[syzbot] [bcachefs?] KASAN: slab-out-of-bounds Read in bch2_btree_node_read_done

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 083fc6d7fa0d Merge tag 'sched-urgent-2025-09-26' of git://..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=16fbe2e2580000
kernel config: https://syzkaller.appspot.com/x/.config?x=f5b21423ca3f0a96
dashboard link: https://syzkaller.appspot.com/bug?extid=ba71155d3eacc8f42477
compiler: Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16433d34580000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=13e79f12580000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/198ae77e2418/disk-083fc6d7.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/3d3f065fd75c/vmlinux-083fc6d7.xz
kernel image: https://storage.googleapis.com/syzbot-assets/e65812e9d7b0/bzImage-083fc6d7.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/5219cf8dcb94/mount_0.gz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+ba7115...@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: slab-out-of-bounds in bch2_btree_node_read_done+0x950/0x5550 fs/bcachefs/btree_io.c:1128
Read of size 8 at addr ffff88805874a010 by task syz.0.17/6055

CPU: 0 UID: 0 PID: 6055 Comm: syz.0.17 Not tainted syzkaller #0 PREEMPT_{RT,(full)}
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/18/2025
Call Trace:
<TASK>
dump_stack_lvl+0x189/0x250 lib/dump_stack.c:120
print_address_description mm/kasan/report.c:378 [inline]
print_report+0xca/0x240 mm/kasan/report.c:482
kasan_report+0x118/0x150 mm/kasan/report.c:595
bch2_btree_node_read_done+0x950/0x5550 fs/bcachefs/btree_io.c:1128
btree_node_read_work+0x40e/0xe60 fs/bcachefs/btree_io.c:1440
bch2_btree_node_read+0x887/0x2a00 fs/bcachefs/btree_io.c:-1
__bch2_btree_root_read fs/bcachefs/btree_io.c:1906 [inline]
bch2_btree_root_read+0x5f0/0x760 fs/bcachefs/btree_io.c:1928
read_btree_roots+0x2c6/0x840 fs/bcachefs/recovery.c:615
bch2_fs_recovery+0x261f/0x3a50 fs/bcachefs/recovery.c:1006
bch2_fs_start+0xaaf/0xda0 fs/bcachefs/super.c:1213
bch2_fs_get_tree+0xb39/0x1520 fs/bcachefs/fs.c:2488
vfs_get_tree+0x8f/0x2b0 fs/super.c:1815
do_new_mount+0x2a2/0x9e0 fs/namespace.c:3808
do_mount fs/namespace.c:4136 [inline]
__do_sys_mount fs/namespace.c:4347 [inline]
__se_sys_mount+0x317/0x410 fs/namespace.c:4324
do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
do_syscall_64+0xfa/0x3b0 arch/x86/entry/syscall_64.c:94
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7feb1599066a
Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb a6 e8 de 1a 00 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffe09e19a98 EFLAGS: 00000246 ORIG_RAX: 00000000000000a5
RAX: ffffffffffffffda RBX: 00007ffe09e19b20 RCX: 00007feb1599066a
RDX: 00002000000058c0 RSI: 0000200000000000 RDI: 00007ffe09e19ae0
RBP: 00002000000058c0 R08: 00007ffe09e19b20 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000200000000000
R13: 00007ffe09e19ae0 R14: 00000000000058f2 R15: 0000200000000180
</TASK>

