Cleanup error handling in relocation
cover Cleanup error handling in relocation
Hello,
Relocation is the last place that is not able to handle errors at all, which results in all sorts of lovely panics if you encounter corruptions or IO errors. I'm going to start cleaning up relocation, but before I move code around I want the error handling to be somewhat sane, so I'm not changing behavior and error handling at the same time.
These patches are purely about error handling, there is no behavior changing other than returning errors up the chain properly. There is a lot of room for follow up cleanups, which will happen next. However I wanted to get this series done today and out so we could get it merged ASAP, and then the follow up cleanups can happen later as they are less important and less critical.
The only exception to the above is the patch to add the error injection sites for btrfs_cow_block and btrfs_search_slot, and a lockdep fix that I discovered while running my tests, those are the first two patches in the series.
I tested this with my error injection stress test, where I keep track of all stack traces that have been tested and only inject errors when we have a new stack trace, which means I should have covered all of the various error conditions. With this patchset I'm no longer panicing while stressing the error conditions. Thanks,
Josef
Josef Bacik (42):
btrfs: allow error injection for btrfs_search_slot and btrfs_cow_block
btrfs: fix lockdep splat in btrfs_recover_relocation
btrfs: convert some BUG_ON()'s to ASSERT()'s in do_relocation
btrfs: convert BUG_ON()'s in relocate_tree_block
btrfs: return an error from btrfs_record_root_in_trans
btrfs: handle errors from select_reloc_root()
btrfs: convert BUG_ON()'s in select_reloc_root() to proper errors
btrfs: check record_root_in_trans related failures in
select_reloc_root
btrfs: do proper error handling in record_reloc_root_in_trans
btrfs: handle btrfs_record_root_in_trans failure in
btrfs_rename_exchange
btrfs: handle btrfs_record_root_in_trans failure in btrfs_rename
btrfs: handle btrfs_record_root_in_trans failure in
btrfs_delete_subvolume
btrfs: handle btrfs_record_root_in_trans failure in
btrfs_recover_log_trees
btrfs: handle btrfs_record_root_in_trans failure in create_subvol
btrfs: btrfs: handle btrfs_record_root_in_trans failure in
relocate_tree_block
btrfs: handle btrfs_record_root_in_trans failure in start_transaction
btrfs: handle record_root_in_trans failure in qgroup_account_snapshot
btrfs: handle record_root_in_trans failure in
btrfs_record_root_in_trans
btrfs: handle record_root_in_trans failure in create_pending_snapshot
btrfs: do not panic in __add_reloc_root
btrfs: have proper error handling in btrfs_init_reloc_root
btrfs: do proper error handling in create_reloc_root
btrfs: handle btrfs_update_reloc_root failure in commit_fs_roots
btrfs: change insert_dirty_subvol to return errors
btrfs: handle btrfs_update_reloc_root failure in insert_dirty_subvol
btrfs: handle btrfs_update_reloc_root failure in prepare_to_merge
btrfs: do proper error handling in btrfs_update_reloc_root
btrfs: convert logic BUG_ON()'s in replace_path to ASSERT()'s
btrfs: handle initial btrfs_cow_block error in replace_path
btrfs: handle the loop btrfs_cow_block error in replace_path
btrfs: handle btrfs_search_slot failure in replace_path
btrfs: handle errors in reference count manipulation in replace_path
btrfs: handle extent reference errors in do_relocation
btrfs: check for BTRFS_BLOCK_FLAG_FULL_BACKREF being set improperly
btrfs: remove the extent item sanity checks in relocate_block_group
btrfs: do proper error handling in create_reloc_inode
btrfs: handle __add_reloc_root failure in btrfs_recover_relocation
btrfs: handle __add_reloc_root failure in btrfs_reloc_post_snapshot
btrfs: cleanup error handling in prepare_to_merge
btrfs: handle extent corruption with select_one_root properly
btrfs: do proper error handling in merge_reloc_roots
btrfs: check return value of btrfs_commit_transaction in relocation
fs/btrfs/ctree.c | 2 + fs/btrfs/inode.c | 21 ++- fs/btrfs/ioctl.c | 6 +- fs/btrfs/relocation.c | 372 ++++++++++++++++++++++++++++++---------- fs/btrfs/transaction.c | 37 ++-- fs/btrfs/tree-checker.c | 5 + fs/btrfs/tree-log.c | 8 +- fs/btrfs/volumes.c | 2 + 8 files changed, 342 insertions(+), 111 deletions(-)
Signature · 1 line
2.26.2
PATCH 1/42 btrfs: allow error injection for btrfs_search_slot and btrfs_cow_block
The following patches are going to address error handling in relocation, in order to test those patches I need to be able to inject errors in btrfs_search_slot and btrfs_cow_block, as we call both of these pretty often in different cases during relocation.
--- fs/btrfs/ctree.c | 2 ++ 1 file changed, 2 insertions(+)
fs/btrfs/ctree.c +2 -0
@@ -1493,6 +1493,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,return ret; } +ALLOW_ERROR_INJECTION (btrfs_cow_block,ERRNO );/* * helper function for defrag to decide if two blocks pointed to by a
@@ -2870,6 +2871,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,btrfs_release_path (p);return ret; } +ALLOW_ERROR_INJECTION (btrfs_search_slot,ERRNO );/* * Like btrfs_search_slot, this looks for a key in the given tree. It uses the
Signature · 1 line
2.26.2
Re: [PATCH 01/42] btrfs: allow error injection for btrfs_search_slot and btrfs_cow_block
Quoted 23 lines · Josef Bacik
The following patches are going to address error handling in relocation, in order to test those patches I need to be able to inject errors in
btrfs_search_slotandbtrfs_cow_block, as we call both of these pretty often in different cases during relocation.Signed-off-by Josef Bacik <[email protected]>--- fs/btrfs/ctree.c | 2 ++ 1 file changed, 2 insertions(+)fs/btrfs/ctree.c +1
-0@@ -1493,6 +1493,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,return ret; } +ALLOW_ERROR_INJECTION (btrfs_cow_block,ERRNO );/* * helper function for defrag to decide if two blocks pointed to by a
fs/btrfs/ctree.c int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root, · line 2874
Quoted 4 lines
fs/btrfs/ctree.c +1
-0@@ -2870,6 +2871,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,btrfs_release_path (p);return ret; } +ALLOW_ERROR_INJECTION (btrfs_search_slot,ERRNO );
This concerns me a little.
For error case, wouldn't we also free the path?
But if we just override the error, the path is not freed by anyone, neither caller nor btrfs_search_slot() would free the path.
Or did I miss something?
Thanks,
Qu
Quoted 4 lines
fs/btrfs/ctree.c +0
-0/* * Like btrfs_search_slot, this looks for a key in the given tree. It uses the
Re: [PATCH 01/42] btrfs: allow error injection for btrfs_search_slot and btrfs_cow_block
Quoted 33 lines · Qu Wenruo
Quoted 23 lines · Josef Bacik
The following patches are going to address error handling in relocation, in order to test those patches I need to be able to inject errors in
btrfs_search_slotandbtrfs_cow_block, as we call both of these pretty often in different cases during relocation.Signed-off-by Josef Bacik <[email protected]>--- fs/btrfs/ctree.c | 2 ++ 1 file changed, 2 insertions(+)fs/btrfs/ctree.c +1
-0@@ -1493,6 +1493,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,return ret; } +ALLOW_ERROR_INJECTION (btrfs_cow_block,ERRNO );/* * helper function for defrag to decide if two blocks pointed to by a fs/btrfs/ctree.c int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root, · line 2874
Quoted 4 lines
fs/btrfs/ctree.c +1
-0@@ -2870,6 +2871,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,btrfs_release_path (p);return ret; } +ALLOW_ERROR_INJECTION (btrfs_search_slot,ERRNO );This concerns me a little.
Quoted 6 lines
For error case, wouldn't we also free the path?
But if we just override the error, the path is not freed by anyone, neither caller norbtrfs_search_slot()would free the path.Or did I miss something?
You're missing that the caller is responsible for free'ing the path, failing btrfs_search_slot isn't going to leak anything unless there's a bug with the caller. Thanks,
Josef
PATCH 2/42 btrfs: fix lockdep splat in btrfs_recover_relocation
In reply to Josef BacikWhile testing the error paths of relocation I hit the following lockdep splat
======================================================
WARNING: possible circular locking dependency detected
5.10.0-rc2-btrfs-next-71 #1 Not tainted
------------------------------------------------------
find/324157 is trying to acquire lock:ffff8ebc48d293a0 (btrfs-tree-01#2/3){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x1a0 [btrfs]
but task is already holding lock:ffff8eb9932c5088 (btrfs-tree-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x1a0 [btrfs]
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (btrfs-tree-00){++++}-{3:3}:
lock_acquire+0xd8/0x490
down_write_nested+0x44/0x120
__btrfs_tree_lock+0x27/0x120 [btrfs]
btrfs_search_slot+0x2a3/0xc50 [btrfs]
btrfs_insert_empty_items+0x58/0xa0 [btrfs]
insert_with_overflow+0x44/0x110 [btrfs]
btrfs_insert_xattr_item+0xb8/0x1d0 [btrfs]
btrfs_setxattr+0xd6/0x4c0 [btrfs]
btrfs_setxattr_trans+0x68/0x100 [btrfs]
__vfs_setxattr+0x66/0x80
__vfs_setxattr_noperm+0x70/0x200
vfs_setxattr+0x6b/0x120
setxattr+0x125/0x240
path_setxattr+0xba/0xd0
__x64_sys_setxattr+0x27/0x30
do_syscall_64+0x33/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xa9-> #0 (btrfs-tree-01#2/3){++++}-{3:3}:
check_prev_add+0x91/0xc60
__lock_acquire+0x1689/0x3130
lock_acquire+0xd8/0x490
down_read_nested+0x45/0x220
__btrfs_tree_read_lock+0x32/0x1a0 [btrfs]
btrfs_next_old_leaf+0x27d/0x580 [btrfs]
btrfs_real_readdir+0x1e3/0x4b0 [btrfs]
iterate_dir+0x170/0x1c0
__x64_sys_getdents64+0x83/0x140
do_syscall_64+0x33/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xa9other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(btrfs-tree-00);
lock(btrfs-tree-01#2/3);
lock(btrfs-tree-00);lock(btrfs-tree-01#2/3);
*** DEADLOCK ***
5 locks held by find/324157:
#0: ffff8ebc502c6e00 (&f->f_pos_lock){+.+.}-{3:3}, at: __fdget_pos+0x4d/0x60
#1: ffff8eb97f689980 (&type->i_mutex_dir_key#10){++++}-{3:3}, at: iterate_dir+0x52/0x1c0 #2: ffff8ebaec00ca58 (btrfs-tree-02#2){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x1a0 [btrfs] #3: ffff8eb98f986f78 (btrfs-tree-01#2){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x1a0 [btrfs] #4: ffff8eb9932c5088 (btrfs-tree-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x1a0 [btrfs]
stack backtrace: CPU: 2 PID: 324157 Comm: find Not tainted 5.10.0-rc2-btrfs-next-71 #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack+0x8d/0xb5 check_noncircular+0xff/0x110 ? mark_lock.part.0+0x468/0xe90 check_prev_add+0x91/0xc60 __lock_acquire+0x1689/0x3130 ? kvm_clock_read+0x14/0x30 ? kvm_sched_clock_read+0x5/0x10 lock_acquire+0xd8/0x490 ? __btrfs_tree_read_lock+0x32/0x1a0 [btrfs] down_read_nested+0x45/0x220 ? __btrfs_tree_read_lock+0x32/0x1a0 [btrfs] __btrfs_tree_read_lock+0x32/0x1a0 [btrfs] btrfs_next_old_leaf+0x27d/0x580 [btrfs] btrfs_real_readdir+0x1e3/0x4b0 [btrfs] iterate_dir+0x170/0x1c0 __x64_sys_getdents64+0x83/0x140 ? filldir+0x1d0/0x1d0 do_syscall_64+0x33/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xa9
This is thankfully straightforward to fix, simply release the path before we setup the reloc_ctl.
--- fs/btrfs/volumes.c | 2 ++ 1 file changed, 2 insertions(+)
fs/btrfs/volumes.c +2 -0
@@ -4283,6 +4283,8 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info)btrfs_warn (fs_info,"balance: cannot set exclusive op status, resume manually" ); +btrfs_release_path (path); +mutex_lock (&fs_info->balance_mutex);BUG_ON (fs_info->balance_ctl);spin_lock (&fs_info->balance_lock);
Signature · 1 line
2.26.2
PATCH 3/42 btrfs: convert some BUG_ON()'s to ASSERT()'s in do_relocation
In reply to Josef BacikA few of these are checking for correctness, and won't be triggered by corrupted file systems, so convert them to ASSERT() instead of BUG_ON() and add a comment explaining their existence.
--- fs/btrfs/relocation.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-)
fs/btrfs/relocation.c +16 -3
@@ -2215,7 +2215,11 @@ static int do_relocation(struct btrfs_trans_handle *trans,int ret;int err =0 ; -BUG_ON (lowest && node->eb); +/* +* If we are lowest then this is the first time we're processing this +* block, and thus shouldn't have an eb associated with it yet. +*/ +ASSERT (!lowest || !node->eb); path->lowest_level = node->level +1 ; rc->backref_cache.path[node->level] = node;
@@ -2316,7 +2320,11 @@ static int do_relocation(struct btrfs_trans_handle *trans, err = ret; goto next; } -BUG_ON (node->eb != eb); +/* +* We've just cow'ed this block, it should have updated +* the correct backref node entry. +*/ +ASSERT (node->eb == eb); }else {btrfs_set_node_blockptr (upper->eb, slot, node->eb->start);
@@ -2352,7 +2360,12 @@ static int do_relocation(struct btrfs_trans_handle *trans, } path->lowest_level =0 ; -BUG_ON (err == -ENOSPC ); + +/* +* We should have allocated all of our space in the block rsv and thus +* shouldn't ENOSPC. +*/ +ASSERT (err != -ENOSPC );return err; }
Signature · 1 line
2.26.2
PATCH 4/42 btrfs: convert BUG_ON()'s in relocate_tree_block
In reply to Josef BacikWe have a couple of BUG_ON()'s in relocate_tree_block() that can be tripped if we have file system corruption. Convert these to ASSERT()'s so developers still get yelled at when they break the backref code, but error out nicely for users so the whole box doesn't go down.
--- fs/btrfs/relocation.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +22 -2
@@ -2504,8 +2504,28 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,if (root) {if (test_bit (BTRFS_ROOT_SHAREABLE , &root->state)) { -BUG_ON (node->new_bytenr); -BUG_ON (!list_empty (&node->list)); +/* +* This block was the root block of a root, and this is +* the first time we're processing the block and thus it +* should not have had the ->new_bytenr modified and +* should have not been included on the changed list. +* +* However in the case of corruption we could have +* multiple refs pointing to the same block improperly, +* and thus we would trip over these checks. ASSERT() +* for the developer case, because it could indicate a +* bug in the backref code, however error out for a +* normal user in the case of corruption. +*/ +ASSERT (node->new_bytenr ==0 ); +ASSERT (list_empty (&node->list)); +if (node->new_bytenr || !list_empty (&node->list)) { +btrfs_err (root->fs_info, +"bytenr %llu has improper references to it" , + node->bytenr); + ret = -EUCLEAN ; + goto out; + }btrfs_record_root_in_trans (trans, root); root = root->reloc_root; node->new_bytenr = root->node->start;
Signature · 1 line
2.26.2
PATCH 5/42 btrfs: return an error from btrfs_record_root_in_trans
In reply to Josef BacikWe can create a reloc root when we record the root in the trans, which can fail for all sorts of different reasons. Propagate this error up the chain of callers. Future patches will fix the callers of btrfs_record_root_in_trans() to handle the error.
