6.0.3-rc1 review

@@ -1718,9 +1718,14 @@ static int writeback_single_inode(struct
 	 */
 	if (!(inode->i_state & I_DIRTY_ALL))
 		inode_cgwb_move_to_attached(inode, wb);
-	else if (!(inode->i_state & I_SYNC_QUEUED) &&
-		 (inode->i_state & I_DIRTY))
-		redirty_tail_locked(inode, wb);
+	else if (!(inode->i_state & I_SYNC_QUEUED)) {
+		if ((inode->i_state & I_DIRTY))
+			redirty_tail_locked(inode, wb);
+		else if (inode->i_state & I_DIRTY_TIME) {
+			inode->dirtied_when = jiffies;
+			inode_io_list_move_locked(inode, wb, &wb->b_dirty_time);
+		}
+	}
 
 	spin_unlock(&wb->list_lock);
 	inode_sync_complete(inode);

@@ -2370,6 +2375,20 @@ void __mark_inode_dirty(struct inode *in
 
 	if (flags & I_DIRTY_INODE) {
 		/*
+		 * Inode timestamp update will piggback on this dirtying.
+		 * We tell ->dirty_inode callback that timestamps need to
+		 * be updated by setting I_DIRTY_TIME in flags.
+		 */
+		if (inode->i_state & I_DIRTY_TIME) {
+			spin_lock(&inode->i_lock);
+			if (inode->i_state & I_DIRTY_TIME) {
+				inode->i_state &= ~I_DIRTY_TIME;
+				flags |= I_DIRTY_TIME;
+			}
+			spin_unlock(&inode->i_lock);
+		}
+
+		/*
 		 * Notify the filesystem about the inode being dirtied, so that
 		 * (if needed) it can update on-disk fields and journal the
 		 * inode.  This is only needed when the inode itself is being

@@ -2378,7 +2397,8 @@ void __mark_inode_dirty(struct inode *in
 		 */
 		trace_writeback_dirty_inode_start(inode, flags);
 		if (sb->s_op->dirty_inode)
-			sb->s_op->dirty_inode(inode, flags & I_DIRTY_INODE);
+			sb->s_op->dirty_inode(inode,
+				flags & (I_DIRTY_INODE | I_DIRTY_TIME));
 		trace_writeback_dirty_inode(inode, flags);
 
 		/* I_DIRTY_INODE supersedes I_DIRTY_TIME. */

@@ -2399,21 +2419,15 @@ void __mark_inode_dirty(struct inode *in
 	 */
 	smp_mb();
 
-	if (((inode->i_state & flags) == flags) ||
-	    (dirtytime && (inode->i_state & I_DIRTY_INODE)))
+	if ((inode->i_state & flags) == flags)
 		return;
 
 	spin_lock(&inode->i_lock);
-	if (dirtytime && (inode->i_state & I_DIRTY_INODE))
-		goto out_unlock_inode;
 	if ((inode->i_state & flags) != flags) {
 		const int was_dirty = inode->i_state & I_DIRTY;
 
 		inode_attach_wb(inode, NULL);
 
-		/* I_DIRTY_INODE supersedes I_DIRTY_TIME. */
-		if (flags & I_DIRTY_INODE)
-			inode->i_state &= ~I_DIRTY_TIME;
 		inode->i_state |= flags;
 
 		/*

@@ -2486,7 +2500,6 @@ void __mark_inode_dirty(struct inode *in
 out_unlock:
 	if (wb)
 		spin_unlock(&wb->list_lock);
-out_unlock_inode:
 	spin_unlock(&inode->i_lock);
 }
 EXPORT_SYMBOL(__mark_inode_dirty);

@@ -653,7 +653,7 @@ xfs_fs_destroy_inode(
 static void
 xfs_fs_dirty_inode(
 	struct inode			*inode,
-	int				flag)
+	int				flags)
 {
 	struct xfs_inode		*ip = XFS_I(inode);
 	struct xfs_mount		*mp = ip->i_mount;

@@ -661,7 +661,13 @@ xfs_fs_dirty_inode(
 
 	if (!(inode->i_sb->s_flags & SB_LAZYTIME))
 		return;
-	if (flag != I_DIRTY_SYNC || !(inode->i_state & I_DIRTY_TIME))
+
+	/*
+	 * Only do the timestamp update if the inode is dirty (I_DIRTY_SYNC)
+	 * and has dirty timestamp (I_DIRTY_TIME). I_DIRTY_TIME can be passed
+	 * in flags possibly together with I_DIRTY_SYNC.
+	 */
+	if ((flags & ~I_DIRTY_TIME) != I_DIRTY_SYNC || !(flags & I_DIRTY_TIME))
 		return;
 
 	if (xfs_trans_alloc(mp, &M_RES(mp)->tr_fsyncts, 0, 0, 0, &tp))

@@ -2371,13 +2371,14 @@ static inline void kiocb_clone(struct ki
  *			don't have to write inode on fdatasync() when only
  *			e.g. the timestamps have changed.
  * I_DIRTY_PAGES	Inode has dirty pages.  Inode itself may be clean.
- * I_DIRTY_TIME		The inode itself only has dirty timestamps, and the
+ * I_DIRTY_TIME		The inode itself has dirty timestamps, and the
  *			lazytime mount option is enabled.  We keep track of this
  *			separately from I_DIRTY_SYNC in order to implement
  *			lazytime.  This gets cleared if I_DIRTY_INODE
- *			(I_DIRTY_SYNC and/or I_DIRTY_DATASYNC) gets set.  I.e.
- *			either I_DIRTY_TIME *or* I_DIRTY_INODE can be set in
- *			i_state, but not both.  I_DIRTY_PAGES may still be set.
+ *			(I_DIRTY_SYNC and/or I_DIRTY_DATASYNC) gets set. But
+ *			I_DIRTY_TIME can still be set if I_DIRTY_SYNC is already
+ *			in place because writeback might already be in progress
+ *			and we don't want to lose the time update
  * I_NEW		Serves as both a mutex and completion notification.
  *			New inodes set I_NEW.  If two processes both create
  *			the same inode, one of them will release its inode and

@@ -3810,6 +3810,7 @@ const struct file_operations btrfs_file_
 	.mmap		= btrfs_file_mmap,
 	.open		= btrfs_file_open,
 	.release	= btrfs_release_file,
+	.get_unmapped_area = thp_get_unmapped_area,
 	.fsync		= btrfs_sync_file,
 	.fallocate	= btrfs_fallocate,
 	.unlocked_ioctl	= btrfs_ioctl,

@@ -2190,7 +2190,16 @@ int btrfs_read_block_groups(struct btrfs
 	int need_clear = 0;
 	u64 cache_gen;
 
-	if (!root)
+	/*
+	 * Either no extent root (with ibadroots rescue option) or we have
+	 * unsupported RO options. The fs can never be mounted read-write, so no
+	 * need to waste time searching block group items.
+	 *
+	 * This also allows new extent tree related changes to be RO compat,
+	 * no need for a full incompat flag.
+	 */
+	if (!root || (btrfs_super_compat_ro_flags(info->super_copy) &
+		      ~BTRFS_FEATURE_COMPAT_RO_SUPP))
 		return fill_dummy_bgs(info);
 
 	key.objectid = 0;

@@ -2112,6 +2112,15 @@ static int btrfs_remount(struct super_bl
 			ret = -EINVAL;
 			goto restore;
 		}
+		if (btrfs_super_compat_ro_flags(fs_info->super_copy) &
+		    ~BTRFS_FEATURE_COMPAT_RO_SUPP) {
+			btrfs_err(fs_info,
+		"can not remount read-write due to unsupported optional flags 0x%llx",
+				btrfs_super_compat_ro_flags(fs_info->super_copy) &
+				~BTRFS_FEATURE_COMPAT_RO_SUPP);
+			ret = -EINVAL;
+			goto restore;
+		}
 		if (fs_info->fs_devices->rw_devices == 0) {
 			ret = -EACCES;
 			goto restore;

@@ -1174,6 +1174,21 @@ out_add_root:
 		fs_info->qgroup_rescan_running = true;
 	        btrfs_queue_work(fs_info->qgroup_rescan_workers,
 	                         &fs_info->qgroup_rescan_work);
+	} else {
+		/*
+		 * We have set both BTRFS_FS_QUOTA_ENABLED and
+		 * BTRFS_QGROUP_STATUS_FLAG_ON, so we can only fail with
+		 * -EINPROGRESS. That can happen because someone started the
+		 * rescan worker by calling quota rescan ioctl before we
+		 * attempted to initialize the rescan worker. Failure due to
+		 * quotas disabled in the meanwhile is not possible, because
+		 * we are holding a write lock on fs_info->subvol_sem, which
+		 * is also acquired when disabling quotas.
+		 * Ignore such error, and any other error would need to undo
+		 * everything we did in the transaction we just committed.
+		 */
+		ASSERT(ret == -EINPROGRESS);
+		ret = 0;
 	}
 
 out_free_path:

@@ -523,6 +523,7 @@ void btrfs_drop_extent_cache(struct btrf
 		testend = 0;
 	}
 	while (1) {
+		bool ends_after_range = false;
 		int no_splits = 0;
 
 		modified = false;

@@ -539,10 +540,12 @@ void btrfs_drop_extent_cache(struct btrf
 			write_unlock(&em_tree->lock);
 			break;
 		}
+		if (testend && em->start + em->len > start + len)
+			ends_after_range = true;
 		flags = em->flags;
 		gen = em->generation;
 		if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) {
-			if (testend && em->start + em->len >= start + len) {
+			if (ends_after_range) {
 				free_extent_map(em);
 				write_unlock(&em_tree->lock);
 				break;