Allocated by task 6055:
kasan_save_stack mm/kasan/common.c:47 [inline]
kasan_save_track+0x3e/0x80 mm/kasan/common.c:68
poison_kmalloc_redzone mm/kasan/common.c:388 [inline]
__kasan_kmalloc+0x93/0xb0 mm/kasan/common.c:405
kasan_kmalloc include/linux/kasan.h:260 [inline]
__do_kmalloc_node mm/slub.c:4376 [inline]
__kvmalloc_node_noprof+0x28d/0x550 mm/slub.c:5067
btree_node_data_alloc+0xdc/0x270 fs/bcachefs/btree_cache.c:151
__bch2_btree_node_mem_alloc+0x1dc/0x2e0 fs/bcachefs/btree_cache.c:193
bch2_fs_btree_cache_init+0x314/0x880 fs/bcachefs/btree_cache.c:653
bch2_fs_alloc fs/bcachefs/super.c:998 [inline]
bch2_fs_open+0x1da6/0x26e0 fs/bcachefs/super.c:2433
bch2_fs_get_tree+0x44f/0x1520 fs/bcachefs/fs.c:2472
vfs_get_tree+0x8f/0x2b0 fs/super.c:1815
do_new_mount+0x2a2/0x9e0 fs/namespace.c:3808
do_mount fs/namespace.c:4136 [inline]
__do_sys_mount fs/namespace.c:4347 [inline]
__se_sys_mount+0x317/0x410 fs/namespace.c:4324
do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
do_syscall_64+0xfa/0x3b0 arch/x86/entry/syscall_64.c:94
entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff888058748000
which belongs to the cache kmalloc-rcl-8k of size 8192
The buggy address is located 16 bytes to the right of
allocated 8192-byte region [ffff888058748000, ffff88805874a000)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x58748
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
memcg:ffff8880273c6ce1
flags: 0x80000000000040(head|node=0|zone=1)
page_type: f5(slab)
raw: 0080000000000040 ffff8880198483c0 dead000000000122 0000000000000000
raw: 0000000000000000 0000000080020002 00000000f5000000 ffff8880273c6ce1
head: 0080000000000040 ffff8880198483c0 dead000000000122 0000000000000000
head: 0000000000000000 0000000080020002 00000000f5000000 ffff8880273c6ce1
head: 0080000000000003 ffffea000161d201 00000000ffffffff 00000000ffffffff
head: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000008
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Reclaimable, gfp_mask 0x528d0(GFP_NOWAIT|__GFP_RECLAIMABLE|__GFP_IO|__GFP_FS|__GFP_NORETRY|__GFP_COMP), pid 6055, tgid 6055 (syz.0.17), ts 114481807879, free_ts 0
set_page_owner include/linux/page_owner.h:32 [inline]
post_alloc_hook+0x240/0x2a0 mm/page_alloc.c:1851
prep_new_page mm/page_alloc.c:1859 [inline]
get_page_from_freelist+0x2119/0x21b0 mm/page_alloc.c:3858
__alloc_frozen_pages_noprof+0x181/0x370 mm/page_alloc.c:5148
alloc_pages_mpol+0xd1/0x380 mm/mempolicy.c:2416
alloc_slab_page mm/slub.c:2492 [inline]
allocate_slab+0x8a/0x370 mm/slub.c:2660
new_slab mm/slub.c:2714 [inline]
___slab_alloc+0x8d1/0xdc0 mm/slub.c:3901
__slab_alloc mm/slub.c:3992 [inline]
__slab_alloc_node mm/slub.c:4067 [inline]
slab_alloc_node mm/slub.c:4228 [inline]
__do_kmalloc_node mm/slub.c:4375 [inline]
__kvmalloc_node_noprof+0x1d6/0x550 mm/slub.c:5067
btree_node_data_alloc+0xdc/0x270 fs/bcachefs/btree_cache.c:151
__bch2_btree_node_mem_alloc+0x1dc/0x2e0 fs/bcachefs/btree_cache.c:193
bch2_fs_btree_cache_init+0x314/0x880 fs/bcachefs/btree_cache.c:653
bch2_fs_alloc fs/bcachefs/super.c:998 [inline]
bch2_fs_open+0x1da6/0x26e0 fs/bcachefs/super.c:2433
bch2_fs_get_tree+0x44f/0x1520 fs/bcachefs/fs.c:2472
vfs_get_tree+0x8f/0x2b0 fs/super.c:1815
do_new_mount+0x2a2/0x9e0 fs/namespace.c:3808
do_mount fs/namespace.c:4136 [inline]
__do_sys_mount fs/namespace.c:4347 [inline]
__se_sys_mount+0x317/0x410 fs/namespace.c:4324
do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
do_syscall_64+0xfa/0x3b0 arch/x86/entry/syscall_64.c:94
page_owner free stack trace missing

Memory state around the buggy address:
ffff888058749f00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ffff888058749f80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff88805874a000: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
^
ffff88805874a080: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffff88805874a100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup

[syzbot]