--- fs/btrfs/transaction.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/btrfs/transaction.c +3 -2
@@ -403,6 +403,7 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans, int force) {struct btrfs_fs_info *fs_info = root->fs_info; +int ret =0 ;if ((test_bit (BTRFS_ROOT_SHAREABLE , &root->state) && root->last_trans < trans->transid) || force) {
@@ -451,11 +452,11 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans,* lock. smp_wmb() makes sure that all the writes above are * done before we pop in the zero below */ -btrfs_init_reloc_root (trans, root); + ret =btrfs_init_reloc_root (trans, root);smp_mb__before_atomic ();clear_bit (BTRFS_ROOT_IN_TRANS_SETUP , &root->state); } -return 0 ; +return ret; }
Signature · 1 line
2.26.2
PATCH 6/42 btrfs: handle errors from select_reloc_root()
In reply to Josef BacikCurrently select_reloc_root() doesn't return an error, but followup patches will make it possible for it to return an error. We do have proper error recovery in do_relocation however, so handle the possibility of select_reloc_root() having an error properly instead of BUG_ON(!root). I've also adjusted select_reloc_root() to return ERR_PTR(-ENOENT) if we don't find a root, instead of NULL, to make the error case easier to deal with. I've replaced the BUG_ON(!root) with an ASSERT(ret != -ENOENT), as this indicates we messed up the backref walking code, but could indicate corruption so we do not want to have a BUG_ON() here.
--- fs/btrfs/relocation.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +13 -2
@@ -2056,7 +2056,7 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,break ; }if (!root) -return NULL ; +return ERR_PTR (-ENOENT ); next = node;/* setup backref node path for btrfs_reloc_cow_block */
@@ -2231,7 +2231,18 @@ static int do_relocation(struct btrfs_trans_handle *trans, upper = edge->node[UPPER ]; root =select_reloc_root (trans, rc, upper, edges); -BUG_ON (!root); +if (IS_ERR (root)) { + err =PTR_ERR (root); + +/* +* This can happen if there's fs corruption, but if we +* have ASSERT()'s on then we're developers and we +* likely made a logic mistake in the backref code, so +* check for this error condition. +*/ +ASSERT (err != -ENOENT ); + goto next; + }if (upper->eb && !upper->locked) {if (!lowest) {
Signature · 1 line
2.26.2
PATCH 7/42 btrfs: convert BUG_ON()'s in select_reloc_root() to proper errors
In reply to Josef BacikWe have several BUG_ON()'s in select_reloc_root() that can be tripped if you have extent tree corruption. Convert these to ASSERT()'s, because if we hit it during testing it really is bad, or could indicate a problem with the backref walking code.
However if users hit these problems it generally indicates corruption, I've hit a few machines in the fleet that trip over these with clearly corrupted extent trees, so be nice and spit out an error message and return an error instead of bringing the whole box down.
--- fs/btrfs/relocation.c | 51 +++++++++++++++++++++++++++++++++++++++---- 1 file changed, 47 insertions(+), 4 deletions(-)
fs/btrfs/relocation.c +47 -4
@@ -2025,8 +2025,35 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,cond_resched (); next =walk_up_backref (next, edges, &index); root = next->root; -BUG_ON (!root); -BUG_ON (!test_bit (BTRFS_ROOT_SHAREABLE , &root->state)); + +/* +* If there is no root, then our references for this block are +* incomplete, as we should be able to walk all the way up to a +* block that is owned by a root. +* +* This path is only for SHAREABLE roots, so if we come upon a +* non-SHAREABLE root then we have backrefs that resolve +* improperly. +* +* Both of these cases indicate file system corruption, or a bug +* in the backref walking code. The ASSERT() is to make sure +* developers get bitten as soon as possible, proper error +* handling is for users who may have corrupt file systems. +*/ +ASSERT (root); +ASSERT (test_bit (BTRFS_ROOT_SHAREABLE , &root->state)); +if (!root) { +btrfs_err (trans->fs_info, +"bytenr %llu doesn't have a backref path ending in a root" , + node->bytenr); +return ERR_PTR (-EUCLEAN ); + } +if (!test_bit (BTRFS_ROOT_SHAREABLE , &root->state)) { +btrfs_err (trans->fs_info, +"bytenr %llu has multiple refs with one ending in a non shareable root" , + node->bytenr); +return ERR_PTR (-EUCLEAN ); + }if (root->root_key.objectid ==BTRFS_TREE_RELOC_OBJECTID ) {record_reloc_root_in_trans (trans, root);
@@ -2037,8 +2064,24 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans, root = root->reloc_root;if (next->new_bytenr != root->node->start) { -BUG_ON (next->new_bytenr); -BUG_ON (!list_empty (&next->list)); +/* +* We just created the reloc root, so we shouldn't have +* ->new_bytenr set and this shouldn't be in the changed +* list. If it is then we have multiple roots pointing +* at the same bytenr, or we've made a mistake in the +* backref walking code. ASSERT() for developers, +* error out for users, as it indicates corruption or a +* bad bug. +*/ +ASSERT (next->new_bytenr ==0 ); +ASSERT (list_empty (&next->list)); +if (next->new_bytenr || !list_empty (&next->list)) { +btrfs_err (trans->fs_info, +"bytenr %llu possibly has multiple roots pointing at the same bytenr %llu" , + node->bytenr, next->bytenr); +return ERR_PTR (-EUCLEAN ); + } + next->new_bytenr = root->node->start;btrfs_put_root (next->root); next->root =btrfs_grab_root (root);
Signature · 1 line
2.26.2
PATCH 8/42 btrfs: check record_root_in_trans related failures in select_reloc_root
In reply to Josef BacikWe will record the fs root or the reloc root in the trans in select_reloc_root. These will actually return errors in the following patches, so check their return value here and return it up the stack.
--- fs/btrfs/relocation.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +7 -2
@@ -2019,6 +2019,7 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,struct btrfs_backref_node *next;struct btrfs_root *root;int index =0 ; +int ret; next = node;while (1 ) {
@@ -2056,11 +2057,15 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans, }if (root->root_key.objectid ==BTRFS_TREE_RELOC_OBJECTID ) { -record_reloc_root_in_trans (trans, root); + ret =record_reloc_root_in_trans (trans, root); +if (ret) +return ERR_PTR (ret);break ; } -btrfs_record_root_in_trans (trans, root); + ret =btrfs_record_root_in_trans (trans, root); +if (ret) +return ERR_PTR (ret); root = root->reloc_root;if (next->new_bytenr != root->node->start) {
Signature · 1 line
2.26.2
PATCH 9/42 btrfs: do proper error handling in record_reloc_root_in_trans
In reply to Josef BacikGenerally speaking this shouldn't ever fail, the corresponding fs root for the reloc root will already be in memory, so we won't get -ENOMEM here.
However if there is no corresponding root for the reloc root then we could get -ENOMEM when we try to allocate it or we could get -ENOENT when we look it up and see that it doesn't exist.
Convert these BUG_ON()'s into ASSERT()'s + proper error handling for the case of corruption.
--- fs/btrfs/relocation.c | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +24 -2
@@ -2002,8 +2002,30 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,return 0 ; root =btrfs_get_fs_root (fs_info, reloc_root->root_key.offset, false); -BUG_ON (IS_ERR (root)); -BUG_ON (root->reloc_root != reloc_root); + +/* +* This should succeed, since we can't have a reloc root without having +* already looked up the actual root and created the reloc root for this +* root. +* +* However if there's some sort of corruption where we have a ref to a +* reloc root without a corresponding root this could return -ENOENT. +* +* The ASSERT()'s are to catch this case in testing, because it could +* indicate a bug, but for non-developers it indicates corruption and we +* should error out. +*/ +ASSERT (!IS_ERR (root)); +ASSERT (root->reloc_root == reloc_root); +if (IS_ERR (root)) +return PTR_ERR (root); +if (root->reloc_root != reloc_root) { +btrfs_err (fs_info, +"root %llu has two reloc roots associated with it" , + reloc_root->root_key.offset); +btrfs_put_root (root); +return -EUCLEAN ; + } ret =btrfs_record_root_in_trans (trans, root);btrfs_put_root (root);
Signature · 1 line
2.26.2
PATCH 10/42 btrfs: handle btrfs_record_root_in_trans failure in btrfs_rename_exchange
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in btrfs_rename_exchange.
--- fs/btrfs/inode.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
fs/btrfs/inode.c +5 -2
@@ -8758,8 +8758,11 @@ static int btrfs_rename_exchange(struct inode *old_dir, goto out_notrans; } -if (dest != root) -btrfs_record_root_in_trans (trans, dest); +if (dest != root) { + ret =btrfs_record_root_in_trans (trans, dest); +if (ret) + goto out_fail; + }/* * We need to find a free sequence number both in the source and
Signature · 1 line
2.26.2
PATCH 11/42 btrfs: handle btrfs_record_root_in_trans failure in btrfs_rename
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in btrfs_rename.
--- fs/btrfs/inode.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
fs/btrfs/inode.c +5 -2
@@ -9066,8 +9066,11 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, goto out_notrans; } -if (dest != root) -btrfs_record_root_in_trans (trans, dest); +if (dest != root) { + ret =btrfs_record_root_in_trans (trans, dest); +if (ret) + goto out_fail; + } ret =btrfs_set_inode_index (BTRFS_I (new_dir), &index);if (ret)
Signature · 1 line
2.26.2
PATCH 12/42 btrfs: handle btrfs_record_root_in_trans failure in btrfs_delete_subvolume
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in btrfs_delete_subvolume.
--- fs/btrfs/inode.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/btrfs/inode.c +6 -1
@@ -4049,7 +4049,12 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) goto out_end_trans; } -btrfs_record_root_in_trans (trans, dest); + ret =btrfs_record_root_in_trans (trans, dest); +if (ret) { + err = ret; +btrfs_abort_transaction (trans, ret); + goto out_end_trans; + }memset (&dest->root_item.drop_progress,0 ,sizeof (dest->root_item.drop_progress));
Signature · 1 line
2.26.2
PATCH 13/42 btrfs: handle btrfs_record_root_in_trans failure in btrfs_recover_log_trees
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in btrfs_recover_log_trees.
--- fs/btrfs/tree-log.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
fs/btrfs/tree-log.c +6 -2
@@ -6261,8 +6261,12 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree) } wc.replay_dest->log_root = log; -btrfs_record_root_in_trans (trans, wc.replay_dest); - ret =walk_log_tree (trans, log, &wc); + ret =btrfs_record_root_in_trans (trans, wc.replay_dest); +if (ret) +btrfs_handle_fs_error (fs_info, ret, +"Couldn't record the root in the transaction." ); +else + ret =walk_log_tree (trans, log, &wc);if (!ret && wc.stage ==LOG_WALK_REPLAY_ALL ) { ret = fixup_inode_link_counts(trans, wc.replay_dest,
Signature · 1 line
2.26.2
PATCH 14/42 btrfs: handle btrfs_record_root_in_trans failure in create_subvol
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in create_subvol.
--- fs/btrfs/ioctl.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/btrfs/ioctl.c +5 -1
@@ -702,7 +702,11 @@ static noinline int create_subvol(struct inode *dir,/* Freeing will be done in btrfs_put_root() of new_root */ anon_dev =0 ; -btrfs_record_root_in_trans (trans, new_root); + ret =btrfs_record_root_in_trans (trans, new_root); +if (ret) { +btrfs_abort_transaction (trans, ret); + goto fail; + } ret =btrfs_create_subvol_root (trans, new_root, root, new_dirid);btrfs_put_root (new_root);
Signature · 1 line
2.26.2
PATCH 15/42 btrfs: btrfs: handle btrfs_record_root_in_trans failure in relocate_tree_block
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in relocate_tree_block.
--- fs/btrfs/relocation.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +3 -1
@@ -2607,7 +2607,9 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, ret = -EUCLEAN ; goto out; } -btrfs_record_root_in_trans (trans, root); + ret =btrfs_record_root_in_trans (trans, root); +if (ret) + goto out; root = root->reloc_root; node->new_bytenr = root->node->start;btrfs_put_root (node->root);
Signature · 1 line
2.26.2
PATCH 16/42 btrfs: handle btrfs_record_root_in_trans failure in start_transaction
In reply to Josef Bacikbtrfs_record_root_in_trans will return errors in the future, so handle the error properly in start_transaction.
--- fs/btrfs/transaction.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/btrfs/transaction.c +5 -1
@@ -737,7 +737,11 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,* Thus it need to be called after current->journal_info initialized, * or we can deadlock. */ -btrfs_record_root_in_trans (h, root); + ret =btrfs_record_root_in_trans (h, root); +if (ret) { +btrfs_end_transaction (h); +return ERR_PTR (ret); + }return h;
Signature · 1 line
2.26.2
PATCH 17/42 btrfs: handle record_root_in_trans failure in qgroup_account_snapshot
In reply to Josef Bacikrecord_root_in_trans can fail currently, so handle this failure properly.
--- fs/btrfs/transaction.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
fs/btrfs/transaction.c +4 -2
@@ -1440,7 +1440,9 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,* recorded root will never be updated again, causing an outdated root * item. */ -record_root_in_trans (trans, src,1 ); + ret =record_root_in_trans (trans, src,1 ); +if (ret) +return ret;/* * We are going to commit transaction, see btrfs_commit_transaction()
@@ -1492,7 +1494,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,* insert_dir_item() */ if (!ret) -record_root_in_trans (trans, parent,1 ); + ret =record_root_in_trans (trans, parent,1 );return ret; }
Signature · 1 line
2.26.2
PATCH 18/42 btrfs: handle record_root_in_trans failure in btrfs_record_root_in_trans
In reply to Josef Bacikrecord_root_in_trans can fail currently, handle this failure properly.
--- fs/btrfs/transaction.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/btrfs/transaction.c +3 -2
@@ -483,6 +483,7 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans, struct btrfs_root *root) {struct btrfs_fs_info *fs_info = root->fs_info; +int ret;if (!test_bit (BTRFS_ROOT_SHAREABLE , &root->state))return 0 ;
@@ -497,10 +498,10 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,return 0 ;mutex_lock (&fs_info->reloc_mutex); -record_root_in_trans (trans, root,0 ); + ret =record_root_in_trans (trans, root,0 );mutex_unlock (&fs_info->reloc_mutex); -return 0 ; +return ret; }static inline int is_transaction_blocked (struct btrfs_transaction *trans)
Signature · 1 line
2.26.2
PATCH 19/42 btrfs: handle record_root_in_trans failure in create_pending_snapshot
In reply to Josef Bacikrecord_root_in_trans can currently fail, so handle this failure properly.
