{"cve":"CVE-2025-68178","nvd":{"published":"2025-12-16T14:15:49.760","modified":"2026-10-14T13:10:00.187","status":"Deferred","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nblk-cgroup: fix possible deadlock while configuring policy\n\nFollowing deadlock can be triggered easily by lockdep:\n\nWARNING: possible circular locking dependency detected\n6.17.0-rc3-00124-ga12c2658ced0 #1665 Not tainted\n------------------------------------------------------\ncheck/1334 is trying to acquire lock:\nff1100011d9d0678 (&q->sysfs_lock){+.+.}-{4:4}, at: blk_unregister_queue+0x53/0x180\n\nbut task is already holding lock:\nff1100011d9d00e0 (&q->q_usage_counter(queue)#3){++++}-{0:0}, at: del_gendisk+0xba/0x110\n\nwhich lock already depends on the new lock.\n\nthe existing dependency chain (in reverse order) is:\n\n-> #2 (&q->q_usage_counter(queue)#3){++++}-{0:0}:\n       blk_queue_enter+0x40b/0x470\n       blkg_conf_prep+0x7b/0x3c0\n       tg_set_limit+0x10a/0x3e0\n       cgroup_file_write+0xc6/0x420\n       kernfs_fop_write_iter+0x189/0x280\n       vfs_write+0x256/0x490\n       ksys_write+0x83/0x190\n       __x64_sys_write+0x21/0x30\n       x64_sys_call+0x4608/0x4630\n       do_syscall_64+0xdb/0x6b0\n       entry_SYSCALL_64_after_hwframe+0x76/0x7e\n\n-> #1 (&q->rq_qos_mutex){+.+.}-{4:4}:\n       __mutex_lock+0xd8/0xf50\n       mutex_lock_nested+0x2b/0x40\n       wbt_init+0x17e/0x280\n       wbt_enable_default+0xe9/0x140\n       blk_register_queue+0x1da/0x2e0\n       __add_disk+0x38c/0x5d0\n       add_disk_fwnode+0x89/0x250\n       device_add_disk+0x18/0x30\n       virtblk_probe+0x13a3/0x1800\n       virtio_dev_probe+0x389/0x610\n       really_probe+0x136/0x620\n       __driver_probe_device+0xb3/0x230\n       driver_probe_device+0x2f/0xe0\n       __driver_attach+0x158/0x250\n       bus_for_each_dev+0xa9/0x130\n       driver_attach+0x26/0x40\n       bus_add_driver+0x178/0x3d0\n       driver_register+0x7d/0x1c0\n       __register_virtio_driver+0x2c/0x60\n       virtio_blk_init+0x6f/0xe0\n       do_one_initcall+0x94/0x540\n       kernel_init_freeable+0x56a/0x7b0\n       kernel_init+0x2b/0x270\n       ret_from_fork+0x268/0x4c0\n       ret_from_fork_asm+0x1a/0x30\n\n-> #0 (&q->sysfs_lock){+.+.}-{4:4}:\n       __lock_acquire+0x1835/0x2940\n       lock_acquire+0xf9/0x450\n       __mutex_lock+0xd8/0xf50\n       mutex_lock_nested+0x2b/0x40\n       blk_unregister_queue+0x53/0x180\n       __del_gendisk+0x226/0x690\n       del_gendisk+0xba/0x110\n       sd_remove+0x49/0xb0 [sd_mod]\n       device_remove+0x87/0xb0\n       device_release_driver_internal+0x11e/0x230\n       device_release_driver+0x1a/0x30\n       bus_remove_device+0x14d/0x220\n       device_del+0x1e1/0x5a0\n       __scsi_remove_device+0x1ff/0x2f0\n       scsi_remove_device+0x37/0x60\n       sdev_store_delete+0x77/0x100\n       dev_attr_store+0x1f/0x40\n       sysfs_kf_write+0x65/0x90\n       kernfs_fop_write_iter+0x189/0x280\n       vfs_write+0x256/0x490\n       ksys_write+0x83/0x190\n       __x64_sys_write+0x21/0x30\n       x64_sys_call+0x4608/0x4630\n       do_syscall_64+0xdb/0x6b0\n       entry_SYSCALL_64_after_hwframe+0x76/0x7e\n\nother info that might help us debug this:\n\nChain exists of:\n  &q->sysfs_lock --> &q->rq_qos_mutex --> &q->q_usage_counter(queue)#3\n\n Possible unsafe locking scenario:\n\n       CPU0                    CPU1\n       ----                    ----\n  lock(&q->q_usage_counter(queue)#3);\n                               lock(&q->rq_qos_mutex);\n                               lock(&q->q_usage_counter(queue)#3);\n  lock(&q->sysfs_lock);\n\nRoot cause is that queue_usage_counter is grabbed with rq_qos_mutex\nheld in blkg_conf_prep(), while queue should be freezed before\nrq_qos_mutex from other context.\n\nThe blk_queue_enter() from blkg_conf_prep() is used to protect against\npolicy deactivation, which is already protected with blkcg_mutex, hence\nconvert blk_queue_enter() to blkcg_mutex to fix this problem. Meanwhile,\nconsider that blkcg_mutex is held after queue is freezed from policy\ndeactivation, also convert blkg_alloc() to use GFP_NOIO.","cvss_v4":null,"cvss_v3":null,"cwes":[],"references":["https://git.kernel.org/stable/c/0585b24d71197dd9ee8cf79c168a31628c631960","https://git.kernel.org/stable/c/56ac639d6fa6fbb99caee74ee1c7276fc9bb47ed","https://git.kernel.org/stable/c/5d726c4dbeeddef612e6bed27edd29733f4d13af","https://git.kernel.org/stable/c/e1729523759cda2c0afb76b1c88e0d2f2ef5b7cb"],"products":[],"source":"nvd","synced_at":"2026-10-19T22:17:07.138532058+00:00"},"epss":{"cve":"CVE-2025-68178","score":0.00201,"percentile":0.09893,"date":"2026-10-25"},"kev":null,"osv":[],"attribution":["This product uses the NVD API but is not endorsed or certified by the NVD.","EPSS scores: FIRST Exploit Prediction Scoring System, https://www.first.org/epss","Known exploited vulnerabilities: CISA Known Exploited Vulnerabilities Catalog, https://www.cisa.gov/known-exploited-vulnerabilities-catalog","Package advisories: OSV (https://osv.dev), from the GitHub Advisory Database, PyPA, Go, RustSec and other ecosystem databases, under CC-BY 4.0 or CC0","CVE records: the CVE Program's CVE List (https://www.cve.org), under its terms of use (https://www.cve.org/Legal/TermsOfUse). CVE is a registered trademark of The MITRE Corporation; this product is not affiliated with the CVE Program."]}