Hello,

syzbot tried to test the proposed patch but the build/boot failed:

failed to apply patch:
can't find file to patch at input line 5
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c
|index 590cd29f3e86..ab14fff1452f 100644
|--- a/fs/bcachefs/btree_io.c
|+++ b/fs/bcachefs/btree_io.c
--------------------------
No file to patch. Skipping patch.
1 out of 1 hunk ignored



Tested on:

commit: 30d4efb2 Merge tag 'for-linus-6.18-rc1-tag' of git://g..
git tree: upstream

kernel config: https://syzkaller.appspot.com/x/.config?x=f5b21423ca3f0a96
dashboard link: https://syzkaller.appspot.com/bug?extid=ba71155d3eacc8f42477
compiler:

patch: https://syzkaller.appspot.com/x/patch.diff?x=16afc334580000

[Edward Adam Davis]

#syz test

diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c
index 590cd29f3e86..ab14fff1452f 100644
--- a/fs/bcachefs/btree_io.c
+++ b/fs/bcachefs/btree_io.c

@@ -1087,6 +1087,13 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
"bad magic: want %llx, got %llx",
bset_magic(c), le64_to_cpu(b->data->magic));

+ btree_err_on(ptr_written >= btree_sectors(c),
+ -BCH_ERR_btree_node_read_err_must_retry,
+ c, ca, b, NULL, NULL,
+ btree_node_bad_magic,
+ "wrong written %u, btree sectors is %lu",
+ ptr_written, btree_sectors(c));
+
if (b->key.k.type == KEY_TYPE_btree_ptr_v2) {
struct bch_btree_ptr_v2 *bp =
&bkey_i_to_btree_ptr_v2(&b->key)->v;

[Edward Adam Davis]

#syz test: upstream 083fc6d7fa0d

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+ba7115...@syzkaller.appspotmail.com
Tested-by: syzbot+ba7115...@syzkaller.appspotmail.com

Tested on:

commit: 083fc6d7 Merge tag 'sched-urgent-2025-09-26' of git://..
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
console output: https://syzkaller.appspot.com/x/log.txt?x=12d02334580000

kernel config: https://syzkaller.appspot.com/x/.config?x=f5b21423ca3f0a96
dashboard link: https://syzkaller.appspot.com/bug?extid=ba71155d3eacc8f42477

compiler: Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8

patch: https://syzkaller.appspot.com/x/patch.diff?x=137706e2580000

Note: testing is done by a robot and is best-effort only.

[Edward Adam Davis]

syzbot reported btree node oob in bch2_btree_node_read_done. [1]

Add sanity check for written, avoid exceeding the allowed access limits
for btree node.

[1]

BUG: KASAN: slab-out-of-bounds in bch2_btree_node_read_done+0x950/0x5550 fs/bcachefs/btree_io.c:1128

Call Trace:

bch2_btree_node_read_done+0x950/0x5550 fs/bcachefs/btree_io.c:1128
btree_node_read_work+0x40e/0xe60 fs/bcachefs/btree_io.c:1440

bch2_btree_root_read+0x5f0/0x760 fs/bcachefs/btree_io.c:1928
read_btree_roots+0x2c6/0x840 fs/bcachefs/recovery.c:615
bch2_fs_recovery+0x261f/0x3a50 fs/bcachefs/recovery.c:1006
bch2_fs_start+0xaaf/0xda0 fs/bcachefs/super.c:1213
bch2_fs_get_tree+0xb39/0x1520 fs/bcachefs/fs.c:2488
vfs_get_tree+0x8f/0x2b0 fs/super.c:1815
do_new_mount+0x2a2/0x9e0 fs/namespace.c:3808

Reported-by: syzbot+ba7115...@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=ba71155d3eacc8f42477
Tested-by: syzbot+ba7115...@syzkaller.appspotmail.com
Signed-off-by: Edward Adam Davis <ead...@qq.com>
---
fs/bcachefs/btree_io.c | 7 +++++++
1 file changed, 7 insertions(+)

--
2.43.0

[kernel test robot]

Hi Edward,

kernel test robot noticed the following build warnings:

