{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-74714","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-08-15T05:44:03.928Z","datePublished":"2026-08-22T15:33:09.643Z","dateUpdated":"2026-08-25T05:41:57.341Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-25T05:41:57.341Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: tcp: Fix use-after-free in bpf_iter_tcp_established_batch()\n\nreqsk_queue_hash_req() publishes a TCP_NEW_SYN_RECV request_sock onto\nthe ehash chain, drops the bucket lock, and only afterwards sets\nrsk_refcnt to 3.\n\nLockless readers such as __inet_lookup_established() handle this with\nrefcount_inc_not_zero(), but bpf_iter_tcp_established_batch() uses plain\nsock_hold() while holding the bucket lock, on the assumption that the\nlock guarantees sk_refcnt > 0. That assumption does not hold for\nrequest_sock:\n\n  CPU 0                                CPU 1\n  -----                                -----\n  tcp_conn_request()\n   reqsk_queue_hash_req()\n    inet_ehash_insert(req)\n     spin_lock(bucket)\n     __sk_nulls_add_node_rcu(req)      // rsk_refcnt == 0\n     spin_unlock(bucket)\n                                       bpf_iter_tcp_established_batch()\n                                        spin_lock(bucket)\n                                        sock_hold(req)   <-- addition on 0\n                                        spin_unlock(bucket)\n    refcount_set(&req->rsk_refcnt, 3)  // clobbers saturated value\n\nwhich surfaces as:\n\n  refcount_t: addition on 0; use-after-free.\n  WARNING: lib/refcount.c:25 at refcount_warn_saturate+0x48/0x90, CPU#1\n  Call Trace:\n   bpf_iter_tcp_established_batch+0x14e/0x170\n   bpf_iter_tcp_batch+0x53/0x200\n   bpf_iter_tcp_seq_next+0x27/0x70\n   bpf_seq_read+0x107/0x410\n   vfs_read+0xb9/0x380\n\nThe iterator's stolen reference is lost when the publishing CPU's\nrefcount_set() overwrites the count, leaving the socket one reference\nshort. When the last legitimate owner drops its reference the reqsk is\nfreed while still reachable, leading to use-after-free.\n\nThis reproduces in seconds with tcp_syncookies=0, a handful of threads\ndoing connect()/close() to a local listener while others read an\niter/tcp link in a tight loop.\n\nUse refcount_inc_not_zero() and skip the socket on failure. A skipped\nsocket is still part of the bucket, so keep counting it in expected.\nThe reallocations are sized from expected, and a request sock whose\nrefcount gets published while the lock is held across the last realloc\nmust already have room.\n\nA skipped socket is counted in expected but never batched, so end_sk\ncan be short of expected on a batch that is actually complete. Decide\ncompleteness by whether the walk left any socket behind instead. The\nWARN after the locked realloc checks the same, replacing an\nend_sk == expected check that could not hold on that path since\ncommit cdec67a489d4 (\"bpf: tcp: Make sure iter->batch always\ncontains a full bucket snapshot\").\n\nIf every matching socket in a bucket is mid-init (refcount 0), end_sk\nstays 0. Advance to the next bucket rather than returning a batch entry\nthat was never filled this round."}],"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 - bpf_iter_tcp_established_batch() is reached only via local bpf() to load a BPF_TRACE_ITER TCP program, create an iterator link/fd, and read() it (bpf_seq_read->bpf_iter_tcp_batch); per kernel CNA guidance BPF iterator paths are Local even though concurrent TCP_NEW_SYN_RECV creation can be driven by network SYNs.\nAC:L - The fix commit reproduces in seconds with threads doing connect()/close() while others read iter/tcp in a tight loop; the attacker controls both the BPF iterator and TCP connection churn, so the refcount race between sock_hold() and refcount_set() is reliably winnable without uncontrollable victim state.\nPR:L - Loading BPF_PROG_TYPE_TRACING/BPF_TRACE_ITER requires CAP_BPF and CAP_PERFMON at bpf_prog_load(); per kernel CNA guidance these are Low privileges because BPF tokens and user namespaces can delegate them to non-init-namespace users, not only real root.