Update Linux to v5.4.148

Sourced from [1]

[1] https://cdn.kernel.org/pub/linux/kernel/v5.x/linux-5.4.148.tar.gz

Change-Id: Ib3d26c5ba9b022e2e03533005c4fed4d7c30b61b
Signed-off-by: Olivier Deprez <olivier.deprez@arm.com>
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 49cb26f..5273965 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -32,6 +32,7 @@
 #include "block-rsv.h"
 #include "delalloc-space.h"
 #include "block-group.h"
+#include "rcu-string.h"
 
 #undef SCRAMBLE_DELAYED_REFS
 
@@ -402,12 +403,11 @@
 			if (type == BTRFS_SHARED_BLOCK_REF_KEY) {
 				ASSERT(eb->fs_info);
 				/*
-				 * Every shared one has parent tree
-				 * block, which must be aligned to
-				 * nodesize.
+				 * Every shared one has parent tree block,
+				 * which must be aligned to sector size.
 				 */
 				if (offset &&
-				    IS_ALIGNED(offset, eb->fs_info->nodesize))
+				    IS_ALIGNED(offset, eb->fs_info->sectorsize))
 					return type;
 			}
 		} else if (is_data == BTRFS_REF_TYPE_DATA) {
@@ -416,12 +416,11 @@
 			if (type == BTRFS_SHARED_DATA_REF_KEY) {
 				ASSERT(eb->fs_info);
 				/*
-				 * Every shared one has parent tree
-				 * block, which must be aligned to
-				 * nodesize.
+				 * Every shared one has parent tree block,
+				 * which must be aligned to sector size.
 				 */
 				if (offset &&
-				    IS_ALIGNED(offset, eb->fs_info->nodesize))
+				    IS_ALIGNED(offset, eb->fs_info->sectorsize))
 					return type;
 			}
 		} else {
@@ -431,8 +430,9 @@
 	}
 
 	btrfs_print_leaf((struct extent_buffer *)eb);
-	btrfs_err(eb->fs_info, "eb %llu invalid extent inline ref type %d",
-		  eb->start, type);
+	btrfs_err(eb->fs_info,
+		  "eb %llu iref 0x%lx invalid extent inline ref type %d",
+		  eb->start, (unsigned long)iref, type);
 	WARN_ON(1);
 
 	return BTRFS_REF_TYPE_INVALID;
@@ -1306,8 +1306,10 @@
 int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 			 u64 num_bytes, u64 *actual_bytes)
 {
-	int ret;
+	int ret = 0;
 	u64 discarded_bytes = 0;
+	u64 end = bytenr + num_bytes;
+	u64 cur = bytenr;
 	struct btrfs_bio *bbio = NULL;
 
 
@@ -1316,35 +1318,56 @@
 	 * associated to its stripes that don't go away while we are discarding.
 	 */
 	btrfs_bio_counter_inc_blocked(fs_info);
-	/* Tell the block device(s) that the sectors can be discarded */
-	ret = btrfs_map_block(fs_info, BTRFS_MAP_DISCARD, bytenr, &num_bytes,
-			      &bbio, 0);
-	/* Error condition is -ENOMEM */
-	if (!ret) {
-		struct btrfs_bio_stripe *stripe = bbio->stripes;
+	while (cur < end) {
+		struct btrfs_bio_stripe *stripe;
 		int i;
 
+		num_bytes = end - cur;
+		/* Tell the block device(s) that the sectors can be discarded */
+		ret = btrfs_map_block(fs_info, BTRFS_MAP_DISCARD, cur,
+				      &num_bytes, &bbio, 0);
+		/*
+		 * Error can be -ENOMEM, -ENOENT (no such chunk mapping) or
+		 * -EOPNOTSUPP. For any such error, @num_bytes is not updated,
+		 * thus we can't continue anyway.
+		 */
+		if (ret < 0)
+			goto out;
 
+		stripe = bbio->stripes;
 		for (i = 0; i < bbio->num_stripes; i++, stripe++) {
 			u64 bytes;
 			struct request_queue *req_q;
+			struct btrfs_device *device = stripe->dev;
 