@@ -592,7 +595,7 @@ void btrfs_drop_extent_cache(struct btrf
 			split = split2;
 			split2 = NULL;
 		}
-		if (testend && em->start + em->len > start + len) {
+		if (ends_after_range) {
 			u64 diff = start + len - em->start;
 
 			split->start = start + len;

@@ -630,14 +633,42 @@ void btrfs_drop_extent_cache(struct btrf
 			} else {
 				ret = add_extent_mapping(em_tree, split,
 							 modified);
-				ASSERT(ret == 0); /* Logic error */
+				/* Logic error, shouldn't happen. */
+				ASSERT(ret == 0);
+				if (WARN_ON(ret != 0) && modified)
+					btrfs_set_inode_full_sync(inode);
 			}
 			free_extent_map(split);
 			split = NULL;
 		}
 next:
-		if (extent_map_in_tree(em))
+		if (extent_map_in_tree(em)) {
+			/*
+			 * If the extent map is still in the tree it means that
+			 * either of the following is true:
+			 *
+			 * 1) It fits entirely in our range (doesn't end beyond
+			 *    it or starts before it);
+			 *
+			 * 2) It starts before our range and/or ends after our
+			 *    range, and we were not able to allocate the extent
+			 *    maps for split operations, @split and @split2.
+			 *
+			 * If we are at case 2) then we just remove the entire
+			 * extent map - this is fine since if anyone needs it to
+			 * access the subranges outside our range, will just
+			 * load it again from the subvolume tree's file extent
+			 * item. However if the extent map was in the list of
+			 * modified extents, then we must mark the inode for a
+			 * full fsync, otherwise a fast fsync will miss this
+			 * extent if it's new and needs to be logged.
+			 */
+			if ((em->start < start || ends_after_range) && modified) {
+				ASSERT(no_splits);
+				btrfs_set_inode_full_sync(inode);
+			}
 			remove_extent_mapping(em_tree, em);
+		}
 		write_unlock(&em_tree->lock);
 
 		/* once for us */

@@ -2201,14 +2232,6 @@ int btrfs_sync_file(struct file *file, l
 	atomic_inc(&root->log_batch);
 
 	/*
-	 * Always check for the full sync flag while holding the inode's lock,
-	 * to avoid races with other tasks. The flag must be either set all the
-	 * time during logging or always off all the time while logging.
-	 */
-	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
-			     &BTRFS_I(inode)->runtime_flags);
-
-	/*
 	 * Before we acquired the inode's lock and the mmap lock, someone may
 	 * have dirtied more pages in the target range. We need to make sure
 	 * that writeback for any such pages does not start while we are logging

@@ -2233,6 +2256,17 @@ int btrfs_sync_file(struct file *file, l
 	}
 
 	/*
+	 * Always check for the full sync flag while holding the inode's lock,
+	 * to avoid races with other tasks. The flag must be either set all the
+	 * time during logging or always off all the time while logging.
+	 * We check the flag here after starting delalloc above, because when
+	 * running delalloc the full sync flag may be set if we need to drop
+	 * extra extent map ranges due to temporary memory allocation failures.
+	 */
+	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+			     &BTRFS_I(inode)->runtime_flags);
+
+	/*
 	 * We have to do this here to avoid the priority inversion of waiting on
 	 * IO of a lower priority task while holding a transaction open.
 	 *

@@ -4888,6 +4888,9 @@ btrfs_init_new_buffer(struct btrfs_trans
 	    !test_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &root->state))
 		lockdep_owner = BTRFS_FS_TREE_OBJECTID;
 
+	/* btrfs_clean_tree_block() accesses generation field. */
+	btrfs_set_header_generation(buf, trans->transid);
+
 	/*
 	 * This needs to stay, because we could allocate a freed block from an
 	 * old tree into a new tree, so we need to make sure this new block is

@@ -97,14 +97,10 @@ static int gc_thread_func(void *data)
 		 */
 		if (sbi->gc_mode == GC_URGENT_HIGH) {
 			spin_lock(&sbi->gc_urgent_high_lock);
-			if (sbi->gc_urgent_high_limited) {
-				if (!sbi->gc_urgent_high_remaining) {
-					sbi->gc_urgent_high_limited = false;
-					spin_unlock(&sbi->gc_urgent_high_lock);
-					sbi->gc_mode = GC_NORMAL;
-					continue;
-				}
-				sbi->gc_urgent_high_remaining--;
+			if (sbi->gc_urgent_high_limited &&
+					!sbi->gc_urgent_high_remaining--) {
+				sbi->gc_urgent_high_limited = false;
+				sbi->gc_mode = GC_NORMAL;
 			}
 			spin_unlock(&sbi->gc_urgent_high_lock);
 		}

@@ -2181,6 +2181,9 @@ static void f2fs_enable_checkpoint(struc
 	f2fs_up_write(&sbi->gc_lock);
 
 	f2fs_sync_fs(sbi->sb, 1);
+
+	/* Let's ensure there's no pending checkpoint anymore */
+	f2fs_flush_ckpt_thread(sbi);
 }
 
 static int f2fs_remount(struct super_block *sb, int *flags, char *data)

@@ -2346,6 +2349,9 @@ static int f2fs_remount(struct super_blo
 		f2fs_stop_ckpt_thread(sbi);
 		need_restart_ckpt = true;
 	} else {
+		/* Flush if the prevous checkpoint, if exists. */
+		f2fs_flush_ckpt_thread(sbi);
+
 		err = f2fs_start_ckpt_thread(sbi);
 		if (err) {
 			f2fs_err(sbi,

@@ -1892,15 +1892,27 @@ int f2fs_start_ckpt_thread(struct f2fs_s
 void f2fs_stop_ckpt_thread(struct f2fs_sb_info *sbi)
 {
 	struct ckpt_req_control *cprc = &sbi->cprc_info;
+	struct task_struct *ckpt_task;
 
-	if (cprc->f2fs_issue_ckpt) {
-		struct task_struct *ckpt_task = cprc->f2fs_issue_ckpt;
+	if (!cprc->f2fs_issue_ckpt)
+		return;
 
-		cprc->f2fs_issue_ckpt = NULL;
-		kthread_stop(ckpt_task);
+	ckpt_task = cprc->f2fs_issue_ckpt;
+	cprc->f2fs_issue_ckpt = NULL;
+	kthread_stop(ckpt_task);
 
-		flush_remained_ckpt_reqs(sbi, NULL);
-	}
+	f2fs_flush_ckpt_thread(sbi);
+}
+
+void f2fs_flush_ckpt_thread(struct f2fs_sb_info *sbi)
+{
+	struct ckpt_req_control *cprc = &sbi->cprc_info;
+
+	flush_remained_ckpt_reqs(sbi, NULL);
+
+	/* Let's wait for the previous dispatched checkpoint. */
+	while (atomic_read(&cprc->queued_ckpt))
+		io_schedule_timeout(DEFAULT_IO_TIMEOUT);
 }
 
 void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi)

@@ -3707,6 +3707,7 @@ static inline bool f2fs_need_rand_seg(st
  * checkpoint.c
  */
 void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi, bool end_io);
+void f2fs_flush_ckpt_thread(struct f2fs_sb_info *sbi);
 struct page *f2fs_grab_meta_page(struct f2fs_sb_info *sbi, pgoff_t index);
 struct page *f2fs_get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index);
 struct page *f2fs_get_meta_page_retry(struct f2fs_sb_info *sbi, pgoff_t index);

@@ -1666,9 +1666,8 @@ static int f2fs_freeze(struct super_bloc
 	if (is_sbi_flag_set(F2FS_SB(sb), SBI_IS_DIRTY))
 		return -EINVAL;
 
-	/* ensure no checkpoint required */
-	if (!llist_empty(&F2FS_SB(sb)->cprc_info.issue_list))
-		return -EINVAL;
+	/* Let's flush checkpoints and stop the thread. */
+	f2fs_flush_ckpt_thread(F2FS_SB(sb));
 
 	/* to avoid deadlock on f2fs_evict_inode->SB_FREEZE_FS */
 	set_sbi_flag(F2FS_SB(sb), SBI_IS_FREEZING);

@@ -301,10 +301,10 @@ static void f2fs_destroy_casefold_cache(
 
 static inline void limit_reserve_root(struct f2fs_sb_info *sbi)
 {
-	block_t limit = min((sbi->user_block_count << 1) / 1000,
+	block_t limit = min((sbi->user_block_count >> 3),
 			sbi->user_block_count - sbi->reserved_blocks);
 
-	/* limit is 0.2% */
+	/* limit is 12.5% */
 	if (test_opt(sbi, RESERVE_ROOT) &&
 			F2FS_OPTION(sbi).root_reserved_blocks > limit) {
 		F2FS_OPTION(sbi).root_reserved_blocks = limit;

@@ -140,7 +140,7 @@ static bool __is_bitmap_valid(struct f2f
 	unsigned int segno, offset;
 	bool exist;
 
-	if (type != DATA_GENERIC_ENHANCE && type != DATA_GENERIC_ENHANCE_READ)
+	if (type == DATA_GENERIC)
 		return true;
 
 	segno = GET_SEGNO(sbi, blkaddr);

