{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2025-40123","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2025-04-16T07:20:57.169Z","datePublished":"2025-11-12T10:23:19.589Z","dateUpdated":"2026-08-05T12:07:56.107Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-05T12:07:56.107Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Enforce expected_attach_type for tailcall compatibility\n\nYinhao et al. recently reported:\n\n  Our fuzzer tool discovered an uninitialized pointer issue in the\n  bpf_prog_test_run_xdp() function within the Linux kernel's BPF subsystem.\n  This leads to a NULL pointer dereference when a BPF program attempts to\n  deference the txq member of struct xdp_buff object.\n\nThe test initializes two programs of BPF_PROG_TYPE_XDP: progA acts as the\nentry point for bpf_prog_test_run_xdp() and its expected_attach_type can\nneither be of be BPF_XDP_DEVMAP nor BPF_XDP_CPUMAP. progA calls into a slot\nof a tailcall map it owns. progB's expected_attach_type must be BPF_XDP_DEVMAP\nto pass xdp_is_valid_access() validation. The program returns struct xdp_md's\negress_ifindex, and the latter is only allowed to be accessed under mentioned\nexpected_attach_type. progB is then inserted into the tailcall which progA\ncalls.\n\nThe underlying issue goes beyond XDP though. Another example are programs\nof type BPF_PROG_TYPE_CGROUP_SOCK_ADDR. sock_addr_is_valid_access() as well\nas sock_addr_func_proto() have different logic depending on the programs'\nexpected_attach_type. Similarly, a program attached to BPF_CGROUP_INET4_GETPEERNAME\nshould not be allowed doing a tailcall into a program which calls bpf_bind()\nout of BPF which is only enabled for BPF_CGROUP_INET4_CONNECT.\n\nIn short, specifying expected_attach_type allows to open up additional\nfunctionality or restrictions beyond what the basic bpf_prog_type enables.\nThe use of tailcalls must not violate these constraints. Fix it by enforcing\nexpected_attach_type in __bpf_prog_map_compatible().\n\nNote that we only enforce this for tailcall maps, but not for BPF devmaps or\ncpumaps: There, the programs are invoked through dev_map_bpf_prog_run*() and\ncpu_map_bpf_prog_run*() which set up a new environment / context and therefore\nthese situations are not prone to this issue."}],"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 vulnerability is reached through the bpf(2) syscall — loading two programs with differing expected_attach_type, populating a BPF_MAP_TYPE_PROG_ARRAY, and invoking the tail call via BPF_PROG_TEST_RUN or a live attach. No network-supplied data introduces the incompatible tail call, so local access is required.\nAC:L - Exploitation is fully deterministic: the attacker authors and loads both programs, owns the prog array, and controls exactly when the tail call executes; there is no race and no memory-layout condition needed for the mis-typed context dereference to occur.\nPR:L - A basic local user with BPF program-loading access can create the prog array and the two mismatched programs; this matches the established scoring of comparable BPF verifier-constraint bypasses, and containerized/delegated-token deployments grant exactly this level without full administrative rights.\nUI:N - The attacker performs every step — program load, map update, and program invocation — entirely on their own; no victim action or cooperation is involved.\nS:U - The mis-typed context dereference and any resulting corruption occur within the kernel that is already the vulnerable component, so impact stays inside a single security authority.\nC:H - xdp_init_buff() never initializes xdp_buff->txq, so a tail-called BPF_XDP_DEVMAP program reading egress_ifindex dereferences attacker-groomable uninitialized stack memory twice and returns the resulting kernel dword into BPF, which can be exfiltrated through a map — an effectively arbitrary kernel read primitive.\nI:H - Bypassing expected_attach_type lets a program execute helpers and context writes the running hook never validated — e.g. bpf_bind()/__inet_bind() invoked from a GETPEERNAME hook mutates live socket binding and hash state without the socket lock, and msg_src_ip4/6 or sockopt level/optname/retval writes land on NULL or wrong-type kernel fields.\nA:H - The reported case is an immediate NULL pointer dereference on xdp_buff->txq in bpf_prog_test_run_xdp(), and in driver XDP paths the uninitialized txq produces a wild pointer dereference — either way a kernel oops that the attacker can trigger repeatedly."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/bpf.h","kernel/bpf/core.c"],"versions":[{"version":"5e43f899b03a3492ce5fc44e8900becb04dae9c0","lessThan":"a99de19128aec0913f3d529f529fbbff5edfaff8","status":"affected","versionType":"git"},{"version":"5e43f899b03a3492ce5fc44e8900becb04dae9c0","lessThan":"08cb3dc9d2b44f153d0bcf2cb966e4a94b5d0f32","status":"affected","versionType":"git"},{"version":"5e43f899b03a3492ce5fc44e8900becb04dae9c0","lessThan":"f856c598080ba7ce1252867b8ecd6ad5bdaf9a6a","status":"affected","versionType":"git"},{"version":"5e43f899b03a3492ce5fc44e8900becb04dae9c0","lessThan":"c1ad19b5d8e23123503dcaf2d4342e1b90b923ad","status":"affected","versionType":"git"},{"version":"5e43f899b03a3492ce5fc44e8900becb04dae9c0","lessThan":"4540aed51b12bc13364149bf95f6ecef013197c0","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/bpf.h","kernel/bpf/core.c"],"versions":[{"version":"4.17","status":"affected"},{"version":"0","lessThan":"4.17","status":"unaffected","versionType":"semver"},{"version":"6.1.156","lessThanOrEqual":"6.1.*","status":"unaffected","versionType":"semver"},{"version":"6.6.112","lessThanOrEqual":"6.6.*","status":"unaffected","versionType":"semver"},{"version":"6.12.53","lessThanOrEqual":"6.12.*","status":"unaffected","versionType":"semver"},{"version":"6.17.3","lessThanOrEqual":"6.17.*","status":"unaffected","versionType":"semver"},{"version":"6.18","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":"4.17","versionEndExcluding":"6.1.156"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"4.17","versionEndExcluding":"6.6.112"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"4.17","versionEndExcluding":"6.12.53"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"4.17","versionEndExcluding":"6.17.3"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"4.17","versionEndExcluding":"6.18"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/a99de19128aec0913f3d529f529fbbff5edfaff8"},{"url":"https://git.kernel.org/stable/c/08cb3dc9d2b44f153d0bcf2cb966e4a94b5d0f32"},{"url":"https://git.kernel.org/stable/c/f856c598080ba7ce1252867b8ecd6ad5bdaf9a6a"},{"url":"https://git.kernel.org/stable/c/c1ad19b5d8e23123503dcaf2d4342e1b90b923ad"},{"url":"https://git.kernel.org/stable/c/4540aed51b12bc13364149bf95f6ecef013197c0"}],"title":"bpf: Enforce expected_attach_type for tailcall compatibility","x_generator":{"engine":"bippy-1.2.0"}}}}