summaryrefslogtreecommitdiff
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c5
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c116
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.h7
-rw-r--r--fs/xfs/quota/xfs_qm_syscalls.c3
-rw-r--r--fs/xfs/xfs_ag.h1
-rw-r--r--fs/xfs/xfs_dfrag.c22
-rw-r--r--fs/xfs/xfs_log.c38
-rw-r--r--fs/xfs/xfs_mount.h1
8 files changed, 164 insertions, 29 deletions
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 52e06b487ce..29f1edca76d 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1209,6 +1209,7 @@ xfs_fs_put_super(
xfs_unmountfs(mp);
xfs_freesb(mp);
+ xfs_inode_shrinker_unregister(mp);
xfs_icsb_destroy_counters(mp);
xfs_close_devices(mp);
xfs_dmops_put(mp);
@@ -1622,6 +1623,8 @@ xfs_fs_fill_super(
if (error)
goto fail_vnrele;
+ xfs_inode_shrinker_register(mp);
+
kfree(mtpt);
return 0;
@@ -1867,6 +1870,7 @@ init_xfs_fs(void)
goto out_cleanup_procfs;
vfs_initquota();
+ xfs_inode_shrinker_init();
error = register_filesystem(&xfs_fs_type);
if (error)
@@ -1894,6 +1898,7 @@ exit_xfs_fs(void)
{
vfs_exitquota();
unregister_filesystem(&xfs_fs_type);
+ xfs_inode_shrinker_destroy();
xfs_sysctl_unregister();
xfs_cleanup_procfs();
xfs_buf_terminate();
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 05cd85317f6..a427c638d90 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -95,7 +95,8 @@ xfs_inode_ag_walk(
struct xfs_perag *pag, int flags),
int flags,
int tag,
- int exclusive)
+ int exclusive,
+ int *nr_to_scan)
{
uint32_t first_index;
int last_error = 0;
@@ -134,7 +135,7 @@ restart:
if (error == EFSCORRUPTED)
break;
- } while (1);
+ } while ((*nr_to_scan)--);
if (skipped) {
delay(1);
@@ -150,12 +151,15 @@ xfs_inode_ag_iterator(
struct xfs_perag *pag, int flags),
int flags,
int tag,
- int exclusive)
+ int exclusive,
+ int *nr_to_scan)
{
int error = 0;
int last_error = 0;
xfs_agnumber_t ag;
+ int nr;
+ nr = nr_to_scan ? *nr_to_scan : INT_MAX;
for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
struct xfs_perag *pag;
@@ -165,14 +169,18 @@ xfs_inode_ag_iterator(
continue;
}
error = xfs_inode_ag_walk(mp, pag, execute, flags, tag,
- exclusive);
+ exclusive, &nr);
xfs_perag_put(pag);
if (error) {
last_error = error;
if (error == EFSCORRUPTED)
break;
}
+ if (nr <= 0)
+ break;
}
+ if (nr_to_scan)
+ *nr_to_scan = nr;
return XFS_ERROR(last_error);
}
@@ -291,7 +299,7 @@ xfs_sync_data(
ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
- XFS_ICI_NO_TAG, 0);
+ XFS_ICI_NO_TAG, 0, NULL);
if (error)
return XFS_ERROR(error);
@@ -310,7 +318,7 @@ xfs_sync_attr(
ASSERT((flags & ~SYNC_WAIT) == 0);
return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
- XFS_ICI_NO_TAG, 0);
+ XFS_ICI_NO_TAG, 0, NULL);
}
STATIC int
@@ -673,6 +681,7 @@ __xfs_inode_set_reclaim_tag(
radix_tree_tag_set(&pag->pag_ici_root,
XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino),
XFS_ICI_RECLAIM_TAG);
+ pag->pag_ici_reclaimable++;
}
/*
@@ -705,6 +714,7 @@ __xfs_inode_clear_reclaim_tag(
{
radix_tree_tag_clear(&pag->pag_ici_root,
XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+ pag->pag_ici_reclaimable--;
}
/*
@@ -820,10 +830,10 @@ xfs_reclaim_inode(
* call into reclaim to find it in a clean state instead of waiting for
* it now. We also don't return errors here - if the error is transient
* then the next reclaim pass will flush the inode, and if the error
- * is permanent then the next sync reclaim will relcaim the inode and
+ * is permanent then the next sync reclaim will reclaim the inode and
* pass on the error.
