonly--
xfs_blkdev_get will complain about that if we aren't doing a norecovery
mount.
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NBrian Foster <bfoster@redhat.com>

fs/xfs/xfs_log.c:1062:9-10: WARNING: return of 0/1 in function 'xfs_log_need_covered' with return type bool

 Return statements in functions returning bool should use
 true/false instead of 1/0.
Generated by: scripts/coccinelle/misc/boolreturn.cocci

Fixes: 37444fc4 ("xfs: lift writable fs check up into log worker task")
CC: Brian Foster <bfoster@redhat.com>
Reported-by: Nkernel test robot <lkp@intel.com>
Signed-off-by: Nkernel test robot <lkp@intel.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

When CONFIG_XFS_DEBUG=y, set WQ_SYSFS on all workqueues that we create
so that we (developers) have a means to monitor cpu affinity and whatnot
for background workers.  In the next patchset we'll expose knobs for
more of the workqueues publicly and document it, but not now.
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>

The log variable is only used in kernels with asserts enabled.
Remove it and open code the dereference to avoid unused variable
warnings.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

xfs_log_sbcount() calls xfs_sync_sb() to sync superblock counters to
disk when lazy superblock accounting is enabled. This occurs on
unmount, freeze, and read-only (re)mount and ensures the final
values are calculated and persisted to disk before each form of
quiesce completes.

Now that log covering occurs in all of these contexts and uses the
same xfs_sync_sb() mechanism to update log state, there is no need
to log the superblock separately for any reason. Update the log
quiesce path to sync the superblock at least once for any mount
where lazy superblock accounting is enabled. If the log is already
covered, it will remain in the covered state. Otherwise, the next
sync as part of the normal covering sequence will carry the
associated superblock update with it. Remove xfs_log_sbcount() now
that it is no longer needed.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>

Now that log covering occurs on quiesce, we'd like to reuse the
underlying superblock sync for final superblock updates. This
includes things like lazy superblock counter updates, log feature
incompat bits in the future, etc. One quirk to this approach is that
once the log is in the IDLE (i.e. already covered) state, any
subsequent log write resets the state back to NEED. This means that
a final superblock sync to an already covered log requires two more
sb syncs to return the log back to IDLE again.

For example, if a lazy superblock enabled filesystem is mount cycled
without any modifications, the unmount path syncs the superblock
once and writes an unmount record. With the desired log quiesce
covering behavior, we sync the superblock three times at unmount
time: once for the lazy superblock counter update and twice more to
cover the log. By contrast, if the log is active or only partially
covered at unmount time, a final superblock sync would doubly serve
as the one or two remaining syncs required to cover the log.

This duplicate covering sequence is unnecessary because the
filesystem remains consistent if a crash occurs at any point. The
superblock will either be recovered in the event of a crash or
written back before the log is quiesced and potentially cleaned with
an unmount record.

Update the log covering state machine to remain in the IDLE state if
additional covering checkpoints pass through the log. This
facilitates final superblock updates (such as lazy superblock
counters) via a single sb sync without losing covered status. This
provides some consistency with the active and partially covered
cases and also avoids harmless, but spurious checkpoints when
quiescing the log.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>

The log quiesce mechanism historically terminates by marking the log
clean with an unmount record. The primary objective is to indicate
that log recovery is no longer required after the quiesce has
flushed all in-core changes and written back filesystem metadata.
While this is perfectly fine, it is somewhat hacky as currently used
in certain contexts. For example, filesystem freeze quiesces (i.e.
cleans) the log and immediately redirties it with a dummy superblock
transaction to ensure that log recovery runs in the event of a
crash.

While this functions correctly, cleaning the log from freeze context
is clearly superfluous given the current redirtying behavior.
Instead, the desired behavior can be achieved by simply covering the
log. This effectively retires all on-disk log items from the active
range of the log by issuing two synchronous and sequential dummy
superblock update transactions that serve to update the on-disk log
head and tail. The subtle difference is that the log technically
remains dirty due to the lack of an unmount record, though recovery
is effectively a no-op due to the content of the checkpoints being
clean (i.e. the unmodified on-disk superblock).

Log covering currently runs in the background and only triggers once
the filesystem and log has idled. The purpose of the background
mechanism is to prevent log recovery from replaying the most
recently logged items long after those items may have been written
back. In the quiesce path, the log has been deliberately idled by
forcing the log and pushing the AIL until empty in a context where
no further mutable filesystem operations are allowed. Therefore, we
can cover the log as the final step in the log quiesce codepath to
reflect that all previously active items have been successfully
written back.

