summaryrefslogtreecommitdiff
path: root/fs/ocfs2/quota_global.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ocfs2/quota_global.c')
-rw-r--r--fs/ocfs2/quota_global.c226
1 files changed, 167 insertions, 59 deletions
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index 332a281f217e..e85b1ccf81be 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
/*
* Implementation of operations over global quota file
*/
@@ -10,6 +11,8 @@
#include <linux/jiffies.h>
#include <linux/writeback.h>
#include <linux/workqueue.h>
+#include <linux/llist.h>
+#include <linux/iversion.h>
#include <cluster/masklog.h>
@@ -32,8 +35,8 @@
* Locking of quotas with OCFS2 is rather complex. Here are rules that
* should be obeyed by all the functions:
* - any write of quota structure (either to local or global file) is protected
- * by dqio_mutex or dquot->dq_lock.
- * - any modification of global quota file holds inode cluster lock, i_mutex,
+ * by dqio_sem or dquot->dq_lock.
+ * - any modification of global quota file holds inode cluster lock, i_rwsem,
* and ip_alloc_sem of the global quota file (achieved by
* ocfs2_lock_global_qf). It also has to hold qinfo_lock.
* - an allocation of new blocks for local quota file is protected by
@@ -41,9 +44,9 @@
*
* A rough sketch of locking dependencies (lf = local file, gf = global file):
* Normal filesystem operation:
- * start_trans -> dqio_mutex -> write to lf
+ * start_trans -> dqio_sem -> write to lf
* Syncing of local and global file:
- * ocfs2_lock_global_qf -> start_trans -> dqio_mutex -> qinfo_lock ->
+ * ocfs2_lock_global_qf -> start_trans -> dqio_sem -> qinfo_lock ->
* write to gf
* -> write to lf
* Acquire dquot for the first time:
@@ -59,7 +62,7 @@
* Recovery:
* inode cluster lock of recovered lf
* -> read bitmaps -> ip_alloc_sem of lf
- * -> ocfs2_lock_global_qf -> start_trans -> dqio_mutex -> qinfo_lock ->
+ * -> ocfs2_lock_global_qf -> start_trans -> dqio_sem -> qinfo_lock ->
* write to gf
*/
@@ -122,7 +125,7 @@ static int ocfs2_global_is_id(void *dp, struct dquot *dquot)
dquot->dq_id);
}
-struct qtree_fmt_operations ocfs2_global_ops = {
+const struct qtree_fmt_operations ocfs2_global_ops = {
.mem2disk_dqblk = ocfs2_global_mem2diskdqb,
.disk2mem_dqblk = ocfs2_global_disk2memdqb,
.is_id = ocfs2_global_is_id,
@@ -234,7 +237,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
len = sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE - offset;
}
- if (gqinode->i_size < off + len) {
+ if (i_size_read(gqinode) < off + len) {
loff_t rounded_end =
ocfs2_align_bytes_to_blocks(sb, off + len);
@@ -270,7 +273,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
if (new)
memset(bh->b_data, 0, sb->s_blocksize);
memcpy(bh->b_data + offset, data, len);
- flush_dcache_page(bh->b_page);
+ flush_dcache_folio(bh->b_folio);
set_buffer_uptodate(bh);
unlock_buffer(bh);
ocfs2_set_buffer_uptodate(INODE_CACHE(gqinode), bh);
@@ -287,7 +290,7 @@ out:
mlog_errno(err);
return err;
}
- gqinode->i_version++;
+ inode_inc_iversion(gqinode);
ocfs2_mark_inode_dirty(handle, gqinode, oinfo->dqi_gqi_bh);
return len;
}
@@ -307,7 +310,7 @@ int ocfs2_lock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex)
WARN_ON(bh != oinfo->dqi_gqi_bh);
spin_unlock(&dq_data_lock);
if (ex) {
- mutex_lock(&oinfo->dqi_gqinode->i_mutex);
+ inode_lock(oinfo->dqi_gqinode);
down_write(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
} else {
down_read(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
@@ -319,7 +322,7 @@ void ocfs2_unlock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex)
{
if (ex) {
up_write(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
- mutex_unlock(&oinfo->dqi_gqinode->i_mutex);
+ inode_unlock(oinfo->dqi_gqinode);
} else {
up_read(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
}