--- fs/btrfs/transaction.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)
fs/btrfs/transaction.c +8 -3
@@ -1572,8 +1572,9 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, dentry = pending->dentry; parent_inode = pending->dir; parent_root =BTRFS_I (parent_inode)->root; -record_root_in_trans (trans, parent_root,0 ); - + ret =record_root_in_trans (trans, parent_root,0 ); +if (ret) + goto fail; cur_time =current_time (parent_inode);/*
@@ -1609,7 +1610,11 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, goto fail; } -record_root_in_trans (trans, root,0 ); + ret =record_root_in_trans (trans, root,0 ); +if (ret) { +btrfs_abort_transaction (trans, ret); + goto fail; + }btrfs_set_root_last_snapshot (&root->root_item, trans->transid);memcpy (new_root_item, &root->root_item,sizeof (*new_root_item));btrfs_check_and_init_root_item (new_root_item);
Signature · 1 line
2.26.2
PATCH 20/42 btrfs: do not panic in __add_reloc_root
In reply to Josef BacikIf we have a duplicate entry for a reloc root then we could have fs corruption that resulted in a double allocation. This shouldn't happen generally so leave an ASSERT() for this case, but return an error instead of panicing in the normal user case.
--- fs/btrfs/relocation.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +3 -1
@@ -637,10 +637,12 @@ static int __must_check __add_reloc_root(struct btrfs_root *root) rb_node =rb_simple_insert (&rc->reloc_root_tree.rb_root, node->bytenr, &node->rb_node);spin_unlock (&rc->reloc_root_tree.lock); +ASSERT (rb_node ==NULL );if (rb_node) { -btrfs_panic (fs_info, -EEXIST , +btrfs_err (fs_info,"Duplicate root found for start=%llu while inserting into relocation tree" , node->bytenr); +return -EEXIST ; }list_add_tail (&root->root_list, &rc->reloc_roots);
Signature · 1 line
2.26.2
PATCH 21/42 btrfs: have proper error handling in btrfs_init_reloc_root
In reply to Josef Bacikcreate_reloc_root will return errors in the future, and __add_reloc_root can return -ENOMEM or -EEXIST, so handle these errors properly.
--- fs/btrfs/relocation.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +6 -1
@@ -860,9 +860,14 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, reloc_root =create_reloc_root (trans, root, root->root_key.objectid);if (clear_rsv) trans->block_rsv = rsv; +if (IS_ERR (reloc_root)) +return PTR_ERR (reloc_root); ret =__add_reloc_root (reloc_root); -BUG_ON (ret <0 ); +if (ret) { +btrfs_put_root (reloc_root); +return ret; + } root->reloc_root =btrfs_grab_root (reloc_root);return 0 ; }
Signature · 1 line
2.26.2
PATCH 22/42 btrfs: do proper error handling in create_reloc_root
In reply to Josef BacikWe do memory allocations here, read blocks from disk, all sorts of operations that could easily fail at any given point. Instead of panicing the box, simply return the error back up the chain, all callers at this point have proper error handling.
--- fs/btrfs/relocation.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-)
fs/btrfs/relocation.c +16 -6
@@ -737,10 +737,11 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,struct extent_buffer *eb;struct btrfs_root_item *root_item;struct btrfs_key root_key; -int ret; +int ret =0 ; root_item =kmalloc (sizeof (*root_item),GFP_NOFS ); -BUG_ON (!root_item); +if (!root_item) +return ERR_PTR (-ENOMEM ); root_key.objectid =BTRFS_TREE_RELOC_OBJECTID ; root_key.type =BTRFS_ROOT_ITEM_KEY ;
@@ -752,7 +753,9 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,/* called by btrfs_init_reloc_root */ ret =btrfs_copy_root (trans, root, root->commit_root, &eb,BTRFS_TREE_RELOC_OBJECTID ); -BUG_ON (ret); +if (ret) + goto fail; +/* * Set the last_snapshot field to the generation of the commit * root - like this ctree.c:btrfs_block_can_be_shared() behaves
@@ -773,7 +776,8 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,*/ ret =btrfs_copy_root (trans, root, root->node, &eb,BTRFS_TREE_RELOC_OBJECTID ); -BUG_ON (ret); +if (ret) + goto fail; }memcpy (root_item, &root->root_item,sizeof (*root_item));
@@ -793,14 +797,20 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans, ret =btrfs_insert_root (trans, fs_info->tree_root, &root_key, root_item); -BUG_ON (ret); +if (ret) + goto fail; +kfree (root_item); reloc_root =btrfs_read_tree_root (fs_info->tree_root, &root_key); -BUG_ON (IS_ERR (reloc_root)); +if (IS_ERR (reloc_root)) +return reloc_root;set_bit (BTRFS_ROOT_SHAREABLE , &reloc_root->state); reloc_root->last_trans = trans->transid;return reloc_root; +fail: +kfree (root_item); +return ERR_PTR (ret); }/*
Signature · 1 line
2.26.2
PATCH 23/42 btrfs: handle btrfs_update_reloc_root failure in commit_fs_roots
In reply to Josef Bacikbtrfs_update_reloc_root will will return errors in the future, so handle the error properly in commit_fs_roots.
--- fs/btrfs/transaction.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
fs/btrfs/transaction.c +3 -1
@@ -1346,7 +1346,9 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans)spin_unlock (&fs_info->fs_roots_radix_lock);btrfs_free_log (trans, root); -btrfs_update_reloc_root (trans, root); + err =btrfs_update_reloc_root (trans, root); +if (err) +return err;btrfs_save_ino_cache (root, trans);
Signature · 1 line
2.26.2
PATCH 24/42 btrfs: change insert_dirty_subvol to return errors
In reply to Josef BacikThis will be able to return errors in the future, so change it to return an error and handle the error appropriately.
--- fs/btrfs/relocation.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-)
fs/btrfs/relocation.c +11 -5
@@ -1575,9 +1575,9 @@ static int find_next_key(struct btrfs_path *path, int level,/* * Insert current subvolume into reloc_control::dirty_subvol_roots */ -static void insert_dirty_subvol (struct btrfs_trans_handle *trans, -struct reloc_control *rc, -struct btrfs_root *root) +static int insert_dirty_subvol (struct btrfs_trans_handle *trans, +struct reloc_control *rc, +struct btrfs_root *root) {struct btrfs_root *reloc_root = root->reloc_root;struct btrfs_root_item *reloc_root_item;
@@ -1597,6 +1597,7 @@ static void insert_dirty_subvol(struct btrfs_trans_handle *trans,btrfs_grab_root (root);list_add_tail (&root->reloc_dirty_list, &rc->dirty_subvol_roots); } +return 0 ; }static int clean_dirty_subvols (struct reloc_control *rc)
@@ -1808,8 +1809,13 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc, out:btrfs_free_path (path); -if (err ==0 ) -insert_dirty_subvol (trans, rc, root); +if (err ==0 ) { + ret =insert_dirty_subvol (trans, rc, root); +if (ret) { +btrfs_abort_transaction (trans, ret); + err = ret; + } + }if (trans)btrfs_end_transaction_throttle (trans);
Signature · 1 line
2.26.2
PATCH 25/42 btrfs: handle btrfs_update_reloc_root failure in insert_dirty_subvol
In reply to Josef Bacikbtrfs_update_reloc_root will will return errors in the future, so handle the error properly in insert_dirty_subvol.
--- fs/btrfs/relocation.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +4 -1
@@ -1581,6 +1581,7 @@ static int insert_dirty_subvol(struct btrfs_trans_handle *trans, {struct btrfs_root *reloc_root = root->reloc_root;struct btrfs_root_item *reloc_root_item; +int ret;/* @root must be a subvolume tree root with a valid reloc tree */ ASSERT (root->root_key.objectid !=BTRFS_TREE_RELOC_OBJECTID );
@@ -1591,7 +1592,9 @@ static int insert_dirty_subvol(struct btrfs_trans_handle *trans, sizeof(reloc_root_item->drop_progress));btrfs_set_root_drop_level (reloc_root_item,0 );btrfs_set_root_refs (reloc_root_item,0 ); -btrfs_update_reloc_root (trans, root); + ret =btrfs_update_reloc_root (trans, root); +if (ret) +return ret;if (list_empty (&root->reloc_dirty_list)) {btrfs_grab_root (root);
Signature · 1 line
2.26.2
PATCH 26/42 btrfs: handle btrfs_update_reloc_root failure in prepare_to_merge
In reply to Josef Bacikbtrfs_update_reloc_root will will return errors in the future, so handle an error properly in prepare_to_merge.
--- fs/btrfs/relocation.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +12 -1
@@ -1891,10 +1891,21 @@ int prepare_to_merge(struct reloc_control *rc, int err)*/ if (!err)btrfs_set_root_refs (&reloc_root->root_item,1 ); -btrfs_update_reloc_root (trans, root); + ret =btrfs_update_reloc_root (trans, root); +/* +* Even if we have an error we need this reloc root back on our +* list so we can clean up properly. +*/ list_add (&reloc_root->root_list, &reloc_roots);btrfs_put_root (root); + +if (ret) { +btrfs_abort_transaction (trans, ret); +if (!err) + err = ret; +break ; + } }list_splice (&reloc_roots, &rc->reloc_roots);
Signature · 1 line
2.26.2
PATCH 27/42 btrfs: do proper error handling in btrfs_update_reloc_root
In reply to Josef BacikWe call btrfs_update_root in btrfs_update_reloc_root, which can fail for all sorts of reasons, including IO errors. Instead of panicing the box lets return the error, now that all callers properly handle those errors.
--- fs/btrfs/relocation.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
fs/btrfs/relocation.c +2 -4
@@ -894,7 +894,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,int ret;if (!have_reloc_root (root)) - goto out; +return 0 ; reloc_root = root->reloc_root; root_item = &reloc_root->root_item;
@@ -927,10 +927,8 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans, ret =btrfs_update_root (trans, fs_info->tree_root, &reloc_root->root_key, root_item); -BUG_ON (ret);btrfs_put_root (reloc_root); -out: -return 0 ; +return ret; }/*
Signature · 1 line
2.26.2
PATCH 28/42 btrfs: convert logic BUG_ON()'s in replace_path to ASSERT()'s
In reply to Josef BacikA few BUG_ON()'s in replace_path are purely to keep us from making logical mistakes, so replace them with ASSERT()'s.
--- fs/btrfs/relocation.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
fs/btrfs/relocation.c +3 -3
@@ -1202,8 +1202,8 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,int ret;int slot; -BUG_ON (src->root_key.objectid !=BTRFS_TREE_RELOC_OBJECTID ); -BUG_ON (dest->root_key.objectid ==BTRFS_TREE_RELOC_OBJECTID ); +ASSERT (src->root_key.objectid ==BTRFS_TREE_RELOC_OBJECTID ); +ASSERT (dest->root_key.objectid !=BTRFS_TREE_RELOC_OBJECTID ); last_snapshot =btrfs_root_last_snapshot (&src->root_item); again:
@@ -1236,7 +1236,7 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,struct btrfs_key first_key; level =btrfs_header_level (parent); -BUG_ON (level < lowest_level); +ASSERT (level >= lowest_level); ret =btrfs_bin_search (parent, &key, &slot);if (ret <0 )
Signature · 1 line
2.26.2
PATCH 29/42 btrfs: handle initial btrfs_cow_block error in replace_path
In reply to Josef BacikIf we error out cow'ing the root node when doing a replace_path then we simply unlock and free the buffer and return the error.
--- fs/btrfs/relocation.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +5 -1
@@ -1222,7 +1222,11 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,if (cow) { ret =btrfs_cow_block (trans, dest, eb,NULL ,0 , &eb,BTRFS_NESTING_COW ); -BUG_ON (ret); +if (ret) { +btrfs_tree_unlock (eb); +free_extent_buffer (eb); +return ret; + } }if (next_key) {
Signature · 1 line
2.26.2
PATCH 30/42 btrfs: handle the loop btrfs_cow_block error in replace_path
In reply to Josef BacikAs we loop through the path to replace it, we will have to cow each node we hit on the path down to the lowest_level. If this fails we simply unlock and free the block and break from the loop.
--- fs/btrfs/relocation.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +5 -1
@@ -1294,7 +1294,11 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc, ret =btrfs_cow_block (trans, dest, eb, parent, slot, &eb,BTRFS_NESTING_COW ); -BUG_ON (ret); +if (ret) { +btrfs_tree_unlock (eb); +free_extent_buffer (eb); +break ; + } }btrfs_tree_unlock (parent);
Signature · 1 line
2.26.2
PATCH 31/42 btrfs: handle btrfs_search_slot failure in replace_path
In reply to Josef BacikThis can fail for any number of reasons, why bring the whole box down with it?
--- fs/btrfs/relocation.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +2 -1
@@ -1322,7 +1322,8 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc, path->lowest_level = level; ret =btrfs_search_slot (trans, src, &key, path,0 ,1 ); path->lowest_level =0 ; -BUG_ON (ret); +if (ret) +break ;/* * Info qgroup to trace both subtrees.
Signature · 1 line
2.26.2
PATCH 32/42 btrfs: handle errors in reference count manipulation in replace_path
In reply to Josef BacikIf any of the reference count manipulation stuff fails in replace_path we need to abort the transaction, as we've modified the blocks already. We can simply break at this point and everything will be cleaned up.
--- fs/btrfs/relocation.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-)
fs/btrfs/relocation.c +16 -4
@@ -1363,27 +1363,39 @@ int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc, ref.skip_qgroup = true;btrfs_init_tree_ref (&ref, level -1 , src->root_key.objectid); ret =btrfs_inc_extent_ref (trans, &ref); -BUG_ON (ret); +if (ret) { +btrfs_abort_transaction (trans, ret); +break ; + }btrfs_init_generic_ref (&ref,BTRFS_ADD_DELAYED_REF , new_bytenr, blocksize,0 ); ref.skip_qgroup = true;btrfs_init_tree_ref (&ref, level -1 , dest->root_key.objectid); ret =btrfs_inc_extent_ref (trans, &ref); -BUG_ON (ret); +if (ret) { +btrfs_abort_transaction (trans, ret); +break ; + }btrfs_init_generic_ref (&ref,BTRFS_DROP_DELAYED_REF , new_bytenr, blocksize, path->nodes[level]->start);btrfs_init_tree_ref (&ref, level -1 , src->root_key.objectid); ref.skip_qgroup = true; ret =btrfs_free_extent (trans, &ref); -BUG_ON (ret); +if (ret) { +btrfs_abort_transaction (trans, ret); +break ; + }btrfs_init_generic_ref (&ref,BTRFS_DROP_DELAYED_REF , old_bytenr, blocksize,0 );btrfs_init_tree_ref (&ref, level -1 , dest->root_key.objectid); ref.skip_qgroup = true; ret =btrfs_free_extent (trans, &ref); -BUG_ON (ret); +if (ret) { +btrfs_abort_transaction (trans, ret); +break ; + }btrfs_unlock_up_safe (path,0 );
Signature · 1 line
2.26.2
PATCH 33/42 btrfs: handle extent reference errors in do_relocation
In reply to Josef BacikWe can already deal with errors appropriately from do_relocation, simply handle any errors that come from changing the refs at this point cleanly. We have to abort the transaction if we fail here as we've modified metadata at this point.
--- fs/btrfs/relocation.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
fs/btrfs/relocation.c +6 -4
@@ -2476,10 +2476,12 @@ static int do_relocation(struct btrfs_trans_handle *trans,btrfs_init_tree_ref (&ref, node->level,btrfs_header_owner (upper->eb)); ret =btrfs_inc_extent_ref (trans, &ref); -BUG_ON (ret); - - ret =btrfs_drop_subtree (trans, root, eb, upper->eb); -BUG_ON (ret); +if (!ret) +btrfs_drop_subtree (trans, root, eb, upper->eb); +if (ret) { +btrfs_abort_transaction (trans, ret); + err = ret; + } } next:if (!upper->pending)
Signature · 1 line
2.26.2
PATCH 34/42 btrfs: check for BTRFS_BLOCK_FLAG_FULL_BACKREF being set improperly
In reply to Josef BacikWe need to validate that a data extent item does not have the FULL_BACKREF flag set on it's flags.