[auto build test WARNING on brauner-vfs/vfs.all]
[also build test WARNING on v6.17]
[cannot apply to linus/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url: https://github.com/intel-lab-lkp/linux/commits/Edward-Adam-Davis/bcachefs-Prevent-written-from-exceeding-sectors/20250930-132425
base: https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git vfs.all
patch link: https://lore.kernel.org/r/tencent_68D8E912EEDECFF079226E202DFD6E70950A%40qq.com
patch subject: [PATCH] bcachefs: Prevent written from exceeding sectors
config: sh-randconfig-001-20251001 (https://download.01.org/0day-ci/archive/20251001/202510010446...@intel.com/config)
compiler: sh4-linux-gcc (GCC) 13.4.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251001/202510010446...@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <l...@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202510010446...@intel.com/

All warnings (new ones prefixed by >>):

fs/bcachefs/btree_io.c: In function 'bch2_btree_node_read_done':
>> fs/bcachefs/btree_io.c:1094:22: warning: format '%lu' expects argument of type 'long unsigned int', but argument 13 has type 'size_t' {aka 'unsigned int'} [-Wformat=]
1094 | "wrong written %u, btree sectors is %lu",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1095 | ptr_written, btree_sectors(c));
| ~~~~~~~~~~~~~~~~
| |
| size_t {aka unsigned int}
fs/bcachefs/btree_io.c:661:32: note: in definition of macro 'btree_err'
661 | msg, ##__VA_ARGS__); \
| ^~~
fs/bcachefs/btree_io.c:1090:9: note: in expansion of macro 'btree_err_on'
1090 | btree_err_on(ptr_written >= btree_sectors(c),
| ^~~~~~~~~~~~
fs/bcachefs/btree_io.c:1094:60: note: format string is defined here
1094 | "wrong written %u, btree sectors is %lu",
| ~~^
| |
| long unsigned int
| %u