@@ -334,24 +337,14 @@ void ocfs2_unlock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex)
/* Read information header from global quota file */
int ocfs2_global_read_info(struct super_block *sb, int type)
{
- struct inode *gqinode = NULL;
- unsigned int ino[MAXQUOTAS] = { USER_QUOTA_SYSTEM_INODE,
- GROUP_QUOTA_SYSTEM_INODE };
+ unsigned int ino[OCFS2_MAXQUOTAS] = { USER_QUOTA_SYSTEM_INODE,
+ GROUP_QUOTA_SYSTEM_INODE };
struct ocfs2_global_disk_dqinfo dinfo;
struct mem_dqinfo *info = sb_dqinfo(sb, type);
struct ocfs2_mem_dqinfo *oinfo = info->dqi_priv;
u64 pcount;
int status;
- /* Read global header */
- gqinode = ocfs2_get_system_file_inode(OCFS2_SB(sb), ino[type],
- OCFS2_INVALID_SLOT);
- if (!gqinode) {
- mlog(ML_ERROR, "failed to get global quota inode (type=%d)\n",
- type);
- status = -EINVAL;
- goto out_err;
- }
oinfo->dqi_gi.dqi_sb = sb;
oinfo->dqi_gi.dqi_type = type;
ocfs2_qinfo_lock_res_init(&oinfo->dqi_gqlock, oinfo);
@@ -359,21 +352,35 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
oinfo->dqi_gi.dqi_ops = &ocfs2_global_ops;
oinfo->dqi_gqi_bh = NULL;
oinfo->dqi_gqi_count = 0;
- oinfo->dqi_gqinode = gqinode;
+
+ /* Read global header */
+ oinfo->dqi_gqinode = ocfs2_get_system_file_inode(OCFS2_SB(sb), ino[type],
+ OCFS2_INVALID_SLOT);
+ if (!oinfo->dqi_gqinode) {
+ mlog(ML_ERROR, "failed to get global quota inode (type=%d)\n",
+ type);
+ status = -EINVAL;
+ goto out_err;
+ }
+
status = ocfs2_lock_global_qf(oinfo, 0);
if (status < 0) {
mlog_errno(status);
goto out_err;
}
- status = ocfs2_extent_map_get_blocks(gqinode, 0, &oinfo->dqi_giblk,
+ status = ocfs2_extent_map_get_blocks(oinfo->dqi_gqinode, 0, &oinfo->dqi_giblk,
&pcount, NULL);
- if (status < 0)
+ if (status < 0) {
+ mlog_errno(status);
goto out_unlock;
+ }
status = ocfs2_qinfo_lock(oinfo, 0);
- if (status < 0)
+ if (status < 0) {
+ mlog_errno(status);
goto out_unlock;
+ }
status = sb->s_op->quota_read(sb, type, (char *)&dinfo,
sizeof(struct ocfs2_global_disk_dqinfo),
OCFS2_GLOBAL_INFO_OFF);
@@ -401,15 +408,14 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
schedule_delayed_work(&oinfo->dqi_sync_work,
msecs_to_jiffies(oinfo->dqi_syncms));
-out_err:
- return status;
+ return 0;
out_unlock:
ocfs2_unlock_global_qf(oinfo, 0);
- mlog_errno(status);
- goto out_err;
+out_err:
+ return status;
}
-/* Write information to global quota file. Expects exlusive lock on quota
+/* Write information to global quota file. Expects exclusive lock on quota
* file inode and quota info */
static int __ocfs2_global_write_info(struct super_block *sb, int type)
{
@@ -442,13 +448,20 @@ static int __ocfs2_global_write_info(struct super_block *sb, int type)
int ocfs2_global_write_info(struct super_block *sb, int type)
{
int err;
- struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
+ struct quota_info *dqopt = sb_dqopt(sb);
+ struct ocfs2_mem_dqinfo *info = dqopt->info[type].dqi_priv;
+ unsigned int memalloc;
+ down_write(&dqopt->dqio_sem);
+ memalloc = memalloc_nofs_save();
err = ocfs2_qinfo_lock(info, 1);
if (err < 0)
- return err;
+ goto out_sem;
err = __ocfs2_global_write_info(sb, type);
ocfs2_qinfo_unlock(info, 1);
+out_sem:
+ memalloc_nofs_restore(memalloc);
+ up_write(&dqopt->dqio_sem);
return err;
}
@@ -482,7 +495,7 @@ int __ocfs2_sync_dquot(struct dquot *dquot, int freeing)
struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
struct ocfs2_global_disk_dqblk dqblk;
s64 spacechange, inodechange;
- time_t olditime, oldbtime;
+ time64_t olditime, oldbtime;
err = sb->s_op->quota_read(sb, type, (char *)&dqblk,
sizeof(struct ocfs2_global_disk_dqblk),
@@ -499,7 +512,7 @@ int __ocfs2_sync_dquot(struct dquot *dquot, int freeing)
/* Update space and inode usage. Get also other information from
* global quota file so that we don't overwrite any changes there.
