syzbot

EXT4-fs error (device loop2): ext4_mb_mark_diskspace_used:4047: comm syz.2.15: Allocating blocks 497-513 which overlap fs metadata
EXT4-fs (loop2): pa ffff88805cb99000: logic 128, phys. 385, len 8
EXT4-fs error (device loop2): ext4_mb_release_inode_pa:5388: group 0, free 0, pa_free 1
==================================================================
BUG: KASAN: use-after-free in ext4_ext_rm_leaf fs/ext4/extents.c:2602 [inline]
BUG: KASAN: use-after-free in ext4_ext_remove_space+0x311f/0x42e0 fs/ext4/extents.c:2939
Read of size 4 at addr ffff8880755f6d1c by task syz.2.15/5873

CPU: 1 PID: 5873 Comm: syz.2.15 Not tainted syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/12/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0x18c/0x250 lib/dump_stack.c:106
 print_address_description mm/kasan/report.c:364 [inline]
 print_report+0xa8/0x210 mm/kasan/report.c:468
 kasan_report+0x117/0x150 mm/kasan/report.c:581
 ext4_ext_rm_leaf fs/ext4/extents.c:2602 [inline]
 ext4_ext_remove_space+0x311f/0x42e0 fs/ext4/extents.c:2939
 ext4_ext_truncate+0x17d/0x2d0 fs/ext4/extents.c:4460
 ext4_truncate+0x9b8/0x1070 fs/ext4/inode.c:4271
 ext4_truncate_failed_write fs/ext4/truncate.h:22 [inline]
 ext4_write_begin+0xa6c/0x1070 fs/ext4/inode.c:1263
 ext4_da_write_begin+0x2df/0x9c0 fs/ext4/inode.c:2908
 generic_perform_write+0x2fe/0x5c0 mm/filemap.c:4031
 ext4_buffered_write_iter+0xcc/0x350 fs/ext4/file.c:299
 ext4_file_write_iter+0x1d9/0x1880 fs/ext4/file.c:-1
 call_write_iter include/linux/fs.h:2018 [inline]
 new_sync_write fs/read_write.c:491 [inline]
 vfs_write+0x46c/0x990 fs/read_write.c:584
 ksys_write+0x150/0x260 fs/read_write.c:637
 do_syscall_x64 arch/x86/entry/common.c:46 [inline]
 do_syscall_64+0x55/0xa0 arch/x86/entry/common.c:76
 entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7f293219c799
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f2933094028 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 00007f2932415fa0 RCX: 00007f293219c799
RDX: 0000000000040010 RSI: 0000200000000180 RDI: 0000000000000005
RBP: 00007f2932232bd9 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007f2932416038 R14: 00007f2932415fa0 R15: 00007ffdab39a0a8
 </TASK>

The buggy address belongs to the physical page:
page:ffffea0001d57d80 refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 pfn:0x755f6
flags: 0xfff00000000000(node=0|zone=1|lastcpupid=0x7ff)
page_type: 0xffffffff()
raw: 00fff00000000000 ffffea0001d89848 ffffea0001d57d48 0000000000000000
raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as freed
page last allocated via order 0, migratetype Movable, gfp_mask 0x140cca(GFP_HIGHUSER_MOVABLE|__GFP_COMP), pid 5862, tgid 5861 (syz.2.12), ts 74795278468, free_ts 75688147962
 set_page_owner include/linux/page_owner.h:31 [inline]
 post_alloc_hook+0x1c1/0x200 mm/page_alloc.c:1581
 prep_new_page mm/page_alloc.c:1588 [inline]
 get_page_from_freelist+0x1951/0x19e0 mm/page_alloc.c:3220
 __alloc_pages+0x1f0/0x460 mm/page_alloc.c:4486
 __folio_alloc+0x10/0x20 mm/page_alloc.c:4518
 vma_alloc_folio+0x47a/0x8f0 mm/mempolicy.c:2242
 shmem_alloc_folio+0x1a9/0x2a0 mm/shmem.c:1679
 shmem_alloc_and_acct_folio+0x1e6/0x6d0 mm/shmem.c:1704
 shmem_get_folio_gfp+0xcde/0x2aa0 mm/shmem.c:2041
 shmem_fault+0x1b8/0x810 mm/shmem.c:2248
 __do_fault+0x13b/0x4d0 mm/memory.c:4244
 do_read_fault mm/memory.c:4638 [inline]
 do_fault mm/memory.c:4775 [inline]
 do_pte_missing mm/memory.c:3689 [inline]
 handle_pte_fault mm/memory.c:5047 [inline]
 __handle_mm_fault mm/memory.c:5188 [inline]
 handle_mm_fault+0x2299/0x4c00 mm/memory.c:5353
 faultin_page mm/gup.c:868 [inline]
 __get_user_pages+0x5d0/0x1380 mm/gup.c:1167
 populate_vma_page_range+0x2c1/0x380 mm/gup.c:1593
 __mm_populate+0x260/0x390 mm/gup.c:1696
 mm_populate include/linux/mm.h:3383 [inline]
 vm_mmap_pgoff+0x2da/0x3f0 mm/util.c:561
 do_syscall_x64 arch/x86/entry/common.c:46 [inline]
 do_syscall_64+0x55/0xa0 arch/x86/entry/common.c:76
page last free stack trace:
 reset_page_owner include/linux/page_owner.h:24 [inline]
 free_pages_prepare mm/page_alloc.c:1181 [inline]
 free_unref_page_prepare+0x7b2/0x8c0 mm/page_alloc.c:2365
 free_unref_page_list+0xbe/0x860 mm/page_alloc.c:2504
 release_pages+0x1f7a/0x2200 mm/swap.c:1022
 __folio_batch_release+0x71/0xe0 mm/swap.c:1042
 folio_batch_release include/linux/pagevec.h:83 [inline]
 shmem_undo_range+0x630/0x1b20 mm/shmem.c:1026
 shmem_truncate_range mm/shmem.c:1135 [inline]
 shmem_evict_inode+0x245/0x9e0 mm/shmem.c:1264
 evict+0x4ca/0x8d0 fs/inode.c:705
 __dentry_kill+0x431/0x650 fs/dcache.c:611
 dentry_kill+0xb8/0x290 fs/dcache.c:-1
 dput+0xfe/0x1e0 fs/dcache.c:918
 __fput+0x5e5/0x970 fs/file_table.c:392
 task_work_run+0x1d4/0x260 kernel/task_work.c:245
 exit_task_work include/linux/task_work.h:43 [inline]
 do_exit+0x95a/0x2460 kernel/exit.c:883
 do_group_exit+0x21b/0x2d0 kernel/exit.c:1024
 get_signal+0x12fc/0x13f0 kernel/signal.c:2902
 arch_do_signal_or_restart+0xc2/0x800 arch/x86/kernel/signal.c:310

Memory state around the buggy address:
 ffff8880755f6c00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
 ffff8880755f6c80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>ffff8880755f6d00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
                            ^
 ffff8880755f6d80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
 ffff8880755f6e00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
==================================================================