Message-ID: <6b473885-cb16-4b57-b863-bcba22083c12@redhat.com>
Date: Mon, 29 Sep 2025 10:22:53 +0200
From: David Hildenbrand <david@...hat.com>
To: syzbot <syzbot+f1b91395f2bcbcc8128b@...kaller.appspotmail.com>,
 akpm@...ux-foundation.org, chengming.zhou@...ux.dev,
 linux-kernel@...r.kernel.org, linux-mm@...ck.org,
 syzkaller-bugs@...glegroups.com, xu.xin16@....com.cn
Subject: Re: [syzbot] [mm?] possible deadlock in ksm_do_scan

On 28.09.25 18:15, syzbot wrote:
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    b5a4da2c459f Add linux-next specific files for 20250924
> git tree:       linux-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=15a154e2580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=841973c5ab4f4157
> dashboard link: https://syzkaller.appspot.com/bug?extid=f1b91395f2bcbcc8128b
> compiler:       Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8
> 
> Unfortunately, I don't have any reproducer for this issue yet.
> 
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/2d0e39514585/disk-b5a4da2c.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/c7c8001fe2ea/vmlinux-b5a4da2c.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/807bea872f12/bzImage-b5a4da2c.xz
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+f1b91395f2bcbcc8128b@...kaller.appspotmail.com
> 
> ======================================================
> WARNING: possible circular locking dependency detected
> syzkaller #0 Not tainted
> ------------------------------------------------------
> ksmd/37 is trying to acquire lock:
> ffffffff8e645220 (fs_reclaim){+.+.}-{0:0}, at: might_alloc include/linux/sched/mm.h:318 [inline]
> ffffffff8e645220 (fs_reclaim){+.+.}-{0:0}, at: slab_pre_alloc_hook mm/slub.c:4896 [inline]
> ffffffff8e645220 (fs_reclaim){+.+.}-{0:0}, at: slab_alloc_node mm/slub.c:5220 [inline]
> ffffffff8e645220 (fs_reclaim){+.+.}-{0:0}, at: kmem_cache_alloc_noprof+0x45/0x6e0 mm/slub.c:5251
> 
> but task is already holding lock:
> ffff888028ba01e0 (&mm->mmap_lock){++++}-{4:4}, at: mmap_read_lock include/linux/mmap_lock.h:368 [inline]
> ffff888028ba01e0 (&mm->mmap_lock){++++}-{4:4}, at: scan_get_next_rmap_item mm/ksm.c:2532 [inline]
> ffff888028ba01e0 (&mm->mmap_lock){++++}-{4:4}, at: ksm_do_scan+0x602/0x6540 mm/ksm.c:2659
> 
> which lock already depends on the new lock.
> 
> 
> the existing dependency chain (in reverse order) is:
> 
> -> #7 (&mm->mmap_lock){++++}-{4:4}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         __might_fault+0xcc/0x130 mm/memory.c:7125
>         _copy_from_iter+0xf3/0x1790 lib/iov_iter.c:259
>         copy_from_iter include/linux/uio.h:228 [inline]
>         copy_from_iter_full include/linux/uio.h:245 [inline]
>         skb_do_copy_data_nocache include/net/sock.h:2269 [inline]
>         skb_copy_to_page_nocache include/net/sock.h:2295 [inline]
>         tcp_sendmsg_locked+0x2347/0x5540 net/ipv4/tcp.c:1272
>         tcp_sendmsg+0x2f/0x50 net/ipv4/tcp.c:1413
>         sock_sendmsg_nosec net/socket.c:714 [inline]
>         __sock_sendmsg+0x19c/0x270 net/socket.c:729
>         sock_write_iter+0x279/0x360 net/socket.c:1182
>         new_sync_write fs/read_write.c:593 [inline]
>         vfs_write+0x5c9/0xb30 fs/read_write.c:686
>         ksys_write+0x145/0x250 fs/read_write.c:738
>         do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
>         do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94
>         entry_SYSCALL_64_after_hwframe+0x77/0x7f
> 
> -> #6 (sk_lock-AF_INET){+.+.}-{0:0}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         lock_sock_nested+0x48/0x100 net/core/sock.c:3720
>         lock_sock include/net/sock.h:1679 [inline]
>         inet_shutdown+0x6a/0x390 net/ipv4/af_inet.c:907
>         nbd_mark_nsock_dead+0x2e9/0x560 drivers/block/nbd.c:318
>         sock_shutdown+0x15e/0x260 drivers/block/nbd.c:411
>         nbd_clear_sock drivers/block/nbd.c:1424 [inline]
