4045 zfs write throttle & i/o scheduler performance work
1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync
read, sync write, async read, async write, and scrub/resilver. The scheduler
issues a number of concurrent i/os from each class to the device. Once a class
has been selected, an i/o is selected from this class using either an elevator
algorithem (async, scrub classes) or FIFO (sync classes). The number of
concurrent async write i/os is tuned dynamically based on i/o load, to achieve
good sync i/o latency when there is not a high load of writes, and good write
throughput when there is. See the block comment in vdev_queue.c (reproduced
below) for more details.
2. The write throttle (dsl_pool_tempreserve_space() and
txg_constrain_throughput()) is rewritten to produce much more consistent delays
when under constant load. The new write throttle is based on the amount of
dirty data, rather than guesses about future performance of the system. When
there is a lot of dirty data, each transaction (e.g. write() syscall) will be
delayed by the same small amount. This eliminates the "brick wall of wait"
that the old write throttle could hit, causing all transactions to wait several
seconds until the next txg opens. One of the keys to the new write throttle is
decrementing the amount of dirty data as i/o completes, rather than at the end
of spa_sync(). Note that the write throttle is only applied once the i/o
scheduler is issuing the maximum number of outstanding async writes. See the
block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for
more details.
This diff has several other effects, including:
* the commonly-tuned global variable zfs_vdev_max_pending has been removed;
use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead.
* the size of each txg (meaning the amount of dirty data written, and thus the
time it takes to write out) is now controlled differently. There is no longer
an explicit time goal; the primary determinant is amount of dirty data.
Systems that are under light or medium load will now often see that a txg is
always syncing, but the impact to performance (e.g. read latency) is minimal.
Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this.
* zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression,
checksum, etc. This improves latency by not allowing these CPU-intensive tasks
to consume all CPU (on machines with at least 4 CPU's; the percentage is
rounded up).
--matt
APPENDIX: problems with the current i/o scheduler
The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem
with this is that if there are always i/os pending, then certain classes of
i/os can see very long delays.
For example, if there are always synchronous reads outstanding, then no async
writes will be serviced until they become "past due". One symptom of this
situation is that each pass of the txg sync takes at least several seconds
(typically 3 seconds).
If many i/os become "past due" (their deadline is in the past), then we must
service all of these overdue i/os before any new i/os. This happens when we
enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in
the future. If we can't complete all the i/os in 2.5 seconds (e.g. because
there were always reads pending), then these i/os will become past due. Now we
must service all the "async" writes (which could be hundreds of megabytes)
before we service any reads, introducing considerable latency to synchronous
i/os (reads or ZIL writes).
Notes on porting to ZFS on Linux:
- zio_t gained new members io_physdone and io_phys_children. Because
object caches in the Linux port call the constructor only once at
allocation time, objects may contain residual data when retrieved
from the cache. Therefore zio_create() was updated to zero out the two
new fields.
- vdev_mirror_pending() relied on the depth of the per-vdev pending queue
(vq->vq_pending_tree) to select the least-busy leaf vdev to read from.
This tree has been replaced by vq->vq_active_tree which is now used
for the same purpose.
- vdev_queue_init() used the value of zfs_vdev_max_pending to determine
the number of vdev I/O buffers to pre-allocate. That global no longer
exists, so we instead use the sum of the *_max_active values for each of
the five I/O classes described above.
- The Illumos implementation of dmu_tx_delay() delays a transaction by
sleeping in condition variable embedded in the thread
(curthread->t_delay_cv). We do not have an equivalent CV to use in
Linux, so this change replaced the delay logic with a wrapper called
zfs_sleep_until(). This wrapper could be adopted upstream and in other
downstream ports to abstract away operating system-specific delay logic.
- These tunables are added as module parameters, and descriptions added
to the zfs-module-parameters.5 man page.
spa_asize_inflation
zfs_deadman_synctime_ms
zfs_vdev_max_active
zfs_vdev_async_write_active_min_dirty_percent
zfs_vdev_async_write_active_max_dirty_percent
zfs_vdev_async_read_max_active
zfs_vdev_async_read_min_active
zfs_vdev_async_write_max_active
zfs_vdev_async_write_min_active
zfs_vdev_scrub_max_active
zfs_vdev_scrub_min_active
zfs_vdev_sync_read_max_active
zfs_vdev_sync_read_min_active
zfs_vdev_sync_write_max_active
zfs_vdev_sync_write_min_active
zfs_dirty_data_max_percent
zfs_delay_min_dirty_percent
zfs_dirty_data_max_max_percent
zfs_dirty_data_max
zfs_dirty_data_max_max
zfs_dirty_data_sync
zfs_delay_scale
The latter four have type unsigned long, whereas they are uint64_t in
Illumos. This accommodates Linux's module_param() supported types, but
means they may overflow on 32-bit architectures.
