/* inode is to be reclaimed */ #define XFS_ICI_RECLAIM_TAG 0 /* Inode has speculative preallocations (posteof or cow) to clean. */ #define XFS_ICI_BLOCKGC_TAG 1
/* * The goal for walking incore inodes. These can correspond with incore inode * radix tree tags when convenient. Avoid existing XFS_IWALK namespace.
*/ enum xfs_icwalk_goal { /* Goals directly associated with tagged inodes. */
XFS_ICWALK_BLOCKGC = XFS_ICI_BLOCKGC_TAG,
XFS_ICWALK_RECLAIM = XFS_ICI_RECLAIM_TAG,
};
/* * XXX: If this didn't occur in transactions, we could drop GFP_NOFAIL * and return NULL here on ENOMEM.
*/
ip = alloc_inode_sb(mp->m_super, xfs_inode_cache, GFP_KERNEL | __GFP_NOFAIL);
if (inode_init_always(mp->m_super, VFS_I(ip))) {
kmem_cache_free(xfs_inode_cache, ip); return NULL;
}
switch (VFS_I(ip)->i_mode & S_IFMT) { case S_IFREG: case S_IFDIR: case S_IFLNK:
xfs_idestroy_fork(&ip->i_df); break;
}
xfs_ifork_zap_attr(ip);
if (ip->i_cowfp) {
xfs_idestroy_fork(ip->i_cowfp);
kmem_cache_free(xfs_ifork_cache, ip->i_cowfp);
} if (ip->i_itemp) {
ASSERT(!test_bit(XFS_LI_IN_AIL,
&ip->i_itemp->ili_item.li_flags));
xfs_inode_item_destroy(ip);
ip->i_itemp = NULL;
}
kmem_cache_free(xfs_inode_cache, ip);
}
staticvoid
__xfs_inode_free( struct xfs_inode *ip)
{ /* asserts to verify all state is correct here */
ASSERT(atomic_read(&ip->i_pincount) == 0);
ASSERT(!ip->i_itemp || list_empty(&ip->i_itemp->ili_item.li_bio_list));
XFS_STATS_DEC(ip->i_mount, vn_active);
/* * Because we use RCU freeing we need to ensure the inode always * appears to be reclaimed with an invalid inode number when in the * free state. The ip->i_flags_lock provides the barrier against lookup * races.
*/
spin_lock(&ip->i_flags_lock);
ip->i_flags = XFS_IRECLAIM;
ip->i_ino = 0;
spin_unlock(&ip->i_flags_lock);
__xfs_inode_free(ip);
}
/* * Queue background inode reclaim work if there are reclaimable inodes and there * isn't reclaim work already scheduled or in progress.
*/ staticvoid
xfs_reclaim_work_queue( struct xfs_mount *mp)
{
/* * Background scanning to trim preallocated space. This is queued based on the * 'speculative_prealloc_lifetime' tunable (5m by default).
*/ staticinlinevoid
xfs_blockgc_queue( struct xfs_perag *pag)
{ struct xfs_mount *mp = pag_mount(pag);
if (!xfs_is_blockgc_enabled(mp)) return;
rcu_read_lock(); if (radix_tree_tagged(&pag->pag_ici_root, XFS_ICI_BLOCKGC_TAG))
queue_delayed_work(mp->m_blockgc_wq, &pag->pag_blockgc_work,
secs_to_jiffies(xfs_blockgc_secs));
rcu_read_unlock();
}
/* Set a tag on both the AG incore inode tree and the AG radix tree. */ staticvoid
xfs_perag_set_inode_tag( struct xfs_perag *pag,
xfs_agino_t agino, unsignedint tag)
{ bool was_tagged;
if (tag == XFS_ICI_RECLAIM_TAG)
pag->pag_ici_reclaimable++;
if (was_tagged) return;
/* propagate the tag up into the pag xarray tree */
xfs_group_set_mark(pag_group(pag), ici_tag_to_mark(tag));
/* start background work */ switch (tag) { case XFS_ICI_RECLAIM_TAG:
xfs_reclaim_work_queue(pag_mount(pag)); break; case XFS_ICI_BLOCKGC_TAG:
xfs_blockgc_queue(pag); break;
}
trace_xfs_perag_set_inode_tag(pag, _RET_IP_);
}
/* Clear a tag on both the AG incore inode tree and the AG radix tree. */ staticvoid
xfs_perag_clear_inode_tag( struct xfs_perag *pag,
xfs_agino_t agino, unsignedint tag)
{
lockdep_assert_held(&pag->pag_ici_lock);
/* * Reclaim can signal (with a null agino) that it cleared its own tag * by removing the inode from the radix tree.
*/ if (agino != NULLAGINO)
radix_tree_tag_clear(&pag->pag_ici_root, agino, tag); else
ASSERT(tag == XFS_ICI_RECLAIM_TAG);
if (tag == XFS_ICI_RECLAIM_TAG)
pag->pag_ici_reclaimable--;
if (radix_tree_tagged(&pag->pag_ici_root, tag)) return;
/* clear the tag from the pag xarray */
xfs_group_clear_mark(pag_group(pag), ici_tag_to_mark(tag));
trace_xfs_perag_clear_inode_tag(pag, _RET_IP_);
}
/* * Find the next AG after @pag, or the first AG if @pag is NULL.
*/ staticstruct xfs_perag *
xfs_perag_grab_next_tag( struct xfs_mount *mp, struct xfs_perag *pag, int tag)
{ return to_perag(xfs_group_grab_next_mark(mp,
pag ? pag_group(pag) : NULL,
ici_tag_to_mark(tag), XG_TYPE_AG));
}
/* * When we recycle a reclaimable inode, we need to re-initialise the VFS inode * part of the structure. This is made more complex by the fact we store * information about the on-disk values in the VFS inode and so we can't just * overwrite the values unconditionally. Hence we save the parameters we * need to retain across reinitialisation, and rewrite them into the VFS inode * after reinitialisation even if it fails.
