1 files changed, 205 insertions, 125 deletions

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 3fb98b4569a2..85239807dec7 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -15,6 +15,7 @@
 #include <linux/time.h>
 #include <linux/random.h>
 #include <linux/delay.h>
+#include <linux/writeback.h>
 
 #include <cluster/masklog.h>
 
@@ -89,7 +90,7 @@ enum ocfs2_replay_state {
 struct ocfs2_replay_map {
 	unsigned int rm_slots;
 	enum ocfs2_replay_state rm_state;
-	unsigned char rm_replay_slots[];
+	unsigned char rm_replay_slots[] __counted_by(rm_slots);
 };
 
 static void ocfs2_replay_map_set_state(struct ocfs2_super *osb, int state)
@@ -113,9 +114,9 @@ int ocfs2_compute_replay_slots(struct ocfs2_super *osb)
 	if (osb->replay_map)
 		return 0;
 
-	replay_map = kzalloc(sizeof(struct ocfs2_replay_map) +
-			     (osb->max_slots * sizeof(char)), GFP_KERNEL);
-
+	replay_map = kzalloc(struct_size(replay_map, rm_replay_slots,
+					 osb->max_slots),
+			     GFP_KERNEL);
 	if (!replay_map) {
 		mlog_errno(-ENOMEM);
 		return -ENOMEM;
@@ -173,50 +174,69 @@ int ocfs2_recovery_init(struct ocfs2_super *osb)
 	struct ocfs2_recovery_map *rm;
 
 	mutex_init(&osb->recovery_lock);
-	osb->disable_recovery = 0;
+	osb->recovery_state = OCFS2_REC_ENABLED;
 	osb->recovery_thread_task = NULL;
 	init_waitqueue_head(&osb->recovery_event);
 
-	rm = kzalloc(sizeof(struct ocfs2_recovery_map) +
-		     osb->max_slots * sizeof(unsigned int),
+	rm = kzalloc(struct_size(rm, rm_entries, osb->max_slots),
 		     GFP_KERNEL);
 	if (!rm) {
 		mlog_errno(-ENOMEM);
 		return -ENOMEM;
 	}
 
-	rm->rm_entries = (unsigned int *)((char *)rm +
-					  sizeof(struct ocfs2_recovery_map));
 	osb->recovery_map = rm;
 
 	return 0;
 }
 
-/* we can't grab the goofy sem lock from inside wait_event, so we use
- * memory barriers to make sure that we'll see the null task before
- * being woken up */
 static int ocfs2_recovery_thread_running(struct ocfs2_super *osb)
 {
-	mb();
 	return osb->recovery_thread_task != NULL;
 }
 
-void ocfs2_recovery_exit(struct ocfs2_super *osb)
+static void ocfs2_recovery_disable(struct ocfs2_super *osb,
+				   enum ocfs2_recovery_state state)
 {
-	struct ocfs2_recovery_map *rm;
-
-	/* disable any new recovery threads and wait for any currently
-	 * running ones to exit. Do this before setting the vol_state. */
 	mutex_lock(&osb->recovery_lock);
-	osb->disable_recovery = 1;
+	/*
+	 * If recovery thread is not running, we can directly transition to
+	 * final state.
+	 */
+	if (!ocfs2_recovery_thread_running(osb)) {
+		osb->recovery_state = state + 1;
+		goto out_lock;
+	}
+	osb->recovery_state = state;
+	/* Wait for recovery thread to acknowledge state transition */
+	wait_event_cmd(osb->recovery_event,
+		       !ocfs2_recovery_thread_running(osb) ||
+				osb->recovery_state >= state + 1,
+		       mutex_unlock(&osb->recovery_lock),
+		       mutex_lock(&osb->recovery_lock));
+out_lock:
 	mutex_unlock(&osb->recovery_lock);
-	wait_event(osb->recovery_event, !ocfs2_recovery_thread_running(osb));
 
-	/* At this point, we know that no more recovery threads can be
-	 * launched, so wait for any recovery completion work to
-	 * complete. */
+	/*
+	 * At this point we know that no more recovery work can be queued so
+	 * wait for any recovery completion work to complete.
+	 */
 	if (osb->ocfs2_wq)
 		flush_workqueue(osb->ocfs2_wq);
+}
+
+void ocfs2_recovery_disable_quota(struct ocfs2_super *osb)
+{
+	ocfs2_recovery_disable(osb, OCFS2_REC_QUOTA_WANT_DISABLE);
+}
+
+void ocfs2_recovery_exit(struct ocfs2_super *osb)
+{
+	struct ocfs2_recovery_map *rm;
+
+	/* disable any new recovery threads and wait for any currently
+	 * running ones to exit. Do this before setting the vol_state. */
+	ocfs2_recovery_disable(osb, OCFS2_REC_WANT_DISABLE);
 