The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most
likely to overflow on 32-bit systems, since they express physical RAM
sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to
2^32 which does overflow. To resolve that, this port instead initializes
it in arc_init() to 25% of physical RAM, and adds the tunable
zfs_dirty_data_max_max_percent to override that percentage. While this
solution doesn't completely avoid the overflow issue, it should be a
reasonable default for most systems, and the minority of affected
systems can work around the issue by overriding the defaults.
- Fixed reversed logic in comment above zfs_delay_scale declaration.
- Clarified comments in vdev_queue.c regarding when per-queue minimums take
effect.
- Replaced dmu_tx_write_limit in the dmu_tx kstat file
with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts
how many times a transaction has been delayed because the pool dirty
data has exceeded zfs_delay_min_dirty_percent. The latter counts how
many times the pool dirty data has exceeded zfs_dirty_data_max (which
we expect to never happen).
- The original patch would have regressed the bug fixed in
zfsonlinux/zfs@c418410, which prevented users from setting the
zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE.
A similar fix is added to vdev_queue_aggregate().
- In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the
heap instead of the stack. In Linux we can't afford such large
structures on the stack.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Ned Bass <bass6@llnl.gov>
Reviewed by: Brendan Gregg <brendan.gregg@joyent.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
http://www.illumos.org/issues/4045illumos/illumos-gate@69962b5647
Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1913
Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Christopher Siden <chris.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
NOTES: This patch has been reworked from the original in the
following ways to accomidate Linux ZFS implementation
*) Usage of the cyclic interface was replaced by the delayed taskq
interface. This avoids the need to implement new compatibility
code and allows us to rely on the existing taskq implementation.
*) An extern for zfs_txg_synctime_ms was added to sys/dsl_pool.h
because declaring externs in source files as was done in the
original patch is just plain wrong.
*) Instead of panicing the system when the deadman triggers a
zevent describing the blocked vdev and the first pending I/O
is posted. If the panic behavior is desired Linux provides
other generic methods to panic the system when threads are
observed to hang.
*) For reference, to delay zios by 30 seconds for testing you can
use zinject as follows: 'zinject -d <vdev> -D30 <pool>'
References:
illumos/illumos-gate@283b84606bhttps://www.illumos.org/issues/3246
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1396
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Richard Elling <richard.elling@richardelling.com>
Reviewed by: Gordon Ross <gwr@nexenta.com>
Reviewed by: Garrett D'Amore <garrett@damore.org>
Approved by: Richard Lowe <richlowe@richlowe.net>
Refererces to Illumos issue:
https://www.illumos.org/issues/2671
This patch has been slightly modified from the upstream Illumos
version. In the upstream implementation a warning message is
logged to the console. To prevent pointless console noise this
notification is now posted as a "ereport.fs.zfs.vdev.bad_ashift"
event.
The event indicates a non-optimial (but entirely safe) ashift
value was used to create the pool. Depending on your workload
this may impact pool performance. Unfortunately, the only way
to correct the issue is to recreate the pool with a new ashift.
NOTE: The unrelated fix to the comment in zpool_main.c appears
in the upstream commit and was preserved for consistnecy.
Ported-by: Cyril Plisko <cyril.plisko@mountall.com>
Reworked-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#955
There have been reports of ZFS deadlocking due to what appears to
be a lost IO. This patch addes some debugging to determine the
exact state of the IO which neither 1) completed, 2) failed, or
3) timed out after zio_delay_max (30) seconds.
This information will be logged using the ZFS FMA infrastructure
as a 'delay' event and posted to the internal zevent log. By
default the last 64 events will be kept in the log but the limit
is configurable via the zfs_zevent_len_max module option.
To dump the contents of the log use the 'zpool events -v' command
and look for the resource.fs.zfs.delay event. It will include
various information about the pool, vdev, and zio which may shed
some light on the issue.
In the context of this change the 120 second kernel blocked thread
watchdog has been disabled for synchronous IOs.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #930
Differences between how paging is done on Solaris and Linux can cause
deadlocks if KM_SLEEP is used in any the following contexts.
* The txg_sync thread
* The zvol write/discard threads
* The zpl_putpage() VFS callback
This is because KM_SLEEP will allow for direct reclaim which may result
in the VM calling back in to the filesystem or block layer to write out
pages. If a lock is held over this operation the potential exists to
deadlock the system. To ensure forward progress all memory allocations
in these contexts must us KM_PUSHPAGE which disables performing any I/O
to accomplish the memory allocation.
Previously, this behavior was acheived by setting PF_MEMALLOC on the
thread. However, that resulted in unexpected side effects such as the
exhaustion of pages in ZONE_DMA. This approach touchs more of the zfs
code, but it is more consistent with the right way to handle these cases
under Linux.
This is patch lays the ground work for being able to safely revert the
following commits which used PF_MEMALLOC:
21ade34 Disable direct reclaim for z_wr_* threads
cfc9a5c Fix zpl_writepage() deadlock
eec8164 Fix ASSERTION(!dsl_pool_sync_context(tx->tx_pool))
Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #726
1952 memory leak when adding a file-based l2arc device
1954 leak in ZFS from metaslab_group_create and zfs_ereport_checksum
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Matt Ahrens <matt@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Bill Pijewski <wdp@joyent.com>
Reviewed by: Dan McDonald <danmcd@nexenta.com>
Approved by: Eric Schrock <eric.schrock@delphix.com>
References to Illumos issues:
https://www.illumos.org/issues/1951https://www.illumos.org/issues/1952https://www.illumos.org/issues/1954
Ported-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#650
While there is no right maximum timeout for a disk IO we can start
laying the ground work to measure how long they do take in practice.
This change simply measures the IO time and if it exceeds 30s an
event is posted for 'zpool events'.
This value was carefully selected because for sd devices it implies
that at least one timeout (SD_TIMEOUT) has occured. Unfortunately,
even with FAILFAST set we may retry and request and not get an
error. This behavior is strongly dependant on the device driver
and how it is hooked in to the scsi error handling stack. However
by setting the limit at 30s we can log the event even if no error
was returned.
Slightly longer term we can start recording these delays perhaps
as a simple power-of-two histrogram. This histogram can then be
reported as part of the 'zpool status' command when given an command
line option.
None of this code changes the internal behavior of ZFS. Currently
it is simply for reporting excessively long delays.
By default the Solaris code does not log speculative or soft io errors
in either 'zpool status' or post an event. Under Linux we don't want
to change the expected behavior of 'zpool status' so these io errors
are still suppressed there.
However, since we do need to know about these events for Linux FMA and
the 'zpool events' interface is new we do post the events. With the
addition of the zio_flags field the posted events now contain enough
information that a user space consumer can identify and discard these
events if it sees fit.
This topic branch leverages the Solaris style FMA call points
in ZFS to create a user space visible event notification system
under Linux. This new system is called zevent and it unifies
all previous Solaris style ereports and sysevent notifications.
Under this Linux specific scheme when a sysevent or ereport event
occurs an nvlist describing the event is created which looks almost
exactly like a Solaris ereport. These events are queued up in the
kernel when they occur and conditionally logged to the console.
It is then up to a user space application to consume the events
and do whatever it likes with them.
To make this possible the existing /dev/zfs ABI has been extended
with two new ioctls which behave as follows.
* ZFS_IOC_EVENTS_NEXT
Get the next pending event. The kernel will keep track of the last
event consumed by the file descriptor and provide the next one if
available. If no new events are available the ioctl() will block
waiting for the next event. This ioctl may also be called in a
non-blocking mode by setting zc.zc_guid = ZEVENT_NONBLOCK. In the
non-blocking case if no events are available ENOENT will be returned.
It is possible that ESHUTDOWN will be returned if the ioctl() is
called while module unloading is in progress. And finally ENOMEM
may occur if the provided nvlist buffer is not large enough to
contain the entire event.
* ZFS_IOC_EVENTS_CLEAR
Clear are events queued by the kernel. The kernel will keep a fairly
large number of recent events queued, use this ioctl to clear the
in kernel list. This will effect all user space processes consuming
events.
The zpool command has been extended to use this events ABI with the
'events' subcommand. You may run 'zpool events -v' to output a
verbose log of all recent events. This is very similar to the
Solaris 'fmdump -ev' command with the key difference being it also
includes what would be considered sysevents under Solaris. You
may also run in follow mode with the '-f' option. To clear the
in kernel event queue use the '-c' option.
$ sudo cmd/zpool/zpool events -fv
TIME CLASS
May 13 2010 16:31:15.777711000 ereport.fs.zfs.config.sync
class = "ereport.fs.zfs.config.sync"
ena = 0x40982b7897700001
detector = (embedded nvlist)
version = 0x0
scheme = "zfs"
pool = 0xed976600de75dfa6
(end detector)
time = 0x4bec8bc3 0x2e5aed98
pool = "zpios"
pool_guid = 0xed976600de75dfa6
pool_context = 0x0
While the 'zpool events' command is handy for interactive debugging
it is not expected to be the primary consumer of zevents. This ABI
was primarily added to facilitate the addition of a user space
monitoring daemon. This daemon would consume all events posted by
the kernel and based on the type of event perform an action. For
most events simply forwarding them on to syslog is likely enough.
But this interface also cleanly allows for more sophisticated
actions to be taken such as generating an email for a failed drive.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