*/ staticint
xfs_reinit_inode( struct xfs_mount *mp, struct inode *inode)
{ int error;
uint32_t nlink = inode->i_nlink;
uint32_t generation = inode->i_generation;
uint64_t version = inode_peek_iversion(inode);
umode_t mode = inode->i_mode;
dev_t dev = inode->i_rdev;
kuid_t uid = inode->i_uid;
kgid_t gid = inode->i_gid; unsignedlong state = inode->i_state;
/* * Carefully nudge an inode whose VFS state has been torn down back into a * usable state. Drops the i_flags_lock and the rcu read lock.
*/ staticint
xfs_iget_recycle( struct xfs_perag *pag, struct xfs_inode *ip) __releases(&ip->i_flags_lock)
{ struct xfs_mount *mp = ip->i_mount; struct inode *inode = VFS_I(ip); int error;
trace_xfs_iget_recycle(ip);
if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) return -EAGAIN;
/* * We need to make it look like the inode is being reclaimed to prevent * the actual reclaim workers from stomping over us while we recycle * the inode. We can't clear the radix tree tag yet as it requires * pag_ici_lock to be held exclusive.
*/
ip->i_flags |= XFS_IRECLAIM;
ASSERT(!rwsem_is_locked(&inode->i_rwsem));
error = xfs_reinit_inode(mp, inode);
xfs_iunlock(ip, XFS_ILOCK_EXCL); if (error) { /* * Re-initializing the inode failed, and we are in deep * trouble. Try to re-add it to the reclaim list.
*/
rcu_read_lock();
spin_lock(&ip->i_flags_lock);
ip->i_flags &= ~(XFS_INEW | XFS_IRECLAIM);
ASSERT(ip->i_flags & XFS_IRECLAIMABLE);
spin_unlock(&ip->i_flags_lock);
rcu_read_unlock();
/* * Clear the per-lifetime state in the inode as we are now effectively * a new inode and need to return to the initial state before reuse * occurs.
*/
ip->i_flags &= ~XFS_IRECLAIM_RESET_FLAGS;
ip->i_flags |= XFS_INEW;
xfs_perag_clear_inode_tag(pag, XFS_INO_TO_AGINO(mp, ip->i_ino),
XFS_ICI_RECLAIM_TAG);
inode->i_state = I_NEW;
spin_unlock(&ip->i_flags_lock);
spin_unlock(&pag->pag_ici_lock);
return 0;
}
/* * If we are allocating a new inode, then check what was returned is * actually a free, empty inode. If we are not allocating an inode, * then check we didn't find a free inode. * * Returns: * 0 if the inode free state matches the lookup context * -ENOENT if the inode is free and we are not allocating * -EFSCORRUPTED if there is any state mismatch at all
*/ staticint
xfs_iget_check_free_state( struct xfs_inode *ip, int flags)
{ if (flags & XFS_IGET_CREATE) { /* should be a free inode */ if (VFS_I(ip)->i_mode != 0) {
xfs_warn(ip->i_mount, "Corruption detected! Free inode 0x%llx not marked free! (mode 0x%x)",
ip->i_ino, VFS_I(ip)->i_mode);
xfs_agno_mark_sick(ip->i_mount,
XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
XFS_SICK_AG_INOBT); return -EFSCORRUPTED;
}
/* should be an allocated inode */ if (VFS_I(ip)->i_mode == 0) return -ENOENT;
return 0;
}
/* Make all pending inactivation work start immediately. */ staticbool
xfs_inodegc_queue_all( struct xfs_mount *mp)
{ struct xfs_inodegc *gc; int cpu; bool ret = false;
for_each_cpu(cpu, &mp->m_inodegc_cpumask) {
gc = per_cpu_ptr(mp->m_inodegc, cpu); if (!llist_empty(&gc->list)) {
mod_delayed_work_on(cpu, mp->m_inodegc_wq, &gc->work, 0);
ret = true;
}
}
return ret;
}
/* Wait for all queued work and collect errors */ staticint
xfs_inodegc_wait_all( struct xfs_mount *mp)
{ int cpu; int error = 0;
/* * Check the validity of the inode we just found it the cache
*/ staticint
xfs_iget_cache_hit( struct xfs_perag *pag, struct xfs_inode *ip,
xfs_ino_t ino, int flags, int lock_flags) __releases(RCU)
{ struct inode *inode = VFS_I(ip); struct xfs_mount *mp = ip->i_mount; int error;
/* * check for re-use of an inode within an RCU grace period due to the * radix tree nodes not being updated yet. We monitor for this by * setting the inode number to zero before freeing the inode structure. * If the inode has been reallocated and set up, then the inode number * will not match, so check for that, too.
*/
spin_lock(&ip->i_flags_lock); if (ip->i_ino != ino) goto out_skip;
/* * If we are racing with another cache hit that is currently * instantiating this inode or currently recycling it out of * reclaimable state, wait for the initialisation to complete * before continuing. * * If we're racing with the inactivation worker we also want to wait. * If we're creating a new file, it's possible that the worker * previously marked the inode as free on disk but hasn't finished * updating the incore state yet. The AGI buffer will be dirty and * locked to the icreate transaction, so a synchronous push of the * inodegc workers would result in deadlock. For a regular iget, the * worker is running already, so we might as well wait. * * XXX(hch): eventually we should do something equivalent to * wait_on_inode to wait for these flags to be cleared * instead of polling for it.
*/ if (ip->i_flags & (XFS_INEW | XFS_IRECLAIM | XFS_INACTIVATING)) goto out_skip;
if (ip->i_flags & XFS_NEED_INACTIVE) { /* Unlinked inodes cannot be re-grabbed. */ if (VFS_I(ip)->i_nlink == 0) {
error = -ENOENT; goto out_error;
} goto out_inodegc_flush;
}
/* * Check the inode free state is valid. This also detects lookup * racing with unlinks.