 	/*
 	 * Now that recovery is shut down, and the osb is about to be
@@ -448,6 +468,23 @@ bail:
 }
 
 /*
+ * Make sure handle has at least 'nblocks' credits available. If it does not
+ * have that many credits available, we will try to extend the handle to have
+ * enough credits. If that fails, we will restart transaction to have enough
+ * credits. Similar notes regarding data consistency and locking implications
+ * as for ocfs2_extend_trans() apply here.
+ */
+int ocfs2_assure_trans_credits(handle_t *handle, int nblocks)
+{
+	int old_nblks = jbd2_handle_buffer_credits(handle);
+
+	trace_ocfs2_assure_trans_credits(old_nblks);
+	if (old_nblks >= nblocks)
+		return 0;
+	return ocfs2_extend_trans(handle, nblocks - old_nblks);
+}
+
+/*
  * If we have fewer than thresh credits, extend by OCFS2_MAX_TRANS_DATA.
  * If that fails, restart the transaction & regain write access for the
  * buffer head which is used for metadata modifications.
@@ -481,12 +518,6 @@ bail:
 	return status;
 }
 
-
-struct ocfs2_triggers {
-	struct jbd2_buffer_trigger_type	ot_triggers;
-	int				ot_offset;
-};
-
 static inline struct ocfs2_triggers *to_ocfs2_trigger(struct jbd2_buffer_trigger_type *triggers)
 {
 	return container_of(triggers, struct ocfs2_triggers, ot_triggers);
@@ -550,85 +581,76 @@ static void ocfs2_db_frozen_trigger(struct jbd2_buffer_trigger_type *triggers,
 static void ocfs2_abort_trigger(struct jbd2_buffer_trigger_type *triggers,
 				struct buffer_head *bh)
 {
+	struct ocfs2_triggers *ot = to_ocfs2_trigger(triggers);
+
 	mlog(ML_ERROR,
 	     "ocfs2_abort_trigger called by JBD2.  bh = 0x%lx, "
 	     "bh->b_blocknr = %llu\n",
 	     (unsigned long)bh,
 	     (unsigned long long)bh->b_blocknr);
 
-	ocfs2_error(bh->b_bdev->bd_super,
+	ocfs2_error(ot->sb,
 		    "JBD2 has aborted our journal, ocfs2 cannot continue\n");
 }
 
-static struct ocfs2_triggers di_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_dinode, i_check),
-};
-
-static struct ocfs2_triggers eb_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_extent_block, h_check),
-};
-
-static struct ocfs2_triggers rb_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_refcount_block, rf_check),
-};
-
-static struct ocfs2_triggers gd_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_group_desc, bg_check),
-};
-
-static struct ocfs2_triggers db_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_db_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-};
+static void ocfs2_setup_csum_triggers(struct super_block *sb,
+				      enum ocfs2_journal_trigger_type type,
+				      struct ocfs2_triggers *ot)
+{
+	BUG_ON(type >= OCFS2_JOURNAL_TRIGGER_COUNT);
 
-static struct ocfs2_triggers xb_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_xattr_block, xb_check),
-};
+	switch (type) {
+	case OCFS2_JTR_DI:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_dinode, i_check);
+		break;
+	case OCFS2_JTR_EB:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_extent_block, h_check);
+		break;
+	case OCFS2_JTR_RB:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_refcount_block, rf_check);
+		break;
+	case OCFS2_JTR_GD:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_group_desc, bg_check);
+		break;
+	case OCFS2_JTR_DB:
+		ot->ot_triggers.t_frozen = ocfs2_db_frozen_trigger;
+		break;
+	case OCFS2_JTR_XB:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_xattr_block, xb_check);
+		break;
+	case OCFS2_JTR_DQ:
+		ot->ot_triggers.t_frozen = ocfs2_dq_frozen_trigger;
+		break;
+	case OCFS2_JTR_DR:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_dx_root_block, dr_check);
+		break;
+	case OCFS2_JTR_DL:
+		ot->ot_triggers.t_frozen = ocfs2_frozen_trigger;
+		ot->ot_offset = offsetof(struct ocfs2_dx_leaf, dl_check);
+		break;
+	case OCFS2_JTR_NONE:
+		/* To make compiler happy... */
+		return;
+	}
 
-static struct ocfs2_triggers dq_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_dq_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-};
+	ot->ot_triggers.t_abort = ocfs2_abort_trigger;
+	ot->sb = sb;
+}
 
-static struct ocfs2_triggers dr_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_dx_root_block, dr_check),
-};
+void ocfs2_initialize_journal_triggers(struct super_block *sb,
+				       struct ocfs2_triggers triggers[])
+{
+	enum ocfs2_journal_trigger_type type;
 
