1 files changed, 214 insertions, 14 deletions

diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 48aa9c7401c..9d8fcf2f3b9 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -29,6 +29,7 @@
 #include <linux/highmem.h>
 #include <linux/swap.h>
 #include <linux/quotaops.h>
+#include <linux/blkdev.h>
 
 #include <cluster/masklog.h>
 
@@ -1024,7 +1025,7 @@ static int ocfs2_create_new_meta_bhs(handle_t *handle,
 		for(i = count;  i < (num_got + count); i++) {
 			bhs[i] = sb_getblk(osb->sb, first_blkno);
 			if (bhs[i] == NULL) {
-				status = -EIO;
+				status = -ENOMEM;
 				mlog_errno(status);
 				goto bail;
 			}
@@ -1133,7 +1134,7 @@ static int ocfs2_adjust_rightmost_branch(handle_t *handle,
 	}
 
 	el = path_leaf_el(path);
-	rec = &el->l_recs[le32_to_cpu(el->l_next_free_rec) - 1];
+	rec = &el->l_recs[le16_to_cpu(el->l_next_free_rec) - 1];
 
 	ocfs2_adjust_rightmost_records(handle, et, path, rec);
 
@@ -4741,6 +4742,7 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 				enum ocfs2_alloc_restarted *reason_ret)
 {
 	int status = 0, err = 0;
+	int need_free = 0;
 	int free_extents;
 	enum ocfs2_alloc_restarted reason = RESTART_NONE;
 	u32 bit_off, num_bits;
@@ -4795,7 +4797,8 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 					      OCFS2_JOURNAL_ACCESS_WRITE);
 	if (status < 0) {
 		mlog_errno(status);
-		goto leave;
+		need_free = 1;
+		goto bail;
 	}
 
 	block = ocfs2_clusters_to_blocks(osb->sb, bit_off);
@@ -4806,7 +4809,8 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 				     num_bits, flags, meta_ac);
 	if (status < 0) {
 		mlog_errno(status);
-		goto leave;
+		need_free = 1;
+		goto bail;
 	}
 
 	ocfs2_journal_dirty(handle, et->et_root_bh);
@@ -4820,6 +4824,19 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 		reason = RESTART_TRANS;
 	}
 
+bail:
+	if (need_free) {
+		if (data_ac->ac_which == OCFS2_AC_USE_LOCAL)
+			ocfs2_free_local_alloc_bits(osb, handle, data_ac,
+					bit_off, num_bits);
+		else
+			ocfs2_free_clusters(handle,
+					data_ac->ac_inode,
+					data_ac->ac_bh,
+					ocfs2_clusters_to_blocks(osb->sb, bit_off),
+					num_bits);
+	}
+
 leave:
 	if (reason_ret)
 		*reason_ret = reason;
@@ -5654,7 +5671,7 @@ int ocfs2_remove_btree_range(struct inode *inode,
 					       &ref_tree, NULL);
 		if (ret) {
 			mlog_errno(ret);
-			goto out;
+			goto bail;
 		}
 
 		ret = ocfs2_prepare_refcount_change_for_del(inode,
@@ -5665,7 +5682,7 @@ int ocfs2_remove_btree_range(struct inode *inode,
 							    &extra_blocks);
 		if (ret < 0) {
 			mlog_errno(ret);
-			goto out;
+			goto bail;
 		}
 	}
 
@@ -5673,7 +5690,7 @@ int ocfs2_remove_btree_range(struct inode *inode,
 						 extra_blocks);
 	if (ret) {
 		mlog_errno(ret);
-		return ret;
+		goto bail;
 	}
 
 	mutex_lock(&tl_inode->i_mutex);
@@ -5698,7 +5715,7 @@ int ocfs2_remove_btree_range(struct inode *inode,
 					   OCFS2_JOURNAL_ACCESS_WRITE);
 	if (ret) {
 		mlog_errno(ret);
-		goto out;
+		goto out_commit;
 	}
 
 	dquot_free_space_nodirty(inode,
@@ -5711,6 +5728,7 @@ int ocfs2_remove_btree_range(struct inode *inode,
 	}
 
 	ocfs2_et_update_clusters(et, -len);
+	ocfs2_update_inode_fsync_trans(handle, inode, 1);
 
 	ocfs2_journal_dirty(handle, et->et_root_bh);
 
@@ -5733,7 +5751,7 @@ out_commit:
 	ocfs2_commit_trans(osb, handle);
 out:
 	mutex_unlock(&tl_inode->i_mutex);
-
+bail:
 	if (meta_ac)
 		ocfs2_free_alloc_context(meta_ac);
 
@@ -6028,7 +6046,8 @@ static void ocfs2_truncate_log_worker(struct work_struct *work)
 void ocfs2_schedule_truncate_log_flush(struct ocfs2_super *osb,
 				       int cancel)
 {
-	if (osb->osb_tl_inode) {
+	if (osb->osb_tl_inode &&
+			atomic_read(&osb->osb_tl_disable) == 0) {
 		/* We want to push off log flushes while truncates are
 		 * still running. */
 		if (cancel)
@@ -6205,6 +6224,8 @@ void ocfs2_truncate_log_shutdown(struct ocfs2_super *osb)
 	int status;
 	struct inode *tl_inode = osb->osb_tl_inode;
 