*/
error = xfs_iget_check_free_state(ip, flags); if (error) goto out_error;
/* Skip inodes that have no vfs state. */ if ((flags & XFS_IGET_INCORE) &&
(ip->i_flags & XFS_IRECLAIMABLE)) goto out_skip;
/* The inode fits the selection criteria; process it. */ if (ip->i_flags & XFS_IRECLAIMABLE) { /* Drops i_flags_lock and RCU read lock. */
error = xfs_iget_recycle(pag, ip); if (error == -EAGAIN) goto out_skip; if (error) return error;
} else { /* If the VFS inode is being torn down, pause and try again. */ if (!igrab(inode)) goto out_skip;
/* We've got a live one. */
spin_unlock(&ip->i_flags_lock);
rcu_read_unlock();
trace_xfs_iget_hit(ip);
}
if (lock_flags != 0)
xfs_ilock(ip, lock_flags);
if (!(flags & XFS_IGET_INCORE))
xfs_iflags_clear(ip, XFS_ISTALE);
XFS_STATS_INC(mp, xs_ig_found);
out_inodegc_flush:
spin_unlock(&ip->i_flags_lock);
rcu_read_unlock(); /* * Do not wait for the workers, because the caller could hold an AGI * buffer lock. We're just going to sleep in a loop anyway.
*/ if (xfs_is_inodegc_enabled(mp))
xfs_inodegc_queue_all(mp); return -EAGAIN;
}
staticint
xfs_iget_cache_miss( struct xfs_mount *mp, struct xfs_perag *pag,
xfs_trans_t *tp,
xfs_ino_t ino, struct xfs_inode **ipp, int flags, int lock_flags)
{ struct xfs_inode *ip; int error;
xfs_agino_t agino = XFS_INO_TO_AGINO(mp, ino);
ip = xfs_inode_alloc(mp, ino); if (!ip) return -ENOMEM;
/* * For version 5 superblocks, if we are initialising a new inode and we * are not utilising the XFS_FEAT_IKEEP inode cluster mode, we can * simply build the new inode core with a random generation number. * * For version 4 (and older) superblocks, log recovery is dependent on * the i_flushiter field being initialised from the current on-disk * value and hence we must also read the inode off disk even when * initializing new inodes.
*/ if (xfs_has_v3inodes(mp) &&
(flags & XFS_IGET_CREATE) && !xfs_has_ikeep(mp)) {
VFS_I(ip)->i_generation = get_random_u32();
} else { struct xfs_buf *bp;
error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp); if (error) goto out_destroy;
/* * Check the inode free state is valid. This also detects lookup * racing with unlinks.
*/
error = xfs_iget_check_free_state(ip, flags); if (error) goto out_destroy;
/* * Preload the radix tree so we can insert safely under the * write spinlock. Note that we cannot sleep inside the preload * region.
*/ if (radix_tree_preload(GFP_KERNEL | __GFP_NOLOCKDEP)) {
error = -EAGAIN; goto out_destroy;
}
/* * Because the inode hasn't been added to the radix-tree yet it can't * be found by another thread, so we can do the non-sleeping lock here.
*/ if (lock_flags) { if (!xfs_ilock_nowait(ip, lock_flags))
BUG();
}
/* * These values must be set before inserting the inode into the radix * tree as the moment it is inserted a concurrent lookup (allowed by the * RCU locking mechanism) can find it and that lookup must see that this * is an inode currently under construction (i.e. that XFS_INEW is set). * The ip->i_flags_lock that protects the XFS_INEW flag forms the * memory barrier that ensures this detection works correctly at lookup * time.
*/ if (flags & XFS_IGET_DONTCACHE)
d_mark_dontcache(VFS_I(ip));
ip->i_udquot = NULL;
ip->i_gdquot = NULL;
ip->i_pdquot = NULL;
xfs_iflags_set(ip, XFS_INEW);
/* insert the new inode */
spin_lock(&pag->pag_ici_lock);
error = radix_tree_insert(&pag->pag_ici_root, agino, ip); if (unlikely(error)) {
WARN_ON(error != -EEXIST);
XFS_STATS_INC(mp, xs_ig_dup);
error = -EAGAIN; goto out_preload_end;
}
spin_unlock(&pag->pag_ici_lock);
radix_tree_preload_end();
/* * Look up an inode by number in the given file system. The inode is looked up * in the cache held in each AG. If the inode is found in the cache, initialise * the vfs inode if necessary. * * If it is not in core, read it in from the file system's device, add it to the * cache and initialise the vfs inode. * * The inode is locked according to the value of the lock_flags parameter. * Inode lookup is only done during metadata operations and not as part of the * data IO path. Hence we only allow locking of the XFS_ILOCK during lookup.
*/ int
xfs_iget( struct xfs_mount *mp, struct xfs_trans *tp,
xfs_ino_t ino,
uint flags,
uint lock_flags, struct xfs_inode **ipp)
{ struct xfs_inode *ip; struct xfs_perag *pag;
xfs_agino_t agino; int error;
/* * If we have a real type for an on-disk inode, we can setup the inode * now. If it's a new inode being created, xfs_init_new_inode will * handle it.
*/ if (xfs_iflags_test(ip, XFS_INEW) && VFS_I(ip)->i_mode != 0)
xfs_setup_existing_inode(ip); return 0;
/* * Get a metadata inode. * * The metafile type must match the file mode exactly, and for files in the * metadata directory tree, it must match the inode's metatype exactly.