-static struct ocfs2_triggers dl_triggers = {
-	.ot_triggers = {
-		.t_frozen = ocfs2_frozen_trigger,
-		.t_abort = ocfs2_abort_trigger,
-	},
-	.ot_offset	= offsetof(struct ocfs2_dx_leaf, dl_check),
-};
+	for (type = OCFS2_JTR_DI; type < OCFS2_JOURNAL_TRIGGER_COUNT; type++)
+		ocfs2_setup_csum_triggers(sb, type, &triggers[type]);
+}
 
 static int __ocfs2_journal_access(handle_t *handle,
 				  struct ocfs2_caching_info *ci,
@@ -710,56 +732,91 @@ static int __ocfs2_journal_access(handle_t *handle,
 int ocfs2_journal_access_di(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &di_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				      &osb->s_journal_triggers[OCFS2_JTR_DI],
+				      type);
 }
 
 int ocfs2_journal_access_eb(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &eb_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				      &osb->s_journal_triggers[OCFS2_JTR_EB],
+				      type);
 }
 
 int ocfs2_journal_access_rb(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &rb_triggers,
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				      &osb->s_journal_triggers[OCFS2_JTR_RB],
 				      type);
 }
 
 int ocfs2_journal_access_gd(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &gd_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_GD],
+				     type);
 }
 
 int ocfs2_journal_access_db(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &db_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_DB],
+				     type);
 }
 
 int ocfs2_journal_access_xb(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &xb_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_XB],
+				     type);
 }
 
 int ocfs2_journal_access_dq(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &dq_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_DQ],
+				     type);
 }
 
 int ocfs2_journal_access_dr(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &dr_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_DR],
+				     type);
 }
 
 int ocfs2_journal_access_dl(handle_t *handle, struct ocfs2_caching_info *ci,
 			    struct buffer_head *bh, int type)
 {
-	return __ocfs2_journal_access(handle, ci, bh, &dl_triggers, type);
+	struct ocfs2_super *osb = OCFS2_SB(ocfs2_metadata_cache_get_super(ci));
+
+	return __ocfs2_journal_access(handle, ci, bh,
+				     &osb->s_journal_triggers[OCFS2_JTR_DL],
+				     type);
 }
 
 int ocfs2_journal_access(handle_t *handle, struct ocfs2_caching_info *ci,
@@ -779,14 +836,16 @@ void ocfs2_journal_dirty(handle_t *handle, struct buffer_head *bh)
 		mlog_errno(status);
 		if (!is_handle_aborted(handle)) {
 			journal_t *journal = handle->h_transaction->t_journal;
-			struct super_block *sb = bh->b_bdev->bd_super;
 
-			mlog(ML_ERROR, "jbd2_journal_dirty_metadata failed. "
-					"Aborting transaction and journal.\n");
+			mlog(ML_ERROR, "jbd2_journal_dirty_metadata failed: "
+			     "handle type %u started at line %u, credits %u/%u "
+			     "errcode %d. Aborting transaction and journal.\n",
+			     handle->h_type, handle->h_line_no,
+			     handle->h_requested_credits,
+			     jbd2_handle_buffer_credits(handle), status);
 			handle->h_err = status;
 			jbd2_journal_abort_handle(handle);
 			jbd2_journal_abort(journal, status);
-			ocfs2_abort(sb, "Journal already aborted.\n");
 		}
 	}
 }
@@ -841,6 +900,12 @@ bail:
 	return status;
 }
 
+static int ocfs2_journal_submit_inode_data_buffers(struct jbd2_inode *jinode)
+{
+	return filemap_fdatawrite_range(jinode->i_vfs_inode->i_mapping,
+			jinode->i_dirty_start, jinode->i_dirty_end);
+}
+
 int ocfs2_journal_init(struct ocfs2_super *osb, int *dirty)
 {
 	int status = -1;
@@ -897,9 +962,9 @@ int ocfs2_journal_init(struct ocfs2_super *osb, int *dirty)
 
 	/* call the kernels journal init function now */
 	j_journal = jbd2_journal_init_inode(inode);
-	if (j_journal == NULL) {
+	if (IS_ERR(j_journal)) {
 		mlog(ML_ERROR, "Linux journal layer error\n");
-		status = -EINVAL;
+		status = PTR_ERR(j_journal);
 		goto done;
 	}
 