-			if (!stripe->dev->bdev) {
+			if (!device->bdev) {
 				ASSERT(btrfs_test_opt(fs_info, DEGRADED));
 				continue;
 			}
-			req_q = bdev_get_queue(stripe->dev->bdev);
+			req_q = bdev_get_queue(device->bdev);
 			if (!blk_queue_discard(req_q))
 				continue;
 
-			ret = btrfs_issue_discard(stripe->dev->bdev,
+			if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
+				continue;
+
+			ret = btrfs_issue_discard(device->bdev,
 						  stripe->physical,
 						  stripe->length,
 						  &bytes);
-			if (!ret)
+			if (!ret) {
 				discarded_bytes += bytes;
-			else if (ret != -EOPNOTSUPP)
-				break; /* Logic errors or -ENOMEM, or -EIO but I don't know how that could happen JDM */
+			} else if (ret != -EOPNOTSUPP) {
+				/*
+				 * Logic errors or -ENOMEM, or -EIO, but
+				 * unlikely to happen.
+				 *
+				 * And since there are two loops, explicitly
+				 * go to out to avoid confusion.
+				 */
+				btrfs_put_bbio(bbio);
+				goto out;
+			}
 
 			/*
 			 * Just in case we get back EOPNOTSUPP for some reason,
@@ -1354,7 +1377,9 @@
 			ret = 0;
 		}
 		btrfs_put_bbio(bbio);
+		cur += num_bytes;
 	}
+out:
 	btrfs_bio_counter_dec(fs_info);
 
 	if (actual_bytes)
@@ -1561,7 +1586,7 @@
 	int err = 0;
 	int metadata = !extent_op->is_data;
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 
 	if (metadata && !btrfs_fs_incompat(fs_info, SKINNY_METADATA))
@@ -1681,7 +1706,7 @@
 {
 	int ret = 0;
 
-	if (trans->aborted) {
+	if (TRANS_ABORTED(trans)) {
 		if (insert_reserved)
 			btrfs_pin_extent(trans->fs_info, node->bytenr,
 					 node->num_bytes, 1);
@@ -1848,8 +1873,8 @@
 		btrfs_pin_extent(fs_info, head->bytenr,
 				 head->num_bytes, 1);
 		if (head->is_data) {
-			ret = btrfs_del_csums(trans, fs_info, head->bytenr,
-					      head->num_bytes);
+			ret = btrfs_del_csums(trans, fs_info->csum_root,
+					      head->bytenr, head->num_bytes);
 		}
 	}
 
@@ -1858,7 +1883,7 @@
 	trace_run_delayed_ref_head(fs_info, head, 0);
 	btrfs_delayed_ref_unlock(head);
 	btrfs_put_delayed_ref_head(head);
-	return 0;
+	return ret;
 }
 
 static struct btrfs_delayed_ref_head *btrfs_obtain_ref_head(
@@ -2169,7 +2194,7 @@
 	int run_all = count == (unsigned long)-1;
 
 	/* We'll clean this up in btrfs_cleanup_transaction */
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 
 	if (test_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags))
@@ -2320,7 +2345,8 @@
 
 static noinline int check_committed_ref(struct btrfs_root *root,
 					struct btrfs_path *path,
-					u64 objectid, u64 offset, u64 bytenr)
+					u64 objectid, u64 offset, u64 bytenr,
+					bool strict)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_root *extent_root = fs_info->extent_root;
@@ -2362,9 +2388,13 @@
 	    btrfs_extent_inline_ref_size(BTRFS_EXTENT_DATA_REF_KEY))
 		goto out;
 
-	/* If extent created before last snapshot => it's definitely shared */
-	if (btrfs_extent_generation(leaf, ei) <=
-	    btrfs_root_last_snapshot(&root->root_item))
+	/*
+	 * If extent created before last snapshot => it's shared unless the
+	 * snapshot has been deleted. Use the heuristic if strict is false.
+	 */
+	if (!strict &&
+	    (btrfs_extent_generation(leaf, ei) <=
+	     btrfs_root_last_snapshot(&root->root_item)))
 		goto out;
 
 	iref = (struct btrfs_extent_inline_ref *)(ei + 1);
@@ -2389,7 +2419,7 @@
 }
 
 int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
-			  u64 bytenr)
+			  u64 bytenr, bool strict)
 {
 	struct btrfs_path *path;
 	int ret;
@@ -2400,7 +2430,7 @@
 
 	do {
 		ret = check_committed_ref(root, path, objectid,
-					  offset, bytenr);
+					  offset, bytenr, strict);
 		if (ret && ret != -ENOENT)
 			goto out;
 