\nUI:N - No victim interaction is required; the attacker loads their own iter/tcp BPF program and link, then triggers the race with self-generated connect/close traffic or SYN activity to listeners in the iterated network namespace.\nS:U - Impact is a kernel heap use-after-free and memory corruption within the same host kernel security authority; this is standard local kernel compromise/crash, not a VM escape, sandbox boundary cross, or IOMMU bypass.\nC:H - The iterator retains dangling request_sock pointers after refcount_set() clobbers a sock_hold() taken on sk_refcnt==0; subsequent bpf_iter_tcp_seq_show/put_batch dereference freed kernel objects, giving UAF read primitives per kernel CNA UAF guidance.\nI:H - Freed request_sock slabs can be reallocated while the iterator batch still references them; UAF enables corrupted socket metadata, refcount abuse, and control-flow hijacking via heap grooming, meeting High integrity under kernel CNA memory-corruption guidance.\nA:H - The bug surfaces as refcount_warn_saturate and leaves the reqsk one reference short so it is freed while still reachable from the iterator batch, causing kernel oops/panic; the tight-loop repro shows reliable, repeatable host denial of service."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["net/ipv4/tcp_ipv4.c"],"versions":[{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"cc0295f89296ed351fc4b0b48fee887ba02c5d24","status":"affected","versionType":"git"},{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"ddbe966b5d1fe212ada749bc3d0b410f1a7dea74","status":"affected","versionType":"git"},{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"7d2b60a4bc0499f62ff8520af6309bbe170882fd","status":"affected","versionType":"git"},{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"cefcbbe20846a45f9a7dae868f7ef1000953e2df","status":"affected","versionType":"git"},{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"97e74d3e45d653c07c2d406fc530a9bbe3df8396","status":"affected","versionType":"git"},{"version":"04c7820b776f1c4b48698574c47de9e940d368e8","lessThan":"e5fd3f514e27db1f05fbd72ba615d74941e23c51","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["net/ipv4/tcp_ipv4.c"],"versions":[{"version":"5.15","status":"affected"},{"version":"0","lessThan":"5.15","status":"unaffected","versionType":"semver"},{"version":"6.1.183","lessThanOrEqual":"6.1.*","status":"unaffected","versionType":"semver"},{"version":"6.6.152","lessThanOrEqual":"6.6.*","status":"unaffected","versionType":"semver"},{"version":"6.12.104","lessThanOrEqual":"6.12.*","status":"unaffected","versionType":"semver"},{"version":"6.18.45","lessThanOrEqual":"6.18.*","status":"unaffected","versionType":"semver"},{"version":"7.1.9","lessThanOrEqual":"7.1.*","status":"unaffected","versionType":"semver"},{"version":"7.2","lessThanOrEqual":"*","status":"unaffected","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"6.1.183"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"6.6.152"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"6.12.104"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"6.18.45"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"7.1.9"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.15","versionEndExcluding":"7.2"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/cc0295f89296ed351fc4b0b48fee887ba02c5d24"},{"url":"https://git.kernel.org/stable/c/ddbe966b5d1fe212ada749bc3d0b410f1a7dea74"},{"url":"https://git.kernel.org/stable/c/7d2b60a4bc0499f62ff8520af6309bbe170882fd"},{"url":"https://git.kernel.org/stable/c/cefcbbe20846a45f9a7dae868f7ef1000953e2df"},{"url":"https://git.kernel.org/stable/c/97e74d3e45d653c07c2d406fc530a9bbe3df8396"},{"url":"https://git.kernel.org/stable/c/e5fd3f514e27db1f05fbd72ba615d74941e23c51"}],"title":"bpf: tcp: Fix use-after-free in bpf_iter_tcp_established_batch()","x_generator":{"engine":"bippy-1.2.0"}}}}