+	atomic_set(&osb->osb_tl_disable, 1);
+
 	if (tl_inode) {
 		cancel_delayed_work(&osb->osb_truncate_log_wq);
 		flush_workqueue(ocfs2_wq);
@@ -6236,6 +6257,7 @@ int ocfs2_truncate_log_init(struct ocfs2_super *osb)
 	 * until we're sure all is well. */
 	INIT_DELAYED_WORK(&osb->osb_truncate_log_wq,
 			  ocfs2_truncate_log_worker);
+	atomic_set(&osb->osb_tl_disable, 0);
 	osb->osb_tl_bh    = tl_bh;
 	osb->osb_tl_inode = tl_inode;
 
@@ -6750,8 +6772,7 @@ int ocfs2_zero_range_for_truncate(struct inode *inode, handle_t *handle,
 		mlog_errno(ret);
 
 out:
-	if (pages)
-		kfree(pages);
+	kfree(pages);
 
 	return ret;
 }
@@ -6805,6 +6826,8 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 					 struct buffer_head *di_bh)
 {
 	int ret, i, has_data, num_pages = 0;
+	int need_free = 0;
+	u32 bit_off, num;
 	handle_t *handle;
 	u64 uninitialized_var(block);
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
@@ -6850,7 +6873,6 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 	}
 
 	if (has_data) {
-		u32 bit_off, num;
 		unsigned int page_end;
 		u64 phys;
 
@@ -6886,6 +6908,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 		ret = ocfs2_grab_eof_pages(inode, 0, end, pages, &num_pages);
 		if (ret) {
 			mlog_errno(ret);
+			need_free = 1;
 			goto out_commit;
 		}
 
@@ -6896,6 +6919,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 		ret = ocfs2_read_inline_data(inode, pages[0], di_bh);
 		if (ret) {
 			mlog_errno(ret);
+			need_free = 1;
 			goto out_commit;
 		}
 
@@ -6913,6 +6937,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 	di->i_dyn_features = cpu_to_le16(oi->ip_dyn_features);
 	spin_unlock(&oi->ip_lock);
 
+	ocfs2_update_inode_fsync_trans(handle, inode, 1);
 	ocfs2_dinode_new_extent_list(inode, di);
 
 	ocfs2_journal_dirty(handle, di_bh);
@@ -6927,6 +6952,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode,
 		ret = ocfs2_insert_extent(handle, &et, 0, block, 1, 0, NULL);
 		if (ret) {
 			mlog_errno(ret);
+			need_free = 1;
 			goto out_commit;
 		}
 
@@ -6938,6 +6964,18 @@ out_commit:
 		dquot_free_space_nodirty(inode,
 					  ocfs2_clusters_to_bytes(osb->sb, 1));
 
+	if (need_free) {
+		if (data_ac->ac_which == OCFS2_AC_USE_LOCAL)
+			ocfs2_free_local_alloc_bits(osb, handle, data_ac,
+					bit_off, num);
+		else
+			ocfs2_free_clusters(handle,
+					data_ac->ac_inode,
+					data_ac->ac_bh,
+					ocfs2_clusters_to_blocks(osb->sb, bit_off),
+					num);
+	}
+
 	ocfs2_commit_trans(osb, handle);
 
 out_unlock:
@@ -7126,7 +7164,7 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
 	if (end > i_size_read(inode))
 		end = i_size_read(inode);
 
-	BUG_ON(start >= end);
+	BUG_ON(start > end);
 
 	if (!(OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) ||
 	    !(le16_to_cpu(di->i_dyn_features) & OCFS2_INLINE_DATA_FL) ||
@@ -7176,6 +7214,7 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
 	di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
 	di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
 
+	ocfs2_update_inode_fsync_trans(handle, inode, 1);
 	ocfs2_journal_dirty(handle, di_bh);
 