*/ int
xfs_trans_metafile_iget( struct xfs_trans *tp,
xfs_ino_t ino, enum xfs_metafile_type metafile_type, struct xfs_inode **ipp)
{ struct xfs_mount *mp = tp->t_mountp; struct xfs_inode *ip;
umode_t mode; int error;
/* Grab a metadata file if the caller doesn't already have a transaction. */ int
xfs_metafile_iget( struct xfs_mount *mp,
xfs_ino_t ino, enum xfs_metafile_type metafile_type, struct xfs_inode **ipp)
{ struct xfs_trans *tp; int error;
/* * Grab the inode for reclaim exclusively. * * We have found this inode via a lookup under RCU, so the inode may have * already been freed, or it may be in the process of being recycled by * xfs_iget(). In both cases, the inode will have XFS_IRECLAIM set. If the inode * has been fully recycled by the time we get the i_flags_lock, XFS_IRECLAIMABLE * will not be set. Hence we need to check for both these flag conditions to * avoid inodes that are no longer reclaim candidates. * * Note: checking for other state flags here, under the i_flags_lock or not, is * racy and should be avoided. Those races should be resolved only after we have * ensured that we are able to reclaim this inode and the world can see that we * are going to reclaim it. * * Return true if we grabbed it, false otherwise.
*/ staticbool
xfs_reclaim_igrab( struct xfs_inode *ip, struct xfs_icwalk *icw)
{
ASSERT(rcu_read_lock_held());
spin_lock(&ip->i_flags_lock); if (!__xfs_iflags_test(ip, XFS_IRECLAIMABLE) ||
__xfs_iflags_test(ip, XFS_IRECLAIM)) { /* not a reclaim candidate. */
spin_unlock(&ip->i_flags_lock); returnfalse;
}
/* Don't reclaim a sick inode unless the caller asked for it. */ if (ip->i_sick &&
(!icw || !(icw->icw_flags & XFS_ICWALK_FLAG_RECLAIM_SICK))) {
spin_unlock(&ip->i_flags_lock); returnfalse;
}
/* * Inode reclaim is non-blocking, so the default action if progress cannot be * made is to "requeue" the inode for reclaim by unlocking it and clearing the * XFS_IRECLAIM flag. If we are in a shutdown state, we don't care about * blocking anymore and hence we can wait for the inode to be able to reclaim * it. * * We do no IO here - if callers require inodes to be cleaned they must push the * AIL first to trigger writeback of dirty inodes. This enables writeback to be * done in the background in a non-blocking manner, and enables memory reclaim * to make progress without blocking.
*/ staticvoid
xfs_reclaim_inode( struct xfs_inode *ip, struct xfs_perag *pag)
{
xfs_ino_t ino = ip->i_ino; /* for radix_tree_delete */
if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) goto out; if (xfs_iflags_test_and_set(ip, XFS_IFLUSHING)) goto out_iunlock;
/* * Check for log shutdown because aborting the inode can move the log * tail and corrupt in memory state. This is fine if the log is shut * down, but if the log is still active and only the mount is shut down * then the in-memory log tail movement caused by the abort can be * incorrectly propagated to disk.
*/ if (xlog_is_shutdown(ip->i_mount->m_log)) {
xfs_iunpin_wait(ip); /* * Avoid a ABBA deadlock on the inode cluster buffer vs * concurrent xfs_ifree_cluster() trying to mark the inode * stale. We don't need the inode locked to run the flush abort * code, but the flush abort needs to lock the cluster buffer.
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_iflush_shutdown_abort(ip);
xfs_ilock(ip, XFS_ILOCK_EXCL); goto reclaim;
} if (xfs_ipincount(ip)) goto out_clear_flush; if (!xfs_inode_clean(ip)) goto out_clear_flush;
/* * Because we use RCU freeing we need to ensure the inode always appears * to be reclaimed with an invalid inode number when in the free state. * We do this as early as possible under the ILOCK so that * xfs_iflush_cluster() and xfs_ifree_cluster() can be guaranteed to * detect races with us here. By doing this, we guarantee that once * xfs_iflush_cluster() or xfs_ifree_cluster() has locked XFS_ILOCK that * it will see either a valid inode that will serialise correctly, or it * will see an invalid inode that it can skip.
*/
spin_lock(&ip->i_flags_lock);
ip->i_flags = XFS_IRECLAIM;
ip->i_ino = 0;
ip->i_sick = 0;
ip->i_checked = 0;
spin_unlock(&ip->i_flags_lock);
XFS_STATS_INC(ip->i_mount, xs_ig_reclaims); /* * Remove the inode from the per-AG radix tree. * * Because radix_tree_delete won't complain even if the item was never * added to the tree assert that it's been there before to catch * problems with the inode life time early on.
*/
spin_lock(&pag->pag_ici_lock); if (!radix_tree_delete(&pag->pag_ici_root,
XFS_INO_TO_AGINO(ip->i_mount, ino)))
ASSERT(0);
xfs_perag_clear_inode_tag(pag, NULLAGINO, XFS_ICI_RECLAIM_TAG);
spin_unlock(&pag->pag_ici_lock);
/* * Here we do an (almost) spurious inode lock in order to coordinate * with inode cache radix tree lookups. This is because the lookup * can reference the inodes in the cache without taking references. * * We make that OK here by ensuring that we wait until the inode is * unlocked after the lookup before we go ahead and free it.
*/
xfs_ilock(ip, XFS_ILOCK_EXCL);
ASSERT(!ip->i_udquot && !ip->i_gdquot && !ip->i_pdquot);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
ASSERT(xfs_inode_clean(ip));
/* * The shrinker infrastructure determines how many inodes we should scan for * reclaim. We want as many clean inodes ready to reclaim as possible, so we * push the AIL here. We also want to proactively free up memory if we can to * minimise the amount of work memory reclaim has to do so we kick the * background reclaim if it isn't already scheduled.
*/ long
xfs_reclaim_inodes_nr( struct xfs_mount *mp, unsignedlong nr_to_scan)
{ struct xfs_icwalk icw = {
.icw_flags = XFS_ICWALK_FLAG_SCAN_LIMIT,
.icw_scan_limit = min_t(unsignedlong, LONG_MAX, nr_to_scan),
};
if (xfs_want_reclaim_sick(mp))
icw.icw_flags |= XFS_ICWALK_FLAG_RECLAIM_SICK;
/* kick background reclaimer and push the AIL */
xfs_reclaim_work_queue(mp);
xfs_ail_push_all(mp->m_ail);
/* * Return the number of reclaimable inodes in the filesystem for * the shrinker to determine how much to reclaim.