@@ -2813,10 +2843,10 @@
 		len = cache->key.objectid + cache->key.offset - start;
 		len = min(len, end + 1 - start);
 
-		if (start < cache->last_byte_to_unpin) {
-			len = min(len, cache->last_byte_to_unpin - start);
-			if (return_free_space)
-				btrfs_add_free_space(cache, start, len);
+		if (start < cache->last_byte_to_unpin && return_free_space) {
+			u64 add_len = min(len, cache->last_byte_to_unpin - start);
+
+			btrfs_add_free_space(cache, start, add_len);
 		}
 
 		start += len;
@@ -2892,7 +2922,7 @@
 	else
 		unpin = &fs_info->freed_extents[0];
 
-	while (!trans->aborted) {
+	while (!TRANS_ABORTED(trans)) {
 		struct extent_state *cached_state = NULL;
 
 		mutex_lock(&fs_info->unused_bg_unpin_mutex);
@@ -2924,7 +2954,7 @@
 		u64 trimmed = 0;
 
 		ret = -EROFS;
-		if (!trans->aborted)
+		if (!TRANS_ABORTED(trans))
 			ret = btrfs_discard_extent(fs_info,
 						   block_group->key.objectid,
 						   block_group->key.offset,
@@ -3155,7 +3185,8 @@
 		btrfs_release_path(path);
 
 		if (is_data) {
-			ret = btrfs_del_csums(trans, info, bytenr, num_bytes);
+			ret = btrfs_del_csums(trans, info->csum_root, bytenr,
+					      num_bytes);
 			if (ret) {
 				btrfs_abort_transaction(trans, ret);
 				goto out;
@@ -3774,12 +3805,14 @@
  * |- Push harder to find free extents
  *    |- If not found, re-iterate all block groups
  */
-static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
+static noinline int find_free_extent(struct btrfs_root *root,
 				u64 ram_bytes, u64 num_bytes, u64 empty_size,
 				u64 hint_byte, struct btrfs_key *ins,
 				u64 flags, int delalloc)
 {
+	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret = 0;
+	int cache_block_group_error = 0;
 	struct btrfs_free_cluster *last_ptr = NULL;
 	struct btrfs_block_group_cache *block_group = NULL;
 	struct find_free_extent_ctl ffe_ctl = {0};
@@ -3806,7 +3839,7 @@
 	ins->objectid = 0;
 	ins->offset = 0;
 
-	trace_find_free_extent(fs_info, num_bytes, empty_size, flags);
+	trace_find_free_extent(root, num_bytes, empty_size, flags);
 
 	space_info = btrfs_find_space_info(fs_info, flags);
 	if (!space_info) {
@@ -3939,7 +3972,20 @@
 		if (unlikely(!ffe_ctl.cached)) {
 			ffe_ctl.have_caching_bg = true;
 			ret = btrfs_cache_block_group(block_group, 0);
-			BUG_ON(ret < 0);
+
+			/*
+			 * If we get ENOMEM here or something else we want to
+			 * try other block groups, because it may not be fatal.
+			 * However if we can't find anything else we need to
+			 * save our return here so that we return the actual
+			 * error that caused problems, not ENOSPC.
+			 */
+			if (ret < 0) {
+				if (!cache_block_group_error)
+					cache_block_group_error = ret;
+				ret = 0;
+				goto loop;
+			}
 			ret = 0;
 		}
 
@@ -4026,7 +4072,7 @@
 	if (ret > 0)
 		goto search;
 