@@ -148,6 +148,13 @@ static bool __is_bitmap_valid(struct f2f
 	se = get_seg_entry(sbi, segno);
 
 	exist = f2fs_test_bit(offset, se->cur_valid_map);
+	if (exist && type == DATA_GENERIC_ENHANCE_UPDATE) {
+		f2fs_err(sbi, "Inconsistent error blkaddr:%u, sit bitmap:%d",
+			 blkaddr, exist);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		return exist;
+	}
+
 	if (!exist && type == DATA_GENERIC_ENHANCE) {
 		f2fs_err(sbi, "Inconsistent error blkaddr:%u, sit bitmap:%d",
 			 blkaddr, exist);

@@ -185,6 +192,7 @@ bool f2fs_is_valid_blkaddr(struct f2fs_s
 	case DATA_GENERIC:
 	case DATA_GENERIC_ENHANCE:
 	case DATA_GENERIC_ENHANCE_READ:
+	case DATA_GENERIC_ENHANCE_UPDATE:
 		if (unlikely(blkaddr >= MAX_BLKADDR(sbi) ||
 				blkaddr < MAIN_BLKADDR(sbi))) {
 			f2fs_warn(sbi, "access invalid blkaddr:%u",

@@ -266,6 +266,10 @@ enum {
 					 * condition of read on truncated area
 					 * by extent_cache
 					 */
+	DATA_GENERIC_ENHANCE_UPDATE,	/*
+					 * strong check on range and segment
+					 * bitmap for update case
+					 */
 	META_GENERIC,
 };
 

@@ -698,6 +698,14 @@ retry_prev:
 				goto err;
 			}
 
+			if (f2fs_is_valid_blkaddr(sbi, dest,
+					DATA_GENERIC_ENHANCE_UPDATE)) {
+				f2fs_err(sbi, "Inconsistent dest blkaddr:%u, ino:%lu, ofs:%u",
+					dest, inode->i_ino, dn.ofs_in_node);
+				err = -EFSCORRUPTED;
+				goto err;
+			}
+
 			/* write dummy data page */
 			f2fs_replace_block(sbi, &dn, src, dest,
 						ni.version, false, false);

@@ -1078,7 +1078,7 @@ static bool is_alive(struct f2fs_sb_info
 {
 	struct page *node_page;
 	nid_t nid;
-	unsigned int ofs_in_node;
+	unsigned int ofs_in_node, max_addrs;
 	block_t source_blkaddr;
 
 	nid = le32_to_cpu(sum->nid);

@@ -1104,6 +1104,14 @@ static bool is_alive(struct f2fs_sb_info
 		return false;
 	}
 
+	max_addrs = IS_INODE(node_page) ? DEF_ADDRS_PER_INODE :
+						DEF_ADDRS_PER_BLOCK;
+	if (ofs_in_node >= max_addrs) {
+		f2fs_err(sbi, "Inconsistent ofs_in_node:%u in summary, ino:%u, nid:%u, max:%u",
+			ofs_in_node, dni->ino, dni->nid, max_addrs);
+		return false;
+	}
+
 	*nofs = ofs_of_node(node_page);
 	source_blkaddr = data_blkaddr(NULL, node_page, ofs_in_node);
 	f2fs_put_page(node_page, 1);

@@ -474,7 +474,7 @@ static int check_index_in_prev_nodes(str
 	struct dnode_of_data tdn = *dn;
 	nid_t ino, nid;
 	struct inode *inode;
-	unsigned int offset;
+	unsigned int offset, ofs_in_node, max_addrs;
 	block_t bidx;
 	int i;
 

@@ -501,15 +501,24 @@ static int check_index_in_prev_nodes(str
 got_it:
 	/* Use the locked dnode page and inode */
 	nid = le32_to_cpu(sum.nid);
+	ofs_in_node = le16_to_cpu(sum.ofs_in_node);
+
+	max_addrs = ADDRS_PER_PAGE(dn->node_page, dn->inode);
+	if (ofs_in_node >= max_addrs) {
+		f2fs_err(sbi, "Inconsistent ofs_in_node:%u in summary, ino:%lu, nid:%u, max:%u",
+			ofs_in_node, dn->inode->i_ino, nid, max_addrs);
+		return -EFSCORRUPTED;
+	}
+
 	if (dn->inode->i_ino == nid) {
 		tdn.nid = nid;
 		if (!dn->inode_page_locked)
 			lock_page(dn->inode_page);
 		tdn.node_page = dn->inode_page;
-		tdn.ofs_in_node = le16_to_cpu(sum.ofs_in_node);
+		tdn.ofs_in_node = ofs_in_node;
 		goto truncate_out;
 	} else if (dn->nid == nid) {
-		tdn.ofs_in_node = le16_to_cpu(sum.ofs_in_node);
+		tdn.ofs_in_node = ofs_in_node;
 		goto truncate_out;
 	}
 

@@ -4513,7 +4513,12 @@ static inline bool f2fs_force_buffered_i
 	/* disallow direct IO if any of devices has unaligned blksize */
 	if (f2fs_is_multi_device(sbi) && !sbi->aligned_blksize)
 		return true;
-
+	/*
+	 * for blkzoned device, fallback direct IO to buffered IO, so
+	 * all IOs can be serialized by log-structured write.
+	 */
+	if (f2fs_sb_has_blkzoned(sbi) && (rw == WRITE))
+		return true;
 	if (f2fs_lfs_mode(sbi) && (rw == WRITE)) {
 		if (block_unaligned_IO(inode, iocb, iter))
 			return true;

@@ -570,7 +570,7 @@ void jbd2_journal_commit_transaction(jou
 	journal->j_running_transaction = NULL;
 	start_time = ktime_get();
 	commit_transaction->t_log_start = journal->j_head;
-	wake_up(&journal->j_wait_transaction_locked);
+	wake_up_all(&journal->j_wait_transaction_locked);
 	write_unlock(&journal->j_state_lock);
 
 	jbd2_debug(3, "JBD2: commit phase 2a\n");

@@ -168,7 +168,7 @@ static void wait_transaction_locked(jour
 	int need_to_start;
 	tid_t tid = journal->j_running_transaction->t_tid;
 
-	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
+	prepare_to_wait_exclusive(&journal->j_wait_transaction_locked, &wait,
 			TASK_UNINTERRUPTIBLE);
 	need_to_start = !tid_geq(journal->j_commit_request, tid);
 	read_unlock(&journal->j_state_lock);

@@ -194,7 +194,7 @@ static void wait_transaction_switching(j
 		read_unlock(&journal->j_state_lock);
 		return;
 	}
-	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
+	prepare_to_wait_exclusive(&journal->j_wait_transaction_locked, &wait,
 			TASK_UNINTERRUPTIBLE);
 	read_unlock(&journal->j_state_lock);
 	/*

@@ -920,7 +920,7 @@ void jbd2_journal_unlock_updates (journa
 	write_lock(&journal->j_state_lock);
 	--journal->j_barrier_count;
 	write_unlock(&journal->j_state_lock);
-	wake_up(&journal->j_wait_transaction_locked);
+	wake_up_all(&journal->j_wait_transaction_locked);
 }
 
 static void warn_dirty_buffer(struct buffer_head *bh)

@@ -925,8 +925,14 @@ int jbd2_fc_wait_bufs(journal_t *journal
 		wait_on_buffer(bh);
 		put_bh(bh);
 		journal->j_fc_wbuf[i] = NULL;
-		if (unlikely(!buffer_uptodate(bh)))
+		/*
+		 * Update j_fc_off so jbd2_fc_release_bufs can release remain
+		 * buffer head.
+		 */
+		if (unlikely(!buffer_uptodate(bh))) {
+			journal->j_fc_off = i;
 			return -EIO;
+		}
 	}
 
 	return 0;

@@ -923,16 +923,16 @@ int jbd2_fc_wait_bufs(journal_t *journal
 	for (i = j_fc_off - 1; i >= j_fc_off - num_blks; i--) {
 		bh = journal->j_fc_wbuf[i];
 		wait_on_buffer(bh);
-		put_bh(bh);
-		journal->j_fc_wbuf[i] = NULL;
 		/*
 		 * Update j_fc_off so jbd2_fc_release_bufs can release remain
 		 * buffer head.
 		 */
 		if (unlikely(!buffer_uptodate(bh))) {
-			journal->j_fc_off = i;
+			journal->j_fc_off = i + 1;
 			return -EIO;
 		}
+		put_bh(bh);
+		journal->j_fc_wbuf[i] = NULL;
 	}
 
 	return 0;

@@ -256,6 +256,7 @@ static int fc_do_one_pass(journal_t *jou
 		err = journal->j_fc_replay_callback(journal, bh, pass,
 					next_fc_block - journal->j_fc_first,
 					expected_commit_id);
+		brelse(bh);
 		next_fc_block++;
 		if (err < 0 || err == JBD2_FC_REPLAY_STOP)
 			break;

@@ -1052,6 +1052,13 @@ static int ext2_fill_super(struct super_
 			sbi->s_blocks_per_group);
 		goto failed_mount;
 	}
+	/* At least inode table, bitmaps, and sb have to fit in one group */
+	if (sbi->s_blocks_per_group <= sbi->s_itb_per_group + 3) {
+		ext2_msg(sb, KERN_ERR,
+			"error: #blocks per group smaller than metadata size: %lu <= %lu",
+			sbi->s_blocks_per_group, sbi->s_inodes_per_group + 3);
+		goto failed_mount;
+	}
 	if (sbi->s_frags_per_group > sb->s_blocksize * 8) {
 		ext2_msg(sb, KERN_ERR,
 			"error: #fragments per group too big: %lu",