*/ long
xfs_reclaim_inodes_count( struct xfs_mount *mp)
{
XA_STATE (xas, &mp->m_groups[XG_TYPE_AG].xa, 0); long reclaimable = 0; struct xfs_perag *pag;
if ((icw->icw_flags & XFS_ICWALK_FLAG_GID) &&
!gid_eq(VFS_I(ip)->i_gid, icw->icw_gid)) returnfalse;
if ((icw->icw_flags & XFS_ICWALK_FLAG_PRID) &&
ip->i_projid != icw->icw_prid) returnfalse;
returntrue;
}
/* * A union-based inode filtering algorithm. Process the inode if any of the * criteria match. This is for global/internal scans only.
*/ STATICbool
xfs_icwalk_match_id_union( struct xfs_inode *ip, struct xfs_icwalk *icw)
{ if ((icw->icw_flags & XFS_ICWALK_FLAG_UID) &&
uid_eq(VFS_I(ip)->i_uid, icw->icw_uid)) returntrue;
if ((icw->icw_flags & XFS_ICWALK_FLAG_GID) &&
gid_eq(VFS_I(ip)->i_gid, icw->icw_gid)) returntrue;
if ((icw->icw_flags & XFS_ICWALK_FLAG_PRID) &&
ip->i_projid == icw->icw_prid) returntrue;
returnfalse;
}
/* * Is this inode @ip eligible for eof/cow block reclamation, given some * filtering parameters @icw? The inode is eligible if @icw is null or * if the predicate functions match.
*/ staticbool
xfs_icwalk_match( struct xfs_inode *ip, struct xfs_icwalk *icw)
{ bool match;
if (!icw) returntrue;
if (icw->icw_flags & XFS_ICWALK_FLAG_UNION)
match = xfs_icwalk_match_id_union(ip, icw); else
match = xfs_icwalk_match_id(ip, icw); if (!match) returnfalse;
/* skip the inode if the file size is too small */ if ((icw->icw_flags & XFS_ICWALK_FLAG_MINFILESIZE) &&
XFS_ISIZE(ip) < icw->icw_min_file_size) returnfalse;
returntrue;
}
/* * This is a fast pass over the inode cache to try to get reclaim moving on as * many inodes as possible in a short period of time. It kicks itself every few * seconds, as well as being kicked by the inode cache shrinker when memory * goes low.
*/ void
xfs_reclaim_worker( struct work_struct *work)
{ struct xfs_mount *mp = container_of(to_delayed_work(work), struct xfs_mount, m_reclaim_work);
if (!xfs_iflags_test(ip, XFS_IEOFBLOCKS)) return 0;
/* * If the mapping is dirty the operation can block and wait for some * time. Unless we are waiting, skip it.
*/ if (!wait && mapping_tagged(VFS_I(ip)->i_mapping, PAGECACHE_TAG_DIRTY)) return 0;
if (!xfs_icwalk_match(ip, icw)) return 0;
/* * If the caller is waiting, return -EAGAIN to keep the background * scanner moving and revisit the inode in a subsequent pass.
*/ if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { if (wait) return -EAGAIN; return 0;
}
*lockflags |= XFS_IOLOCK_EXCL;
if (xfs_can_free_eofblocks(ip)) return xfs_free_eofblocks(ip);
/* inode could be preallocated */
trace_xfs_inode_free_eofblocks_invalid(ip);
xfs_inode_clear_eofblocks_tag(ip); return 0;
}
/* * Don't bother locking the AG and looking up in the radix trees * if we already know that we have the tag set.
*/ if (ip->i_flags & iflag) return;
spin_lock(&ip->i_flags_lock);
ip->i_flags |= iflag;
spin_unlock(&ip->i_flags_lock);
/* * Just clear the tag if we have an empty cow fork or none at all. It's * possible the inode was fully unshared since it was originally tagged.
*/ if (!xfs_inode_has_cow_data(ip)) {
trace_xfs_inode_free_cowblocks_invalid(ip);
xfs_inode_clear_cowblocks_tag(ip); returnfalse;
}
/* * A cowblocks trim of an inode can have a significant effect on * fragmentation even when a reasonable COW extent size hint is set. * Therefore, we prefer to not process cowblocks unless they are clean * and idle. We can never process a cowblocks inode that is dirty or has * in-flight I/O under any circumstances, because outstanding writeback * or dio expects targeted COW fork blocks exist through write * completion where they can be remapped into the data fork. * * Therefore, the heuristic used here is to never process inodes * currently opened for write from background (i.e. non-sync) scans. For * sync scans, use the pagecache/dio state of the inode to ensure we * never free COW fork blocks out from under pending I/O.
*/ if (!sync && inode_is_open_for_write(VFS_I(ip))) returnfalse; return xfs_can_free_cowblocks(ip);
}
/* * Automatic CoW Reservation Freeing * * These functions automatically garbage collect leftover CoW reservations * that were made on behalf of a cowextsize hint when we start to run out * of quota or when the reservations sit around for too long. If the file * has dirty pages or is undergoing writeback, its CoW reservations will * be retained. * * The actual garbage collection piggybacks off the same code that runs * the speculative EOF preallocation garbage collector.
*/ STATICint
xfs_inode_free_cowblocks( struct xfs_inode *ip, struct xfs_icwalk *icw, unsignedint *lockflags)
{ bool wait; int ret = 0;
if (!xfs_iflags_test(ip, XFS_ICOWBLOCKS)) return 0;
if (!xfs_prep_free_cowblocks(ip, icw)) return 0;
if (!xfs_icwalk_match(ip, icw)) return 0;
/* * If the caller is waiting, return -EAGAIN to keep the background * scanner moving and revisit the inode in a subsequent pass.