*/
- if (error && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
xfs_fs_cmn_err(CE_WARN, ip->i_mount,
"inode 0x%llx background reclaim flush failed with %d",
(long long)ip->i_ino, error);
@@ -854,5 +864,93 @@ xfs_reclaim_inodes(
int mode)
{
return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
- XFS_ICI_RECLAIM_TAG, 1);
+ XFS_ICI_RECLAIM_TAG, 1, NULL);
+}
+
+/*
+ * Shrinker infrastructure.
+ *
+ * This is all far more complex than it needs to be. It adds a global list of
+ * mounts because the shrinkers can only call a global context. We need to make
+ * the shrinkers pass a context to avoid the need for global state.
+ */
+static LIST_HEAD(xfs_mount_list);
+static struct rw_semaphore xfs_mount_list_lock;
+
+static int
+xfs_reclaim_inode_shrink(
+ int nr_to_scan,
+ gfp_t gfp_mask)
+{
+ struct xfs_mount *mp;
+ struct xfs_perag *pag;
+ xfs_agnumber_t ag;
+ int reclaimable = 0;
+
+ if (nr_to_scan) {
+ if (!(gfp_mask & __GFP_FS))
+ return -1;
+
+ down_read(&xfs_mount_list_lock);
+ list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
+ xfs_inode_ag_iterator(mp, xfs_reclaim_inode, 0,
+ XFS_ICI_RECLAIM_TAG, 1, &nr_to_scan);
+ if (nr_to_scan <= 0)
+ break;
+ }
+ up_read(&xfs_mount_list_lock);
+ }
+
+ down_read(&xfs_mount_list_lock);
+ list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
+ for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
+
+ pag = xfs_perag_get(mp, ag);
+ if (!pag->pag_ici_init) {
+ xfs_perag_put(pag);
+ continue;
+ }
+ reclaimable += pag->pag_ici_reclaimable;
+ xfs_perag_put(pag);
+ }
+ }
+ up_read(&xfs_mount_list_lock);
+ return reclaimable;
+}
+
+static struct shrinker xfs_inode_shrinker = {
+ .shrink = xfs_reclaim_inode_shrink,
+ .seeks = DEFAULT_SEEKS,
+};
+
+void __init
+xfs_inode_shrinker_init(void)
+{
+ init_rwsem(&xfs_mount_list_lock);
+ register_shrinker(&xfs_inode_shrinker);
+}
+
+void
+xfs_inode_shrinker_destroy(void)
+{
+ ASSERT(list_empty(&xfs_mount_list));
+ unregister_shrinker(&xfs_inode_shrinker);
+}
+
+void
+xfs_inode_shrinker_register(
+ struct xfs_mount *mp)
+{
+ down_write(&xfs_mount_list_lock);
+ list_add_tail(&mp->m_mplist, &xfs_mount_list);
+ up_write(&xfs_mount_list_lock);
+}
+
+void
+xfs_inode_shrinker_unregister(
+ struct xfs_mount *mp)
+{
+ down_write(&xfs_mount_list_lock);
+ list_del(&mp->m_mplist);
+ up_write(&xfs_mount_list_lock);
}
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index d480c346cab..cdcbaaca988 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -53,6 +53,11 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
int xfs_inode_ag_iterator(struct xfs_mount *mp,
int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
- int flags, int tag, int write_lock);
+ int flags, int tag, int write_lock, int *nr_to_scan);
+
+void xfs_inode_shrinker_init(void);
+void xfs_inode_shrinker_destroy(void);
+void xfs_inode_shrinker_register(struct xfs_mount *mp);
+void xfs_inode_shrinker_unregister(struct xfs_mount *mp);
#endif
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 5d0ee8d492d..50bee07d6b0 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -891,7 +891,8 @@ xfs_qm_dqrele_all_inodes(
uint flags)
{
ASSERT(mp->m_quotainfo);
- xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
+ xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags,
+ XFS_ICI_NO_TAG, 0, NULL);
}
/*------------------------------------------------------------------------*/
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index b1a5a1ff88e..abb8222b88c 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -223,6 +223,7 @@ typedef struct xfs_perag {
int pag_ici_init; /* incore inode cache initialised */
rwlock_t pag_ici_lock; /* incore inode lock */
struct radix_tree_root pag_ici_root; /* incore inode cache root */
+ int pag_ici_reclaimable; /* reclaimable inodes */
#endif
int pagb_count; /* pagb slots in use */
xfs_perag_busy_t pagb_list[XFS_PAGB_NUM_SLOTS]; /* unstable blocks */
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index cd27c9d6c71..5bba29a0781 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -177,16 +177,26 @@ xfs_swap_extents_check_format(
XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
return EINVAL;
- /* Check root block of temp in btree form to max in target */
+ /*
+ * If we are in a btree format, check that the temp root block will fit
+ * in the target and that it has enough extents to be in btree format
+ * in the target.