This facilitates selective log covering from certain contexts (i.e.
freeze) that only seek to quiesce, but not necessarily clean the
log. Note that as a side effect of this change, log covering now
occurs when cleaning the log as well. This is harmless, facilitates
subsequent cleanups, and is mostly temporary as various operations
switch to use explicit log covering.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>

Log quiesce is currently associated with cleaning the log, which is
accomplished by writing an unmount record as the last step of the
quiesce sequence. The quiesce codepath is a bit convoluted in this
regard due to how it is reused from various contexts. In preparation
to create separate log cleaning and log covering interfaces, lift
the write of the unmount record into a new cleaning helper and call
that wherever xfs_log_quiesce() is currently invoked. No functional
changes.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

The log covering helper checks whether the filesystem is writable to
determine whether to cover the log. The helper is currently only
called from the background log worker. In preparation to reuse the
helper from freezing contexts, lift the check into xfs_log_worker().
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

xfs_log_sbcount() syncs the superblock specifically to accumulate
the in-core percpu superblock counters and commit them to disk. This
is required to maintain filesystem consistency across quiesce
(freeze, read-only mount/remount) or unmount when lazy superblock
accounting is enabled because individual transactions do not update
the superblock directly.

This mechanism works as expected for writable mounts, but
xfs_log_sbcount() skips the update for read-only mounts. Read-only
mounts otherwise still allow log recovery and write out an unmount
record during log quiesce. If a read-only mount performs log
recovery, it can modify the in-core superblock counters and write an
unmount record when the filesystem unmounts without ever syncing the
in-core counters. This leaves the filesystem with a clean log but in
an inconsistent state with regard to lazy sb counters.

Update xfs_log_sbcount() to use the same logic
xfs_log_unmount_write() uses to determine when to write an unmount
record. This ensures that lazy accounting is always synced before
the log is cleaned. Refactor this logic into a new helper to
distinguish between a writable filesystem and a writable log.
Specifically, the log is writable unless the filesystem is mounted
with the norecovery mount option, the underlying log device is
read-only, or the filesystem is shutdown. Drop the freeze state
check because the update is already allowed during the freezing
process and no context calls this function on an already frozen fs.
Also, retain the shutdown check in xfs_log_unmount_write() to catch
the case where the preceding log force might have triggered a
shutdown.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NGao Xiang <hsiangkao@redhat.com>
Reviewed-by: NAllison Henderson <allison.henderson@oracle.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NBill O'Donnell <billodo@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

xfs_buftarg_drain() is called from xfs_log_quiesce() to ensure the
buffer cache is reclaimed during unmount. xfs_log_quiesce() is also
called from xfs_quiesce_attr(), however, which means that cache
state is completely drained for filesystem freeze and read-only
remount. While technically harmless, this is unnecessarily
heavyweight. Both freeze and read-only mounts allow reads and thus
allow population of the buffer cache. Therefore, the transitional
sequence in either case really only needs to quiesce outstanding
writes to return the filesystem in a generally read-only state.

Additionally, some users have reported that attempts to freeze a
filesystem concurrent with a read-heavy workload causes the freeze
process to stall for a significant amount of time. This occurs
because, as mentioned above, the read workload repopulates the
buffer LRU while the freeze task attempts to drain it.

To improve this situation, replace the drain in xfs_log_quiesce()
with a buffer I/O quiesce and lift the drain into the unmount path.
This removes buffer LRU reclaim from freeze and read-only [re]mount,
but ensures the LRU is still drained before the filesystem unmounts.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

xfs_wait_buftarg() is vaguely named and somewhat overloaded. Its
primary purpose is to reclaim all buffers from the provided buffer
target LRU. In preparation to refactor xfs_wait_buftarg() into
serialization and LRU draining components, rename the function and
associated helpers to something more descriptive. This patch has no
functional changes with the minor exception of renaming a
tracepoint.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <djwong@kernel.org>
Signed-off-by: NDarrick J. Wong <djwong@kernel.org>

Separate the computation of the log push threshold and the push logic in
xlog_grant_push_ail.  This enables higher level code to determine (for
example) that it is holding on to a logged intent item and the log is so
busy that it is more than 75% full.  In that case, it would be desirable
to move the log item towards the head to release the tail, which we will
cover in the next patch.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>

Let's use DIV_ROUND_UP() to calculate log record header
blocks as what did in xlog_get_iclog_buffer_size() and
wrap up a common helper for log recovery.
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Signed-off-by: NGao Xiang <hsiangkao@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

xlog_ticket_alloc() is always called under NOFS context, except from
unmount path, which eitherway is holding many FS locks, so, there is no
need for its callers to keep passing allocation flags into it.

change xlog_ticket_alloc() to use default kmem_cache_zalloc(), remove
its alloc_flags argument, and always use GFP_NOFS | __GFP_NOFAIL flags.
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NCarlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>

In commit f467cad9, I added the ability to force a recalculation of
the filesystem summary counters if they seemed incorrect.  This was done
(not entirely correctly) by tweaking the log code to write an unmount
record without the UMOUNT_TRANS flag set.  At next mount, the log
recovery code will fail to find the unmount record and go into recovery,
which triggers the recalculation.

What actually gets written to the log is what ought to be an unmount
record, but without any flags set to indicate what kind of record it
actually is.  This worked to trigger the recalculation, but we shouldn't
write bogus log records when we could simply write nothing.