>         nbd_config_put+0x342/0x790 drivers/block/nbd.c:1448
>         nbd_release+0xfe/0x140 drivers/block/nbd.c:1753
>         bdev_release+0x536/0x650 block/bdev.c:-1
>         blkdev_release+0x15/0x20 block/fops.c:702
>         __fput+0x44c/0xa70 fs/file_table.c:468
>         task_work_run+0x1d4/0x260 kernel/task_work.c:227
>         resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
>         exit_to_user_mode_loop+0xe9/0x130 kernel/entry/common.c:43
>         exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline]
>         syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline]
>         syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline]
>         do_syscall_64+0x2bd/0xfa0 arch/x86/entry/syscall_64.c:100
>         entry_SYSCALL_64_after_hwframe+0x77/0x7f
> 
> -> #5 (&nsock->tx_lock){+.+.}-{4:4}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         __mutex_lock_common kernel/locking/mutex.c:598 [inline]
>         __mutex_lock+0x187/0x1350 kernel/locking/mutex.c:760
>         nbd_handle_cmd drivers/block/nbd.c:1140 [inline]
>         nbd_queue_rq+0x257/0xf10 drivers/block/nbd.c:1204
>         blk_mq_dispatch_rq_list+0x4c0/0x1900 block/blk-mq.c:2129
>         __blk_mq_do_dispatch_sched block/blk-mq-sched.c:168 [inline]
>         blk_mq_do_dispatch_sched block/blk-mq-sched.c:182 [inline]
>         __blk_mq_sched_dispatch_requests+0xda4/0x1570 block/blk-mq-sched.c:307
>         blk_mq_sched_dispatch_requests+0xd7/0x190 block/blk-mq-sched.c:329
>         blk_mq_run_hw_queue+0x348/0x4f0 block/blk-mq.c:2367
>         blk_mq_dispatch_list+0xd0c/0xe00 include/linux/spinlock.h:-1
>         blk_mq_flush_plug_list+0x469/0x550 block/blk-mq.c:2976
>         __blk_flush_plug+0x3d3/0x4b0 block/blk-core.c:1225
>         blk_finish_plug block/blk-core.c:1252 [inline]
>         __submit_bio+0x2d3/0x5a0 block/blk-core.c:651
>         __submit_bio_noacct_mq block/blk-core.c:724 [inline]
>         submit_bio_noacct_nocheck+0x2fb/0xa50 block/blk-core.c:755
>         submit_bh fs/buffer.c:2829 [inline]
>         block_read_full_folio+0x599/0x830 fs/buffer.c:2447
>         filemap_read_folio+0x117/0x380 mm/filemap.c:2444
>         do_read_cache_folio+0x350/0x590 mm/filemap.c:4009
>         read_mapping_folio include/linux/pagemap.h:999 [inline]
>         read_part_sector+0xb6/0x2b0 block/partitions/core.c:722
>         adfspart_check_ICS+0xa4/0xa50 block/partitions/acorn.c:360
>         check_partition block/partitions/core.c:141 [inline]
>         blk_add_partitions block/partitions/core.c:589 [inline]
>         bdev_disk_changed+0x75f/0x14b0 block/partitions/core.c:693
>         blkdev_get_whole+0x380/0x510 block/bdev.c:748
>         bdev_open+0x31e/0xd30 block/bdev.c:957
>         blkdev_open+0x457/0x600 block/fops.c:694
>         do_dentry_open+0x953/0x13f0 fs/open.c:965
>         vfs_open+0x3b/0x340 fs/open.c:1097
>         do_open fs/namei.c:3975 [inline]
>         path_openat+0x2ee5/0x3830 fs/namei.c:4134
>         do_filp_open+0x1fa/0x410 fs/namei.c:4161
>         do_sys_openat2+0x121/0x1c0 fs/open.c:1437
>         do_sys_open fs/open.c:1452 [inline]
>         __do_sys_openat fs/open.c:1468 [inline]
>         __se_sys_openat fs/open.c:1463 [inline]
>         __x64_sys_openat+0x138/0x170 fs/open.c:1463
>         do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
>         do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94
>         entry_SYSCALL_64_after_hwframe+0x77/0x7f
> 
> -> #4 (&cmd->lock){+.+.}-{4:4}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         __mutex_lock_common kernel/locking/mutex.c:598 [inline]
>         __mutex_lock+0x187/0x1350 kernel/locking/mutex.c:760
>         nbd_queue_rq+0xc8/0xf10 drivers/block/nbd.c:1196
>         blk_mq_dispatch_rq_list+0x4c0/0x1900 block/blk-mq.c:2129
>         __blk_mq_do_dispatch_sched block/blk-mq-sched.c:168 [inline]