* We are */
- spin_lock(&dq_data_lock);
+ spin_lock(&dquot->dq_dqb_lock);
spacechange = dquot->dq_dqb.dqb_curspace -
OCFS2_DQUOT(dquot)->dq_origspace;
inodechange = dquot->dq_dqb.dqb_curinodes -
@@ -555,7 +568,7 @@ int __ocfs2_sync_dquot(struct dquot *dquot, int freeing)
__clear_bit(DQ_LASTSET_B + QIF_ITIME_B, &dquot->dq_flags);
OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace;
OCFS2_DQUOT(dquot)->dq_originodes = dquot->dq_dqb.dqb_curinodes;
- spin_unlock(&dq_data_lock);
+ spin_unlock(&dquot->dq_dqb_lock);
err = ocfs2_qinfo_lock(info, freeing);
if (err < 0) {
mlog(ML_ERROR, "Failed to lock quota info, losing quota write"
@@ -594,6 +607,7 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv;
struct ocfs2_super *osb = OCFS2_SB(sb);
int status = 0;
+ unsigned int memalloc;
trace_ocfs2_sync_dquot_helper(from_kqid(&init_user_ns, dquot->dq_id),
dquot->dq_id.type,
@@ -610,7 +624,8 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
mlog_errno(status);
goto out_ilock;
}
- mutex_lock(&sb_dqopt(sb)->dqio_mutex);
+ down_write(&sb_dqopt(sb)->dqio_sem);
+ memalloc = memalloc_nofs_save();
status = ocfs2_sync_dquot(dquot);
if (status < 0)
mlog_errno(status);
@@ -618,7 +633,8 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
status = ocfs2_local_write_dquot(dquot);
if (status < 0)
mlog_errno(status);
- mutex_unlock(&sb_dqopt(sb)->dqio_mutex);
+ memalloc_nofs_restore(memalloc);
+ up_write(&sb_dqopt(sb)->dqio_sem);
ocfs2_commit_trans(osb, handle);
out_ilock:
ocfs2_unlock_global_qf(oinfo, 1);
@@ -633,7 +649,15 @@ static void qsync_work_fn(struct work_struct *work)
dqi_sync_work.work);
struct super_block *sb = oinfo->dqi_gqinode->i_sb;
- dquot_scan_active(sb, ocfs2_sync_dquot_helper, oinfo->dqi_type);
+ /*
+ * We have to be careful here not to deadlock on s_umount as umount
+ * disabling quotas may be in progress and it waits for this work to
+ * complete. If trylock fails, we'll do the sync next time...