@@ -910,7 +975,7 @@ int ocfs2_journal_init(struct ocfs2_super *osb, int *dirty)
 
 	journal->j_journal = j_journal;
 	journal->j_journal->j_submit_inode_data_buffers =
-		jbd2_journal_submit_inode_data_buffers;
+		ocfs2_journal_submit_inode_data_buffers;
 	journal->j_journal->j_finish_inode_data_buffers =
 		jbd2_journal_finish_inode_data_buffers;
 	journal->j_inode = inode;
@@ -1005,7 +1070,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 	if (!igrab(inode))
 		BUG();
 
-	num_running_trans = atomic_read(&(osb->journal->j_num_trans));
+	num_running_trans = atomic_read(&(journal->j_num_trans));
 	trace_ocfs2_journal_shutdown(num_running_trans);
 
 	/* Do a commit_cache here. It will flush our journal, *and*
@@ -1024,9 +1089,10 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 		osb->commit_task = NULL;
 	}
 
-	BUG_ON(atomic_read(&(osb->journal->j_num_trans)) != 0);
+	BUG_ON(atomic_read(&(journal->j_num_trans)) != 0);
 
-	if (ocfs2_mount_local(osb)) {
+	if (ocfs2_mount_local(osb) &&
+	    (journal->j_journal->j_flags & JBD2_LOADED)) {
 		jbd2_journal_lock_updates(journal->j_journal);
 		status = jbd2_journal_flush(journal->j_journal, 0);
 		jbd2_journal_unlock_updates(journal->j_journal);
@@ -1198,7 +1264,7 @@ static int ocfs2_force_read_journal(struct inode *inode)
 		}
 
 		for (i = 0; i < p_blocks; i++, p_blkno++) {
-			bh = __find_get_block(osb->sb->s_bdev, p_blkno,
+			bh = __find_get_block_nonatomic(osb->sb->s_bdev, p_blkno,
 					osb->sb->s_blocksize);
 			/* block not cached. */
 			if (!bh)
@@ -1421,6 +1487,18 @@ static int __ocfs2_recovery_thread(void *arg)
 		}
 	}
 restart:
+	if (quota_enabled) {
+		mutex_lock(&osb->recovery_lock);
+		/* Confirm that recovery thread will no longer recover quotas */
+		if (osb->recovery_state == OCFS2_REC_QUOTA_WANT_DISABLE) {
+			osb->recovery_state = OCFS2_REC_QUOTA_DISABLED;
+			wake_up(&osb->recovery_event);
+		}
+		if (osb->recovery_state >= OCFS2_REC_QUOTA_DISABLED)
+			quota_enabled = 0;
+		mutex_unlock(&osb->recovery_lock);
+	}
+
 	status = ocfs2_super_lock(osb, 1);
 	if (status < 0) {
 		mlog_errno(status);
@@ -1518,27 +1596,29 @@ bail:
 
 	ocfs2_free_replay_slots(osb);
 	osb->recovery_thread_task = NULL;
-	mb(); /* sync with ocfs2_recovery_thread_running */
+	if (osb->recovery_state == OCFS2_REC_WANT_DISABLE)
+		osb->recovery_state = OCFS2_REC_DISABLED;
 	wake_up(&osb->recovery_event);
 
 	mutex_unlock(&osb->recovery_lock);
 
-	if (quota_enabled)
-		kfree(rm_quota);
+	kfree(rm_quota);
 
 	return status;
 }
 
 void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num)
 {
+	int was_set = -1;
+
 	mutex_lock(&osb->recovery_lock);
+	if (osb->recovery_state < OCFS2_REC_WANT_DISABLE)
+		was_set = ocfs2_recovery_map_set(osb, node_num);
 
 	trace_ocfs2_recovery_thread(node_num, osb->node_num,
-		osb->disable_recovery, osb->recovery_thread_task,
-		osb->disable_recovery ?
-		-1 : ocfs2_recovery_map_set(osb, node_num));
+		osb->recovery_state, osb->recovery_thread_task, was_set);
 
-	if (osb->disable_recovery)
+	if (osb->recovery_state >= OCFS2_REC_WANT_DISABLE)
 		goto out;
 
 	if (osb->recovery_thread_task)
@@ -1673,9 +1753,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 	}
 
 	journal = jbd2_journal_init_inode(inode);
-	if (journal == NULL) {
+	if (IS_ERR(journal)) {
 		mlog(ML_ERROR, "Linux journal layer error\n");
-		status = -EIO;
+		status = PTR_ERR(journal);
 		goto done;
 	}
 
@@ -1905,7 +1985,7 @@ bail:
 
 /*
  * Scan timer should get fired every ORPHAN_SCAN_SCHEDULE_TIMEOUT. Add some
- * randomness to the timeout to minimize multple nodes firing the timer at the
+ * randomness to the timeout to minimize multiple nodes firing the timer at the
  * same time.
  */
 static inline unsigned long ocfs2_orphan_scan_timeout(void)