*/ if (!(*lockflags & XFS_IOLOCK_EXCL) &&
!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { if (wait) return -EAGAIN; return 0;
}
*lockflags |= XFS_IOLOCK_EXCL;
if (!xfs_ilock_nowait(ip, XFS_MMAPLOCK_EXCL)) { if (wait) return -EAGAIN; return 0;
}
*lockflags |= XFS_MMAPLOCK_EXCL;
/* * Check again, nobody else should be able to dirty blocks or change * the reflink iflag now that we have the first two locks held.
*/ if (xfs_prep_free_cowblocks(ip, icw))
ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, false); return ret;
}
trace_xfs_blockgc_start(mp, __return_address); while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
xfs_blockgc_queue(pag);
}
/* Don't try to run block gc on an inode that's in any of these states. */ #define XFS_BLOCKGC_NOGRAB_IFLAGS (XFS_INEW | \
XFS_NEED_INACTIVE | \
XFS_INACTIVATING | \
XFS_IRECLAIMABLE | \
XFS_IRECLAIM) /* * Decide if the given @ip is eligible for garbage collection of speculative * preallocations, and grab it if so. Returns true if it's ready to go or * false if we should just ignore it.
*/ staticbool
xfs_blockgc_igrab( struct xfs_inode *ip)
{ struct inode *inode = VFS_I(ip);
ASSERT(rcu_read_lock_held());
/* Check for stale RCU freed inode */
spin_lock(&ip->i_flags_lock); if (!ip->i_ino) goto out_unlock_noent;
if (ip->i_flags & XFS_BLOCKGC_NOGRAB_IFLAGS) goto out_unlock_noent;
spin_unlock(&ip->i_flags_lock);
/* nothing to sync during shutdown */ if (xfs_is_shutdown(ip->i_mount)) returnfalse;
/* If we can't grab the inode, it must on it's way to reclaim. */ if (!igrab(inode)) returnfalse;
/* Scan one incore inode for block preallocations that we can remove. */ staticint
xfs_blockgc_scan_inode( struct xfs_inode *ip, struct xfs_icwalk *icw)
{ unsignedint lockflags = 0; int error;
error = xfs_inode_free_eofblocks(ip, icw, &lockflags); if (error) goto unlock;
/* * Try to free space in the filesystem by purging inactive inodes, eofblocks * and cowblocks.
*/ int
xfs_blockgc_free_space( struct xfs_mount *mp, struct xfs_icwalk *icw)
{ int error;
trace_xfs_blockgc_free_space(mp, icw, _RET_IP_);
error = xfs_icwalk(mp, XFS_ICWALK_BLOCKGC, icw); if (error) return error;
return xfs_inodegc_flush(mp);
}
/* * Reclaim all the free space that we can by scheduling the background blockgc * and inodegc workers immediately and waiting for them all to clear.
*/ int
xfs_blockgc_flush_all( struct xfs_mount *mp)
{ struct xfs_perag *pag = NULL;
/* * For each blockgc worker, move its queue time up to now. If it wasn't * queued, it will not be requeued. Then flush whatever is left.
*/ while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
mod_delayed_work(mp->m_blockgc_wq, &pag->pag_blockgc_work, 0);
while ((pag = xfs_perag_grab_next_tag(mp, pag, XFS_ICI_BLOCKGC_TAG)))
flush_delayed_work(&pag->pag_blockgc_work);
return xfs_inodegc_flush(mp);
}
/* * Run cow/eofblocks scans on the supplied dquots. We don't know exactly which * quota caused an allocation failure, so we make a best effort by including * each quota under low free space conditions (less than 1% free space) in the * scan. * * Callers must not hold any inode's ILOCK. If requesting a synchronous scan * (XFS_ICWALK_FLAG_SYNC), the caller also must not hold any inode's IOLOCK or * MMAPLOCK.
*/ int
xfs_blockgc_free_dquots( struct xfs_mount *mp, struct xfs_dquot *udqp, struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, unsignedint iwalk_flags)
{ struct xfs_icwalk icw = {0}; bool do_work = false;
if (!udqp && !gdqp && !pdqp) return 0;
/* * Run a scan to free blocks using the union filter to cover all * applicable quotas in a single scan.
*/
icw.icw_flags = XFS_ICWALK_FLAG_UNION | iwalk_flags;
/* Run cow/eofblocks scans on the quotas attached to the inode. */ int
xfs_blockgc_free_quota( struct xfs_inode *ip, unsignedint iwalk_flags)
{ return xfs_blockgc_free_dquots(ip->i_mount,
xfs_inode_dquot(ip, XFS_DQTYPE_USER),
xfs_inode_dquot(ip, XFS_DQTYPE_GROUP),
xfs_inode_dquot(ip, XFS_DQTYPE_PROJ), iwalk_flags);
}
/* XFS Inode Cache Walking Code */
/* * The inode lookup is done in batches to keep the amount of lock traffic and * radix tree lookups to a minimum. The batch size is a trade off between * lookup reduction and stack usage. This is in the reclaim path, so we can't * be too greedy.
*/ #define XFS_LOOKUP_BATCH 32
/* * Decide if we want to grab this inode in anticipation of doing work towards * the goal.
*/ staticinlinebool
xfs_icwalk_igrab( enum xfs_icwalk_goal goal, struct xfs_inode *ip, struct xfs_icwalk *icw)
{ switch (goal) { case XFS_ICWALK_BLOCKGC: return xfs_blockgc_igrab(ip); case XFS_ICWALK_RECLAIM: return xfs_reclaim_igrab(ip, icw); default: returnfalse;
}
}
/* * Process an inode. Each processing function must handle any state changes * made by the icwalk igrab function. Return -EAGAIN to skip an inode.