+ */
+ if (down_read_trylock(&sb->s_umount)) {
+ dquot_scan_active(sb, ocfs2_sync_dquot_helper, oinfo->dqi_type);
+ up_read(&sb->s_umount);
+ }
schedule_delayed_work(&oinfo->dqi_sync_work,
msecs_to_jiffies(oinfo->dqi_syncms));
}
@@ -647,6 +671,7 @@ static int ocfs2_write_dquot(struct dquot *dquot)
handle_t *handle;
struct ocfs2_super *osb = OCFS2_SB(dquot->dq_sb);
int status = 0;
+ unsigned int memalloc;
trace_ocfs2_write_dquot(from_kqid(&init_user_ns, dquot->dq_id),
dquot->dq_id.type);
@@ -657,9 +682,11 @@ static int ocfs2_write_dquot(struct dquot *dquot)
mlog_errno(status);
goto out;
}
- mutex_lock(&sb_dqopt(dquot->dq_sb)->dqio_mutex);
+ down_write(&sb_dqopt(dquot->dq_sb)->dqio_sem);
+ memalloc = memalloc_nofs_save();
status = ocfs2_local_write_dquot(dquot);
- mutex_unlock(&sb_dqopt(dquot->dq_sb)->dqio_mutex);
+ memalloc_nofs_restore(memalloc);
+ up_write(&sb_dqopt(dquot->dq_sb)->dqio_sem);
ocfs2_commit_trans(osb, handle);
out:
return status;
@@ -679,6 +706,27 @@ static int ocfs2_calc_qdel_credits(struct super_block *sb, int type)
OCFS2_INODE_UPDATE_CREDITS;
}
+void ocfs2_drop_dquot_refs(struct work_struct *work)
+{
+ struct ocfs2_super *osb = container_of(work, struct ocfs2_super,
+ dquot_drop_work);
+ struct llist_node *list;
+ struct ocfs2_dquot *odquot, *next_odquot;
+
+ list = llist_del_all(&osb->dquot_drop_list);
+ llist_for_each_entry_safe(odquot, next_odquot, list, list) {
+ /* Drop the reference we acquired in ocfs2_dquot_release() */
+ dqput(&odquot->dq_dquot);
+ }
+}
+
+/*
+ * Called when the last reference to dquot is dropped. If we are called from
+ * downconvert thread, we cannot do all the handling here because grabbing
+ * quota lock could deadlock (the node holding the quota lock could need some
+ * other cluster lock to proceed but with blocked downconvert thread we cannot
+ * release any lock).
+ */
static int ocfs2_release_dquot(struct dquot *dquot)
{
handle_t *handle;
@@ -692,14 +740,32 @@ static int ocfs2_release_dquot(struct dquot *dquot)
mutex_lock(&dquot->dq_lock);
/* Check whether we are not racing with some other dqget() */
- if (atomic_read(&dquot->dq_count) > 1)
+ if (dquot_is_busy(dquot))
+ goto out;
+ /* Running from downconvert thread? Postpone quota processing to wq */
+ if (current == osb->dc_task) {
+ /*
+ * Grab our own reference to dquot and queue it for delayed
+ * dropping. Quota code rechecks after calling
+ * ->release_dquot() and won't free dquot structure.
+ */
+ dqgrab(dquot);
+ /* First entry on list -> queue work */
+ if (llist_add(&OCFS2_DQUOT(dquot)->list, &osb->dquot_drop_list))
+ queue_work(osb->ocfs2_wq, &osb->dquot_drop_work);
goto out;
+ }
status = ocfs2_lock_global_qf(oinfo, 1);
if (status < 0)
goto out;
handle = ocfs2_start_trans(osb,
ocfs2_calc_qdel_credits(dquot->dq_sb, dquot->dq_id.type));
if (IS_ERR(handle)) {
+ /*
+ * Mark dquot as inactive to avoid endless cycle in
+ * quota_release_workfn().
+ */
+ clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
status = PTR_ERR(handle);
mlog_errno(status);
goto out_ilock;
@@ -717,6 +783,12 @@ static int ocfs2_release_dquot(struct dquot *dquot)
*/
if (status < 0)
mlog_errno(status);
+ /*
+ * Clear dq_off so that we search for the structure in quota file next
+ * time we acquire it. The structure might be deleted and reallocated
+ * elsewhere by another node while our dquot structure is on freelist.
+ */
+ dquot->dq_off = 0;
clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
out_trans:
ocfs2_commit_trans(osb, handle);
@@ -756,16 +828,17 @@ static int ocfs2_acquire_dquot(struct dquot *dquot)
status = ocfs2_lock_global_qf(info, 1);
if (status < 0)
goto out;
- if (!test_bit(DQ_READ_B, &dquot->dq_flags)) {
- status = ocfs2_qinfo_lock(info, 0);
- if (status < 0)
- goto out_dq;
- status = qtree_read_dquot(&info->dqi_gi, dquot);
- ocfs2_qinfo_unlock(info, 0);
- if (status < 0)
- goto out_dq;
- }
- set_bit(DQ_READ_B, &dquot->dq_flags);
+ status = ocfs2_qinfo_lock(info, 0);
+ if (status < 0)
+ goto out_dq;
+ /*
+ * We always want to read dquot structure from disk because we don't
+ * know what happened with it while it was on freelist.