@@ -1065,9 +1072,14 @@ static int ext2_fill_super(struct super_
 			sbi->s_inodes_per_group);
 		goto failed_mount;
 	}
+	if (sb_bdev_nr_blocks(sb) < le32_to_cpu(es->s_blocks_count)) {
+		ext2_msg(sb, KERN_ERR,
+			 "bad geometry: block count %u exceeds size of device (%u blocks)",
+			 le32_to_cpu(es->s_blocks_count),
+			 (unsigned)sb_bdev_nr_blocks(sb));
+		goto failed_mount;
+	}
 
-	if (EXT2_BLOCKS_PER_GROUP(sb) == 0)
-		goto cantfind_ext2;
 	sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) -
 				le32_to_cpu(es->s_first_data_block) - 1)
 					/ EXT2_BLOCKS_PER_GROUP(sb)) + 1;

@@ -528,6 +528,12 @@ static ssize_t ext4_dio_write_iter(struc
 		ret = -EAGAIN;
 		goto out;
 	}
+	/*
+	 * Make sure inline data cannot be created anymore since we are going
+	 * to allocate blocks for DIO. We know the inode does not have any
+	 * inline data now because ext4_dio_supported() checked for that.
+	 */
+	ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
 
 	offset = iocb->ki_pos;
 	count = ret;

@@ -6653,7 +6653,7 @@ static int ext4_write_info(struct super_
 	handle_t *handle;
 
 	/* Data block + inode block */
-	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
+	handle = ext4_journal_start_sb(sb, EXT4_HT_QUOTA, 2);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ret = dquot_commit_info(sb, type);

@@ -3767,6 +3767,7 @@ static int ext4_lazyinit_thread(void *ar
 	unsigned long next_wakeup, cur;
 
 	BUG_ON(NULL == eli);
+	set_freezable();
 
 cont_thread:
 	while (true) {

@@ -126,7 +126,7 @@ static struct buffer_head *__ext4_read_d
 	struct ext4_dir_entry *dirent;
 	int is_dx_block = 0;
 
-	if (block >= inode->i_size) {
+	if (block >= inode->i_size >> inode->i_blkbits) {
 		ext4_error_inode(inode, func, line, block,
 		       "Attempting to read directory block (%u) that is past i_size (%llu)",
 		       block, inode->i_size);

@@ -5731,7 +5731,12 @@ int ext4_mark_iloc_dirty(handle_t *handl
 	}
 	ext4_fc_track_inode(handle, inode);
 
-	if (IS_I_VERSION(inode))
+	/*
+	 * ea_inodes are using i_version for storing reference count, don't
+	 * mess with it
+	 */
+	if (IS_I_VERSION(inode) &&
+	    !(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
 		inode_inc_iversion(inode);
 
 	/* the do_update_inode consumes one bh->b_count */

@@ -5425,7 +5425,7 @@ int ext4_setattr(struct user_namespace *
 			return -EINVAL;
 		}
 
-		if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size)
+		if (attr->ia_size != inode->i_size)
 			inode_inc_iversion(inode);
 
 		if (shrink) {

@@ -5735,8 +5735,7 @@ int ext4_mark_iloc_dirty(handle_t *handl
 	 * ea_inodes are using i_version for storing reference count, don't
 	 * mess with it
 	 */
-	if (IS_I_VERSION(inode) &&
-	    !(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
+	if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
 		inode_inc_iversion(inode);
 
 	/* the do_update_inode consumes one bh->b_count */

@@ -1585,7 +1585,7 @@ enum {
 	Opt_inlinecrypt,
 	Opt_usrjquota, Opt_grpjquota, Opt_quota,
 	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
-	Opt_usrquota, Opt_grpquota, Opt_prjquota, Opt_i_version,
+	Opt_usrquota, Opt_grpquota, Opt_prjquota,
 	Opt_dax, Opt_dax_always, Opt_dax_inode, Opt_dax_never,
 	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_warn_on_error,
 	Opt_nowarn_on_error, Opt_mblk_io_submit, Opt_debug_want_extra_isize,

@@ -1694,7 +1694,7 @@ static const struct fs_parameter_spec ex
 	fsparam_flag	("barrier",		Opt_barrier),
 	fsparam_u32	("barrier",		Opt_barrier),
 	fsparam_flag	("nobarrier",		Opt_nobarrier),
-	fsparam_flag	("i_version",		Opt_i_version),
+	fsparam_flag	("i_version",		Opt_removed),
 	fsparam_flag	("dax",			Opt_dax),
 	fsparam_enum	("dax",			Opt_dax_type, ext4_param_dax),
 	fsparam_u32	("stripe",		Opt_stripe),

@@ -2140,11 +2140,6 @@ static int ext4_parse_param(struct fs_co
 	case Opt_abort:
 		ctx_set_mount_flag(ctx, EXT4_MF_FS_ABORTED);
 		return 0;
-	case Opt_i_version:
-		ext4_msg(NULL, KERN_WARNING, deprecated_msg, param->key, "5.20");
-		ext4_msg(NULL, KERN_WARNING, "Use iversion instead\n");
-		ctx_set_flags(ctx, SB_I_VERSION);
-		return 0;
 	case Opt_inlinecrypt:
 #ifdef CONFIG_FS_ENCRYPTION_INLINE_CRYPT
 		ctx_set_flags(ctx, SB_INLINECRYPT);

@@ -2814,14 +2809,6 @@ static void ext4_apply_options(struct fs
 	sb->s_flags &= ~ctx->mask_s_flags;
 	sb->s_flags |= ctx->vals_s_flags;
 
-	/*
-	 * i_version differs from common mount option iversion so we have
-	 * to let vfs know that it was set, otherwise it would get cleared
-	 * on remount
-	 */
-	if (ctx->mask_s_flags & SB_I_VERSION)
-		fc->sb_flags |= SB_I_VERSION;
-
 #define APPLY(X) ({ if (ctx->spec & EXT4_SPEC_##X) sbi->X = ctx->X; })
 	APPLY(s_commit_interval);
 	APPLY(s_stripe);

@@ -2970,8 +2957,6 @@ static int _ext4_show_options(struct seq
 		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
 	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
 		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
-	if (sb->s_flags & SB_I_VERSION)
-		SEQ_OPTS_PUTS("i_version");
 	if (nodefs || sbi->s_stripe)
 		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
 	if (nodefs || EXT4_MOUNT_DATA_FLAGS &

@@ -4641,6 +4626,9 @@ static int __ext4_fill_super(struct fs_c
 	sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
 		(test_opt(sb, POSIX_ACL) ? SB_POSIXACL : 0);
 
+	/* i_version is always enabled now */
+	sb->s_flags |= SB_I_VERSION;
+
 	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
 	    (ext4_has_compat_features(sb) ||
 	     ext4_has_ro_compat_features(sb) ||

@@ -205,19 +205,12 @@ int ext4_read_bh(struct buffer_head *bh,
 
 int ext4_read_bh_lock(struct buffer_head *bh, blk_opf_t op_flags, bool wait)
 {
-	if (trylock_buffer(bh)) {
-		if (wait)
-			return ext4_read_bh(bh, op_flags, NULL);
+	lock_buffer(bh);
+	if (!wait) {
 		ext4_read_bh_nowait(bh, op_flags, NULL);
 		return 0;
 	}
-	if (wait) {
-		wait_on_buffer(bh);
-		if (buffer_uptodate(bh))
-			return 0;
-		return -EIO;
-	}
-	return 0;
+	return ext4_read_bh(bh, op_flags, NULL);
 }
 
 /*

@@ -264,7 +257,8 @@ void ext4_sb_breadahead_unmovable(struct
 	struct buffer_head *bh = sb_getblk_gfp(sb, block, 0);
 
 	if (likely(bh)) {
-		ext4_read_bh_lock(bh, REQ_RAHEAD, false);
+		if (trylock_buffer(bh))
+			ext4_read_bh_nowait(bh, REQ_RAHEAD, NULL);
 		brelse(bh);
 	}
 }

@@ -1188,6 +1188,13 @@ retry_grab:
 	page = grab_cache_page_write_begin(mapping, index);
 	if (!page)
 		return -ENOMEM;
+	/*
+	 * The same as page allocation, we prealloc buffer heads before
+	 * starting the handle.
+	 */
+	if (!page_has_buffers(page))
+		create_empty_buffers(page, inode->i_sb->s_blocksize, 0);
+
 	unlock_page(page);
 
 retry_journal:

@@ -5349,6 +5349,7 @@ int ext4_setattr(struct user_namespace *
 	int error, rc = 0;
 	int orphan = 0;
 	const unsigned int ia_valid = attr->ia_valid;
+	bool inc_ivers = true;
 
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;

@@ -5432,8 +5433,8 @@ int ext4_setattr(struct user_namespace *
 			return -EINVAL;
 		}
 
-		if (attr->ia_size != inode->i_size)
-			inode_inc_iversion(inode);
+		if (attr->ia_size == inode->i_size)
+			inc_ivers = false;
 
 		if (shrink) {
 			if (ext4_should_order_data(inode)) {

@@ -5535,6 +5536,8 @@ out_mmap_sem:
 	}
 
 	if (!error) {
+		if (inc_ivers)
+			inode_inc_iversion(inode);
 		setattr_copy(mnt_userns, inode, attr);
 		mark_inode_dirty(inode);
 	}

@@ -5738,13 +5741,6 @@ int ext4_mark_iloc_dirty(handle_t *handl
 	}
 	ext4_fc_track_inode(handle, inode);
 
-	/*
-	 * ea_inodes are using i_version for storing reference count, don't
-	 * mess with it
-	 */
-	if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
-		inode_inc_iversion(inode);
-
 	/* the do_update_inode consumes one bh->b_count */
 	get_bh(iloc->bh);
 