vim +1094 fs/bcachefs/btree_io.c

1050
1051 int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
1052 struct btree *b,
1053 struct bch_io_failures *failed,
1054 struct printbuf *err_msg)
1055 {
1056 struct btree_node_entry *bne;
1057 struct sort_iter *iter;
1058 struct btree_node *sorted;
1059 struct bkey_packed *k;
1060 struct bset *i;
1061 bool used_mempool, blacklisted;
1062 bool updated_range = b->key.k.type == KEY_TYPE_btree_ptr_v2 &&
1063 BTREE_PTR_RANGE_UPDATED(&bkey_i_to_btree_ptr_v2(&b->key)->v);
1064 unsigned ptr_written = btree_ptr_sectors_written(bkey_i_to_s_c(&b->key));
1065 u64 max_journal_seq = 0;
1066 struct printbuf buf = PRINTBUF;
1067 int ret = 0, write = READ;
1068 u64 start_time = local_clock();
1069
1070 b->version_ondisk = U16_MAX;
1071 /* We might get called multiple times on read retry: */
1072 b->written = 0;
1073
1074 iter = mempool_alloc(&c->fill_iter, GFP_NOFS);
1075 sort_iter_init(iter, b, (btree_blocks(c) + 1) * 2);
1076
1077 if (bch2_meta_read_fault("btree"))
1078 btree_err(-BCH_ERR_btree_node_read_err_must_retry,
1079 c, ca, b, NULL, NULL,
1080 btree_node_fault_injected,
1081 "dynamic fault");
1082
1083 btree_err_on(le64_to_cpu(b->data->magic) != bset_magic(c),
1084 -BCH_ERR_btree_node_read_err_must_retry,
1085 c, ca, b, NULL, NULL,
1086 btree_node_bad_magic,
1087 "bad magic: want %llx, got %llx",
1088 bset_magic(c), le64_to_cpu(b->data->magic));
1089
1090 btree_err_on(ptr_written >= btree_sectors(c),
1091 -BCH_ERR_btree_node_read_err_must_retry,
1092 c, ca, b, NULL, NULL,
1093 btree_node_bad_magic,
> 1094 "wrong written %u, btree sectors is %lu",
1095 ptr_written, btree_sectors(c));
1096
1097 if (b->key.k.type == KEY_TYPE_btree_ptr_v2) {
1098 struct bch_btree_ptr_v2 *bp =
1099 &bkey_i_to_btree_ptr_v2(&b->key)->v;
1100
1101 bch2_bpos_to_text(&buf, b->data->min_key);
1102 prt_str(&buf, "-");
1103 bch2_bpos_to_text(&buf, b->data->max_key);
1104
1105 btree_err_on(b->data->keys.seq != bp->seq,
1106 -BCH_ERR_btree_node_read_err_must_retry,
1107 c, ca, b, NULL, NULL,
1108 btree_node_bad_seq,
1109 "got wrong btree node: got\n%s",
1110 (printbuf_reset(&buf),
1111 bch2_btree_node_header_to_text(&buf, b->data),
1112 buf.buf));
1113 } else {
1114 btree_err_on(!b->data->keys.seq,
1115 -BCH_ERR_btree_node_read_err_must_retry,
1116 c, ca, b, NULL, NULL,
1117 btree_node_bad_seq,
1118 "bad btree header: seq 0\n%s",
1119 (printbuf_reset(&buf),
1120 bch2_btree_node_header_to_text(&buf, b->data),
1121 buf.buf));
1122 }
1123
1124 while (b->written < (ptr_written ?: btree_sectors(c))) {
1125 unsigned sectors;
1126 bool first = !b->written;
1127
1128 if (first) {
1129 bne = NULL;
1130 i = &b->data->keys;
1131 } else {
1132 bne = write_block(b);
1133 i = &bne->keys;
1134
1135 if (i->seq != b->data->keys.seq)
1136 break;
1137 }
1138
1139 struct nonce nonce = btree_nonce(i, b->written << 9);
1140 bool good_csum_type = bch2_checksum_type_valid(c, BSET_CSUM_TYPE(i));
1141
1142 btree_err_on(!good_csum_type,
1143 bch2_csum_type_is_encryption(BSET_CSUM_TYPE(i))
1144 ? -BCH_ERR_btree_node_read_err_must_retry
1145 : -BCH_ERR_btree_node_read_err_want_retry,
1146 c, ca, b, i, NULL,
1147 bset_unknown_csum,
1148 "unknown checksum type %llu", BSET_CSUM_TYPE(i));
1149
1150 if (first) {
1151 sectors = vstruct_sectors(b->data, c->block_bits);
1152 if (btree_err_on(b->written + sectors > (ptr_written ?: btree_sectors(c)),
1153 -BCH_ERR_btree_node_read_err_fixable,
1154 c, ca, b, i, NULL,
1155 bset_past_end_of_btree_node,
1156 "bset past end of btree node (offset %u len %u but written %zu)",
1157 b->written, sectors, ptr_written ?: btree_sectors(c)))
1158 i->u64s = 0;
1159 if (good_csum_type) {
1160 struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, b->data);
1161 bool csum_bad = bch2_crc_cmp(b->data->csum, csum);
1162 if (csum_bad)
1163 bch2_io_error(ca, BCH_MEMBER_ERROR_checksum);
1164