--- fs/btrfs/tree-checker.c | 5 +++++ 1 file changed, 5 insertions(+)
fs/btrfs/tree-checker.c +5 -0
@@ -1272,6 +1272,11 @@ static int check_extent_item(struct extent_buffer *leaf, key->offset, fs_info->sectorsize);return -EUCLEAN ; } +if (flags &BTRFS_BLOCK_FLAG_FULL_BACKREF ) { +extent_err (leaf, slot, +"invalid extent flag, data has full backref set" ); +return -EUCLEAN ; + } } ptr = (unsigned long )(struct btrfs_extent_item *)(ei +1 );
Signature · 1 line
2.26.2
PATCH 35/42 btrfs: remove the extent item sanity checks in relocate_block_group
In reply to Josef BacikThese checks are all taken care of for us by the tree checker code.
--- fs/btrfs/relocation.c | 28 ---------------------------- 1 file changed, 28 deletions(-)
fs/btrfs/relocation.c +0 -28
@@ -3365,20 +3365,6 @@ static void unset_reloc_control(struct reloc_control *rc)mutex_unlock (&fs_info->reloc_mutex); } -static int check_extent_flags (u64 flags) -{ -if ((flags &BTRFS_EXTENT_FLAG_DATA ) && - (flags &BTRFS_EXTENT_FLAG_TREE_BLOCK )) -return 1 ; -if (!(flags &BTRFS_EXTENT_FLAG_DATA ) && - !(flags &BTRFS_EXTENT_FLAG_TREE_BLOCK )) -return 1 ; -if ((flags &BTRFS_EXTENT_FLAG_DATA ) && - (flags &BTRFS_BLOCK_FLAG_FULL_BACKREF )) -return 1 ; -return 0 ; -} -static noinline_for_stack int prepare_to_relocate(struct reloc_control *rc) {
@@ -3430,7 +3416,6 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)struct btrfs_path *path;struct btrfs_extent_item *ei;u64 flags; -u32 item_size;int ret;int err =0 ;int progress =0 ;
@@ -3479,19 +3464,6 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc) ei =btrfs_item_ptr (path->nodes[0 ], path->slots[0 ], struct btrfs_extent_item); - item_size =btrfs_item_size_nr (path->nodes[0 ], path->slots[0 ]); -if (item_size >=sizeof (*ei)) { - flags =btrfs_extent_flags (path->nodes[0 ], ei); - ret =check_extent_flags (flags); -BUG_ON (ret); - }else if (unlikely (item_size ==sizeof (struct btrfs_extent_item_v0 ))) { - err = -EINVAL ; -btrfs_print_v0_err (trans->fs_info); -btrfs_abort_transaction (trans, err); -break ; - }else { -BUG (); - }if (flags &BTRFS_EXTENT_FLAG_TREE_BLOCK ) { ret =add_tree_block (rc, &key, path, &blocks);
Signature · 1 line
2.26.2
PATCH 36/42 btrfs: do proper error handling in create_reloc_inode
In reply to Josef BacikWe already handle some errors in this function, and the callers do the correct error handling, so clean up the rest of the function to do the appropriate error handling.
--- fs/btrfs/relocation.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +7 -2
@@ -3628,10 +3628,15 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info, goto out; err =__insert_orphan_inode (trans, root, objectid); -BUG_ON (err); +if (err) + goto out; inode =btrfs_iget (fs_info->sb, objectid, root); -BUG_ON (IS_ERR (inode)); +if (IS_ERR (inode)) { + err =PTR_ERR (inode); + inode =NULL ; + goto out; + }BTRFS_I (inode)->index_cnt = group->start; err =btrfs_orphan_add (trans,BTRFS_I (inode));
Signature · 1 line
2.26.2
PATCH 37/42 btrfs: handle __add_reloc_root failure in btrfs_recover_relocation
In reply to Josef BacikWe can already handle errors appropriately from this function, deal with an error coming from __add_reloc_root appropriately.
--- fs/btrfs/relocation.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +6 -1
@@ -3978,7 +3978,12 @@ int btrfs_recover_relocation(struct btrfs_root *root) } err =__add_reloc_root (reloc_root); -BUG_ON (err <0 );/* -ENOMEM or logic error */ +if (err) { +list_add_tail (&reloc_root->root_list, &reloc_roots); +btrfs_put_root (fs_root); +btrfs_end_transaction (trans); + goto out_unset; + } fs_root->reloc_root =btrfs_grab_root (reloc_root);btrfs_put_root (fs_root); }
Signature · 1 line
2.26.2
PATCH 38/42 btrfs: handle __add_reloc_root failure in btrfs_reloc_post_snapshot
In reply to Josef BacikIf we fail to add the reloc root, drop it and return the error. All callers of this function already handle errors appropriately.
--- fs/btrfs/relocation.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
fs/btrfs/relocation.c +4 -1
@@ -4198,7 +4198,10 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,return PTR_ERR (reloc_root); ret =__add_reloc_root (reloc_root); -BUG_ON (ret <0 ); +if (ret) { +btrfs_put_root (reloc_root); +return ret; + } new_root->reloc_root =btrfs_grab_root (reloc_root);if (rc->create_reloc_tree)
Signature · 1 line
2.26.2
PATCH 39/42 btrfs: cleanup error handling in prepare_to_merge
In reply to Josef BacikThis probably can't happen even with a corrupt file system, because we would have failed much earlier on than here. However there's no reason we can't just check and bail out as appropriate, so do that and convert the correctness BUG_ON() to an ASSERT().
--- fs/btrfs/relocation.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +8 -2
@@ -1901,8 +1901,14 @@ int prepare_to_merge(struct reloc_control *rc, int err) root =btrfs_get_fs_root (fs_info, reloc_root->root_key.offset, false); -BUG_ON (IS_ERR (root)); -BUG_ON (root->reloc_root != reloc_root); +if (IS_ERR (root)) { +list_add (&reloc_root->root_list, &reloc_roots); +btrfs_abort_transaction (trans, (int )PTR_ERR (root)); +if (!err) + err =PTR_ERR (root); +break ; + } +ASSERT (root->reloc_root == reloc_root);/* * set reference count to 1, so btrfs_recover_relocation
Signature · 1 line
2.26.2
PATCH 40/42 btrfs: handle extent corruption with select_one_root properly
In reply to Josef BacikIn corruption cases we could have paths from a block up to no root at all, and thus we'll BUG_ON(!root) in select_one_root. Handle this by adding an ASSERT() for developers, and returning an error for normal users.
--- fs/btrfs/relocation.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-)
fs/btrfs/relocation.c +16 -3
@@ -2224,7 +2224,16 @@ struct btrfs_root *select_one_root(struct btrfs_backref_node *node)cond_resched (); next =walk_up_backref (next, edges, &index); root = next->root; -BUG_ON (!root); + +/* +* This can occur if we have incomplete extent refs leading all +* the way up a particular path, in this case return -EUCLEAN. +* However leave as an ASSERT() for developers, because it could +* indicate a bug in the backref code. +*/ +ASSERT (root); +if (!root) +return ERR_PTR (-EUCLEAN );/* No other choice for non-shareable tree */ if (!test_bit (BTRFS_ROOT_SHAREABLE , &root->state))
@@ -2642,8 +2651,12 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,BUG_ON (node->processed); root =select_one_root (node); -if (root ==ERR_PTR (-ENOENT )) { -update_processed_blocks (rc, node); +if (IS_ERR (root)) { + ret =PTR_ERR (root); +if (ret == -ENOENT ) { + ret =0 ; +update_processed_blocks (rc, node); + } goto out; }
Signature · 1 line
2.26.2
PATCH 41/42 btrfs: do proper error handling in merge_reloc_roots
In reply to Josef BacikWe have a BUG_ON() if we get an error back from btrfs_get_fs_root(). This honestly should never fail, as at this point we have a solid coordination of fs root to reloc root, and these roots will all be in memory. But in the name of killing BUG_ON()'s remove this one and handle the error properly. Change the remaining BUG_ON() to an ASSERT().
--- fs/btrfs/relocation.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-)
fs/btrfs/relocation.c +11 -2
@@ -1980,9 +1980,18 @@ void merge_reloc_roots(struct reloc_control *rc) root =btrfs_get_fs_root (fs_info, reloc_root->root_key.offset, false); +if (IS_ERR (root)) { +/* +* This likely won't happen, since we would have failed +* at a higher level. However for correctness sake +* handle the error anyway. +*/ + ret =PTR_ERR (root); + goto out; + } +if (btrfs_root_refs (&reloc_root->root_item) >0 ) { -BUG_ON (IS_ERR (root)); -BUG_ON (root->reloc_root != reloc_root); +ASSERT (root->reloc_root == reloc_root); ret =merge_reloc_root (rc, root);btrfs_put_root (root);if (ret) {
Signature · 1 line
2.26.2
PATCH 42/42 btrfs: check return value of btrfs_commit_transaction in relocation
In reply to Josef BacikThere's a few places where we don't check the return value of btrfs_commit_transaction in relocation.c. Thankfully all these places have straightforward error handling, so simply change all of the sites at once.
--- fs/btrfs/relocation.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
fs/btrfs/relocation.c +5 -4
@@ -1936,7 +1936,7 @@ int prepare_to_merge(struct reloc_control *rc, int err)list_splice (&reloc_roots, &rc->reloc_roots);if (!err) -btrfs_commit_transaction (trans); + err =btrfs_commit_transaction (trans);else btrfs_end_transaction (trans);return err;
@@ -3431,8 +3431,7 @@ int prepare_to_relocate(struct reloc_control *rc)*/ return PTR_ERR (trans); } -btrfs_commit_transaction (trans); -return 0 ; +return btrfs_commit_transaction (trans); } static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
@@ -3590,7 +3589,9 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc) err =PTR_ERR (trans); goto out_free; } -btrfs_commit_transaction (trans); + ret =btrfs_commit_transaction (trans); +if (ret && !err) + err = ret; out_free: ret =clean_dirty_subvols (rc);if (ret <0 && !err)
Signature · 1 line
2.26.2
Quoted 18 lines · Josef Bacik
Hello,
Relocation is the last place that is not able to handle errors at all, which results in all sorts of lovely panics if you encounter corruptions or IO errors. I'm going to start cleaning up relocation, but before I move code around I want the error handling to be somewhat sane, so I'm not changing behavior and error handling at the same time.
These patches are purely about error handling, there is no behavior changing other than returning errors up the chain properly. There is a lot of room for follow up cleanups, which will happen next. However I wanted to get this series done today and out so we could get it merged ASAP, and then the follow up cleanups can happen later as they are less important and less critical.
The only exception to the above is the patch to add the error injection sites for
btrfs_cow_blockandbtrfs_search_slot, and a lockdep fix that I discovered while running my tests, those are the first two patches in the series.
Quoted 5 lines
I tested this with my error injection stress test, where I keep track of all stack traces that have been tested and only inject errors when we have a new stack trace, which means I should have covered all of the various error conditions. With this patchset I'm no longer panicing while stressing the error conditions. Thanks,
I just threw this patch set on top of kdave/for-next (a12315094469d573e41fe3eee91c99a83cec02df) and I got something that looks like runaway balances:
[Thu Nov 12 22:16:37 2020] BTRFS info (device dm-0): balance: start -dlimit=9 [Thu Nov 12 22:16:37 2020] BTRFS info (device dm-0): relocating block group 37256442150912 flags data [Thu Nov 12 22:16:38 2020] BTRFS info (device dm-0): found 19 extents, loops 1, stage: move data extents [Thu Nov 12 22:16:41 2020] BTRFS info (device dm-0): found 19 extents, loops 2, stage: move data extents [Thu Nov 12 22:16:45 2020] BTRFS info (device dm-0): found 19 extents, loops 3, stage: move data extents [Thu Nov 12 22:16:46 2020] BTRFS info (device dm-0): found 19 extents, loops 4, stage: move data extents [Thu Nov 12 22:16:47 2020] BTRFS info (device dm-0): found 19 extents, loops 5, stage: move data extents [Thu Nov 12 22:16:47 2020] avg_delayed_ref_runtime = 3492558, time = 750175264685, count = 214792 [Thu Nov 12 22:16:50 2020] BTRFS info (device dm-0): found 19 extents, loops 6, stage: move data extents [Thu Nov 12 22:16:55 2020] BTRFS info (device dm-0): found 19 extents, loops 7, stage: move data extents [Thu Nov 12 22:16:58 2020] BTRFS info (device dm-0): found 19 extents, loops 8, stage: move data extents [Thu Nov 12 22:17:02 2020] BTRFS info (device dm-0): found 19 extents, loops 9, stage: move data extents [Thu Nov 12 22:17:07 2020] BTRFS info (device dm-0): found 19 extents, loops 10, stage: move data extents [Thu Nov 12 22:17:11 2020] BTRFS info (device dm-0): found 19 extents, loops 11, stage: move data extents [Thu Nov 12 22:17:15 2020] BTRFS info (device dm-0): found 19 extents, loops 12, stage: move data extents [Thu Nov 12 22:17:19 2020] BTRFS info (device dm-0): found 19 extents, loops 13, stage: move data extents [Thu Nov 12 22:17:23 2020] BTRFS info (device dm-0): found 19 extents, loops 14, stage: move data extents [Thu Nov 12 22:17:27 2020] BTRFS info (device dm-0): found 19 extents, loops 15, stage: move data extents [Thu Nov 12 22:17:31 2020] BTRFS info (device dm-0): found 19 extents, loops 16, stage: move data extents [Thu Nov 12 22:17:35 2020] BTRFS info (device dm-0): found 19 extents, loops 17, stage: move data extents [Thu Nov 12 22:17:40 2020] BTRFS info (device dm-0): found 19 extents, loops 18, stage: move data extents [Thu Nov 12 22:17:47 2020] BTRFS info (device dm-0): found 19 extents, loops 19, stage: move data extents [Thu Nov 12 22:17:51 2020] BTRFS info (device dm-0): found 19 extents, loops 20, stage: move data extents [Thu Nov 12 22:17:55 2020] BTRFS info (device dm-0): found 19 extents, loops 21, stage: move data extents [Thu Nov 12 22:17:59 2020] BTRFS info (device dm-0): found 19 extents, loops 22, stage: move data extents [Thu Nov 12 22:18:04 2020] BTRFS info (device dm-0): found 19 extents, loops 23, stage: move data extents [Thu Nov 12 22:18:07 2020] BTRFS info (device dm-0): found 19 extents, loops 24, stage: move data extents [Thu Nov 12 22:18:11 2020] BTRFS info (device dm-0): found 19 extents, loops 25, stage: move data extents [Thu Nov 12 22:18:15 2020] BTRFS info (device dm-0): found 19 extents, loops 26, stage: move data extents [Thu Nov 12 22:18:20 2020] BTRFS info (device dm-0): found 19 extents, loops 27, stage: move data extents [Thu Nov 12 22:18:24 2020] BTRFS info (device dm-0): found 19 extents, loops 28, stage: move data extents [Thu Nov 12 22:18:28 2020] BTRFS info (device dm-0): found 19 extents, loops 29, stage: move data extents [Thu Nov 12 22:18:32 2020] BTRFS info (device dm-0): found 19 extents, loops 30, stage: move data extents [Thu Nov 12 22:18:38 2020] BTRFS info (device dm-0): found 19 extents, loops 31, stage: move data extents [Thu Nov 12 22:18:45 2020] BTRFS info (device dm-0): found 19 extents, loops 32, stage: move data extents [Thu Nov 12 22:18:49 2020] BTRFS info (device dm-0): found 19 extents, loops 33, stage: move data extents [Thu Nov 12 22:18:53 2020] BTRFS info (device dm-0): found 19 extents, loops 34, stage: move data extents [Thu Nov 12 22:18:59 2020] BTRFS info (device dm-0): found 19 extents, loops 35, stage: move data extents [Thu Nov 12 22:19:02 2020] BTRFS info (device dm-0): found 19 extents, loops 36, stage: move data extents [Thu Nov 12 22:19:06 2020] BTRFS info (device dm-0): found 19 extents, loops 37, stage: move data extents [Thu Nov 12 22:19:12 2020] BTRFS info (device dm-0): found 19 extents, loops 38, stage: move data extents [Thu Nov 12 22:19:15 2020] BTRFS info (device dm-0): found 19 extents, loops 39, stage: move data extents [Thu Nov 12 22:19:20 2020] BTRFS info (device dm-0): found 19 extents, loops 40, stage: move data extents [Thu Nov 12 22:19:23 2020] BTRFS info (device dm-0): found 19 extents, loops 41, stage: move data extents [Thu Nov 12 22:19:27 2020] BTRFS info (device dm-0): found 19 extents, loops 42, stage: move data extents [Thu Nov 12 22:19:31 2020] BTRFS info (device dm-0): found 19 extents, loops 43, stage: move data extents [Thu Nov 12 22:19:35 2020] BTRFS info (device dm-0): found 19 extents, loops 44, stage: move data extents [Thu Nov 12 22:19:38 2020] BTRFS info (device dm-0): found 19 extents, loops 45, stage: move data extents [Thu Nov 12 22:19:42 2020] BTRFS info (device dm-0): found 19 extents, loops 46, stage: move data extents [Thu Nov 12 22:19:46 2020] BTRFS info (device dm-0): found 19 extents, loops 47, stage: move data extents [Thu Nov 12 22:19:51 2020] BTRFS info (device dm-0): found 19 extents, loops 48, stage: move data extents [Thu Nov 12 22:19:54 2020] BTRFS info (device dm-0): found 19 extents, loops 49, stage: move data extents [Thu Nov 12 22:19:57 2020] BTRFS info (device dm-0): found 19 extents, loops 50, stage: move data extents [Thu Nov 12 22:19:58 2020] BTRFS info (device dm-0): found 19 extents, loops 51, stage: move data extents [Thu Nov 12 22:19:58 2020] BTRFS info (device dm-0): found 19 extents, loops 52, stage: move data extents [Thu Nov 12 22:20:02 2020] BTRFS info (device dm-0): found 19 extents, loops 53, stage: move data extents [Thu Nov 12 22:20:06 2020] BTRFS info (device dm-0): found 19 extents, loops 54, stage: move data extents [Thu Nov 12 22:20:10 2020] BTRFS info (device dm-0): found 19 extents, loops 55, stage: move data extents [Thu Nov 12 22:20:14 2020] BTRFS info (device dm-0): found 19 extents, loops 56, stage: move data extents [Thu Nov 12 22:20:26 2020] BTRFS info (device dm-0): found 19 extents, loops 57, stage: move data extents [Thu Nov 12 22:21:05 2020] BTRFS info (device dm-0): found 19 extents, loops 58, stage: move data extents [Thu Nov 12 22:21:16 2020] BTRFS info (device dm-0): found 19 extents, loops 59, stage: move data extents [Thu Nov 12 22:21:20 2020] BTRFS info (device dm-0): found 19 extents, loops 60, stage: move data extents [Thu Nov 12 22:21:25 2020] BTRFS info (device dm-0): found 19 extents, loops 61, stage: move data extents [Thu Nov 12 22:21:31 2020] BTRFS info (device dm-0): found 19 extents, loops 62, stage: move data extents [Thu Nov 12 22:21:37 2020] BTRFS info (device dm-0): found 19 extents, loops 63, stage: move data extents
The "loops N" is something I added for testing previous runaway balance fixes as I got tired of counting them from scripts all the time.