+ */
+ status = qtree_read_dquot(&info->dqi_gi, dquot);
+ ocfs2_qinfo_unlock(info, 0);
+ if (status < 0)
+ goto out_dq;
OCFS2_DQUOT(dquot)->dq_use_count++;
OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace;
@@ -778,8 +851,8 @@ static int ocfs2_acquire_dquot(struct dquot *dquot)
*/
WARN_ON(journal_current_handle());
status = ocfs2_extend_no_holes(gqinode, NULL,
- gqinode->i_size + (need_alloc << sb->s_blocksize_bits),
- gqinode->i_size);
+ i_size_read(gqinode) + (need_alloc << sb->s_blocksize_bits),
+ i_size_read(gqinode));
if (status < 0)
goto out_dq;
}
@@ -818,6 +891,37 @@ out:
return status;
}
+static int ocfs2_get_next_id(struct super_block *sb, struct kqid *qid)
+{
+ int type = qid->type;
+ struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
+ int status = 0;
+
+ trace_ocfs2_get_next_id(from_kqid(&init_user_ns, *qid), type);
+ if (!sb_has_quota_active(sb, type)) {
+ status = -ESRCH;
+ goto out;
+ }
+ status = ocfs2_lock_global_qf(info, 0);
+ if (status < 0)
+ goto out;
+ status = ocfs2_qinfo_lock(info, 0);
+ if (status < 0)
+ goto out_global;
+ status = qtree_get_next_id(&info->dqi_gi, qid);
+ ocfs2_qinfo_unlock(info, 0);
+out_global:
+ ocfs2_unlock_global_qf(info, 0);
+out:
+ /*
+ * Avoid logging ENOENT since it just means there isn't next ID and
+ * ESRCH which means quota isn't enabled for the filesystem.
+ */
+ if (status && status != -ENOENT && status != -ESRCH)
+ mlog_errno(status);
+ return status;
+}
+
static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
{
unsigned long mask = (1 << (DQ_LASTSET_B + QIF_ILIMITS_B)) |
@@ -833,16 +937,17 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv;
handle_t *handle;
struct ocfs2_super *osb = OCFS2_SB(sb);
+ unsigned int memalloc;
trace_ocfs2_mark_dquot_dirty(from_kqid(&init_user_ns, dquot->dq_id),
type);
/* In case user set some limits, sync dquot immediately to global
* quota file so that information propagates quicker */
- spin_lock(&dq_data_lock);
+ spin_lock(&dquot->dq_dqb_lock);
if (dquot->dq_flags & mask)
sync = 1;
- spin_unlock(&dq_data_lock);
+ spin_unlock(&dquot->dq_dqb_lock);
/* This is a slight hack but we can't afford getting global quota
* lock if we already have a transaction started. */
if (!sync || journal_current_handle()) {
@@ -858,7 +963,8 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
mlog_errno(status);
goto out_ilock;
}
- mutex_lock(&sb_dqopt(sb)->dqio_mutex);
+ down_write(&sb_dqopt(sb)->dqio_sem);
+ memalloc = memalloc_nofs_save();
status = ocfs2_sync_dquot(dquot);
if (status < 0) {
mlog_errno(status);
@@ -867,7 +973,8 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
/* Now write updated local dquot structure */
status = ocfs2_local_write_dquot(dquot);
out_dlock:
- mutex_unlock(&sb_dqopt(sb)->dqio_mutex);
+ memalloc_nofs_restore(memalloc);
+ up_write(&sb_dqopt(sb)->dqio_sem);
ocfs2_commit_trans(osb, handle);
out_ilock:
ocfs2_unlock_global_qf(oinfo, 1);
@@ -926,4 +1033,5 @@ const struct dquot_operations ocfs2_quota_operations = {
.write_info = ocfs2_write_info,
.alloc_dquot = ocfs2_alloc_dquot,
.destroy_dquot = ocfs2_destroy_dquot,
+ .get_next_id = ocfs2_get_next_id,
};
generated by cgit (git 2.34.1) at 2025-12-25 14:43:28 +0000