-	if (ret == -ENOSPC) {
+	if (ret == -ENOSPC && !cache_block_group_error) {
 		/*
 		 * Use ffe_ctl->total_free_space as fallback if we can't find
 		 * any contiguous hole.
@@ -4037,6 +4083,8 @@
 		space_info->max_extent_size = ffe_ctl.max_extent_size;
 		spin_unlock(&space_info->lock);
 		ins->offset = ffe_ctl.max_extent_size;
+	} else if (ret == -ENOSPC) {
+		ret = cache_block_group_error;
 	}
 	return ret;
 }
@@ -4099,7 +4147,7 @@
 	flags = get_alloc_profile_by_root(root, is_data);
 again:
 	WARN_ON(num_bytes < fs_info->sectorsize);
-	ret = find_free_extent(fs_info, ram_bytes, num_bytes, empty_size,
+	ret = find_free_extent(root, ram_bytes, num_bytes, empty_size,
 			       hint_byte, ins, flags, delalloc);
 	if (!ret && !is_data) {
 		btrfs_dec_block_group_reservations(fs_info, ins->objectid);
@@ -4394,6 +4442,8 @@
 
 	ret = alloc_reserved_file_extent(trans, 0, root_objectid, 0, owner,
 					 offset, ins, 1);
+	if (ret)
+		btrfs_pin_extent(fs_info, ins->objectid, ins->offset, 1);
 	btrfs_put_block_group(block_group);
 	return ret;
 }
@@ -4422,7 +4472,7 @@
 		return ERR_PTR(-EUCLEAN);
 	}
 
-	btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
+	btrfs_set_buffer_lockdep_class(owner, buf, level);
 	btrfs_tree_lock(buf);
 	btrfs_clean_tree_block(buf);
 	clear_bit(EXTENT_BUFFER_STALE, &buf->bflags);
@@ -5202,7 +5252,14 @@
 		goto out;
 	}
 
-	trans = btrfs_start_transaction(tree_root, 0);
+	/*
+	 * Use join to avoid potential EINTR from transaction start. See
+	 * wait_reserve_ticket and the whole reservation callchain.
+	 */
+	if (for_reloc)
+		trans = btrfs_join_transaction(tree_root);
+	else
+		trans = btrfs_start_transaction(tree_root, 0);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
 		goto out_free;
@@ -5337,7 +5394,15 @@
 				goto out_free;
 			}
 
-			trans = btrfs_start_transaction(tree_root, 0);
+		       /*
+			* Use join to avoid potential EINTR from transaction
+			* start. See wait_reserve_ticket and the whole
+			* reservation callchain.
+			*/
+			if (for_reloc)
+				trans = btrfs_join_transaction(tree_root);
+			else
+				trans = btrfs_start_transaction(tree_root, 0);
 			if (IS_ERR(trans)) {
 				err = PTR_ERR(trans);
 				goto out_free;
@@ -5398,8 +5463,6 @@
 	 */
 	if (!for_reloc && !root_dropped)
 		btrfs_add_dead_root(root);
-	if (err && err != -EAGAIN)
-		btrfs_handle_fs_error(fs_info, err, NULL);
 	return err;
 }
 
@@ -5568,6 +5631,19 @@
 					    &start, &end,
 					    CHUNK_TRIMMED | CHUNK_ALLOCATED);
 
+		/* Check if there are any CHUNK_* bits left */
+		if (start > device->total_bytes) {
+			WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
+			btrfs_warn_in_rcu(fs_info,
+"ignoring attempt to trim beyond device size: offset %llu length %llu device %s device size %llu",
+					  start, end - start + 1,
+					  rcu_str_deref(device->name),
+					  device->total_bytes);
+			mutex_unlock(&fs_info->chunk_mutex);
+			ret = 0;
+			break;
+		}
+
 		/* Ensure we skip the reserved area in the first 1M */
 		start = max_t(u64, start, SZ_1M);
 
@@ -5692,6 +5768,9 @@
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	devices = &fs_info->fs_devices->devices;
 	list_for_each_entry(device, devices, dev_list) {
+		if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
+			continue;
+
 		ret = btrfs_trim_free_extents(device, &group_trimmed);
 		if (ret) {
 			dev_failed++;