The same for-next kernel without patches had been running for 48 hours before this with no strange balance loop counts, so this looks like new behavior from this patch series.
On a metadata block group the extent count does change over time:
[Thu Nov 12 22:45:23 2020] BTRFS info (device dm-0): balance: start -mlimit=1 -slimit=1 [Thu Nov 12 22:45:23 2020] BTRFS info (device dm-0): relocating block group 37252147183616 flags metadata|raid1 [Thu Nov 12 22:45:31 2020] BTRFS info (device dm-0): found 21944 extents, loops 1, stage: move data extents [Thu Nov 12 22:45:51 2020] BTRFS info (device dm-0): found 21482 extents, loops 2, stage: move data extents [Thu Nov 12 22:46:00 2020] BTRFS info (device dm-0): found 20066 extents, loops 3, stage: move data extents [Thu Nov 12 22:46:02 2020] BTRFS info (device dm-0): found 20000 extents, loops 4, stage: move data extents [Thu Nov 12 22:46:06 2020] BTRFS info (device dm-0): found 19986 extents, loops 5, stage: move data extents [Thu Nov 12 22:46:08 2020] BTRFS info (device dm-0): found 19980 extents, loops 6, stage: move data extents [Thu Nov 12 22:46:12 2020] BTRFS info (device dm-0): found 19978 extents, loops 7, stage: move data extents [Thu Nov 12 22:46:15 2020] BTRFS info (device dm-0): found 19971 extents, loops 8, stage: move data extents [Thu Nov 12 22:46:18 2020] BTRFS info (device dm-0): found 19958 extents, loops 9, stage: move data extents [Thu Nov 12 22:46:21 2020] BTRFS info (device dm-0): found 19955 extents, loops 10, stage: move data extents [Thu Nov 12 22:46:24 2020] BTRFS info (device dm-0): found 19943 extents, loops 11, stage: move data extents [Thu Nov 12 22:46:27 2020] BTRFS info (device dm-0): found 19939 extents, loops 12, stage: move data extents [Thu Nov 12 22:46:30 2020] BTRFS info (device dm-0): found 19927 extents, loops 13, stage: move data extents [Thu Nov 12 22:46:33 2020] BTRFS info (device dm-0): found 19925 extents, loops 14, stage: move data extents [Thu Nov 12 22:46:48 2020] BTRFS info (device dm-0): found 19891 extents, loops 15, stage: move data extents [Thu Nov 12 22:47:06 2020] BTRFS info (device dm-0): found 19808 extents, loops 16, stage: move data extents [Thu Nov 12 22:47:10 2020] BTRFS info (device dm-0): found 19805 extents, loops 17, stage: move data extents [Thu Nov 12 22:47:13 2020] BTRFS info (device dm-0): found 19803 extents, loops 18, stage: move data extents [Thu Nov 12 22:47:15 2020] BTRFS info (device dm-0): found 19802 extents, loops 19, stage: move data extents [Thu Nov 12 22:47:18 2020] BTRFS info (device dm-0): found 19802 extents, loops 20, stage: move data extents [Thu Nov 12 22:47:21 2020] BTRFS info (device dm-0): found 19802 extents, loops 21, stage: move data extents [Thu Nov 12 22:47:24 2020] BTRFS info (device dm-0): found 19802 extents, loops 22, stage: move data extents [Thu Nov 12 22:47:28 2020] BTRFS info (device dm-0): found 19802 extents, loops 23, stage: move data extents [Thu Nov 12 22:47:30 2020] BTRFS info (device dm-0): found 19802 extents, loops 24, stage: move data extents [Thu Nov 12 22:47:34 2020] BTRFS info (device dm-0): found 19802 extents, loops 25, stage: move data extents [Thu Nov 12 22:47:36 2020] BTRFS info (device dm-0): found 19800 extents, loops 26, stage: move data extents [Thu Nov 12 22:47:40 2020] BTRFS info (device dm-0): found 19800 extents, loops 27, stage: move data extents [Thu Nov 12 22:47:43 2020] BTRFS info (device dm-0): found 19800 extents, loops 28, stage: move data extents [Thu Nov 12 22:47:56 2020] BTRFS info (device dm-0): found 19800 extents, loops 29, stage: move data extents [Thu Nov 12 22:48:10 2020] BTRFS info (device dm-0): found 19770 extents, loops 30, stage: move data extents [Thu Nov 12 22:48:15 2020] BTRFS info (device dm-0): found 19765 extents, loops 31, stage: move data extents [Thu Nov 12 22:48:17 2020] BTRFS info (device dm-0): found 19764 extents, loops 32, stage: move data extents [Thu Nov 12 22:48:19 2020] BTRFS info (device dm-0): found 19764 extents, loops 33, stage: move data extents [Thu Nov 12 22:48:23 2020] BTRFS info (device dm-0): found 19764 extents, loops 34, stage: move data extents [Thu Nov 12 22:48:25 2020] BTRFS info (device dm-0): found 19764 extents, loops 35, stage: move data extents [Thu Nov 12 22:48:29 2020] BTRFS info (device dm-0): found 19764 extents, loops 36, stage: move data extents [Thu Nov 12 22:48:32 2020] BTRFS info (device dm-0): found 19763 extents, loops 37, stage: move data extents [Thu Nov 12 22:48:35 2020] BTRFS info (device dm-0): found 19763 extents, loops 38, stage: move data extents [Thu Nov 12 22:48:38 2020] BTRFS info (device dm-0): found 19763 extents, loops 39, stage: move data extents [Thu Nov 12 22:48:41 2020] BTRFS info (device dm-0): found 19763 extents, loops 40, stage: move data extents [Thu Nov 12 22:48:44 2020] BTRFS info (device dm-0): found 19763 extents, loops 41, stage: move data extents
Quoted 65 lines
Josef
Josef Bacik (42):
btrfs: allow error injection forbtrfs_search_slotandbtrfs_cow_block
btrfs: fix lockdep splat inbtrfs_recover_relocation
btrfs: convert some BUG_ON()'s to ASSERT()'s indo_relocation
btrfs: convert BUG_ON()'s inrelocate_tree_block
btrfs: return an error frombtrfs_record_root_in_trans
btrfs: handle errors fromselect_reloc_root()
btrfs: convert BUG_ON()'s inselect_reloc_root()to proper errors
btrfs: checkrecord_root_in_transrelated failures in
select_reloc_root
btrfs: do proper error handling inrecord_reloc_root_in_trans
btrfs: handlebtrfs_record_root_in_transfailure in
btrfs_rename_exchange
btrfs: handlebtrfs_record_root_in_transfailure inbtrfs_rename
btrfs: handlebtrfs_record_root_in_transfailure in
btrfs_delete_subvolume
btrfs: handlebtrfs_record_root_in_transfailure in
btrfs_recover_log_trees
btrfs: handlebtrfs_record_root_in_transfailure increate_subvol
btrfs: btrfs: handlebtrfs_record_root_in_transfailure in
relocate_tree_block
btrfs: handlebtrfs_record_root_in_transfailure instart_transaction
btrfs: handlerecord_root_in_transfailure inqgroup_account_snapshot
btrfs: handlerecord_root_in_transfailure in
btrfs_record_root_in_trans
btrfs: handlerecord_root_in_transfailure increate_pending_snapshot
btrfs: do not panic in__add_reloc_root
btrfs: have proper error handling inbtrfs_init_reloc_root
btrfs: do proper error handling increate_reloc_root
btrfs: handlebtrfs_update_reloc_rootfailure incommit_fs_roots
btrfs: changeinsert_dirty_subvolto return errors
btrfs: handlebtrfs_update_reloc_rootfailure ininsert_dirty_subvol
btrfs: handlebtrfs_update_reloc_rootfailure inprepare_to_merge
btrfs: do proper error handling inbtrfs_update_reloc_root
btrfs: convert logic BUG_ON()'s inreplace_pathto ASSERT()'s
btrfs: handle initialbtrfs_cow_blockerror inreplace_path
btrfs: handle the loopbtrfs_cow_blockerror inreplace_path
btrfs: handlebtrfs_search_slotfailure inreplace_path
btrfs: handle errors in reference count manipulation inreplace_path
btrfs: handle extent reference errors indo_relocation
btrfs: check forBTRFS_BLOCK_FLAG_FULL_BACKREFbeing set improperly
btrfs: remove the extent item sanity checks inrelocate_block_group
btrfs: do proper error handling increate_reloc_inode
btrfs: handle__add_reloc_rootfailure inbtrfs_recover_relocation
btrfs: handle__add_reloc_rootfailure inbtrfs_reloc_post_snapshot
btrfs: cleanup error handling inprepare_to_merge
btrfs: handle extent corruption withselect_one_rootproperly
btrfs: do proper error handling inmerge_reloc_roots
btrfs: check return value ofbtrfs_commit_transactionin relocationfs/btrfs/ctree.c | 2 + fs/btrfs/inode.c | 21 ++- fs/btrfs/ioctl.c | 6 +- fs/btrfs/relocation.c | 372 ++++++++++++++++++++++++++++++---------- fs/btrfs/transaction.c | 37 ++-- fs/btrfs/tree-checker.c | 5 + fs/btrfs/tree-log.c | 8 +- fs/btrfs/volumes.c | 2 + 8 files changed, 342 insertions(+), 111 deletions(-)Signature · 1 line
2.26.2
Quoted 25 lines · Zygo Blaxell
Quoted 18 lines · Josef Bacik
Hello,
Relocation is the last place that is not able to handle errors at all, which results in all sorts of lovely panics if you encounter corruptions or IO errors. I'm going to start cleaning up relocation, but before I move code around I want the error handling to be somewhat sane, so I'm not changing behavior and error handling at the same time.
These patches are purely about error handling, there is no behavior changing other than returning errors up the chain properly. There is a lot of room for follow up cleanups, which will happen next. However I wanted to get this series done today and out so we could get it merged ASAP, and then the follow up cleanups can happen later as they are less important and less critical.
The only exception to the above is the patch to add the error injection sites for
btrfs_cow_blockandbtrfs_search_slot, and a lockdep fix that I discovered while running my tests, those are the first two patches in the series.Quoted 5 lines
I tested this with my error injection stress test, where I keep track of all stack traces that have been tested and only inject errors when we have a new stack trace, which means I should have covered all of the various error conditions. With this patchset I'm no longer panicing while stressing the error conditions. Thanks,
Quoted 4 lines
I just threw this patch set on top of kdave/for-next (
a12315094469d573e41fe3eee91c99a83cec02df) and I got something that looks like runaway balances:
Yup I hit this with my xfstests run that I started after I sent these out, I got a little happy with deleting things for one of the patches, this time I'm running xfstests _before_ I send the next version. Thanks,
Josef
Quoted 26 lines · Josef Bacik
Quoted 25 lines · Zygo Blaxell
Quoted 23 lines · Josef Bacik
Hello,
Relocation is the last place that is not able to handle errors at all, which results in all sorts of lovely panics if you encounter corruptions or IO errors. I'm going to start cleaning up relocation, but before I move code around I want the error handling to be somewhat sane, so I'm not changing behavior and error handling at the same time.
These patches are purely about error handling, there is no behavior changing other than returning errors up the chain properly. There is a lot of room for follow up cleanups, which will happen next. However I wanted to get this series done today and out so we could get it merged ASAP, and then the follow up cleanups can happen later as they are less important and less critical.