1165 btree_err_on(csum_bad,
1166 -BCH_ERR_btree_node_read_err_want_retry,
1167 c, ca, b, i, NULL,
1168 bset_bad_csum,
1169 "%s",
1170 (printbuf_reset(&buf),
1171 bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), b->data->csum, csum),
1172 buf.buf));
1173
1174 ret = bset_encrypt(c, i, b->written << 9);
1175 if (bch2_fs_fatal_err_on(ret, c,
1176 "decrypting btree node: %s", bch2_err_str(ret)))
1177 goto fsck_err;
1178 }
1179
1180 btree_err_on(btree_node_type_is_extents(btree_node_type(b)) &&
1181 !BTREE_NODE_NEW_EXTENT_OVERWRITE(b->data),
1182 -BCH_ERR_btree_node_read_err_incompatible,
1183 c, NULL, b, NULL, NULL,
1184 btree_node_unsupported_version,
1185 "btree node does not have NEW_EXTENT_OVERWRITE set");
1186 } else {
1187 sectors = vstruct_sectors(bne, c->block_bits);
1188 if (btree_err_on(b->written + sectors > (ptr_written ?: btree_sectors(c)),
1189 -BCH_ERR_btree_node_read_err_fixable,
1190 c, ca, b, i, NULL,
1191 bset_past_end_of_btree_node,
1192 "bset past end of btree node (offset %u len %u but written %zu)",
1193 b->written, sectors, ptr_written ?: btree_sectors(c)))
1194 i->u64s = 0;
1195 if (good_csum_type) {
1196 struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne);
1197 bool csum_bad = bch2_crc_cmp(bne->csum, csum);
1198 if (ca && csum_bad)
1199 bch2_io_error(ca, BCH_MEMBER_ERROR_checksum);
1200
1201 btree_err_on(csum_bad,
1202 -BCH_ERR_btree_node_read_err_want_retry,
1203 c, ca, b, i, NULL,
1204 bset_bad_csum,
1205 "%s",
1206 (printbuf_reset(&buf),
1207 bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), bne->csum, csum),
1208 buf.buf));
1209
1210 ret = bset_encrypt(c, i, b->written << 9);
1211 if (bch2_fs_fatal_err_on(ret, c,
1212 "decrypting btree node: %s", bch2_err_str(ret)))
1213 goto fsck_err;
1214 }
1215 }
1216
1217 b->version_ondisk = min(b->version_ondisk,
1218 le16_to_cpu(i->version));
1219
1220 ret = validate_bset(c, ca, b, i, b->written, READ, failed, err_msg);
1221 if (ret)
1222 goto fsck_err;
1223
1224 if (!b->written)
1225 btree_node_set_format(b, b->data->format);
1226
1227 ret = validate_bset_keys(c, b, i, READ, failed, err_msg);
1228 if (ret)
1229 goto fsck_err;
1230
1231 SET_BSET_BIG_ENDIAN(i, CPU_BIG_ENDIAN);
1232
1233 blacklisted = bch2_journal_seq_is_blacklisted(c,
1234 le64_to_cpu(i->journal_seq),
1235 true);
1236
1237 btree_err_on(blacklisted && first,
1238 -BCH_ERR_btree_node_read_err_fixable,
1239 c, ca, b, i, NULL,
1240 bset_blacklisted_journal_seq,
1241 "first btree node bset has blacklisted journal seq (%llu)",
1242 le64_to_cpu(i->journal_seq));
1243
1244 btree_err_on(blacklisted && ptr_written,
1245 -BCH_ERR_btree_node_read_err_fixable,
1246 c, ca, b, i, NULL,
1247 first_bset_blacklisted_journal_seq,
1248 "found blacklisted bset (journal seq %llu) in btree node at offset %u-%u/%u",
1249 le64_to_cpu(i->journal_seq),
1250 b->written, b->written + sectors, ptr_written);
1251
1252 b->written = min(b->written + sectors, btree_sectors(c));
1253
1254 if (blacklisted && !first)
1255 continue;
1256
1257 sort_iter_add(iter,
1258 vstruct_idx(i, 0),
1259 vstruct_last(i));
1260
1261 max_journal_seq = max(max_journal_seq, le64_to_cpu(i->journal_seq));
1262 }
1263
1264 if (ptr_written) {
1265 btree_err_on(b->written < ptr_written,
1266 -BCH_ERR_btree_node_read_err_want_retry,
1267 c, ca, b, NULL, NULL,
1268 btree_node_data_missing,
1269 "btree node data missing: expected %u sectors, found %u",
1270 ptr_written, b->written);
1271 } else {
1272 for (bne = write_block(b);
1273 bset_byte_offset(b, bne) < btree_buf_bytes(b);
1274 bne = (void *) bne + block_bytes(c))
1275 btree_err_on(bne->keys.seq == b->data->keys.seq &&
1276 !bch2_journal_seq_is_blacklisted(c,
1277 le64_to_cpu(bne->keys.journal_seq),
1278 true),
1279 -BCH_ERR_btree_node_read_err_want_retry,
1280 c, ca, b, NULL, NULL,
1281 btree_node_bset_after_end,
1282 "found bset signature after last bset");
1283 }
1284
