{"id":"CVE-2023-52903","state":"PUBLISHED","title":"io_uring: lock overflowing for IOPOLL","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nio_uring: lock overflowing for IOPOLL\n\nsyzbot reports an issue with overflow filling for IOPOLL:\n\nWARNING: CPU: 0 PID: 28 at io_uring/io_uring.c:734 io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734\nCPU: 0 PID: 28 Comm: kworker/u4:1 Not tainted 6.2.0-rc3-syzkaller-16369-g358a161a6a9e #0\nWorkqueue: events_unbound io_ring_exit_work\nCall trace:\n io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734\n io_req_cqe_overflow+0x5c/0x70 io_uring/io_uring.c:773\n io_fill_cqe_req io_uring/io_uring.h:168 [inline]\n io_do_iopoll+0x474/0x62c io_uring/rw.c:1065\n io_iopoll_try_reap_events+0x6c/0x108 io_uring/io_uring.c:1513\n io_uring_try_cancel_requests+0x13c/0x258 io_uring/io_uring.c:3056\n io_ring_exit_work+0xec/0x390 io_uring/io_uring.c:2869\n process_one_work+0x2d8/0x504 kernel/workqueue.c:2289\n worker_thread+0x340/0x610 kernel/workqueue.c:2436\n kthread+0x12c/0x158 kernel/kthread.c:376\n ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:863\n\nThere is no real problem for normal IOPOLL as flush is also called with\nuring_lock taken, but it's getting more complicated for IOPOLL|SQPOLL,\nfor which __io_cqring_overflow_flush() happens from the CQ waiting path.","publishedAt":"2024-08-21T06:10:43.857Z","updatedAt":"2026-08-05T09:11:53.830Z","severity":{"score":7.8,"label":"High","version":"3.1","vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","source":"Linux"},"assigner":"Linux","products":["Linux Linux"],"affected":[{"vendor":"Linux","product":"Linux","defaultStatus":"unaffected","versions":[{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"de77faee280163ff03b7ab64af6c9d779a43d4c4","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"ed4629d1e968359fbb91d0a3780b1e86a2c08845","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"7fc3990dad04a677606337ebc61964094d6cb41b","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"544d163d659d45a206d8929370d5a2984e546cb7","versionType":"git"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["io_uring/rw.c"]},{"vendor":"Linux","product":"Linux","defaultStatus":"affected","versions":[{"version":"5.1","status":"affected"},{"version":"0","status":"unaffected","lessThan":"5.1","versionType":"semver"},{"version":"5.10.165","status":"unaffected","lessThanOrEqual":"5.10.*","versionType":"semver"},{"version":"5.15.89","status":"unaffected","lessThanOrEqual":"5.15.*","versionType":"semver"},{"version":"6.1.7","status":"unaffected","lessThanOrEqual":"6.1.*","versionType":"semver"},{"version":"6.2","status":"unaffected","lessThanOrEqual":"*","versionType":"original_commit_for_fix"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["io_uring/rw.c"]}],"references":["https://git.kernel.org/stable/c/de77faee280163ff03b7ab64af6c9d779a43d4c4","https://git.kernel.org/stable/c/ed4629d1e968359fbb91d0a3780b1e86a2c08845","https://git.kernel.org/stable/c/7fc3990dad04a677606337ebc61964094d6cb41b","https://git.kernel.org/stable/c/544d163d659d45a206d8929370d5a2984e546cb7"],"weaknesses":[],"ingestedAt":"2026-10-29T20:06:33.064Z","source":{"repository":"CVEProject/cvelistV5","path":"cves/2023/52xxx/CVE-2023-52903.json","commit":"556c9d3a8cd17e218bb8e452ba7cd619a384ff47"},"raw":{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2023-52903","state":"PUBLISHED","datePublished":"2024-08-21T06:10:43.857Z","dateUpdated":"2026-08-05T09:11:53.830Z","assignerShortName":"Linux","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","dateReserved":"2024-08-21T06:07:11.014Z"},"containers":{"cna":{"title":"io_uring: lock overflowing for IOPOLL","descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nio_uring: lock overflowing for IOPOLL\n\nsyzbot reports an issue with overflow filling for IOPOLL:\n\nWARNING: CPU: 0 PID: 28 at io_uring/io_uring.c:734 io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734\nCPU: 0 PID: 28 Comm: kworker/u4:1 Not tainted 6.2.0-rc3-syzkaller-16369-g358a161a6a9e #0\nWorkqueue: events_unbound io_ring_exit_work\nCall trace:\n io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734\n io_req_cqe_overflow+0x5c/0x70 io_uring/io_uring.c:773\n io_fill_cqe_req io_uring/io_uring.h:168 [inline]\n io_do_iopoll+0x474/0x62c io_uring/rw.c:1065\n io_iopoll_try_reap_events+0x6c/0x108 io_uring/io_uring.c:1513\n io_uring_try_cancel_requests+0x13c/0x258 io_uring/io_uring.c:3056\n io_ring_exit_work+0xec/0x390 io_uring/io_uring.c:2869\n process_one_work+0x2d8/0x504 kernel/workqueue.c:2289\n worker_thread+0x340/0x610 kernel/workqueue.c:2436\n kthread+0x12c/0x158 kernel/kthread.c:376\n ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:863\n\nThere is no real problem for normal IOPOLL as flush is also called with\nuring_lock taken, but it's getting more complicated for IOPOLL|SQPOLL,\nfor which __io_cqring_overflow_flush() happens from the CQ waiting path."