@@ -452,6 +452,7 @@ static long swap_inode_boot_loader(struc
 	swap_inode_data(inode, inode_bl);
 
 	inode->i_ctime = inode_bl->i_ctime = current_time(inode);
+	inode_inc_iversion(inode);
 
 	inode->i_generation = prandom_u32();
 	inode_bl->i_generation = prandom_u32();

@@ -665,6 +666,7 @@ static int ext4_ioctl_setflags(struct in
 	ext4_set_inode_flags(inode, false);
 
 	inode->i_ctime = current_time(inode);
+	inode_inc_iversion(inode);
 
 	err = ext4_mark_iloc_dirty(handle, inode, &iloc);
 flags_err:

@@ -775,6 +777,7 @@ static int ext4_ioctl_setproject(struct
 
 	EXT4_I(inode)->i_projid = kprojid;
 	inode->i_ctime = current_time(inode);
+	inode_inc_iversion(inode);
 out_dirty:
 	rc = ext4_mark_iloc_dirty(handle, inode, &iloc);
 	if (!err)

@@ -1257,6 +1260,7 @@ static long __ext4_ioctl(struct file *fi
 		err = ext4_reserve_inode_write(handle, inode, &iloc);
 		if (err == 0) {
 			inode->i_ctime = current_time(inode);
+			inode_inc_iversion(inode);
 			inode->i_generation = generation;
 			err = ext4_mark_iloc_dirty(handle, inode, &iloc);
 		}

@@ -2412,6 +2412,7 @@ retry_inode:
 	if (!error) {
 		ext4_xattr_update_super_block(handle, inode->i_sb);
 		inode->i_ctime = current_time(inode);
+		inode_inc_iversion(inode);
 		if (!value)
 			no_expand = 0;
 		error = ext4_mark_iloc_dirty(handle, inode, &is.iloc);

@@ -85,15 +85,20 @@ static struct buffer_head *ext4_append(h
 		return bh;
 	inode->i_size += inode->i_sb->s_blocksize;
 	EXT4_I(inode)->i_disksize = inode->i_size;
+	err = ext4_mark_inode_dirty(handle, inode);
+	if (err)
+		goto out;
 	BUFFER_TRACE(bh, "get_write_access");
 	err = ext4_journal_get_write_access(handle, inode->i_sb, bh,
 					    EXT4_JTR_NONE);
-	if (err) {
-		brelse(bh);
-		ext4_std_error(inode->i_sb, err);
-		return ERR_PTR(err);
-	}
+	if (err)
+		goto out;
 	return bh;
+
+out:
+	brelse(bh);
+	ext4_std_error(inode->i_sb, err);
+	return ERR_PTR(err);
 }
 
 static int ext4_dx_csum_verify(struct inode *inode,

@@ -874,22 +874,25 @@ static int ext4_fc_write_inode(struct in
 	tl.fc_tag = cpu_to_le16(EXT4_FC_TAG_INODE);
 	tl.fc_len = cpu_to_le16(inode_len + sizeof(fc_inode.fc_ino));
 
+	ret = -ECANCELED;
 	dst = ext4_fc_reserve_space(inode->i_sb,
 			sizeof(tl) + inode_len + sizeof(fc_inode.fc_ino), crc);
 	if (!dst)
-		return -ECANCELED;
+		goto err;
 
 	if (!ext4_fc_memcpy(inode->i_sb, dst, &tl, sizeof(tl), crc))
-		return -ECANCELED;
+		goto err;
 	dst += sizeof(tl);
 	if (!ext4_fc_memcpy(inode->i_sb, dst, &fc_inode, sizeof(fc_inode), crc))
-		return -ECANCELED;
+		goto err;
 	dst += sizeof(fc_inode);
 	if (!ext4_fc_memcpy(inode->i_sb, dst, (u8 *)ext4_raw_inode(&iloc),
 					inode_len, crc))
-		return -ECANCELED;
-
-	return 0;
+		goto err;
+	ret = 0;
+err:
+	brelse(iloc.bh);
+	return ret;
 }
 
 /*

@@ -1494,13 +1494,15 @@ static int ext4_fc_record_modified_inode
 		if (state->fc_modified_inodes[i] == ino)
 			return 0;
 	if (state->fc_modified_inodes_used == state->fc_modified_inodes_size) {
-		state->fc_modified_inodes = krealloc(
-				state->fc_modified_inodes,
+		int *fc_modified_inodes;
+
+		fc_modified_inodes = krealloc(state->fc_modified_inodes,
 				sizeof(int) * (state->fc_modified_inodes_size +
 				EXT4_FC_REPLAY_REALLOC_INCREMENT),
 				GFP_KERNEL);
-		if (!state->fc_modified_inodes)
+		if (!fc_modified_inodes)
 			return -ENOMEM;
+		state->fc_modified_inodes = fc_modified_inodes;
 		state->fc_modified_inodes_size +=
 			EXT4_FC_REPLAY_REALLOC_INCREMENT;
 	}

@@ -1687,15 +1687,17 @@ int ext4_fc_record_regions(struct super_
 	if (replay && state->fc_regions_used != state->fc_regions_valid)
 		state->fc_regions_used = state->fc_regions_valid;
 	if (state->fc_regions_used == state->fc_regions_size) {
+		struct ext4_fc_alloc_region *fc_regions;
+
 		state->fc_regions_size +=
 			EXT4_FC_REPLAY_REALLOC_INCREMENT;
-		state->fc_regions = krealloc(
-					state->fc_regions,
-					state->fc_regions_size *
-					sizeof(struct ext4_fc_alloc_region),
-					GFP_KERNEL);
-		if (!state->fc_regions)
+		fc_regions = krealloc(state->fc_regions,
+				      state->fc_regions_size *
+				      sizeof(struct ext4_fc_alloc_region),
+				      GFP_KERNEL);
+		if (!fc_regions)
 			return -ENOMEM;
+		state->fc_regions = fc_regions;
 	}
 	region = &state->fc_regions[state->fc_regions_used++];
 	region->ino = ino;

@@ -1689,14 +1689,15 @@ int ext4_fc_record_regions(struct super_
 	if (state->fc_regions_used == state->fc_regions_size) {
 		struct ext4_fc_alloc_region *fc_regions;
 
-		state->fc_regions_size +=
-			EXT4_FC_REPLAY_REALLOC_INCREMENT;
 		fc_regions = krealloc(state->fc_regions,
-				      state->fc_regions_size *
-				      sizeof(struct ext4_fc_alloc_region),
+				      sizeof(struct ext4_fc_alloc_region) *
+				      (state->fc_regions_size +
+				       EXT4_FC_REPLAY_REALLOC_INCREMENT),
 				      GFP_KERNEL);
 		if (!fc_regions)
 			return -ENOMEM;
+		state->fc_regions_size +=
+			EXT4_FC_REPLAY_REALLOC_INCREMENT;
 		state->fc_regions = fc_regions;
 	}
 	region = &state->fc_regions[state->fc_regions_used++];

@@ -610,9 +610,23 @@ void klp_reverse_transition(void)
 /* Called from copy_process() during fork */
 void klp_copy_process(struct task_struct *child)
 {
-	child->patch_state = current->patch_state;
 
-	/* TIF_PATCH_PENDING gets copied in setup_thread_stack() */
+	/*
+	 * The parent process may have gone through a KLP transition since
+	 * the thread flag was copied in setup_thread_stack earlier. Bring
+	 * the task flag up to date with the parent here.
+	 *
+	 * The operation is serialized against all klp_*_transition()
+	 * operations by the tasklist_lock. The only exception is
+	 * klp_update_patch_state(current), but we cannot race with
+	 * that because we are current.
+	 */
+	if (test_tsk_thread_flag(current, TIF_PATCH_PENDING))
+		set_tsk_thread_flag(child, TIF_PATCH_PENDING);
+	else
+		clear_tsk_thread_flag(child, TIF_PATCH_PENDING);
+
+	child->patch_state = current->patch_state;
 }
 
 /*

@@ -6081,8 +6081,12 @@ int ftrace_regex_release(struct inode *i
 
 		if (filter_hash) {
 			orig_hash = &iter->ops->func_hash->filter_hash;
-			if (iter->tr && !list_empty(&iter->tr->mod_trace))
-				iter->hash->flags |= FTRACE_HASH_FL_MOD;
+			if (iter->tr) {
+				if (list_empty(&iter->tr->mod_trace))
+					iter->hash->flags &= ~FTRACE_HASH_FL_MOD;
+				else
+					iter->hash->flags |= FTRACE_HASH_FL_MOD;
+			}
 		} else
 			orig_hash = &iter->ops->func_hash->notrace_hash;
 

@@ -1644,6 +1644,18 @@ ftrace_find_tramp_ops_any_other(struct d
 static struct ftrace_ops *
 ftrace_find_tramp_ops_next(struct dyn_ftrace *rec, struct ftrace_ops *ops);
 
+static bool skip_record(struct dyn_ftrace *rec)
+{
+	/*
+	 * At boot up, weak functions are set to disable. Function tracing
+	 * can be enabled before they are, and they still need to be disabled now.
+	 * If the record is disabled, still continue if it is marked as already
+	 * enabled (this is needed to keep the accounting working).
+	 */
+	return rec->flags & FTRACE_FL_DISABLED &&
+		!(rec->flags & FTRACE_FL_ENABLED);
+}
+
 static bool __ftrace_hash_rec_update(struct ftrace_ops *ops,
 				     int filter_hash,
 				     bool inc)

