bisecting fixing commit since a1b977b49b66c75e6c51a515f6700371ae720217 building syzkaller on 4a77ae0bdc5cd75ebe88ce7c896aae6bbf457a29 testing commit a1b977b49b66c75e6c51a515f6700371ae720217 with gcc (GCC) 8.4.1 20210217 kernel signature: d16b0542e0da4714ba5701e0e479b81a96a1273d2b78fc79604a0737cb170256 all runs: crashed: KASAN: use-after-free Read in ntfs_are_names_equal testing current HEAD 125222814e7b8f84df767d6ab622aff2a6d2f234 testing commit 125222814e7b8f84df767d6ab622aff2a6d2f234 with gcc (GCC) 8.4.1 20210217 kernel signature: a8fde72e6006ca59bdbed88f6e223d33a9269ea1d7c01b05955a56f71e9f756a all runs: crashed: KASAN: use-after-free Read in ntfs_are_names_equal revisions tested: 2, total time: 23m20.798231146s (build: 16m23.131747505s, test: 6m34.745145667s) the crash still happens on HEAD commit msg: Linux 4.19.182 crash: KASAN: use-after-free Read in ntfs_are_names_equal ntfs: (device loop5): ntfs_read_locked_inode(): Failed with error code -5. Marking corrupt inode 0xa as bad. Run chkdsk. ntfs: (device loop2): map_mft_record_page(): Mft record 0x1 is corrupt. Run chkdsk. ntfs: volume version 3.1. ================================================================== BUG: KASAN: use-after-free in ntfs_ucsncmp fs/ntfs/unistr.c:156 [inline] BUG: KASAN: use-after-free in ntfs_are_names_equal+0x178/0x1a0 fs/ntfs/unistr.c:75 Read of size 2 at addr ffff8881c9d4eee8 by task syz-executor.5/7425 CPU: 0 PID: 7425 Comm: syz-executor.5 Not tainted 4.19.182-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x17c/0x226 lib/dump_stack.c:118 print_address_description.cold.6+0x9/0x211 mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report mm/kasan/report.c:412 [inline] kasan_report.cold.7+0x242/0x2fe mm/kasan/report.c:396 __asan_report_load2_noabort+0x14/0x20 mm/kasan/report.c:431 ntfs_ucsncmp fs/ntfs/unistr.c:156 [inline] ntfs_are_names_equal+0x178/0x1a0 fs/ntfs/unistr.c:75 ntfs_attr_find+0x36c/0xb60 fs/ntfs/attrib.c:628 ntfs_external_attr_find fs/ntfs/attrib.c:1107 [inline] ntfs_attr_lookup+0x1077/0x2590 fs/ntfs/attrib.c:1205 ntfs_read_locked_attr_inode fs/ntfs/inode.c:1264 [inline] ntfs_attr_iget+0x702/0x2370 fs/ntfs/inode.c:251 ntfs_read_locked_inode+0x24b9/0x5030 fs/ntfs/inode.c:985 ntfs_iget+0xe6/0x120 fs/ntfs/inode.c:190 load_system_files fs/ntfs/super.c:1988 [inline] ntfs_fill_super+0x4b4f/0x79c0 fs/ntfs/super.c:2908 mount_bdev+0x26f/0x330 fs/super.c:1158 ntfs_mount+0x10/0x20 fs/ntfs/super.c:3065 mount_fs+0x7f/0x2b0 fs/super.c:1261 vfs_kern_mount.part.11+0x58/0x3d0 fs/namespace.c:961 vfs_kern_mount fs/namespace.c:951 [inline] do_new_mount fs/namespace.c:2469 [inline] do_mount+0x376/0x2710 fs/namespace.c:2799 ksys_mount+0xb1/0xd0 fs/namespace.c:3015 __do_sys_mount fs/namespace.c:3029 [inline] __se_sys_mount fs/namespace.c:3026 [inline] __x64_sys_mount+0xb9/0x150 fs/namespace.c:3026 do_syscall_64+0xd0/0x4e0 arch/x86/entry/common.c:293 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x466d6a Code: 48 c7 c2 bc ff ff ff f7 d8 64 89 02 b8 ff ff ff ff eb d2 e8 b8 04 00 00 0f 1f 84 00 00 00 00 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007fce19a2afa8 EFLAGS: 00000246 ORIG_RAX: 00000000000000a5 RAX: ffffffffffffffda RBX: 0000000020000200 RCX: 0000000000466d6a RDX: 0000000020000000 RSI: 0000000020000100 RDI: 00007fce19a2b000 RBP: 00007fce19a2b040 R08: 00007fce19a2b040 R09: 0000000020000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000020000000 R13: 0000000020000100 R14: 00007fce19a2b000 R15: 000000002007dc00 The buggy address belongs to the page: page:ffffea0007275380 count:0 mapcount:0 mapping:0000000000000000 index:0x1 flags: 0x17ffe0000000000() raw: 017ffe0000000000 ffffea0007275488 ffffea00072753c8 0000000000000000 raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff8881c9d4ed80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ffff8881c9d4ee00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff >ffff8881c9d4ee80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ^ ffff8881c9d4ef00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ffff8881c9d4ef80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ================================================================== ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. __ntfs_warning: 656 callbacks suppressed ntfs: (device loop4): is_boot_sector_ntfs(): Invalid end of sector marker. ntfs: (device loop4): map_mft_record_page(): Mft record 0x1 is corrupt. Run chkdsk. ntfs: (device loop4): map_mft_record(): Failed with error code 5. ntfs: (device loop4): ntfs_read_locked_inode(): Failed with error code -5. Marking corrupt inode 0x1 as bad. Run chkdsk. ntfs: (device loop4): load_system_files(): Failed to load $MFTMirr. Mounting read-only. Run ntfsfix and/or chkdsk. ntfs: (device loop4): map_mft_record_page(): Mft record 0xa is corrupt. Run chkdsk. ntfs: (device loop4): map_mft_record(): Failed with error code 5. ntfs: (device loop4): ntfs_read_locked_inode(): Failed with error code -5. Marking corrupt inode 0xa as bad. Run chkdsk. ntfs: (device loop4): load_and_init_upcase(): Failed to load $UpCase from the volume. Using default. ntfs: volume version 3.1. ntfs: (device loop4): map_mft_record_page(): Mft record 0x2 is corrupt. Run chkdsk. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1. ntfs: volume version 3.1.