Fixes: f467cad9 ("xfs: force summary counter recalc at next mount")
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>

Running metadata intensive workloads, I've been seeing the AIL
pushing getting stuck on pinned buffers and triggering log forces.
The log force is taking a long time to run because the log IO is
getting throttled by wbt_wait() - the block layer writeback
throttle. It's being throttled because there is a huge amount of
metadata writeback going on which is filling the request queue.

IOWs, we have a priority inversion problem here.

Mark the log IO bios with REQ_IDLE so they don't get throttled
by the block layer writeback throttle. When we are forcing the CIL,
we are likely to need to to tens of log IOs, and they are issued as
fast as they can be build and IO completed. Hence REQ_IDLE is
appropriate - it's an indication that more IO will follow shortly.

And because we also set REQ_SYNC, the writeback throttle will now
treat log IO the same way it treats direct IO writes - it will not
throttle them at all. Hence we solve the priority inversion problem
caused by the writeback throttle being unable to distinguish between
high priority log IO and background metadata writeback.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NAllison Collins <allison.henderson@oracle.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Separate out the unmount record writing from the rest of the
ticket and log state futzing necessary to make it work. This is
a no-op, just makes the code cleaner and places the unmount record
formatting and writing alongside the commit record formatting and
writing code.

We can also get rid of the ticket flag clearing before the
xlog_write() call because it no longer cares about the state of
XLOG_TIC_INITED.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

xlog_write_done() is just a thin wrapper around xlog_commit_record(), so
they can be merged together easily.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Remove xlog_ticket_done and just call the renamed low-level helpers for
ungranting or regranting log space directly.  To make that a little
the reference put on the ticket and all tracing is moved into the actual
helpers.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

It is not longer used or checked by anything, so remove the last
traces from the log ticket code.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

xfs_log_done() does two separate things. Firstly, it triggers commit
records to be written for permanent transactions, and secondly it
releases or regrants transaction reservation space.

Since delayed logging was introduced, transactions no longer write
directly to the log, hence they never have the XLOG_TIC_INITED flag
cleared on them. Hence transactions never write commit records to
the log and only need to modify reservation space.

Split up xfs_log_done into two parts, and only call the parts of the
operation needed for the context xfs_log_done() is currently being
called from.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Commit and unmount records records do not need start records to be
written, so rearrange the logic in xlog_write() to remove the need
to check for XLOG_TIC_INITED to determine if we should account for
the space used by a start record.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

The xlog_write() function iterates over iclogs until it completes
writing all the log vectors passed in. The ticket tracks whether
a start record has been written or not, so only the first iclog gets
a start record. We only ever pass single use tickets to
xlog_write() so we only ever need to write a start record once per
xlog_write() call.

Hence we don't need to store whether we should write a start record
in the ticket as the callers provide all the information we need to
determine if a start record should be written. For the moment, we
have to ensure that we clear the XLOG_TIC_INITED appropriately so
the code in xfs_log_done() still works correctly for committing
transactions.

(darrick: Note the slight behavior change that we always deduct the
size of the op header from the ticket, even for unmount records)
Signed-off-by: NDave Chinner <dchinner@redhat.com>
[hch: pass an explicit need_start_rec argument]
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

If the bio_add_page() call fails, we proceed to write out a
partially constructed log buffer. This corrupts the physical log
such that log recovery is not possible. Worse, persistent
occurrences of this error eventually lead to a BUG_ON() failure in
bio_split() as iclogs wrap the end of the physical log, which
triggers log recovery on subsequent mount.

Rather than warn about writing out a corrupted log buffer, shutdown
the fs as is done for any log I/O related error. This preserves the
consistency of the physical log such that log recovery succeeds on a
subsequent mount. Note that this was observed on a 64k page debug
kernel without upstream commit 59bb4798 ("mm, sl[aou]b:
guarantee natural alignment for kmalloc(power-of-two)"), which
demonstrated frequent iclog bio overflows due to unaligned (slab
allocated) iclog data buffers.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Open code the xlog_state_want_sync logic in its two callers given that
this function is a trivial wrapper around xlog_state_switch_iclogs.

Move the lockdep assert into xlog_state_switch_iclogs to not lose this
debugging aid, and improve the comment that documents
xlog_state_switch_iclogs as well.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Use the shutdown flag in the log to bypass xlog_state_clean_iclog
entirely in case of a shut down log.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

Factor out a few self-contained helpers from xlog_state_clean_iclog, and
update the documentation so it primarily documents why things happens
instead of how.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

We can just check for a shut down log all the way down in
xlog_cil_committed instead of passing the parameter.  This means a
slight behavior change in that we now also abort log items if the
shutdown came in halfway into the I/O completion processing, which
actually is the right thing to do.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

There is no need to check for the ioerror state before the lock, as
the shutdown case is not a fast path.  Also remove the call to force
shutdown the file system, as it must have been shut down already
for an iclog to be in the ioerror state.  Also clean up the flow of
the function a bit.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>

The only caller of xfs_log_release_iclog doesn't care about the return
value, so remove it.  Also don't bother passing the mount pointer,
given that we can trivially derive it from the iclog.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDarrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>