>         blk_mq_do_dispatch_sched block/blk-mq-sched.c:182 [inline]
>         __blk_mq_sched_dispatch_requests+0xda4/0x1570 block/blk-mq-sched.c:307
>         blk_mq_sched_dispatch_requests+0xd7/0x190 block/blk-mq-sched.c:329
>         blk_mq_run_hw_queue+0x348/0x4f0 block/blk-mq.c:2367
>         blk_mq_dispatch_list+0xd0c/0xe00 include/linux/spinlock.h:-1
>         blk_mq_flush_plug_list+0x469/0x550 block/blk-mq.c:2976
>         __blk_flush_plug+0x3d3/0x4b0 block/blk-core.c:1225
>         blk_finish_plug block/blk-core.c:1252 [inline]
>         __submit_bio+0x2d3/0x5a0 block/blk-core.c:651
>         __submit_bio_noacct_mq block/blk-core.c:724 [inline]
>         submit_bio_noacct_nocheck+0x2fb/0xa50 block/blk-core.c:755
>         submit_bh fs/buffer.c:2829 [inline]
>         block_read_full_folio+0x599/0x830 fs/buffer.c:2447
>         filemap_read_folio+0x117/0x380 mm/filemap.c:2444
>         do_read_cache_folio+0x350/0x590 mm/filemap.c:4009
>         read_mapping_folio include/linux/pagemap.h:999 [inline]
>         read_part_sector+0xb6/0x2b0 block/partitions/core.c:722
>         adfspart_check_ICS+0xa4/0xa50 block/partitions/acorn.c:360
>         check_partition block/partitions/core.c:141 [inline]
>         blk_add_partitions block/partitions/core.c:589 [inline]
>         bdev_disk_changed+0x75f/0x14b0 block/partitions/core.c:693
>         blkdev_get_whole+0x380/0x510 block/bdev.c:748
>         bdev_open+0x31e/0xd30 block/bdev.c:957
>         blkdev_open+0x457/0x600 block/fops.c:694
>         do_dentry_open+0x953/0x13f0 fs/open.c:965
>         vfs_open+0x3b/0x340 fs/open.c:1097
>         do_open fs/namei.c:3975 [inline]
>         path_openat+0x2ee5/0x3830 fs/namei.c:4134
>         do_filp_open+0x1fa/0x410 fs/namei.c:4161
>         do_sys_openat2+0x121/0x1c0 fs/open.c:1437
>         do_sys_open fs/open.c:1452 [inline]
>         __do_sys_openat fs/open.c:1468 [inline]
>         __se_sys_openat fs/open.c:1463 [inline]
>         __x64_sys_openat+0x138/0x170 fs/open.c:1463
>         do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
>         do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94
>         entry_SYSCALL_64_after_hwframe+0x77/0x7f
> 
> -> #3 (set->srcu){.+.+}-{0:0}:
>         lock_sync+0xba/0x160 kernel/locking/lockdep.c:5916
>         srcu_lock_sync include/linux/srcu.h:173 [inline]
>         __synchronize_srcu+0x96/0x3a0 kernel/rcu/srcutree.c:1429
>         elevator_switch+0x12b/0x640 block/elevator.c:588
>         elevator_change+0x315/0x4c0 block/elevator.c:691
>         elevator_set_default+0x186/0x260 block/elevator.c:767
>         blk_register_queue+0x34e/0x3f0 block/blk-sysfs.c:942
>         __add_disk+0x677/0xd50 block/genhd.c:528
>         add_disk_fwnode+0xfc/0x480 block/genhd.c:597
>         add_disk include/linux/blkdev.h:775 [inline]
>         nbd_dev_add+0x717/0xae0 drivers/block/nbd.c:1981
>         nbd_init+0x168/0x1f0 drivers/block/nbd.c:2688
>         do_one_initcall+0x236/0x820 init/main.c:1283
>         do_initcall_level+0x104/0x190 init/main.c:1345
>         do_initcalls+0x59/0xa0 init/main.c:1361
>         kernel_init_freeable+0x334/0x4b0 init/main.c:1593
>         kernel_init+0x1d/0x1d0 init/main.c:1483
>         ret_from_fork+0x4bc/0x870 arch/x86/kernel/process.c:158
>         ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
> 
> -> #2 (&q->elevator_lock){+.+.}-{4:4}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         __mutex_lock_common kernel/locking/mutex.c:598 [inline]
>         __mutex_lock+0x187/0x1350 kernel/locking/mutex.c:760
>         elevator_change+0x1e5/0x4c0 block/elevator.c:689
>         elevator_set_none+0x42/0xb0 block/elevator.c:782
>         blk_mq_elv_switch_none block/blk-mq.c:5032 [inline]
>         __blk_mq_update_nr_hw_queues block/blk-mq.c:5075 [inline]
>         blk_mq_update_nr_hw_queues+0x598/0x1ab0 block/blk-mq.c:5133