@@ -1693,7 +1705,7 @@ static bool __ftrace_hash_rec_update(str
 		int in_hash = 0;
 		int match = 0;
 
-		if (rec->flags & FTRACE_FL_DISABLED)
+		if (skip_record(rec))
 			continue;
 
 		if (all) {

@@ -2126,7 +2138,7 @@ static int ftrace_check_record(struct dy
 
 	ftrace_bug_type = FTRACE_BUG_UNKNOWN;
 
-	if (rec->flags & FTRACE_FL_DISABLED)
+	if (skip_record(rec))
 		return FTRACE_UPDATE_IGNORE;
 
 	/*

@@ -2241,7 +2253,7 @@ static int ftrace_check_record(struct dy
 	if (update) {
 		/* If there's no more users, clear all flags */
 		if (!ftrace_rec_count(rec))
-			rec->flags = 0;
+			rec->flags &= FTRACE_FL_DISABLED;
 		else
 			/*
 			 * Just disable the record, but keep the ops TRAMP

@@ -2634,7 +2646,7 @@ void __weak ftrace_replace_code(int mod_
 
 	do_for_each_ftrace_rec(pg, rec) {
 
-		if (rec->flags & FTRACE_FL_DISABLED)
+		if (skip_record(rec))
 			continue;
 
 		failed = __ftrace_replace_code(rec, enable);

@@ -5616,7 +5616,15 @@ int ring_buffer_read_page(struct trace_b
 		unsigned int pos = 0;
 		unsigned int size;
 
-		if (full)
+		/*
+		 * If a full page is expected, this can still be returned
+		 * if there's been a previous partial read and the
+		 * rest of the page can be read and the commit page is off
+		 * the reader page.
+		 */
+		if (full &&
+		    (!read || (len < (commit - read)) ||
+		     cpu_buffer->reader_page == cpu_buffer->commit_page))
 			goto out_unlock;
 
 		if (len > (commit - read))

@@ -1011,7 +1011,7 @@ int ring_buffer_wait(struct trace_buffer
 			nr_pages = cpu_buffer->nr_pages;
 			dirty = ring_buffer_nr_dirty_pages(buffer, cpu);
 			if (!cpu_buffer->shortest_full ||
-			    cpu_buffer->shortest_full < full)
+			    cpu_buffer->shortest_full > full)
 				cpu_buffer->shortest_full = full;
 			raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
 			if (!pagebusy &&

@@ -917,8 +917,9 @@ static void rb_wake_up_waiters(struct ir
 	struct rb_irq_work *rbwork = container_of(work, struct rb_irq_work, work);
 
 	wake_up_all(&rbwork->waiters);
-	if (rbwork->wakeup_full) {
+	if (rbwork->full_waiters_pending || rbwork->wakeup_full) {
 		rbwork->wakeup_full = false;
+		rbwork->full_waiters_pending = false;
 		wake_up_all(&rbwork->full_waiters);
 	}
 }

@@ -101,7 +101,7 @@ __ring_buffer_alloc(unsigned long size,
 int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full);
 __poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu,
 			  struct file *filp, poll_table *poll_table);
-
+void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu);
 
 #define RING_BUFFER_ALL_CPUS -1
 

@@ -413,6 +413,7 @@ struct rb_irq_work {
 	struct irq_work			work;
 	wait_queue_head_t		waiters;
 	wait_queue_head_t		full_waiters;
+	long				wait_index;
 	bool				waiters_pending;
 	bool				full_waiters_pending;
 	bool				wakeup_full;

@@ -925,6 +926,37 @@ static void rb_wake_up_waiters(struct ir
 }
 
 /**
+ * ring_buffer_wake_waiters - wake up any waiters on this ring buffer
+ * @buffer: The ring buffer to wake waiters on
+ *
+ * In the case of a file that represents a ring buffer is closing,
+ * it is prudent to wake up any waiters that are on this.
+ */
+void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu)
+{
+	struct ring_buffer_per_cpu *cpu_buffer;
+	struct rb_irq_work *rbwork;
+
+	if (cpu == RING_BUFFER_ALL_CPUS) {
+
+		/* Wake up individual ones too. One level recursion */
+		for_each_buffer_cpu(buffer, cpu)
+			ring_buffer_wake_waiters(buffer, cpu);
+
+		rbwork = &buffer->irq_work;
+	} else {
+		cpu_buffer = buffer->buffers[cpu];
+		rbwork = &cpu_buffer->irq_work;
+	}
+
+	rbwork->wait_index++;
+	/* make sure the waiters see the new index */
+	smp_wmb();
+
+	rb_wake_up_waiters(&rbwork->work);
+}
+
+/**
  * ring_buffer_wait - wait for input to the ring buffer
  * @buffer: buffer to wait on
  * @cpu: the cpu buffer to wait on

@@ -939,6 +971,7 @@ int ring_buffer_wait(struct trace_buffer
 	struct ring_buffer_per_cpu *cpu_buffer;
 	DEFINE_WAIT(wait);
 	struct rb_irq_work *work;
+	long wait_index;
 	int ret = 0;
 
 	/*

@@ -957,6 +990,7 @@ int ring_buffer_wait(struct trace_buffer
 		work = &cpu_buffer->irq_work;
 	}
 
+	wait_index = READ_ONCE(work->wait_index);
 
 	while (true) {
 		if (full)

@@ -1021,6 +1055,11 @@ int ring_buffer_wait(struct trace_buffer
 		}
 
 		schedule();
+
+		/* Make sure to see the new wait index */
+		smp_rmb();
+		if (wait_index != work->wait_index)
+			break;
 	}
 
 	if (full)

@@ -2648,6 +2648,9 @@ rb_reset_tail(struct ring_buffer_per_cpu
 		/* Mark the rest of the page with padding */
 		rb_event_set_padding(event);
 
+		/* Make sure the padding is visible before the write update */
+		smp_wmb();
+
 		/* Set the write back to the previous setting */
 		local_sub(length, &tail_page->write);
 		return;

@@ -2659,6 +2662,9 @@ rb_reset_tail(struct ring_buffer_per_cpu
 	/* time delta must be non zero */
 	event->time_delta = 1;
 
+	/* Make sure the padding is visible before the tail_page->write update */
+	smp_wmb();
+
 	/* Set write to end of buffer */
 	length = (tail + length) - BUF_PAGE_SIZE;
 	local_sub(length, &tail_page->write);

@@ -4627,6 +4633,33 @@ rb_get_reader_page(struct ring_buffer_pe
 	arch_spin_unlock(&cpu_buffer->lock);
 	local_irq_restore(flags);
 
+	/*
+	 * The writer has preempt disable, wait for it. But not forever
+	 * Although, 1 second is pretty much "forever"
+	 */
+#define USECS_WAIT	1000000
+        for (nr_loops = 0; nr_loops < USECS_WAIT; nr_loops++) {
+		/* If the write is past the end of page, a writer is still updating it */
+		if (likely(!reader || rb_page_write(reader) <= BUF_PAGE_SIZE))
+			break;
+
+		udelay(1);
+
+		/* Get the latest version of the reader write value */
+		smp_rmb();
+	}
+
+	/* The writer is not moving forward? Something is wrong */
+	if (RB_WARN_ON(cpu_buffer, nr_loops == USECS_WAIT))
+		reader = NULL;
+
+	/*
+	 * Make sure we see any padding after the write update
+	 * (see rb_reset_tail())
+	 */
+	smp_rmb();
+
+
 	return reader;
 }
 

@@ -968,8 +968,7 @@ static int __trace_eprobe_create(int arg
 	}
 
 	if (!event) {
-		strscpy(buf1, argv[1], MAX_EVENT_NAME_LEN);
-		sanitize_event_name(buf1);
+		strscpy(buf1, sys_event, MAX_EVENT_NAME_LEN);
 		event = buf1;
 	}
 

@@ -1193,12 +1193,14 @@ void *tracing_cond_snapshot_data(struct
 {
 	void *cond_data = NULL;
 
+	local_irq_disable();
 	arch_spin_lock(&tr->max_lock);
 
 	if (tr->cond_snapshot)
 		cond_data = tr->cond_snapshot->cond_data;
 
 	arch_spin_unlock(&tr->max_lock);
+	local_irq_enable();
 
 	return cond_data;
 }

@@ -1334,9 +1336,11 @@ int tracing_snapshot_cond_enable(struct
 		goto fail_unlock;
 	}
 
+	local_irq_disable();
 	arch_spin_lock(&tr->max_lock);
 	tr->cond_snapshot = cond_snapshot;
 	arch_spin_unlock(&tr->max_lock);
+	local_irq_enable();
 
 	mutex_unlock(&trace_types_lock);
 

@@ -1363,6 +1367,7 @@ int tracing_snapshot_cond_disable(struct
 {
 	int ret = 0;
 
+	local_irq_disable();
 	arch_spin_lock(&tr->max_lock);
 
 	if (!tr->cond_snapshot)

@@ -1373,6 +1378,7 @@ int tracing_snapshot_cond_disable(struct
 	}
 
 	arch_spin_unlock(&tr->max_lock);
+	local_irq_enable();
 
 	return ret;
 }

@@ -2200,6 +2206,11 @@ static size_t tgid_map_max;
 
 #define SAVED_CMDLINES_DEFAULT 128
 #define NO_CMDLINE_MAP UINT_MAX
+/*
+ * Preemption must be disabled before acquiring trace_cmdline_lock.
+ * The various trace_arrays' max_lock must be acquired in a context
+ * where interrupt is disabled.
+ */
 static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
 struct saved_cmdlines_buffer {
 	unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];