 out_commit:
@@ -7184,3 +7223,164 @@ out_commit:
 out:
 	return ret;
 }
+
+static int ocfs2_trim_extent(struct super_block *sb,
+			     struct ocfs2_group_desc *gd,
+			     u32 start, u32 count)
+{
+	u64 discard, bcount;
+
+	bcount = ocfs2_clusters_to_blocks(sb, count);
+	discard = le64_to_cpu(gd->bg_blkno) +
+			ocfs2_clusters_to_blocks(sb, start);
+
+	trace_ocfs2_trim_extent(sb, (unsigned long long)discard, bcount);
+
+	return sb_issue_discard(sb, discard, bcount, GFP_NOFS, 0);
+}
+
+static int ocfs2_trim_group(struct super_block *sb,
+			    struct ocfs2_group_desc *gd,
+			    u32 start, u32 max, u32 minbits)
+{
+	int ret = 0, count = 0, next;
+	void *bitmap = gd->bg_bitmap;
+
+	if (le16_to_cpu(gd->bg_free_bits_count) < minbits)
+		return 0;
+
+	trace_ocfs2_trim_group((unsigned long long)le64_to_cpu(gd->bg_blkno),
+			       start, max, minbits);
+
+	while (start < max) {
+		start = ocfs2_find_next_zero_bit(bitmap, max, start);
+		if (start >= max)
+			break;
+		next = ocfs2_find_next_bit(bitmap, max, start);
+
+		if ((next - start) >= minbits) {
+			ret = ocfs2_trim_extent(sb, gd,
+						start, next - start);
+			if (ret < 0) {
+				mlog_errno(ret);
+				break;
+			}
+			count += next - start;
+		}
+		start = next + 1;
+
+		if (fatal_signal_pending(current)) {
+			count = -ERESTARTSYS;
+			break;
+		}
+
+		if ((le16_to_cpu(gd->bg_free_bits_count) - count) < minbits)
+			break;
+	}
+
+	if (ret < 0)
+		count = ret;
+
+	return count;
+}
+
+int ocfs2_trim_fs(struct super_block *sb, struct fstrim_range *range)
+{
+	struct ocfs2_super *osb = OCFS2_SB(sb);
+	u64 start, len, trimmed, first_group, last_group, group;
+	int ret, cnt;
+	u32 first_bit, last_bit, minlen;
+	struct buffer_head *main_bm_bh = NULL;
+	struct inode *main_bm_inode = NULL;
+	struct buffer_head *gd_bh = NULL;
+	struct ocfs2_dinode *main_bm;
+	struct ocfs2_group_desc *gd = NULL;
+
+	start = range->start >> osb->s_clustersize_bits;
+	len = range->len >> osb->s_clustersize_bits;
+	minlen = range->minlen >> osb->s_clustersize_bits;
+
+	if (minlen >= osb->bitmap_cpg || range->len < sb->s_blocksize)
+		return -EINVAL;
+
+	main_bm_inode = ocfs2_get_system_file_inode(osb,
+						    GLOBAL_BITMAP_SYSTEM_INODE,
+						    OCFS2_INVALID_SLOT);
+	if (!main_bm_inode) {
+		ret = -EIO;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	mutex_lock(&main_bm_inode->i_mutex);
+
+	ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 0);
+	if (ret < 0) {
+		mlog_errno(ret);
+		goto out_mutex;
+	}
+	main_bm = (struct ocfs2_dinode *)main_bm_bh->b_data;
+
+	if (start >= le32_to_cpu(main_bm->i_clusters)) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
+	len = range->len >> osb->s_clustersize_bits;
+	if (start + len > le32_to_cpu(main_bm->i_clusters))
+		len = le32_to_cpu(main_bm->i_clusters) - start;
+
+	trace_ocfs2_trim_fs(start, len, minlen);
+
+	/* Determine first and last group to examine based on start and len */
+	first_group = ocfs2_which_cluster_group(main_bm_inode, start);
+	if (first_group == osb->first_cluster_group_blkno)
+		first_bit = start;
+	else
+		first_bit = start - ocfs2_blocks_to_clusters(sb, first_group);
+	last_group = ocfs2_which_cluster_group(main_bm_inode, start + len - 1);
+	last_bit = osb->bitmap_cpg;
+
+	trimmed = 0;
+	for (group = first_group; group <= last_group;) {
+		if (first_bit + len >= osb->bitmap_cpg)
+			last_bit = osb->bitmap_cpg;
+		else
+			last_bit = first_bit + len;
+
+		ret = ocfs2_read_group_descriptor(main_bm_inode,
+						  main_bm, group,
+						  &gd_bh);
+		if (ret < 0) {
+			mlog_errno(ret);
+			break;
+		}
+
+		gd = (struct ocfs2_group_desc *)gd_bh->b_data;
+		cnt = ocfs2_trim_group(sb, gd, first_bit, last_bit, minlen);
+		brelse(gd_bh);
+		gd_bh = NULL;
+		if (cnt < 0) {
+			ret = cnt;
+			mlog_errno(ret);
+			break;
+		}
+
+		trimmed += cnt;
+		len -= osb->bitmap_cpg - first_bit;
+		first_bit = 0;
+		if (group == osb->first_cluster_group_blkno)
+			group = ocfs2_clusters_to_blocks(sb, osb->bitmap_cpg);
+		else
+			group += ocfs2_clusters_to_blocks(sb, osb->bitmap_cpg);
+	}
+	range->len = trimmed * sb->s_blocksize;
+out_unlock:
+	ocfs2_inode_unlock(main_bm_inode, 0);
+	brelse(main_bm_bh);
+out_mutex:
+	mutex_unlock(&main_bm_inode->i_mutex);
+	iput(main_bm_inode);
+out:
+	return ret;
+}