}],"affected":[{"vendor":"Linux","product":"Linux","defaultStatus":"unaffected","versions":[{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"de77faee280163ff03b7ab64af6c9d779a43d4c4","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"ed4629d1e968359fbb91d0a3780b1e86a2c08845","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"7fc3990dad04a677606337ebc61964094d6cb41b","versionType":"git"},{"version":"2b188cc1bb857a9d4701ae59aa7768b5124e262e","status":"affected","lessThan":"544d163d659d45a206d8929370d5a2984e546cb7","versionType":"git"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["io_uring/rw.c"]},{"vendor":"Linux","product":"Linux","defaultStatus":"affected","versions":[{"version":"5.1","status":"affected"},{"version":"0","status":"unaffected","lessThan":"5.1","versionType":"semver"},{"version":"5.10.165","status":"unaffected","lessThanOrEqual":"5.10.*","versionType":"semver"},{"version":"5.15.89","status":"unaffected","lessThanOrEqual":"5.15.*","versionType":"semver"},{"version":"6.1.7","status":"unaffected","lessThanOrEqual":"6.1.*","versionType":"semver"},{"version":"6.2","status":"unaffected","lessThanOrEqual":"*","versionType":"original_commit_for_fix"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["io_uring/rw.c"]}],"references":[{"url":"https://git.kernel.org/stable/c/de77faee280163ff03b7ab64af6c9d779a43d4c4"},{"url":"https://git.kernel.org/stable/c/ed4629d1e968359fbb91d0a3780b1e86a2c08845"},{"url":"https://git.kernel.org/stable/c/7fc3990dad04a677606337ebc61964094d6cb41b"},{"url":"https://git.kernel.org/stable/c/544d163d659d45a206d8929370d5a2984e546cb7"}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":7.8,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:L - The bug is in io_uring IOPOLL completion overflow (`io_do_iopoll` → `io_fill_cqe_req` → `io_req_cqe_overflow`), reachable only via local `io_uring_setup`/`io_uring_enter` on an attacker-owned ring, not via network packet processing.\nAC:L - An attacker can force CQ overflow on an IOPOLL ring while concurrently posting aux CQEs (e.g. `IORING_MSG_DATA` via msg_ring’s `io_post_aux_cqe`, which took only `completion_lock`) so both sides of the `cq_overflow_list` race are attacker-controlled.\nPR:L - io_uring is usable by unprivileged local users by default (`kernel.io_uring_disabled=0`); IOPOLL and msg_ring require no CAP_SYS_ADMIN or init-namespace root.\nUI:N - The attacker creates the ring(s), submits IOPOLL work, overflows the CQ, and drives concurrent overflow/aux posting themselves; no victim action is required.\nS:U - Impact is kernel memory corruption within the host OS authority (local privilege escalation); no VM escape, IOMMU bypass, or sandbox boundary is crossed.\nC:H - Missing `completion_lock` while mutating `cq_overflow_list` races with other overflow/aux posters holding that lock, enabling linked-list corruption and use-after-free of overflow entries that can be leveraged for arbitrary kernel memory disclosure.\nI:H - Concurrent unlocked list add/delete/kfree on `cq_overflow_list` yields classic heap UAF/double-free primitives exploitable for arbitrary write and control-flow hijacking.\nA:H - The same race can oops/panic the kernel via list corruption, and the observed lockdep path is a `WARN` in `io_cqring_event_overflow` (High under panic_on_warn / oops deployments)."}]}],"providerMetadata":{"shortName":"Linux","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","dateUpdated":"2026-08-05T09:11:53.830Z"},"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"5.10.165"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"5.15.89"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"6.1.7"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"6.2"}]}]}],"x_generator":{"engine":"bippy-1.2.0"}},"adp":[{"title":"CISA ADP Vulnrichment","metrics":[{"other":{"type":"ssvc","content":{"id":"CVE-2023-52903","role":"CISA Coordinator","options":[{"Exploitation":"none"},{"Automatable":"no"},{"Technical Impact":"partial"}],"version":"2.0.3","timestamp":"2024-09-10T16:03:21.069603Z"}}}],"providerMetadata":{"shortName":"CISA-ADP","orgId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","dateUpdated":"2024-09-12T17:33:14.061Z"}}]}}}