*/ staticinlineint
xfs_icwalk_process_inode( enum xfs_icwalk_goal goal, struct xfs_inode *ip, struct xfs_perag *pag, struct xfs_icwalk *icw)
{ int error = 0;
switch (goal) { case XFS_ICWALK_BLOCKGC:
error = xfs_blockgc_scan_inode(ip, icw); break; case XFS_ICWALK_RECLAIM:
xfs_reclaim_inode(ip, pag); break;
} return error;
}
/* * For a given per-AG structure @pag and a goal, grab qualifying inodes and * process them in some manner.
*/ staticint
xfs_icwalk_ag( struct xfs_perag *pag, enum xfs_icwalk_goal goal, struct xfs_icwalk *icw)
{ struct xfs_mount *mp = pag_mount(pag);
uint32_t first_index; int last_error = 0; int skipped; bool done; int nr_found;
restart:
done = false;
skipped = 0; if (goal == XFS_ICWALK_RECLAIM)
first_index = READ_ONCE(pag->pag_ici_reclaim_cursor); else
first_index = 0;
nr_found = 0; do { struct xfs_inode *batch[XFS_LOOKUP_BATCH]; int error = 0; int i;
/* * Grab the inodes before we drop the lock. if we found * nothing, nr == 0 and the loop will be skipped.
*/ for (i = 0; i < nr_found; i++) { struct xfs_inode *ip = batch[i];
if (done || !xfs_icwalk_igrab(goal, ip, icw))
batch[i] = NULL;
/* * Update the index for the next lookup. Catch * overflows into the next AG range which can occur if * we have inodes in the last block of the AG and we * are currently pointing to the last inode. * * Because we may see inodes that are from the wrong AG * due to RCU freeing and reallocation, only update the * index if it lies in this AG. It was a race that lead * us to see this inode, so another lookup from the * same index will not find it again.
*/ if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag_agno(pag)) continue;
first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
done = true;
}
/* unlock now we've grabbed the inodes. */
rcu_read_unlock();
for (i = 0; i < nr_found; i++) { if (!batch[i]) continue;
error = xfs_icwalk_process_inode(goal, batch[i], pag,
icw); if (error == -EAGAIN) {
skipped++; continue;
} if (error && last_error != -EFSCORRUPTED)
last_error = error;
}
/* bail out if the filesystem is corrupted. */ if (error == -EFSCORRUPTED) break;
cond_resched();
if (icw && (icw->icw_flags & XFS_ICWALK_FLAG_SCAN_LIMIT)) {
icw->icw_scan_limit -= XFS_LOOKUP_BATCH; if (icw->icw_scan_limit <= 0) break;
}
} while (nr_found && !done);
if (goal == XFS_ICWALK_RECLAIM) { if (done)
first_index = 0;
WRITE_ONCE(pag->pag_ici_reclaim_cursor, first_index);
}
if (skipped) {
delay(1); goto restart;
} return last_error;
}
/* Walk all incore inodes to achieve a given goal. */ staticint
xfs_icwalk( struct xfs_mount *mp, enum xfs_icwalk_goal goal, struct xfs_icwalk *icw)
{ struct xfs_perag *pag = NULL; int error = 0; int last_error = 0;
/* * Free all speculative preallocations and possibly even the inode itself. * This is the last chance to make changes to an otherwise unreferenced file * before incore reclamation happens.
*/ staticint
xfs_inodegc_inactivate( struct xfs_inode *ip)
{ int error;
/* * Clear the cpu mask bit and ensure that we have seen the latest * update of the gc structure associated with this CPU. This matches * with the release semantics used when setting the cpumask bit in * xfs_inodegc_queue.
*/
cpumask_clear_cpu(gc->cpu, &mp->m_inodegc_cpumask);
smp_mb__after_atomic();
WRITE_ONCE(gc->items, 0);
if (!node) return;
/* * We can allocate memory here while doing writeback on behalf of * memory reclaim. To avoid memory allocation deadlocks set the * task-wide nofs context for the following operations.
*/
nofs_flag = memalloc_nofs_save();
ip = llist_entry(node, struct xfs_inode, i_gclist);
trace_xfs_inodegc_worker(mp, READ_ONCE(gc->shrinker_hits));
WRITE_ONCE(gc->shrinker_hits, 0);
llist_for_each_entry_safe(ip, n, node, i_gclist) { int error;
/* * Expedite all pending inodegc work to run immediately. This does not wait for * completion of the work.
*/ void
xfs_inodegc_push( struct xfs_mount *mp)
{ if (!xfs_is_inodegc_enabled(mp)) return;
trace_xfs_inodegc_push(mp, __return_address);
xfs_inodegc_queue_all(mp);
}
/* * Force all currently queued inode inactivation work to run immediately and * wait for the work to finish.
*/ int
xfs_inodegc_flush( struct xfs_mount *mp)
{
xfs_inodegc_push(mp);
trace_xfs_inodegc_flush(mp, __return_address); return xfs_inodegc_wait_all(mp);
}
/* * Flush all the pending work and then disable the inode inactivation background * workers and wait for them to stop. Caller must hold sb->s_umount to * coordinate changes in the inodegc_enabled state.
*/ void
xfs_inodegc_stop( struct xfs_mount *mp)
{ bool rerun;
if (!xfs_clear_inodegc_enabled(mp)) return;
/* * Drain all pending inodegc work, including inodes that could be * queued by racing xfs_inodegc_queue or xfs_inodegc_shrinker_scan * threads that sample the inodegc state just prior to us clearing it. * The inodegc flag state prevents new threads from queuing more * inodes, so we queue pending work items and flush the workqueue until * all inodegc lists are empty. IOWs, we cannot use drain_workqueue * here because it does not allow other unserialized mechanisms to * reschedule inodegc work while this draining is in progress.
*/
xfs_inodegc_queue_all(mp); do {
flush_workqueue(mp->m_inodegc_wq);
rerun = xfs_inodegc_queue_all(mp);
} while (rerun);
trace_xfs_inodegc_stop(mp, __return_address);
}
/* * Enable the inode inactivation background workers and schedule deferred inode * inactivation work if there is any. Caller must hold sb->s_umount to * coordinate changes in the inodegc_enabled state.