The only exception to the above is the patch to add the error injection sites for
btrfs_cow_blockandbtrfs_search_slot, and a lockdep fix that I discovered while running my tests, those are the first two patches in the series.I tested this with my error injection stress test, where I keep track of all stack traces that have been tested and only inject errors when we have a new stack trace, which means I should have covered all of the various error conditions. With this patchset I'm no longer panicing while stressing the error conditions. Thanks,
Quoted 8 lines
Quoted 4 lines
I just threw this patch set on top of kdave/for-next (
a12315094469d573e41fe3eee91c99a83cec02df) and I got something that looks like runaway balances:Yup I hit this with my xfstests run that I started after I sent these out, I got a little happy with deleting things for one of the patches, this time I'm running xfstests _before_ I send the next version. Thanks,
Well, the good news is you killed the BUG_ON I was hitting every few hours while running a test that sends a SIGINT to balance:
https://lore.kernel.org/linux-btrfs/[email protected]/
so I'm looking forward to the next version.
Quoted 1 line
Josef
Quoted 36 lines · Zygo Blaxell
Quoted 26 lines · Josef Bacik
Quoted 25 lines · Zygo Blaxell
Quoted 23 lines · Josef Bacik
Hello,
Relocation is the last place that is not able to handle errors at all, which results in all sorts of lovely panics if you encounter corruptions or IO errors. I'm going to start cleaning up relocation, but before I move code around I want the error handling to be somewhat sane, so I'm not changing behavior and error handling at the same time.
These patches are purely about error handling, there is no behavior changing other than returning errors up the chain properly. There is a lot of room for follow up cleanups, which will happen next. However I wanted to get this series done today and out so we could get it merged ASAP, and then the follow up cleanups can happen later as they are less important and less critical.
The only exception to the above is the patch to add the error injection sites for
btrfs_cow_blockandbtrfs_search_slot, and a lockdep fix that I discovered while running my tests, those are the first two patches in the series.I tested this with my error injection stress test, where I keep track of all stack traces that have been tested and only inject errors when we have a new stack trace, which means I should have covered all of the various error conditions. With this patchset I'm no longer panicing while stressing the error conditions. Thanks,
Quoted 8 lines
Quoted 4 lines
I just threw this patch set on top of kdave/for-next (
a12315094469d573e41fe3eee91c99a83cec02df) and I got something that looks like runaway balances:Yup I hit this with my xfstests run that I started after I sent these out, I got a little happy with deleting things for one of the patches, this time I'm running xfstests _before_ I send the next version. Thanks,
Quoted 6 lines
Well, the good news is you killed the
BUG_ONI was hitting every few hours while running a test that sends a SIGINT to balance:https://lore.kernel.org/linux-btrfs/[email protected]/so I'm looking forward to the next version.
OK, I've run the next version, and...the SIGINT test is still failing, just not in the original place, because the original place doesn't exist any more.
This happened once:
[48401.882283][T11825] BTRFS info (device dm-0): balance: start -mlimit=1 -slimit=1 [48401.889294][T11825] BTRFS info (device dm-0): relocating block group 37267716440064 flags metadata|raid1 [48423.523596][T11825] BTRFS info (device dm-0): balance: canceled [48553.928742][ T1426] BTRFS info (device dm-0): balance: start -mlimit=1 -slimit=1 [48553.952356][ T1426] BTRFS info (device dm-0): relocating block group 37267716440064 flags metadata|raid1 [48626.179977][ T4548] avg_delayed_ref_runtime = 5084522, time = 780837759114, count = 153571 [48639.557809][ T1426] ================================================================== [48639.559006][ T1426] BUG: KASAN: null-ptr-deref in select_reloc_root+0x1b0/0x6a0 [48639.560146][ T1426] Read of size 8 at addr 0000000000000000 by task btrfs/1426 [48639.561274][ T1426] [48639.561618][ T1426] CPU: 1 PID: 1426 Comm: btrfs Tainted: G W 5.10.0-87fa69f8a71b-for-next+ #14 [48639.563076][ T1426] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 [48639.564382][ T1426] Call Trace: [48639.564909][ T1426] dump_stack+0xbc/0xf9 [48639.565504][ T1426] ? select_reloc_root+0x1b0/0x6a0 [48639.566238][ T1426] ? select_reloc_root+0x1b0/0x6a0 [48639.566980][ T1426] kasan_report.cold.10+0x5/0x37 [48639.567685][ T1426] ? select_reloc_root+0x1b0/0x6a0 [48639.568599][ T1426] __asan_load8+0x69/0x90 [48639.569481][ T1426] select_reloc_root+0x1b0/0x6a0 [48639.570199][ T1426] ? __kasan_slab_free+0xf3/0x140 [48639.570926][ T1426] ? relocate_data_extent+0x1a0/0x1a0 [48639.571710][ T1426] ? btrfs_ioctl+0x24c8/0x4380 [48639.572400][ T1426] ? do_syscall_64+0x37/0x80 [48639.573062][ T1426] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [48639.573934][ T1426] ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 [48639.574953][ T1426] ? free_extent_buffer.part.52+0xb1/0x140 [48639.575786][ T1426] ? do_raw_spin_unlock+0xa8/0x140 [48639.576528][ T1426] do_relocation+0x23c/0xc10 [48639.577187][ T1426] ? free_extent_buffer.part.52+0xd7/0x140 [48639.578020][ T1426] ? __list_add_valid+0x33/0x70 [48639.578711][ T1426] ? select_reloc_root+0x6a0/0x6a0 [48639.579441][ T1426] ? btrfs_backref_finish_upper_links+0x419/0x7d0 [48639.580358][ T1426] ? walk_up_backref+0x91/0xd0 [48639.581039][ T1426] ? __asan_loadN+0xf/0x20 [48639.581664][ T1426] ? select_one_root+0x11d/0x2d0 [48639.582373][ T1426] ? lock_downgrade+0x3f0/0x3f0 [48639.583068][ T1426] ? do_raw_spin_unlock+0xa8/0x140 [48639.583802][ T1426] ? _raw_spin_unlock+0x22/0x30 [48639.584503][ T1426] ? btrfs_block_rsv_refill+0x50/0xa0 [48639.585280][ T1426] relocate_tree_blocks+0x853/0xb60 [48639.586030][ T1426] ? do_relocation+0xc10/0xc10 [48639.586709][ T1426] ? kasan_kmalloc+0x9/0x10 [48639.587356][ T1426] ? kmem_cache_alloc_trace+0x6a3/0xcb0 [48639.588147][ T1426] ? free_extent_buffer.part.52+0xd7/0x140 [48639.588983][ T1426] ? rb_insert_color+0x342/0x360 [48639.589687][ T1426] ? add_tree_block.isra.36+0x236/0x2b0 [48639.590478][ T1426] relocate_block_group+0x2eb/0x780 [48639.591223][ T1426] ? merge_reloc_roots+0x470/0x470 [48639.591972][ T1426] btrfs_relocate_block_group+0x26e/0x4c0 [48639.592945][ T1426] btrfs_relocate_chunk+0x52/0x120 [48639.593668][ T1426] btrfs_balance+0xe2e/0x18f0 [48639.594311][ T1426] ? __kasan_check_read+0x11/0x20 [48639.594962][ T1426] ? lock_acquire+0xd0/0x550 [48639.595597][ T1426] ? btrfs_relocate_chunk+0x120/0x120 [48639.596372][ T1426] ? kasan_unpoison_task_stack+0xf/0x20 [48639.597168][ T1426] ? kmem_cache_alloc_trace+0x6a3/0xcb0 [48639.597969][ T1426] ? _copy_from_user+0x83/0xc0 [48639.598648][ T1426] btrfs_ioctl_balance+0x3a7/0x460 [48639.599383][ T1426] btrfs_ioctl+0x24c8/0x4380 [48639.600060][ T1426] ? __kasan_check_read+0x11/0x20 [48639.600769][ T1426] ? lock_release+0xc8/0x640 [48639.601429][ T1426] ? lru_cache_add+0x178/0x250 [48639.602102][ T1426] ? btrfs_ioctl_get_supported_features+0x30/0x30 [48639.603004][ T1426] ? lock_downgrade+0x3f0/0x3f0 [48639.603681][ T1426] ? handle_mm_fault+0x159e/0x2150 [48639.604408][ T1426] ? __kasan_check_read+0x11/0x20 [48639.605108][ T1426] ? lock_release+0xc8/0x640 [48639.605750][ T1426] ? do_user_addr_fault+0x299/0x5a0 [48639.606481][ T1426] ? do_raw_spin_unlock+0xa8/0x140 [48639.607198][ T1426] ? lock_downgrade+0x3f0/0x3f0 [48639.607880][ T1426] ? _raw_spin_unlock+0x22/0x30 [48639.608557][ T1426] ? handle_mm_fault+0xad6/0x2150 [48639.609262][ T1426] ? do_vfs_ioctl+0xfc/0x9d0 [48639.609907][ T1426] ? ioctl_file_clone+0xe0/0xe0 [48639.610584][ T1426] ? __kasan_check_write+0x14/0x20 [48639.611301][ T1426] ? up_read+0x176/0x4f0 [48639.611901][ T1426] ? down_write_nested+0x2d0/0x2d0 [48639.612612][ T1426] ? vmacache_find+0xc9/0x120 [48639.613274][ T1426] ? __kasan_check_read+0x11/0x20 [48639.613977][ T1426] ? __fget_light+0xae/0x110 [48639.614620][ T1426] __x64_sys_ioctl+0xc3/0x100 [48639.615280][ T1426] do_syscall_64+0x37/0x80 [48639.615905][ T1426] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [48639.616726][ T1426] RIP: 0033:0x7ffb3753c427 [48639.617349][ T1426] Code: 00 00 90 48 8b 05 69 aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 39 aa 0c 00 f7 d8 64 89 01 48 [48639.620089][ T1426] RSP: 002b:00007ffddaf08558 EFLAGS: 00000202 ORIG_RAX: 0000000000000010 [48639.621268][ T1426] RAX: ffffffffffffffda RBX: 00007ffddaf085f8 RCX: 00007ffb3753c427 [48639.622380][ T1426] RDX: 00007ffddaf085f8 RSI: 00000000c4009420 RDI: 0000000000000003 [48639.623492][ T1426] RBP: 0000000000000003 R08: 0000000000000003 R09: 0000000000000078 [48639.624606][ T1426] R10: fffffffffffff59d R11: 0000000000000202 R12: 0000000000000001 [48639.625719][ T1426] R13: 0000000000000000 R14: 00007ffddaf09a34 R15: 0000000000000001 [48639.626841][ T1426] ==================================================================
(gdb) list *(select_reloc_root+0x1b0)
0xffffffff8194db20 is in select_reloc_root (fs/btrfs/relocation.c:2145).
2140 ret = btrfs_record_root_in_trans(trans, root);
2141 if (ret)
2142 return ERR_PTR(ret);
2143 root = root->reloc_root;
2144
2145 if (next->new_bytenr != root->node->start) {
2146 /*
2147 * We just created the reloc root, so we shouldn't have
2148 * ->new_bytenr set and this shouldn't be in the changed
2149 * list. If it is then we have multiple roots pointingThe following sequence has happened 3 times since yesterday:
[ 3252.939619][ T8472] BTRFS info (device dm-0): balance: resume -musage=90,limit=1 -susage=90,limit=1
[ 3253.962086][ T8472] BTRFS info (device dm-0): relocating block group 37252147183616 flags metadata|raid1
[ 3279.938601][ T8472]
[ 3279.938977][ T8472] ======================================================
[ 3279.939993][ T8472] WARNING: possible circular locking dependency detected
[ 3279.941020][ T8472] 5.10.0-87fa69f8a71b-for-next+ #14 Tainted: G W
[ 3279.942201][ T8472] ------------------------------------------------------
[ 3279.943481][ T8472] btrfs/8472 is trying to acquire lock:
[ 3279.944615][ T8472] ffff888040a3ba28 (btrfs-tree-00){++++}-{3:3}, at: __btrfs_tree_lock+0x29/0x190
[ 3279.946631][ T8472]
[ 3279.946631][ T8472] but task is already holding lock:
[ 3279.947819][ T8472] ffff88812c69eeb8 (btrfs-extent-01){++++}-{3:3}, at: __btrfs_tree_read_lock+0x49/0x240
[ 3279.949266][ T8472]
[ 3279.949266][ T8472] which lock already depends on the new lock.
