c/mirror_zfs
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_zfs.git synced 2026-05-22 18:40:43 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fixes in persistent L2ARC

Browse Source 
In l2arc_add_vdev() first decide whether the device is eligible for
L2ARC rebuild or whole device trim and then add it to the list of cache
devices. Otherwise l2arc_feed_thread() might already start writing on
the device invalidating previous content as l2ad_hand = l2ad_start.
However l2arc_rebuild_vdev() needs the device present in the cache
device list to figure out its l2arc_dev_t. Fix this by moving most of
l2arc_rebuild_vdev() in a new function l2arc_rebuild_dev() which does
not need to search in the cache device list.

In contrast to l2arc_add_vdev() we do not have to worry about
l2arc_feed_thread() invalidating previous content when onlining a
cache device. The device parameters (l2ad*) are not cleared when
offlining the device and writing new buffers will not invalidate
all previous content. In worst case only buffers that have not had
their log block written to the device will be lost.

Retire persist_l2arc_00{4,5,8} tests since they cover code already
covered by the remaining ones. Test persist_l2arc_006 is renamed to
persist_l2arc_004 and persist_l2arc_007 is renamed to persist_l2arc_005.

Fix a typo in persist_l2arc_004, and remove an assertion that is not
always true from l2arc_arcstats_pos. Also update an assertion in
persist_l2arc_005 and explain why in a comment.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #12365
This commit is contained in:
George Amanakis
2021-07-26 21:30:24 +02:00
committed by Tony Hutter
parent ac573e3105
commit 406534f807
10 changed files with 166 additions and 495 deletions
Download Patch File Download Diff File Expand all files Collapse all files
module/zfs/arc.c
+105 -78
View File
@@ -9743,85 +9743,12 @@ l2arc_vdev_get(vdev_t *vd)
return (dev);
}
/*
* Add a vdev for use by the L2ARC. By this point the spa has already
* validated the vdev and opened it.
*/
void
l2arc_add_vdev(spa_t *spa, vdev_t *vd)
static void
l2arc_rebuild_dev(l2arc_dev_t *dev, boolean_t reopen)
{
l2arc_dev_t *adddev;
uint64_t l2dhdr_asize;
ASSERT(!l2arc_vdev_present(vd));
/*
* Create a new l2arc device entry.
*/
adddev = vmem_zalloc(sizeof (l2arc_dev_t), KM_SLEEP);
adddev->l2ad_spa = spa;
adddev->l2ad_vdev = vd;
/* leave extra size for an l2arc device header */
l2dhdr_asize = adddev->l2ad_dev_hdr_asize =
MAX(sizeof (*adddev->l2ad_dev_hdr), 1 << vd->vdev_ashift);
adddev->l2ad_start = VDEV_LABEL_START_SIZE + l2dhdr_asize;
adddev->l2ad_end = VDEV_LABEL_START_SIZE + vdev_get_min_asize(vd);
ASSERT3U(adddev->l2ad_start, <, adddev->l2ad_end);
adddev->l2ad_hand = adddev->l2ad_start;
adddev->l2ad_evict = adddev->l2ad_start;
adddev->l2ad_first = B_TRUE;
adddev->l2ad_writing = B_FALSE;
adddev->l2ad_trim_all = B_FALSE;
list_link_init(&adddev->l2ad_node);
adddev->l2ad_dev_hdr = kmem_zalloc(l2dhdr_asize, KM_SLEEP);
mutex_init(&adddev->l2ad_mtx, NULL, MUTEX_DEFAULT, NULL);
/*
* This is a list of all ARC buffers that are still valid on the
* device.
*/
list_create(&adddev->l2ad_buflist, sizeof (arc_buf_hdr_t),
offsetof(arc_buf_hdr_t, b_l2hdr.b_l2node));
/*
* This is a list of pointers to log blocks that are still present
* on the device.
*/
list_create(&adddev->l2ad_lbptr_list, sizeof (l2arc_lb_ptr_buf_t),
offsetof(l2arc_lb_ptr_buf_t, node));
vdev_space_update(vd, 0, 0, adddev->l2ad_end - adddev->l2ad_hand);
zfs_refcount_create(&adddev->l2ad_alloc);
zfs_refcount_create(&adddev->l2ad_lb_asize);
zfs_refcount_create(&adddev->l2ad_lb_count);
/*
* Add device to global list
*/
mutex_enter(&l2arc_dev_mtx);
list_insert_head(l2arc_dev_list, adddev);
atomic_inc_64(&l2arc_ndev);
mutex_exit(&l2arc_dev_mtx);
/*