*/ void
xfs_inodegc_start( struct xfs_mount *mp)
{ if (xfs_set_inodegc_enabled(mp)) return;
/* * Schedule the inactivation worker when: * * - We've accumulated more than one inode cluster buffer's worth of inodes. * - There is less than 5% free space left. * - Any of the quotas for this inode are near an enforcement limit.
*/ staticinlinebool
xfs_inodegc_want_queue_work( struct xfs_inode *ip, unsignedint items)
{ struct xfs_mount *mp = ip->i_mount;
if (items > mp->m_ino_geo.inodes_per_cluster) returntrue;
if (xfs_compare_freecounter(mp, XC_FREE_BLOCKS,
mp->m_low_space[XFS_LOWSP_5_PCNT],
XFS_FDBLOCKS_BATCH) < 0) returntrue;
if (xfs_inodegc_want_queue_rt_file(ip)) returntrue;
if (xfs_inode_near_dquot_enforcement(ip, XFS_DQTYPE_USER)) returntrue;
if (xfs_inode_near_dquot_enforcement(ip, XFS_DQTYPE_GROUP)) returntrue;
if (xfs_inode_near_dquot_enforcement(ip, XFS_DQTYPE_PROJ)) returntrue;
returnfalse;
}
/* * Upper bound on the number of inodes in each AG that can be queued for * inactivation at any given time, to avoid monopolizing the workqueue.
*/ #define XFS_INODEGC_MAX_BACKLOG (4 * XFS_INODES_PER_CHUNK)
/* * Make the frontend wait for inactivations when: * * - Memory shrinkers queued the inactivation worker and it hasn't finished. * - The queue depth exceeds the maximum allowable percpu backlog. * * Note: If we are in a NOFS context here (e.g. current thread is running a * transaction) the we don't want to block here as inodegc progress may require * filesystem resources we hold to make progress and that could result in a * deadlock. Hence we skip out of here if we are in a scoped NOFS context.
*/ staticinlinebool
xfs_inodegc_want_flush_work( struct xfs_inode *ip, unsignedint items, unsignedint shrinker_hits)
{ if (current->flags & PF_MEMALLOC_NOFS) returnfalse;
if (shrinker_hits > 0) returntrue;
if (items > XFS_INODEGC_MAX_BACKLOG) returntrue;
returnfalse;
}
/* * Queue a background inactivation worker if there are inodes that need to be * inactivated and higher level xfs code hasn't disabled the background * workers.
*/ staticvoid
xfs_inodegc_queue( struct xfs_inode *ip)
{ struct xfs_mount *mp = ip->i_mount; struct xfs_inodegc *gc; int items; unsignedint shrinker_hits; unsignedint cpu_nr; unsignedlong queue_delay = 1;
/* * Ensure the list add is always seen by anyone who finds the cpumask * bit set. This effectively gives the cpumask bit set operation * release ordering semantics.
*/
smp_mb__before_atomic(); if (!cpumask_test_cpu(cpu_nr, &mp->m_inodegc_cpumask))
cpumask_test_and_set_cpu(cpu_nr, &mp->m_inodegc_cpumask);
/* * We queue the work while holding the current CPU so that the work * is scheduled to run on this CPU.
*/ if (!xfs_is_inodegc_enabled(mp)) {
put_cpu(); return;
}
if (xfs_inodegc_want_queue_work(ip, items))
queue_delay = 0;
if (xfs_inodegc_want_flush_work(ip, items, shrinker_hits)) {
trace_xfs_inodegc_throttle(mp, __return_address);
flush_delayed_work(&gc->work);
}
}
/* * We set the inode flag atomically with the radix tree tag. Once we get tag * lookups on the radix tree, this inode flag can go away. * * We always use background reclaim here because even if the inode is clean, it * still may be under IO and hence we have wait for IO completion to occur * before we can reclaim the inode. The background reclaim path handles this * more efficiently than we can here, so simply let background reclaim tear down * all inodes.
*/ void
xfs_inode_mark_reclaimable( struct xfs_inode *ip)
{ struct xfs_mount *mp = ip->i_mount; bool need_inactive;
XFS_STATS_INC(mp, vn_reclaim);
/* * We should never get here with any of the reclaim flags already set.
*/
ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_ALL_IRECLAIM_FLAGS));
need_inactive = xfs_inode_needs_inactive(ip); if (need_inactive) {
xfs_inodegc_queue(ip); return;
}
/* Going straight to reclaim, so drop the dquots. */
xfs_qm_dqdetach(ip);
xfs_inodegc_set_reclaimable(ip);
}
/* * Register a phony shrinker so that we can run background inodegc sooner when * there's memory pressure. Inactivation does not itself free any memory but * it does make inodes reclaimable, which eventually frees memory. * * The count function, seek value, and batch value are crafted to trigger the * scan function during the second round of scanning. Hopefully this means * that we reclaimed enough memory that initiating metadata transactions won't * make things worse.
*/ #define XFS_INODEGC_SHRINKER_COUNT (1UL << DEF_PRIORITY) #define XFS_INODEGC_SHRINKER_BATCH ((XFS_INODEGC_SHRINKER_COUNT / 2) + 1)
/* * If there are no inodes to inactivate, we don't want the shrinker * to think there's deferred work to call us back about.
*/ if (no_items) return LONG_MAX;
return SHRINK_STOP;
}
/* Register a shrinker so we can accelerate inodegc and throttle queuing. */ int
xfs_inodegc_register_shrinker( struct xfs_mount *mp)
{
mp->m_inodegc_shrinker = shrinker_alloc(SHRINKER_NONSLAB, "xfs-inodegc:%s",
mp->m_super->s_id); if (!mp->m_inodegc_shrinker) return -ENOMEM;
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.