>         nbd_start_device+0x17f/0xb10 drivers/block/nbd.c:1486
>         nbd_genl_connect+0x135b/0x18f0 drivers/block/nbd.c:2236
>         genl_family_rcv_msg_doit+0x215/0x300 net/netlink/genetlink.c:1115
>         genl_family_rcv_msg net/netlink/genetlink.c:1195 [inline]
>         genl_rcv_msg+0x60e/0x790 net/netlink/genetlink.c:1210
>         netlink_rcv_skb+0x208/0x470 net/netlink/af_netlink.c:2552
>         genl_rcv+0x28/0x40 net/netlink/genetlink.c:1219
>         netlink_unicast_kernel net/netlink/af_netlink.c:1320 [inline]
>         netlink_unicast+0x82f/0x9e0 net/netlink/af_netlink.c:1346
>         netlink_sendmsg+0x805/0xb30 net/netlink/af_netlink.c:1896
>         sock_sendmsg_nosec net/socket.c:714 [inline]
>         __sock_sendmsg+0x21c/0x270 net/socket.c:729
>         ____sys_sendmsg+0x505/0x830 net/socket.c:2617
>         ___sys_sendmsg+0x21f/0x2a0 net/socket.c:2671
>         __sys_sendmsg net/socket.c:2703 [inline]
>         __do_sys_sendmsg net/socket.c:2708 [inline]
>         __se_sys_sendmsg net/socket.c:2706 [inline]
>         __x64_sys_sendmsg+0x19b/0x260 net/socket.c:2706
>         do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
>         do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94
>         entry_SYSCALL_64_after_hwframe+0x77/0x7f
> 
> -> #1 (&q->q_usage_counter(io)#49){++++}-{0:0}:
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         blk_alloc_queue+0x538/0x620 block/blk-core.c:461
>         blk_mq_alloc_queue block/blk-mq.c:4399 [inline]
>         __blk_mq_alloc_disk+0x15c/0x340 block/blk-mq.c:4446
>         nbd_dev_add+0x46c/0xae0 drivers/block/nbd.c:1951
>         nbd_init+0x168/0x1f0 drivers/block/nbd.c:2688
>         do_one_initcall+0x236/0x820 init/main.c:1283
>         do_initcall_level+0x104/0x190 init/main.c:1345
>         do_initcalls+0x59/0xa0 init/main.c:1361
>         kernel_init_freeable+0x334/0x4b0 init/main.c:1593
>         kernel_init+0x1d/0x1d0 init/main.c:1483
>         ret_from_fork+0x4bc/0x870 arch/x86/kernel/process.c:158
>         ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
> 
> -> #0 (fs_reclaim){+.+.}-{0:0}:
>         check_prev_add kernel/locking/lockdep.c:3165 [inline]
>         check_prevs_add kernel/locking/lockdep.c:3284 [inline]
>         validate_chain+0xb9b/0x2140 kernel/locking/lockdep.c:3908
>         __lock_acquire+0xab9/0xd20 kernel/locking/lockdep.c:5237
>         lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5868
>         __fs_reclaim_acquire mm/page_alloc.c:4269 [inline]
>         fs_reclaim_acquire+0x72/0x100 mm/page_alloc.c:4283
>         might_alloc include/linux/sched/mm.h:318 [inline]
>         slab_pre_alloc_hook mm/slub.c:4896 [inline]
>         slab_alloc_node mm/slub.c:5220 [inline]
>         kmem_cache_alloc_noprof+0x45/0x6e0 mm/slub.c:5251
>         alloc_rmap_item mm/ksm.c:565 [inline]
>         get_next_rmap_item mm/ksm.c:2378 [inline]
>         scan_get_next_rmap_item mm/ksm.c:2565 [inline]
>         ksm_do_scan+0xbaf/0x6540 mm/ksm.c:2659
>         ksm_scan_thread+0x10b/0x4b0 mm/ksm.c:2684
>         kthread+0x711/0x8a0 kernel/kthread.c:463
>         ret_from_fork+0x4bc/0x870 arch/x86/kernel/process.c:158
>         ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
> 
> other info that might help us debug this:
> 
> Chain exists of:
>    fs_reclaim --> sk_lock-AF_INET --> &mm->mmap_lock
> 
>   Possible unsafe locking scenario:
> 
>         CPU0                    CPU1
>         ----                    ----
>    rlock(&mm->mmap_lock);
>                                 lock(sk_lock-AF_INET);
>                                 lock(&mm->mmap_lock);
>    lock(fs_reclaim);

KSM simply holds the mmap lock here while performing an allocation that 
might trigger reclaim. Nothing special about that.

So the issue (if true) is likely somewhere in the  sk_lock-AF_INET path.

-- 
Cheers

David / dhildenb

Powered by blists - more mailing lists