[ 3279.949266][ T8472]
[ 3279.950909][ T8472]
[ 3279.950909][ T8472] the existing dependency chain (in reverse order) is:
[ 3279.952357][ T8472]
[ 3279.952357][ T8472] -> #1 (btrfs-extent-01){++++}-{3:3}:
[ 3279.953462][ T8472] lock_release+0x28c/0x640
[ 3279.954184][ T8472] up_read+0x7d/0x4f0
[ 3279.954833][ T8472] btrfs_tree_read_unlock+0xbc/0x1b0
[ 3279.955861][ T8472] btrfs_search_slot+0xd84/0xfb0
[ 3279.956639][ T8472] btrfs_lookup_extent_info+0x165/0x5b0
[ 3279.957508][ T8472] do_walk_down+0x29c/0xc00
[ 3279.958230][ T8472] walk_down_tree+0x1af/0x210
[ 3279.958973][ T8472] btrfs_drop_snapshot+0x44f/0xda0
[ 3279.959779][ T8472] clean_dirty_subvols+0x1bd/0x220
[ 3279.960592][ T8472] btrfs_recover_relocation+0x60d/0x940
[ 3279.961464][ T8472] btrfs_mount_rw+0x12e/0x200
[ 3279.962236][ T8472] open_ctree+0x2149/0x2507
[ 3279.962986][ T8472] btrfs_mount_root.cold.74+0xe/0xea
[ 3279.963816][ T8472] legacy_get_tree+0x89/0xd0
[ 3279.964547][ T8472] vfs_get_tree+0x52/0x150
[ 3279.965257][ T8472] fc_mount+0x14/0x60
[ 3279.965903][ T8472] vfs_kern_mount.part.38+0x61/0xa0
[ 3279.966717][ T8472] vfs_kern_mount+0x13/0x20
[ 3279.967435][ T8472] btrfs_mount+0x1f0/0x5d0
[ 3279.968147][ T8472] legacy_get_tree+0x89/0xd0
[ 3279.968938][ T8472] vfs_get_tree+0x52/0x150
[ 3279.969652][ T8472] path_mount+0xa53/0xf00
[ 3279.970411][ T8472] do_mount+0xd2/0xf0
[ 3279.971063][ T8472] __x64_sys_mount+0x100/0x120
[ 3279.971833][ T8472] do_syscall_64+0x37/0x80
[ 3279.972586][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3279.973564][ T8472]
[ 3279.973564][ T8472] -> #0 (btrfs-tree-00){++++}-{3:3}:
[ 3279.974673][ T8472] __lock_acquire+0x1dce/0x28a0
[ 3279.975509][ T8472] lock_acquire+0x192/0x550
[ 3279.976287][ T8472] down_write_nested+0xa6/0x2d0
[ 3279.977131][ T8472] __btrfs_tree_lock+0x29/0x190
[ 3279.977972][ T8472] btrfs_tree_lock+0x10/0x20
[ 3279.978764][ T8472] btrfs_search_slot+0x462/0xfb0
[ 3279.979614][ T8472] relocate_tree_blocks+0x8cd/0xb60
[ 3279.980506][ T8472] relocate_block_group+0x2eb/0x780
[ 3279.981368][ T8472] btrfs_relocate_block_group+0x26e/0x4c0
[ 3279.982285][ T8472] btrfs_relocate_chunk+0x52/0x120
[ 3279.983093][ T8472] btrfs_balance+0xe2e/0x18f0
[ 3279.983924][ T8472] btrfs_ioctl_balance+0x1f9/0x460
[ 3279.984729][ T8472] btrfs_ioctl+0x24c8/0x4380
[ 3279.985472][ T8472] __x64_sys_ioctl+0xc3/0x100
[ 3279.986327][ T8472] do_syscall_64+0x37/0x80
[ 3279.987058][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3279.987999][ T8472]
[ 3279.987999][ T8472] other info that might help us debug this:
[ 3279.987999][ T8472]
[ 3279.989509][ T8472] Possible unsafe locking scenario:
[ 3279.989509][ T8472]
[ 3279.990577][ T8472] CPU0 CPU1
[ 3279.991346][ T8472] ---- ----
[ 3279.992116][ T8472] lock(btrfs-extent-01);
[ 3279.992751][ T8472] lock(btrfs-tree-00);
[ 3279.993722][ T8472] lock(btrfs-extent-01);
[ 3279.994718][ T8472] lock(btrfs-tree-00);
[ 3279.995333][ T8472]
[ 3279.995333][ T8472] *** DEADLOCK ***
[ 3279.995333][ T8472]
[ 3279.996497][ T8472] 5 locks held by btrfs/8472:
[ 3279.997172][ T8472] #0: ffff888113010498 (sb_writers#13){.+.+}-{0:0}, at: mnt_want_write_file+0x43/0x90
[ 3279.998557][ T8472] #1: ffff888106cc2318 (&fs_info->delete_unused_bgs_mutex){+.+.}-{3:3}, at: btrfs_balance+0xa54/0x18f0
[ 3280.000165][ T8472] #2: ffff888106cc08f0 (&fs_info->cleaner_mutex){+.+.}-{3:3}, at: btrfs_relocate_block_group+0x266/0x4c0
[ 3280.001823][ T8472] #3: ffff8881130106b8 (sb_internal#2){.+.+}-{0:0}, at: start_transaction+0x6a3/0x910
[ 3280.003355][ T8472] #4: ffff88812c69eeb8 (btrfs-extent-01){++++}-{3:3}, at: __btrfs_tree_read_lock+0x49/0x240
[ 3280.004829][ T8472]
[ 3280.004829][ T8472] stack backtrace:
[ 3280.005671][ T8472] CPU: 3 PID: 8472 Comm: btrfs Tainted: G W 5.10.0-87fa69f8a71b-for-next+ #14
[ 3280.007129][ T8472] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[ 3280.008438][ T8472] Call Trace:
[ 3280.008915][ T8472] dump_stack+0xbc/0xf9
[ 3280.009562][ T8472] print_circular_bug.isra.42.cold.67+0x146/0x14b
[ 3280.010491][ T8472] check_noncircular+0x219/0x250
[ 3280.011207][ T8472] ? print_circular_bug.isra.42+0x1c0/0x1c0
[ 3280.012059][ T8472] ? pvclock_clocksource_read+0xeb/0x190
[ 3280.012873][ T8472] ? __kasan_check_write+0x14/0x20
[ 3280.013606][ T8472] ? lockdep_lock+0xa7/0x160
[ 3280.014269][ T8472] __lock_acquire+0x1dce/0x28a0
[ 3280.014973][ T8472] ? register_lock_class+0x8f0/0x8f0
[ 3280.015734][ T8472] ? rcu_read_lock_sched_held+0xa1/0xd0
[ 3280.016535][ T8472] ? rcu_read_lock_bh_held+0xb0/0xb0
[ 3280.017300][ T8472] lock_acquire+0x192/0x550
[ 3280.017950][ T8472] ? __btrfs_tree_lock+0x29/0x190
[ 3280.018725][ T8472] ? check_flags+0x30/0x30
[ 3280.019366][ T8472] ? ___might_sleep+0x10f/0x1e0
[ 3280.020062][ T8472] ? __might_sleep+0x71/0xe0
[ 3280.020716][ T8472] down_write_nested+0xa6/0x2d0
[ 3280.021413][ T8472] ? __btrfs_tree_lock+0x29/0x190
[ 3280.022135][ T8472] ? _down_write_nest_lock+0x2c0/0x2c0
[ 3280.022911][ T8472] ? balance_node_right+0x3c0/0x3c0
[ 3280.023649][ T8472] ? rcu_read_lock_bh_held+0x70/0xb0
[ 3280.024402][ T8472] ? rcu_read_lock_bh_held+0xb0/0xb0
[ 3280.025157][ T8472] __btrfs_tree_lock+0x29/0x190
[ 3280.025852][ T8472] btrfs_tree_lock+0x10/0x20
[ 3280.026524][ T8472] btrfs_search_slot+0x462/0xfb0
[ 3280.027261][ T8472] ? split_leaf+0x9a0/0x9a0
[ 3280.027941][ T8472] ? walk_up_backref+0xd0/0xd0
[ 3280.028652][ T8472] ? do_raw_spin_lock+0x1e0/0x1e0
[ 3280.029445][ T8472] ? do_raw_spin_unlock+0xa8/0x140
[ 3280.030222][ T8472] ? _raw_spin_unlock+0x22/0x30
[ 3280.030960][ T8472] ? btrfs_block_rsv_refill+0x50/0xa0
[ 3280.031772][ T8472] relocate_tree_blocks+0x8cd/0xb60
[ 3280.032567][ T8472] ? do_relocation+0xc10/0xc10
[ 3280.033294][ T8472] ? kmem_cache_alloc_trace+0xa06/0xcb0
[ 3280.034144][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.035039][ T8472] ? rb_insert_color+0x342/0x360
[ 3280.035842][ T8472] ? add_tree_block.isra.36+0x236/0x2b0
[ 3280.036707][ T8472] relocate_block_group+0x2eb/0x780
[ 3280.037544][ T8472] ? merge_reloc_roots+0x470/0x470
[ 3280.038371][ T8472] btrfs_relocate_block_group+0x26e/0x4c0
[ 3280.039289][ T8472] btrfs_relocate_chunk+0x52/0x120
[ 3280.040114][ T8472] btrfs_balance+0xe2e/0x18f0
[ 3280.040867][ T8472] ? pvclock_clocksource_read+0xeb/0x190
[ 3280.041775][ T8472] ? btrfs_relocate_chunk+0x120/0x120
[ 3280.042655][ T8472] ? lock_contended+0x620/0x6e0
[ 3280.043490][ T8472] ? do_raw_spin_lock+0x1e0/0x1e0
[ 3280.044303][ T8472] ? do_raw_spin_unlock+0xa8/0x140
[ 3280.045051][ T8472] btrfs_ioctl_balance+0x1f9/0x460
[ 3280.045980][ T8472] btrfs_ioctl+0x24c8/0x4380
[ 3280.047029][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.047851][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.048644][ T8472] ? __asan_loadN+0xf/0x20
[ 3280.049360][ T8472] ? btrfs_ioctl_get_supported_features+0x30/0x30
[ 3280.050469][ T8472] ? kvm_sched_clock_read+0x18/0x30
[ 3280.051598][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.052378][ T8472] ? lock_downgrade+0x3f0/0x3f0
[ 3280.053141][ T8472] ? handle_mm_fault+0xad6/0x2150
[ 3280.053934][ T8472] ? do_vfs_ioctl+0xfc/0x9d0
[ 3280.054654][ T8472] ? ioctl_file_clone+0xe0/0xe0
[ 3280.055421][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.056240][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.057060][ T8472] ? check_flags+0x26/0x30
[ 3280.057749][ T8472] ? lock_is_held_type+0xc3/0xf0
[ 3280.058573][ T8472] ? syscall_enter_from_user_mode+0x1b/0x60
[ 3280.059553][ T8472] ? do_syscall_64+0x13/0x80
[ 3280.060294][ T8472] ? rcu_read_lock_sched_held+0xa1/0xd0
[ 3280.061149][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.061924][ T8472] ? __fget_light+0xae/0x110
[ 3280.062631][ T8472] __x64_sys_ioctl+0xc3/0x100
[ 3280.063352][ T8472] do_syscall_64+0x37/0x80
[ 3280.064036][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3280.064941][ T8472] RIP: 0033:0x7f18ca415427
[ 3280.065619][ T8472] Code: 00 00 90 48 8b 05 69 aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 39 aa 0c 00 f7 d8 64 89 01 48
[ 3280.068691][ T8472] RSP: 002b:00007ffcd1fed6b8 EFLAGS: 00000206 ORIG_RAX: 0000000000000010
[ 3280.070070][ T8472] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f18ca415427
[ 3280.071348][ T8472] RDX: 00007ffcd1fed6c8 RSI: 00000000c4009420 RDI: 0000000000000003
[ 3280.072750][ T8472] RBP: 0000000000000003 R08: 0000000000000003 R09: 0000000000000078
[ 3280.073998][ T8472] R10: fffffffffffff59d R11: 0000000000000206 R12: 0000000000000002
[ 3280.075238][ T8472] R13: 00007ffcd1fefa39 R14: 0000560042d94119 R15: 0000000000000000
[ 3280.222230][ T8472] ------------[ cut here ]------------
[ 3280.223530][ T8472] WARNING: CPU: 3 PID: 8472 at fs/btrfs/backref.c:2627 btrfs_backref_add_tree_node+0xfa4/0x11d0
[ 3280.226859][ T8472] Modules linked in:
[ 3280.228422][ T8472] CPU: 3 PID: 8472 Comm: btrfs Tainted: G W 5.10.0-87fa69f8a71b-for-next+ #14
[ 3280.231521][ T8472] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[ 3280.233746][ T8472] RIP: 0010:btrfs_backref_add_tree_node+0xfa4/0x11d0
[ 3280.234776][ T8472] Code: 89 e7 4d 8d 66 30 e8 db 3e f3 ff 49 8d 46 2c 48 89 45 c0 e9 da f1 ff ff 49 8d 46 2c bb 8b ff ff ff 48 89 45 c0 e9 c8 f1 ff ff <0f> 0b 49 8d 46 2c bb fe ff ff ff 4d 8d 66 30 48 89 45 c0 e9 b0 f1
[ 3280.237631][ T8472] RSP: 0018:ffffc9000769f4c0 EFLAGS: 00010246
[ 3280.238626][ T8472] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 1ffff92000ed3e85
[ 3280.239826][ T8472] RDX: 0000000000000000 RSI: dffffc0000000000 RDI: ffff88805ef475e8
[ 3280.241013][ T8472] RBP: ffffc9000769f598 R08: ffffed100bde8ebe R09: ffffed100bde8ebe
[ 3280.242197][ T8472] R10: ffff88805ef475eb R11: ffffed100bde8ebd R12: ffff88805ef47020
[ 3280.243428][ T8472] R13: 0000217411380000 R14: ffff888126976b80 R15: 0000000000000000
[ 3280.244679][ T8472] FS: 00007f18ca3228c0(0000) GS:ffff8881f5e00000(0000) knlGS:0000000000000000
[ 3280.246099][ T8472] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 3280.247102][ T8472] CR2: 000055b5c456ab00 CR3: 000000012c08e002 CR4: 00000000001706e0
[ 3280.248455][ T8472] Call Trace:
[ 3280.249018][ T8472] build_backref_tree+0xc5/0x700
[ 3280.249845][ T8472] ? _raw_spin_unlock+0x22/0x30
[ 3280.250576][ T8472] ? release_extent_buffer+0x225/0x280
[ 3280.251381][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.252253][ T8472] relocate_tree_blocks+0x2a6/0xb60
[ 3280.253044][ T8472] ? kasan_unpoison_shadow+0x35/0x50
[ 3280.253841][ T8472] ? do_relocation+0xc10/0xc10
[ 3280.254559][ T8472] ? kasan_kmalloc+0x9/0x10
[ 3280.255227][ T8472] ? kmem_cache_alloc_trace+0x6a3/0xcb0
[ 3280.256063][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.256939][ T8472] ? rb_insert_color+0x342/0x360
[ 3280.257685][ T8472] ? add_tree_block.isra.36+0x236/0x2b0
[ 3280.258519][ T8472] relocate_block_group+0x2eb/0x780
[ 3280.259332][ T8472] ? merge_reloc_roots+0x470/0x470
[ 3280.260223][ T8472] btrfs_relocate_block_group+0x26e/0x4c0
[ 3280.261169][ T8472] btrfs_relocate_chunk+0x52/0x120
[ 3280.262199][ T8472] btrfs_balance+0xe2e/0x18f0
[ 3280.263146][ T8472] ? pvclock_clocksource_read+0xeb/0x190
[ 3280.264035][ T8472] ? btrfs_relocate_chunk+0x120/0x120
[ 3280.264851][ T8472] ? lock_contended+0x620/0x6e0
[ 3280.265587][ T8472] ? do_raw_spin_lock+0x1e0/0x1e0
[ 3280.266334][ T8472] ? do_raw_spin_unlock+0xa8/0x140
[ 3280.267103][ T8472] btrfs_ioctl_balance+0x1f9/0x460
[ 3280.267896][ T8472] btrfs_ioctl+0x24c8/0x4380
[ 3280.268599][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.269377][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.270149][ T8472] ? __asan_loadN+0xf/0x20
[ 3280.270875][ T8472] ? btrfs_ioctl_get_supported_features+0x30/0x30
[ 3280.271894][ T8472] ? kvm_sched_clock_read+0x18/0x30
[ 3280.272716][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.273503][ T8472] ? lock_downgrade+0x3f0/0x3f0
[ 3280.274258][ T8472] ? handle_mm_fault+0xad6/0x2150
[ 3280.275053][ T8472] ? do_vfs_ioctl+0xfc/0x9d0
[ 3280.275796][ T8472] ? ioctl_file_clone+0xe0/0xe0
[ 3280.276575][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.277393][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.278222][ T8472] ? check_flags+0x26/0x30
[ 3280.278943][ T8472] ? lock_is_held_type+0xc3/0xf0