@@ -2412,7 +2423,11 @@ static int trace_save_cmdline(struct tas
 	 * the lock, but we also don't want to spin
 	 * nor do we want to disable interrupts,
 	 * so if we miss here, then better luck next time.
+	 *
+	 * This is called within the scheduler and wake up, so interrupts
+	 * had better been disabled and run queue lock been held.
 	 */
+	lockdep_assert_preemption_disabled();
 	if (!arch_spin_trylock(&trace_cmdline_lock))
 		return 0;
 

@@ -5890,9 +5905,11 @@ tracing_saved_cmdlines_size_read(struct
 	char buf[64];
 	int r;
 
+	preempt_disable();
 	arch_spin_lock(&trace_cmdline_lock);
 	r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num);
 	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
 
 	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
 }

@@ -5917,10 +5934,12 @@ static int tracing_resize_saved_cmdlines
 		return -ENOMEM;
 	}
 
+	preempt_disable();
 	arch_spin_lock(&trace_cmdline_lock);
 	savedcmd_temp = savedcmd;
 	savedcmd = s;
 	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
 	free_saved_cmdlines_buffer(savedcmd_temp);
 
 	return 0;

@@ -6373,10 +6392,12 @@ int tracing_set_tracer(struct trace_arra
 
 #ifdef CONFIG_TRACER_SNAPSHOT
 	if (t->use_max_tr) {
+		local_irq_disable();
 		arch_spin_lock(&tr->max_lock);
 		if (tr->cond_snapshot)
 			ret = -EBUSY;
 		arch_spin_unlock(&tr->max_lock);
+		local_irq_enable();
 		if (ret)
 			goto out;
 	}

@@ -7436,10 +7457,12 @@ tracing_snapshot_write(struct file *filp
 		goto out;
 	}
 
+	local_irq_disable();
 	arch_spin_lock(&tr->max_lock);
 	if (tr->cond_snapshot)
 		ret = -EBUSY;
 	arch_spin_unlock(&tr->max_lock);
+	local_irq_enable();
 	if (ret)
 		goto out;
 

@@ -92,6 +92,7 @@ struct trace_iterator {
 	unsigned int		temp_size;
 	char			*fmt;	/* modified format holder */
 	unsigned int		fmt_size;
+	long			wait_index;
 
 	/* trace_seq for __print_flags() and __print_symbolic() etc. */
 	struct trace_seq	tmp_seq;

@@ -8160,6 +8160,12 @@ static int tracing_buffers_release(struc
 
 	__trace_array_put(iter->tr);
 
+	iter->wait_index++;
+	/* Make sure the waiters see the new wait_index */
+	smp_wmb();
+
+	ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file);
+
 	if (info->spare)
 		ring_buffer_free_read_page(iter->array_buffer->buffer,
 					   info->spare_cpu, info->spare);

@@ -8313,6 +8319,8 @@ tracing_buffers_splice_read(struct file
 
 	/* did we read anything? */
 	if (!spd.nr_pages) {
+		long wait_index;
+
 		if (ret)
 			goto out;
 

@@ -8320,10 +8328,17 @@ tracing_buffers_splice_read(struct file
 		if ((file->f_flags & O_NONBLOCK) || (flags & SPLICE_F_NONBLOCK))
 			goto out;
 
+		wait_index = READ_ONCE(iter->wait_index);
+
 		ret = wait_on_pipe(iter, iter->tr->buffer_percent);
 		if (ret)
 			goto out;
 
+		/* Make sure we see the new wait_index */
+		smp_rmb();
+		if (wait_index != iter->wait_index)
+			goto out;
+
 		goto again;
 	}
 

@@ -8334,6 +8334,10 @@ tracing_buffers_splice_read(struct file
 		if (ret)
 			goto out;
 
+		/* No need to wait after waking up when tracing is off */
+		if (!tracer_tracing_is_on(iter->tr))
+			goto out;
+
 		/* Make sure we see the new wait_index */
 		smp_rmb();
 		if (wait_index != iter->wait_index)

@@ -9043,6 +9047,8 @@ rb_simple_write(struct file *filp, const
 			tracer_tracing_off(tr);
 			if (tr->current_trace->stop)
 				tr->current_trace->stop(tr);
+			/* Wake up any waiters */
+			ring_buffer_wake_waiters(buffer, RING_BUFFER_ALL_CPUS);
 		}
 		mutex_unlock(&trace_types_lock);
 	}

@@ -8353,12 +8353,34 @@ out:
 	return ret;
 }
 
+/* An ioctl call with cmd 0 to the ring buffer file will wake up all waiters */
+static long tracing_buffers_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct ftrace_buffer_info *info = file->private_data;
+	struct trace_iterator *iter = &info->iter;
+
+	if (cmd)
+		return -ENOIOCTLCMD;
+
+	mutex_lock(&trace_types_lock);
+
+	iter->wait_index++;
+	/* Make sure the waiters see the new wait_index */
+	smp_wmb();
+
+	ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file);
+
+	mutex_unlock(&trace_types_lock);
+	return 0;
+}
+
 static const struct file_operations tracing_buffers_fops = {
 	.open		= tracing_buffers_open,
 	.read		= tracing_buffers_read,
 	.poll		= tracing_buffers_poll,
 	.release	= tracing_buffers_release,
 	.splice_read	= tracing_buffers_splice_read,
+	.unlocked_ioctl = tracing_buffers_ioctl,
 	.llseek		= no_llseek,
 };
 

@@ -6428,12 +6428,12 @@ int tracing_set_tracer(struct trace_arra
 	if (tr->current_trace->reset)
 		tr->current_trace->reset(tr);
 
+#ifdef CONFIG_TRACER_MAX_TRACE
+	had_max_tr = tr->current_trace->use_max_tr;
+
 	/* Current trace needs to be nop_trace before synchronize_rcu */
 	tr->current_trace = &nop_trace;
 
-#ifdef CONFIG_TRACER_MAX_TRACE
-	had_max_tr = tr->allocated_snapshot;
-
 	if (had_max_tr && !t->use_max_tr) {
 		/*
 		 * We need to make sure that the update_max_tr sees that

@@ -6446,11 +6446,13 @@ int tracing_set_tracer(struct trace_arra
 		free_snapshot(tr);
 	}
 
-	if (t->use_max_tr && !had_max_tr) {
+	if (t->use_max_tr && !tr->allocated_snapshot) {
 		ret = tracing_alloc_snapshot_instance(tr);
 		if (ret < 0)
 			goto out;
 	}
+#else
+	tr->current_trace = &nop_trace;
 #endif
 
 	if (t->init) {

@@ -16,6 +16,7 @@
 #include "trace_dynevent.h"
 #include "trace_probe.h"
 #include "trace_probe_tmpl.h"
+#include "trace_probe_kernel.h"
 
 #define EPROBE_EVENT_SYSTEM "eprobes"
 

@@ -453,29 +454,14 @@ NOKPROBE_SYMBOL(process_fetch_insn)
 static nokprobe_inline int
 fetch_store_strlen_user(unsigned long addr)
 {
-	const void __user *uaddr =  (__force const void __user *)addr;
-
-	return strnlen_user_nofault(uaddr, MAX_STRING_SIZE);
+	return kern_fetch_store_strlen_user(addr);
 }
 
 /* Return the length of string -- including null terminal byte */
 static nokprobe_inline int
 fetch_store_strlen(unsigned long addr)
 {
-	int ret, len = 0;
-	u8 c;
-
-#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
-	if (addr < TASK_SIZE)
-		return fetch_store_strlen_user(addr);
-#endif
-
-	do {
-		ret = copy_from_kernel_nofault(&c, (u8 *)addr + len, 1);
-		len++;
-	} while (c && ret == 0 && len < MAX_STRING_SIZE);
-
-	return (ret < 0) ? ret : len;
+	return kern_fetch_store_strlen(addr);
 }
 
 /*

@@ -485,21 +471,7 @@ fetch_store_strlen(unsigned long addr)
 static nokprobe_inline int
 fetch_store_string_user(unsigned long addr, void *dest, void *base)
 {
-	const void __user *uaddr =  (__force const void __user *)addr;
-	int maxlen = get_loc_len(*(u32 *)dest);
-	void *__dest;
-	long ret;
-
-	if (unlikely(!maxlen))
-		return -ENOMEM;
-
-	__dest = get_loc_data(dest, base);
-
-	ret = strncpy_from_user_nofault(__dest, uaddr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
-
-	return ret;
+	return kern_fetch_store_string_user(addr, dest, base);
 }
 
 /*

@@ -509,29 +481,7 @@ fetch_store_string_user(unsigned long ad
 static nokprobe_inline int
 fetch_store_string(unsigned long addr, void *dest, void *base)
 {
-	int maxlen = get_loc_len(*(u32 *)dest);
-	void *__dest;
-	long ret;
-
-#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
-	if ((unsigned long)addr < TASK_SIZE)
-		return fetch_store_string_user(addr, dest, base);
-#endif
-
-	if (unlikely(!maxlen))
-		return -ENOMEM;
-
-	__dest = get_loc_data(dest, base);
-
-	/*
-	 * Try to get string again, since the string can be changed while
-	 * probing.
-	 */
-	ret = strncpy_from_kernel_nofault(__dest, (void *)addr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
-
-	return ret;
+	return kern_fetch_store_string(addr, dest, base);
 }
 
 static nokprobe_inline int

90 lines not on this page · raw message

96 lines not on this page · raw message

@@ -2,6 +2,8 @@
 #ifndef __TRACE_PROBE_KERNEL_H_
 #define __TRACE_PROBE_KERNEL_H_
 
+#define FAULT_STRING "(fault)"
+
 /*
  * This depends on trace_probe.h, but can not include it due to
  * the way trace_probe_tmpl.h is used by trace_kprobe.c and trace_eprobe.c.