+ *
+ * Note that we have to be careful to allow btree->extent conversions
+ * (a common defrag case) which will occur when the temp inode is in
+ * extent format...
+ */
if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_BOFF(ip) &&
- tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+ ((XFS_IFORK_BOFF(ip) &&
+ tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip)) ||
+ XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= ip->i_df.if_ext_max))
return EINVAL;
- /* Check root block of target in btree form to max in temp */
+ /* Reciprocal target->temp btree format checks */
if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_BOFF(tip) &&
- ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+ ((XFS_IFORK_BOFF(tip) &&
+ ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip)) ||
+ XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= tip->i_df.if_ext_max))
return EINVAL;
return 0;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index e8fba92d7cd..2be01913628 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -745,9 +745,16 @@ xfs_log_move_tail(xfs_mount_t *mp,
/*
* Determine if we have a transaction that has gone to disk
- * that needs to be covered. Log activity needs to be idle (no AIL and
- * nothing in the iclogs). And, we need to be in the right state indicating
- * something has gone out.
+ * that needs to be covered. To begin the transition to the idle state
+ * firstly the log needs to be idle (no AIL and nothing in the iclogs).
+ * If we are then in a state where covering is needed, the caller is informed
+ * that dummy transactions are required to move the log into the idle state.
+ *
+ * Because this is called as part of the sync process, we should also indicate
+ * that dummy transactions should be issued in anything but the covered or
+ * idle states. This ensures that the log tail is accurately reflected in
+ * the log at the end of the sync, hence if a crash occurrs avoids replay
+ * of transactions where the metadata is already on disk.
*/
int
xfs_log_need_covered(xfs_mount_t *mp)
@@ -759,17 +766,24 @@ xfs_log_need_covered(xfs_mount_t *mp)
return 0;
spin_lock(&log->l_icloglock);
- if (((log->l_covered_state == XLOG_STATE_COVER_NEED) ||
- (log->l_covered_state == XLOG_STATE_COVER_NEED2))
- && !xfs_trans_ail_tail(log->l_ailp)
- && xlog_iclogs_empty(log)) {
- if (log->l_covered_state == XLOG_STATE_COVER_NEED)
- log->l_covered_state = XLOG_STATE_COVER_DONE;
- else {
- ASSERT(log->l_covered_state == XLOG_STATE_COVER_NEED2);
- log->l_covered_state = XLOG_STATE_COVER_DONE2;
+ switch (log->l_covered_state) {
+ case XLOG_STATE_COVER_DONE:
+ case XLOG_STATE_COVER_DONE2:
+ case XLOG_STATE_COVER_IDLE:
+ break;
+ case XLOG_STATE_COVER_NEED:
+ case XLOG_STATE_COVER_NEED2:
+ if (!xfs_trans_ail_tail(log->l_ailp) &&
+ xlog_iclogs_empty(log)) {
+ if (log->l_covered_state == XLOG_STATE_COVER_NEED)
+ log->l_covered_state = XLOG_STATE_COVER_DONE;
+ else
+ log->l_covered_state = XLOG_STATE_COVER_DONE2;
}
+ /* FALLTHRU */
+ default:
needed = 1;
+ break;
}
spin_unlock(&log->l_icloglock);
return needed;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 4fa0bc7b983..9ff48a16a7e 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -259,6 +259,7 @@ typedef struct xfs_mount {
wait_queue_head_t m_wait_single_sync_task;
__int64_t m_update_flags; /* sb flags we need to update
on the next remount,rw */
+ struct list_head m_mplist; /* inode shrinker mount list */
} xfs_mount_t;
/*
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-17 13:57:30 +0000