[ 3280.279734][ T8472] ? syscall_enter_from_user_mode+0x1b/0x60
[ 3280.280672][ T8472] ? do_syscall_64+0x13/0x80
[ 3280.281394][ T8472] ? rcu_read_lock_sched_held+0xa1/0xd0
[ 3280.282276][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.283084][ T8472] ? __fget_light+0xae/0x110
[ 3280.283826][ T8472] __x64_sys_ioctl+0xc3/0x100
[ 3280.284576][ T8472] do_syscall_64+0x37/0x80
[ 3280.285268][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3280.286198][ T8472] RIP: 0033:0x7f18ca415427
[ 3280.286912][ T8472] Code: 00 00 90 48 8b 05 69 aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 39 aa 0c 00 f7 d8 64 89 01 48
[ 3280.289939][ T8472] RSP: 002b:00007ffcd1fed6b8 EFLAGS: 00000206 ORIG_RAX: 0000000000000010
[ 3280.291265][ T8472] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f18ca415427
[ 3280.292520][ T8472] RDX: 00007ffcd1fed6c8 RSI: 00000000c4009420 RDI: 0000000000000003
[ 3280.293776][ T8472] RBP: 0000000000000003 R08: 0000000000000003 R09: 0000000000000078
[ 3280.295056][ T8472] R10: fffffffffffff59d R11: 0000000000000206 R12: 0000000000000002
[ 3280.296340][ T8472] R13: 00007ffcd1fefa39 R14: 0000560042d94119 R15: 0000000000000000
[ 3280.297670][ T8472] CPU: 3 PID: 8472 Comm: btrfs Tainted: G W 5.10.0-87fa69f8a71b-for-next+ #14
[ 3280.299320][ T8472] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[ 3280.300804][ T8472] Call Trace:
[ 3280.301338][ T8472] dump_stack+0xbc/0xf9
[ 3280.302022][ T8472] ? btrfs_backref_add_tree_node+0xfa4/0x11d0
[ 3280.302979][ T8472] __warn.cold.11+0xe/0x53
[ 3280.303672][ T8472] ? btrfs_backref_add_tree_node+0xfa4/0x11d0
[ 3280.304627][ T8472] report_bug+0xf4/0x130
[ 3280.305301][ T8472] handle_bug+0x41/0x80
[ 3280.305959][ T8472] exc_invalid_op+0x18/0x40
[ 3280.306666][ T8472] asm_exc_invalid_op+0x12/0x20
[ 3280.307429][ T8472] RIP: 0010:btrfs_backref_add_tree_node+0xfa4/0x11d0
[ 3280.308481][ T8472] Code: 89 e7 4d 8d 66 30 e8 db 3e f3 ff 49 8d 46 2c 48 89 45 c0 e9 da f1 ff ff 49 8d 46 2c bb 8b ff ff ff 48 89 45 c0 e9 c8 f1 ff ff <0f> 0b 49 8d 46 2c bb fe ff ff ff 4d 8d 66 30 48 89 45 c0 e9 b0 f1
[ 3280.311594][ T8472] RSP: 0018:ffffc9000769f4c0 EFLAGS: 00010246
[ 3280.312577][ T8472] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 1ffff92000ed3e85
[ 3280.313872][ T8472] RDX: 0000000000000000 RSI: dffffc0000000000 RDI: ffff88805ef475e8
[ 3280.315142][ T8472] RBP: ffffc9000769f598 R08: ffffed100bde8ebe R09: ffffed100bde8ebe
[ 3280.316417][ T8472] R10: ffff88805ef475eb R11: ffffed100bde8ebd R12: ffff88805ef47020
[ 3280.317664][ T8472] R13: 0000217411380000 R14: ffff888126976b80 R15: 0000000000000000
[ 3280.318924][ T8472] build_backref_tree+0xc5/0x700
[ 3280.319696][ T8472] ? _raw_spin_unlock+0x22/0x30
[ 3280.320459][ T8472] ? release_extent_buffer+0x225/0x280
[ 3280.321321][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.322237][ T8472] relocate_tree_blocks+0x2a6/0xb60
[ 3280.323056][ T8472] ? kasan_unpoison_shadow+0x35/0x50
[ 3280.323897][ T8472] ? do_relocation+0xc10/0xc10
[ 3280.324707][ T8472] ? kasan_kmalloc+0x9/0x10
[ 3280.325406][ T8472] ? kmem_cache_alloc_trace+0x6a3/0xcb0
[ 3280.326258][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.327176][ T8472] ? rb_insert_color+0x342/0x360
[ 3280.327970][ T8472] ? add_tree_block.isra.36+0x236/0x2b0
[ 3280.328825][ T8472] relocate_block_group+0x2eb/0x780
[ 3280.329621][ T8472] ? merge_reloc_roots+0x470/0x470
[ 3280.330443][ T8472] btrfs_relocate_block_group+0x26e/0x4c0
[ 3280.331348][ T8472] btrfs_relocate_chunk+0x52/0x120
[ 3280.332179][ T8472] btrfs_balance+0xe2e/0x18f0
[ 3280.332951][ T8472] ? pvclock_clocksource_read+0xeb/0x190
[ 3280.333844][ T8472] ? btrfs_relocate_chunk+0x120/0x120
[ 3280.334711][ T8472] ? lock_contended+0x620/0x6e0
[ 3280.335562][ T8472] ? do_raw_spin_lock+0x1e0/0x1e0
[ 3280.336383][ T8472] ? do_raw_spin_unlock+0xa8/0x140
[ 3280.337215][ T8472] btrfs_ioctl_balance+0x1f9/0x460
[ 3280.338026][ T8472] btrfs_ioctl+0x24c8/0x4380
[ 3280.338743][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.339530][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.340326][ T8472] ? __asan_loadN+0xf/0x20
[ 3280.341065][ T8472] ? btrfs_ioctl_get_supported_features+0x30/0x30
[ 3280.342069][ T8472] ? kvm_sched_clock_read+0x18/0x30
[ 3280.342890][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.343670][ T8472] ? lock_downgrade+0x3f0/0x3f0
[ 3280.344440][ T8472] ? handle_mm_fault+0xad6/0x2150
[ 3280.345257][ T8472] ? do_vfs_ioctl+0xfc/0x9d0
[ 3280.346010][ T8472] ? ioctl_file_clone+0xe0/0xe0
[ 3280.346801][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.347640][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.348542][ T8472] ? check_flags+0x26/0x30
[ 3280.349322][ T8472] ? lock_is_held_type+0xc3/0xf0
[ 3280.350180][ T8472] ? syscall_enter_from_user_mode+0x1b/0x60
[ 3280.351128][ T8472] ? do_syscall_64+0x13/0x80
[ 3280.351872][ T8472] ? rcu_read_lock_sched_held+0xa1/0xd0
[ 3280.352819][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.353604][ T8472] ? __fget_light+0xae/0x110
[ 3280.354321][ T8472] __x64_sys_ioctl+0xc3/0x100
[ 3280.355050][ T8472] do_syscall_64+0x37/0x80
[ 3280.355733][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3280.356645][ T8472] RIP: 0033:0x7f18ca415427
[ 3280.357332][ T8472] Code: 00 00 90 48 8b 05 69 aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 39 aa 0c 00 f7 d8 64 89 01 48
[ 3280.360434][ T8472] RSP: 002b:00007ffcd1fed6b8 EFLAGS: 00000206 ORIG_RAX: 0000000000000010
[ 3280.361741][ T8472] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f18ca415427
[ 3280.362962][ T8472] RDX: 00007ffcd1fed6c8 RSI: 00000000c4009420 RDI: 0000000000000003
[ 3280.364165][ T8472] RBP: 0000000000000003 R08: 0000000000000003 R09: 0000000000000078
[ 3280.365366][ T8472] R10: fffffffffffff59d R11: 0000000000000206 R12: 0000000000000002
[ 3280.366570][ T8472] R13: 00007ffcd1fefa39 R14: 0000560042d94119 R15: 0000000000000000
[ 3280.367904][ T8472] irq event stamp: 4498529
[ 3280.368608][ T8472] hardirqs last enabled at (4498529): [<ffffffff83a60905>] irqentry_exit+0x35/0x80
[ 3280.370128][ T8472] hardirqs last disabled at (4498527): [<ffffffff83e00506>] __do_softirq+0x506/0x5a7
[ 3280.371656][ T8472] softirqs last enabled at (4498528): [<ffffffff83e0044f>] __do_softirq+0x44f/0x5a7
[ 3280.373173][ T8472] softirqs last disabled at (4498521): [<ffffffff83c01122>] asm_call_irq_on_stack+0x12/0x20
[ 3280.374833][ T8472] ---[ end trace 0689b357dfaddeab ]--- (gdb) list *(btrfs_backref_add_tree_node+0xfa4)
0xffffffff81970194 is in btrfs_backref_add_tree_node (fs/btrfs/backref.c:2627).
2622
2623 cur->is_reloc_root = 1;
2624 /* Only reloc backref cache cares about a specific root */
2625 if (cache->is_reloc) {
2626 root = find_reloc_root(cache->fs_info, cur->bytenr);
2627 if (WARN_ON(!root))
2628 return -ENOENT;
2629 cur->root = root;
2630 } else {
2631 /*
[ 3280.375749][ T8472] ------------[ cut here ]------------
[ 3280.376711][ T8472] kernel BUG at fs/btrfs/backref.c:2545!
[ 3280.377576][ T8472] invalid opcode: 0000 [#1] SMP KASAN PTI
[ 3280.378401][ T8472] CPU: 3 PID: 8472 Comm: btrfs Tainted: G W 5.10.0-87fa69f8a71b-for-next+ #14
[ 3280.379966][ T8472] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[ 3280.381287][ T8472] RIP: 0010:btrfs_backref_cleanup_node+0x5f7/0x600
[ 3280.382376][ T8472] Code: 5f 5d c3 48 89 df e8 08 e2 fa ff 49 8d 7f 70 e8 0f c8 b9 ff 41 80 67 71 fb 4c 89 e7 e8 b2 c9 b9 ff 49 8b 5f 68 e9 b2 fc ff ff <0f> 0b 0f 0b 0f 0b 0f 0b 90 0f 1f 44 00 00 55 48 89 e5 41 57 41 56
[ 3280.385743][ T8472] RSP: 0018:ffffc9000769f4c8 EFLAGS: 00010283
[ 3280.386689][ T8472] RAX: ffff8880436bc880 RBX: ffff8880436bc180 RCX: ffffffff8296ec9b
[ 3280.387966][ T8472] RDX: dffffc0000000000 RSI: dffffc0000000000 RDI: ffff8881ebd43f40
[ 3280.389242][ T8472] RBP: ffffc9000769f520 R08: 0000000000000000 R09: 0000000000000000
[ 3280.390503][ T8472] R10: ffffffff8533ba03 R11: fffffbfff0a67740 R12: ffff8880436bc190
[ 3280.391763][ T8472] R13: ffff8881ebd43900 R14: ffff8880436bc190 R15: ffff8881ebd43f00
[ 3280.393123][ T8472] FS: 00007f18ca3228c0(0000) GS:ffff8881f5e00000(0000) knlGS:0000000000000000
[ 3280.394562][ T8472] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 3280.395950][ T8472] CR2: 000055b5c456ab00 CR3: 000000012c08e002 CR4: 00000000001706e0
[ 3280.397253][ T8472] Call Trace:
[ 3280.399029][ T8472] btrfs_backref_error_cleanup+0x4df/0x530
[ 3280.399997][ T8472] build_backref_tree+0x1a5/0x700
[ 3280.400814][ T8472] ? _raw_spin_unlock+0x22/0x30
[ 3280.401597][ T8472] ? release_extent_buffer+0x225/0x280
[ 3280.402532][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.403470][ T8472] relocate_tree_blocks+0x2a6/0xb60
[ 3280.404295][ T8472] ? kasan_unpoison_shadow+0x35/0x50
[ 3280.405146][ T8472] ? do_relocation+0xc10/0xc10
[ 3280.405987][ T8472] ? kasan_kmalloc+0x9/0x10
[ 3280.407291][ T8472] ? kmem_cache_alloc_trace+0x6a3/0xcb0
[ 3280.408229][ T8472] ? free_extent_buffer.part.52+0xd7/0x140
[ 3280.409188][ T8472] ? rb_insert_color+0x342/0x360
[ 3280.409969][ T8472] ? add_tree_block.isra.36+0x236/0x2b0
[ 3280.410985][ T8472] relocate_block_group+0x2eb/0x780
[ 3280.411805][ T8472] ? merge_reloc_roots+0x470/0x470
[ 3280.412546][ T8472] btrfs_relocate_block_group+0x26e/0x4c0
[ 3280.413436][ T8472] btrfs_relocate_chunk+0x52/0x120
[ 3280.414302][ T8472] btrfs_balance+0xe2e/0x18f0
[ 3280.415022][ T8472] ? pvclock_clocksource_read+0xeb/0x190
[ 3280.415922][ T8472] ? btrfs_relocate_chunk+0x120/0x120
[ 3280.416842][ T8472] ? lock_contended+0x620/0x6e0
[ 3280.417598][ T8472] ? do_raw_spin_lock+0x1e0/0x1e0
[ 3280.418394][ T8472] ? do_raw_spin_unlock+0xa8/0x140
[ 3280.419208][ T8472] btrfs_ioctl_balance+0x1f9/0x460
[ 3280.420016][ T8472] btrfs_ioctl+0x24c8/0x4380
[ 3280.420734][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.421523][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.423328][ T8472] ? __asan_loadN+0xf/0x20
[ 3280.424033][ T8472] ? btrfs_ioctl_get_supported_features+0x30/0x30
[ 3280.425042][ T8472] ? kvm_sched_clock_read+0x18/0x30
[ 3280.425858][ T8472] ? check_chain_key+0x1f4/0x2f0
[ 3280.426632][ T8472] ? lock_downgrade+0x3f0/0x3f0
[ 3280.427401][ T8472] ? handle_mm_fault+0xad6/0x2150
[ 3280.428192][ T8472] ? do_vfs_ioctl+0xfc/0x9d0
[ 3280.428920][ T8472] ? ioctl_file_clone+0xe0/0xe0
[ 3280.429667][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.430463][ T8472] ? check_flags.part.50+0x6c/0x1e0
[ 3280.431252][ T8472] ? check_flags+0x26/0x30
[ 3280.431928][ T8472] ? lock_is_held_type+0xc3/0xf0
[ 3280.432681][ T8472] ? syscall_enter_from_user_mode+0x1b/0x60
[ 3280.433574][ T8472] ? do_syscall_64+0x13/0x80
[ 3280.434278][ T8472] ? rcu_read_lock_sched_held+0xa1/0xd0
[ 3280.435122][ T8472] ? __kasan_check_read+0x11/0x20
[ 3280.435886][ T8472] ? __fget_light+0xae/0x110
[ 3280.436586][ T8472] __x64_sys_ioctl+0xc3/0x100
[ 3280.437300][ T8472] do_syscall_64+0x37/0x80
[ 3280.438040][ T8472] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 3280.438932][ T8472] RIP: 0033:0x7f18ca415427
[ 3280.439595][ T8472] Code: 00 00 90 48 8b 05 69 aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 39 aa 0c 00 f7 d8 64 89 01 48
[ 3280.442689][ T8472] RSP: 002b:00007ffcd1fed6b8 EFLAGS: 00000206 ORIG_RAX: 0000000000000010
[ 3280.444023][ T8472] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f18ca415427
[ 3280.445292][ T8472] RDX: 00007ffcd1fed6c8 RSI: 00000000c4009420 RDI: 0000000000000003
[ 3280.446601][ T8472] RBP: 0000000000000003 R08: 0000000000000003 R09: 0000000000000078
[ 3280.448014][ T8472] R10: fffffffffffff59d R11: 0000000000000206 R12: 0000000000000002
[ 3280.449397][ T8472] R13: 00007ffcd1fefa39 R14: 0000560042d94119 R15: 0000000000000000
[ 3280.450805][ T8472] Modules linked in:
[ 3280.451530][ T8472] ---[ end trace 0689b357dfaddeac ]---
(gdb) list *(btrfs_backref_cleanup_node+0x5f7)
0xffffffff8196f037 is in btrfs_backref_cleanup_node (fs/btrfs/backref.c:2545).
2540 list_del(&edge->list[LOWER]);
2541 list_del(&edge->list[UPPER]);
2542 btrfs_backref_free_edge(cache, edge);
2543
2544 if (RB_EMPTY_NODE(&upper->rb_node)) {
2545 BUG_ON(!list_empty(&node->upper));
2546 btrfs_backref_drop_node(cache, node);
2547 node = upper;
2548 node->lowest = 1;
2549 continue;Quoted 1 line
Quoted 1 line
Josef