* Decide if vdev is eligible for L2ARC rebuild
*/
l2arc_rebuild_vdev(adddev->l2ad_vdev, B_FALSE);
}
void
l2arc_rebuild_vdev(vdev_t *vd, boolean_t reopen)
{
l2arc_dev_t *dev = NULL;
l2arc_dev_hdr_phys_t *l2dhdr;
uint64_t l2dhdr_asize;
spa_t *spa;
dev = l2arc_vdev_get(vd);
ASSERT3P(dev, !=, NULL);
spa = dev->l2ad_spa;
l2dhdr = dev->l2ad_dev_hdr;
l2dhdr_asize = dev->l2ad_dev_hdr_asize;
l2arc_dev_hdr_phys_t *l2dhdr = dev->l2ad_dev_hdr;
uint64_t l2dhdr_asize = dev->l2ad_dev_hdr_asize;
spa_t *spa = dev->l2ad_spa;
/*
* The L2ARC has to hold at least the payload of one log block for
@@ -9890,6 +9817,106 @@ l2arc_rebuild_vdev(vdev_t *vd, boolean_t reopen)
}
}
/*
* Add a vdev for use by the L2ARC. By this point the spa has already
* validated the vdev and opened it.
*/
void
l2arc_add_vdev(spa_t *spa, vdev_t *vd)
{
l2arc_dev_t *adddev;
uint64_t l2dhdr_asize;
ASSERT(!l2arc_vdev_present(vd));
/*
* Create a new l2arc device entry.
*/
adddev = vmem_zalloc(sizeof (l2arc_dev_t), KM_SLEEP);
adddev->l2ad_spa = spa;
adddev->l2ad_vdev = vd;
/* leave extra size for an l2arc device header */
l2dhdr_asize = adddev->l2ad_dev_hdr_asize =
MAX(sizeof (*adddev->l2ad_dev_hdr), 1 << vd->vdev_ashift);
adddev->l2ad_start = VDEV_LABEL_START_SIZE + l2dhdr_asize;
adddev->l2ad_end = VDEV_LABEL_START_SIZE + vdev_get_min_asize(vd);
ASSERT3U(adddev->l2ad_start, <, adddev->l2ad_end);
adddev->l2ad_hand = adddev->l2ad_start;
adddev->l2ad_evict = adddev->l2ad_start;
adddev->l2ad_first = B_TRUE;
adddev->l2ad_writing = B_FALSE;
adddev->l2ad_trim_all = B_FALSE;
list_link_init(&adddev->l2ad_node);
adddev->l2ad_dev_hdr = kmem_zalloc(l2dhdr_asize, KM_SLEEP);
mutex_init(&adddev->l2ad_mtx, NULL, MUTEX_DEFAULT, NULL);
/*
* This is a list of all ARC buffers that are still valid on the
* device.
*/
list_create(&adddev->l2ad_buflist, sizeof (arc_buf_hdr_t),
offsetof(arc_buf_hdr_t, b_l2hdr.b_l2node));
/*
* This is a list of pointers to log blocks that are still present
* on the device.
*/
list_create(&adddev->l2ad_lbptr_list, sizeof (l2arc_lb_ptr_buf_t),
offsetof(l2arc_lb_ptr_buf_t, node));
vdev_space_update(vd, 0, 0, adddev->l2ad_end - adddev->l2ad_hand);
zfs_refcount_create(&adddev->l2ad_alloc);
zfs_refcount_create(&adddev->l2ad_lb_asize);
zfs_refcount_create(&adddev->l2ad_lb_count);
/*
* Decide if dev is eligible for L2ARC rebuild or whole device
* trimming. This has to happen before the device is added in the
* cache device list and l2arc_dev_mtx is released. Otherwise
* l2arc_feed_thread() might already start writing on the
* device.
*/
l2arc_rebuild_dev(adddev, B_FALSE);
/*
* Add device to global list
*/
mutex_enter(&l2arc_dev_mtx);
list_insert_head(l2arc_dev_list, adddev);
atomic_inc_64(&l2arc_ndev);
mutex_exit(&l2arc_dev_mtx);
}
/*
* Decide if a vdev is eligible for L2ARC rebuild, called from vdev_reopen()
* in case of onlining a cache device.
*/
void
l2arc_rebuild_vdev(vdev_t *vd, boolean_t reopen)
{
l2arc_dev_t *dev = NULL;
dev = l2arc_vdev_get(vd);
ASSERT3P(dev, !=, NULL);
/*
* In contrast to l2arc_add_vdev() we do not have to worry about
* l2arc_feed_thread() invalidating previous content when onlining a
* cache device. The device parameters (l2ad*) are not cleared when
* offlining the device and writing new buffers will not invalidate
* all previous content. In worst case only buffers that have not had
* their log block written to the device will be lost.
* When onlining the cache device (ie offline->online without exporting
* the pool in between) this happens:
* vdev_reopen() -> vdev_open() -> l2arc_rebuild_vdev()
* | |
* vdev_is_dead() = B_FALSE l2ad_rebuild = B_TRUE
* During the time where vdev_is_dead = B_FALSE and until l2ad_rebuild
* is set to B_TRUE we might write additional buffers to the device.
*/
l2arc_rebuild_dev(dev, reopen);
}
/*
* Remove a vdev from the L2ARC.
*/