@@ -13,8 +15,16 @@ static nokprobe_inline int
 kern_fetch_store_strlen_user(unsigned long addr)
 {
 	const void __user *uaddr =  (__force const void __user *)addr;
+	int ret;
 
-	return strnlen_user_nofault(uaddr, MAX_STRING_SIZE);
+	ret = strnlen_user_nofault(uaddr, MAX_STRING_SIZE);
+	/*
+	 * strnlen_user_nofault returns zero on fault, insert the
+	 * FAULT_STRING when that occurs.
+	 */
+	if (ret <= 0)
+		return strlen(FAULT_STRING) + 1;
+	return ret;
 }
 
 /* Return the length of string -- including null terminal byte */

@@ -34,7 +44,18 @@ kern_fetch_store_strlen(unsigned long ad
 		len++;
 	} while (c && ret == 0 && len < MAX_STRING_SIZE);
 
-	return (ret < 0) ? ret : len;
+	/* For faults, return enough to hold the FAULT_STRING */
+	return (ret < 0) ? strlen(FAULT_STRING) + 1 : len;
+}
+
+static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void *base, int len)
+{
+	if (ret >= 0) {
+		*(u32 *)dest = make_data_loc(ret, __dest - base);
+	} else {
+		strscpy(__dest, FAULT_STRING, len);
+		ret = strlen(__dest) + 1;
+	}
 }
 
 /*

@@ -55,8 +76,7 @@ kern_fetch_store_string_user(unsigned lo
 	__dest = get_loc_data(dest, base);
 
 	ret = strncpy_from_user_nofault(__dest, uaddr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
+	set_data_loc(ret, dest, __dest, base, maxlen);
 
 	return ret;
 }

@@ -87,8 +107,7 @@ kern_fetch_store_string(unsigned long ad
 	 * probing.
 	 */
 	ret = strncpy_from_kernel_nofault(__dest, (void *)addr, maxlen);
-	if (ret >= 0)
-		*(u32 *)dest = make_data_loc(ret, __dest - base);
+	set_data_loc(ret, dest, __dest, base, maxlen);
 
 	return ret;
 }

@@ -17,6 +17,8 @@
 /* for gfp flag names */
 #include <linux/trace_events.h>
 #include <trace/events/mmflags.h>
+#include "trace_probe.h"
+#include "trace_probe_kernel.h"
 
 #include "trace_synth.h"
 

@@ -409,6 +411,7 @@ static unsigned int trace_string(struct
 {
 	unsigned int len = 0;
 	char *str_field;
+	int ret;
 
 	if (is_dynamic) {
 		u32 data_offset;

@@ -417,19 +420,27 @@ static unsigned int trace_string(struct
 		data_offset += event->n_u64 * sizeof(u64);
 		data_offset += data_size;
 
-		str_field = (char *)entry + data_offset;
-
-		len = strlen(str_val) + 1;
-		strscpy(str_field, str_val, len);
+		len = kern_fetch_store_strlen((unsigned long)str_val);
 
 		data_offset |= len << 16;
 		*(u32 *)&entry->fields[*n_u64] = data_offset;
 
+		ret = kern_fetch_store_string((unsigned long)str_val, &entry->fields[*n_u64], entry);
+
 		(*n_u64)++;
 	} else {
 		str_field = (char *)&entry->fields[*n_u64];
 
-		strscpy(str_field, str_val, STR_VAR_LEN_MAX);
+#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
+		if ((unsigned long)str_val < TASK_SIZE)
+			ret = strncpy_from_user_nofault(str_field, str_val, STR_VAR_LEN_MAX);
+		else
+#endif
+			ret = strncpy_from_kernel_nofault(str_field, str_val, STR_VAR_LEN_MAX);
+
+		if (ret < 0)
+			strcpy(str_field, FAULT_STRING);
+
 		(*n_u64) += STR_VAR_LEN_MAX / sizeof(u64);
 	}
 

@@ -462,7 +473,7 @@ static notrace void trace_event_raw_even
 		val_idx = var_ref_idx[field_pos];
 		str_val = (char *)(long)var_ref_vals[val_idx];
 
-		len = strlen(str_val) + 1;
+		len = kern_fetch_store_strlen((unsigned long)str_val);
 
 		fields_size += len;
 	}

@@ -76,7 +76,9 @@ int rpmsg_chrdev_eptdev_destroy(struct d
 
 	mutex_lock(&eptdev->ept_lock);
 	if (eptdev->ept) {
-		rpmsg_destroy_ept(eptdev->ept);
+		/* The default endpoint is released by the rpmsg core */
+		if (!eptdev->default_ept)
+			rpmsg_destroy_ept(eptdev->ept);
 		eptdev->ept = NULL;
 	}
 	mutex_unlock(&eptdev->ept_lock);

@@ -2822,6 +2822,26 @@ static void tb_switch_credits_init(struc
 		tb_sw_info(sw, "failed to determine preferred buffer allocation, using defaults\n");
 }
 
+static int tb_switch_port_hotplug_enable(struct tb_switch *sw)
+{
+	struct tb_port *port;
+
+	if (tb_switch_is_icm(sw))
+		return 0;
+
+	tb_switch_for_each_port(sw, port) {
+		int res;
+
+		if (!port->cap_usb4)
+			continue;
+
+		res = usb4_port_hotplug_enable(port);
+		if (res)
+			return res;
+	}
+	return 0;
+}
+
 /**
  * tb_switch_add() - Add a switch to the domain
  * @sw: Switch to add

@@ -2891,6 +2911,10 @@ int tb_switch_add(struct tb_switch *sw)
 			return ret;
 	}
 
+	ret = tb_switch_port_hotplug_enable(sw);
+	if (ret)
+		return ret;
+
 	ret = device_add(&sw->dev);
 	if (ret) {
 		dev_err(&sw->dev, "failed to add device: %d\n", ret);

@@ -1174,6 +1174,7 @@ int usb4_switch_add_ports(struct tb_swit
 void usb4_switch_remove_ports(struct tb_switch *sw);
 
 int usb4_port_unlock(struct tb_port *port);
+int usb4_port_hotplug_enable(struct tb_port *port);
 int usb4_port_configure(struct tb_port *port);
 void usb4_port_unconfigure(struct tb_port *port);
 int usb4_port_configure_xdomain(struct tb_port *port);

@@ -308,6 +308,7 @@ struct tb_regs_port_header {
 #define ADP_CS_5				0x05
 #define ADP_CS_5_LCA_MASK			GENMASK(28, 22)
 #define ADP_CS_5_LCA_SHIFT			22
+#define ADP_CS_5_DHP				BIT(31)
 
 /* TMU adapter registers */
 #define TMU_ADP_CS_3				0x03

@@ -1046,6 +1046,26 @@ int usb4_port_unlock(struct tb_port *por
 	return tb_port_write(port, &val, TB_CFG_PORT, ADP_CS_4, 1);
 }
 
+/**
+ * usb4_port_hotplug_enable() - Enables hotplug for a port
+ * @port: USB4 port to operate on
+ *
+ * Enables hot plug events on a given port. This is only intended
+ * to be used on lane, DP-IN, and DP-OUT adapters.
+ */
+int usb4_port_hotplug_enable(struct tb_port *port)
+{
+	int ret;
+	u32 val;
+
+	ret = tb_port_read(port, &val, TB_CFG_PORT, ADP_CS_5, 1);
+	if (ret)
+		return ret;
+
+	val &= ~ADP_CS_5_DHP;
+	return tb_port_write(port, &val, TB_CFG_PORT, ADP_CS_5, 1);
+}
+
 static int usb4_port_set_configured(struct tb_port *port, bool configured)
 {
 	int ret;

@@ -280,14 +280,6 @@ efi_status_t allocate_new_fdt_and_exit_b
 		goto fail;
 	}
 
-	/*
-	 * Now that we have done our final memory allocation (and free)
-	 * we can get the memory map key needed for exit_boot_services().
-	 */
-	status = efi_get_memory_map(&map);
-	if (status != EFI_SUCCESS)
-		goto fail_free_new_fdt;
-
 	status = update_fdt((void *)fdt_addr, fdt_size,
 			    (void *)*new_fdt_addr, MAX_FDT_SIZE, cmdline_ptr,
 			    initrd_addr, initrd_size);

@@ -106,11 +106,11 @@ void cedrus_device_run(void *priv)
 
 	/* Trigger decoding if setup went well, bail out otherwise. */
 	if (!error) {
-		dev->dec_ops[ctx->current_codec]->trigger(ctx);
-
 		/* Start the watchdog timer. */
 		schedule_delayed_work(&dev->watchdog_work,
 				      msecs_to_jiffies(2000));
+
+		dev->dec_ops[ctx->current_codec]->trigger(ctx);
 	} else {
 		v4l2_m2m_buf_done_and_job_finish(ctx->dev->m2m_dev,
 						 ctx->fh.m2m_ctx,

@@ -448,6 +448,8 @@ static int cedrus_probe(struct platform_
 	if (!dev)
 		return -ENOMEM;
 
+	platform_set_drvdata(pdev, dev);
+
 	dev->vfd = cedrus_video_device;
 	dev->dev = &pdev->dev;
 	dev->pdev = pdev;