1285 sorted = btree_bounce_alloc(c, btree_buf_bytes(b), &used_mempool);
1286 sorted->keys.u64s = 0;
1287
1288 b->nr = bch2_key_sort_fix_overlapping(c, &sorted->keys, iter);
1289 memset((uint8_t *)(sorted + 1) + b->nr.live_u64s * sizeof(u64), 0,
1290 btree_buf_bytes(b) -
1291 sizeof(struct btree_node) -
1292 b->nr.live_u64s * sizeof(u64));
1293
1294 b->data->keys.u64s = sorted->keys.u64s;
1295 *sorted = *b->data;
1296 swap(sorted, b->data);
1297 set_btree_bset(b, b->set, &b->data->keys);
1298 b->nsets = 1;
1299 b->data->keys.journal_seq = cpu_to_le64(max_journal_seq);
1300
1301 BUG_ON(b->nr.live_u64s != le16_to_cpu(b->data->keys.u64s));
1302
1303 btree_bounce_free(c, btree_buf_bytes(b), used_mempool, sorted);
1304
1305 i = &b->data->keys;
1306 for (k = i->start; k != vstruct_last(i);) {
1307 struct bkey tmp;
1308 struct bkey_s u = __bkey_disassemble(b, k, &tmp);
1309
1310 ret = btree_node_bkey_val_validate(c, b, u.s_c, READ);
1311 if (ret == -BCH_ERR_fsck_delete_bkey ||
1312 (static_branch_unlikely(&bch2_inject_invalid_keys) &&
1313 !bversion_cmp(u.k->bversion, MAX_VERSION))) {
1314 btree_keys_account_key_drop(&b->nr, 0, k);
1315
1316 i->u64s = cpu_to_le16(le16_to_cpu(i->u64s) - k->u64s);
1317 memmove_u64s_down(k, bkey_p_next(k),
1318 (u64 *) vstruct_end(i) - (u64 *) k);
1319 set_btree_bset_end(b, b->set);
1320 set_btree_node_need_rewrite(b);
1321 set_btree_node_need_rewrite_error(b);
1322 continue;
1323 }
1324 if (ret)
1325 goto fsck_err;
1326
1327 if (u.k->type == KEY_TYPE_btree_ptr_v2) {
1328 struct bkey_s_btree_ptr_v2 bp = bkey_s_to_btree_ptr_v2(u);
1329
1330 bp.v->mem_ptr = 0;
1331 }
1332
1333 k = bkey_p_next(k);
1334 }
1335
1336 bch2_bset_build_aux_tree(b, b->set, false);
1337
1338 set_needs_whiteout(btree_bset_first(b), true);
1339
1340 btree_node_reset_sib_u64s(b);
1341
1342 if (updated_range)
1343 bch2_btree_node_drop_keys_outside_node(b);
1344
1345 /*
1346 * XXX:
1347 *
1348 * We deadlock if too many btree updates require node rewrites while
1349 * we're still in journal replay.
1350 *
1351 * This is because btree node rewrites generate more updates for the
1352 * interior updates (alloc, backpointers), and if those updates touch
1353 * new nodes and generate more rewrites - well, you see the problem.
1354 *
1355 * The biggest cause is that we don't use the btree write buffer (for
1356 * the backpointer updates - this needs some real thought on locking in
1357 * order to fix.
1358 *
1359 * The problem with this workaround (not doing the rewrite for degraded
1360 * nodes in journal replay) is that those degraded nodes persist, and we
1361 * don't want that (this is a real bug when a btree node write completes
1362 * with fewer replicas than we wanted and leaves a degraded node due to
1363 * device _removal_, i.e. the device went away mid write).
1364 *
1365 * It's less of a bug here, but still a problem because we don't yet
1366 * have a way of tracking degraded data - we another index (all
1367 * extents/btree nodes, by replicas entry) in order to fix properly
1368 * (re-replicate degraded data at the earliest possible time).
1369 */
1370 if (c->recovery.passes_complete & BIT_ULL(BCH_RECOVERY_PASS_journal_replay)) {
1371 scoped_guard(rcu)
1372 bkey_for_each_ptr(bch2_bkey_ptrs(bkey_i_to_s(&b->key)), ptr) {
1373 struct bch_dev *ca2 = bch2_dev_rcu(c, ptr->dev);
1374
1375 if (!ca2 || ca2->mi.state != BCH_MEMBER_STATE_rw) {
1376 set_btree_node_need_rewrite(b);
1377 set_btree_node_need_rewrite_degraded(b);
1378 }
1379 }
1380 }
1381
1382 if (!ptr_written) {
1383 set_btree_node_need_rewrite(b);
1384 set_btree_node_need_rewrite_ptr_written_zero(b);
1385 }
1386 fsck_err:
1387 mempool_free(iter, &c->fill_iter);
1388 printbuf_exit(&buf);
1389 bch2_time_stats_update(&c->times[BCH_TIME_btree_node_read_done], start_time);
1390 return ret;
1391 }
1392

--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki