Commit f83d436a authored by Darrick J. Wong's avatar Darrick J. Wong
Browse files

xfs: increase the default parallelism levels of pwork clients 



Increase the parallelism level for pwork clients to the workqueue
defaults so that we can take advantage of computers with a lot of CPUs
and a lot of hardware.  On fast systems this will speed up quotacheck by
a large factor, and the following posteof/cowblocks cleanup series will
use the functionality presented in this patch to run garbage collection
as quickly as possible.

We do this by switching the pwork workqueue to unbounded, since the
current user (quotacheck) runs lengthy scans for each work item and we
don't care about dispatching the work on a warm cpu cache or anything
like that.  Also set WQ_SYSFS so that we can monitor where the wq is
running.

Signed-off-by: default avatarDarrick J. Wong <djwong@kernel.org>
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
Reviewed-by: default avatarBrian Foster <bfoster@redhat.com>
parent a1a7d05a

Documentation/admin-guide/xfs.rst

+38 −0
Original line number Diff line number Diff line
@@ -495,3 +495,41 @@ the class and error context. For example, the default values for 
"metadata/ENODEV" are "0" rather than "-1" so that this error handler defaults

to "fail immediately" behaviour. This is done because ENODEV is a fatal,

unrecoverable error no matter how many times the metadata IO is retried.



Workqueue Concurrency

=====================



XFS uses kernel workqueues to parallelize metadata update processes.  This

enables it to take advantage of storage hardware that can service many IO

operations simultaneously.  This interface exposes internal implementation

details of XFS, and as such is explicitly not part of any userspace API/ABI

guarantee the kernel may give userspace.  These are undocumented features of

the generic workqueue implementation XFS uses for concurrency, and they are

provided here purely for diagnostic and tuning purposes and may change at any

time in the future.



The control knobs for a filesystem's workqueues are organized by task at hand

and the short name of the data device.  They all can be found in:



  /sys/bus/workqueue/devices/${task}!${device}



================  ===========

  Task            Description

================  ===========

  xfs_iwalk-$pid  Inode scans of the entire filesystem. Currently limited to

                  mount time quotacheck.

================  ===========



For example, the knobs for the quotacheck workqueue for /dev/nvme0n1 would be

found in /sys/bus/workqueue/devices/xfs_iwalk-1111!nvme0n1/.



The interesting knobs for XFS workqueues are as follows:



============     ===========

  Knob           Description

============     ===========

  max_active     Maximum number of background threads that can be started to

                 run the work.

  cpumask        CPUs upon which the threads are allowed to run.

  nice           Relative priority of scheduling the threads.  These are the

                 same nice levels that can be applied to userspace processes.

fs/xfs/xfs_iwalk.c

+1 −4
Original line number Diff line number Diff line
@@ -618,15 +618,12 @@ xfs_iwalk_threaded( 
{

	struct xfs_pwork_ctl	pctl;

	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);

	unsigned int		nr_threads;

	int			error;



	ASSERT(agno < mp->m_sb.sb_agcount);

	ASSERT(!(flags & ~XFS_IWALK_FLAGS_ALL));



	nr_threads = xfs_pwork_guess_datadev_parallelism(mp);

	error = xfs_pwork_init(mp, &pctl, xfs_iwalk_ag_work, "xfs_iwalk",

			nr_threads);

	error = xfs_pwork_init(mp, &pctl, xfs_iwalk_ag_work, "xfs_iwalk");

	if (error)

		return error;

fs/xfs/xfs_pwork.c

+5 −20
Original line number Diff line number Diff line
@@ -61,16 +61,18 @@ xfs_pwork_init( 
	struct xfs_mount	*mp,

	struct xfs_pwork_ctl	*pctl,

	xfs_pwork_work_fn	work_fn,

	const char		*tag,

	unsigned int		nr_threads)

	const char		*tag)

{

	unsigned int		nr_threads = 0;



#ifdef DEBUG

	if (xfs_globals.pwork_threads >= 0)

		nr_threads = xfs_globals.pwork_threads;

#endif

	trace_xfs_pwork_init(mp, nr_threads, current->pid);



	pctl->wq = alloc_workqueue("%s-%d", WQ_FREEZABLE, nr_threads, tag,

	pctl->wq = alloc_workqueue("%s-%d",

			WQ_UNBOUND | WQ_SYSFS | WQ_FREEZABLE, nr_threads, tag,

			current->pid);

	if (!pctl->wq)

		return -ENOMEM;
@@ -117,20 +119,3 @@ xfs_pwork_poll( 
				atomic_read(&pctl->nr_work) == 0, HZ) == 0)

		touch_softlockup_watchdog();

}
 


/*

 * Return the amount of parallelism that the data device can handle, or 0 for

 * no limit.

 */

unsigned int

xfs_pwork_guess_datadev_parallelism(

	struct xfs_mount	*mp)

{

	struct xfs_buftarg	*btp = mp->m_ddev_targp;



	/*

	 * For now we'll go with the most conservative setting possible,

	 * which is two threads for an SSD and 1 thread everywhere else.

	 */

	return blk_queue_nonrot(btp->bt_bdev->bd_disk->queue) ? 2 : 1;

}

fs/xfs/xfs_pwork.h

+1 −3
Original line number Diff line number Diff line
@@ -51,11 +51,9 @@ xfs_pwork_want_abort( 
}



int xfs_pwork_init(struct xfs_mount *mp, struct xfs_pwork_ctl *pctl,

		xfs_pwork_work_fn work_fn, const char *tag,

		unsigned int nr_threads);

		xfs_pwork_work_fn work_fn, const char *tag);

void xfs_pwork_queue(struct xfs_pwork_ctl *pctl, struct xfs_pwork *pwork);

int xfs_pwork_destroy(struct xfs_pwork_ctl *pctl);

void xfs_pwork_poll(struct xfs_pwork_ctl *pctl);

unsigned int xfs_pwork_guess_datadev_parallelism(struct xfs_mount *mp);



#endif /* __XFS_PWORK_H__ */