{"id":"CVE-2024-58060","state":"PUBLISHED","title":"bpf: Reject struct_ops registration that uses module ptr and the module btf_id is missing","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Reject struct_ops registration that uses module ptr and the module btf_id is missing\n\nThere is a UAF report in the bpf_struct_ops when CONFIG_MODULES=n.\nIn particular, the report is on tcp_congestion_ops that has\na \"struct module *owner\" member.\n\nFor struct_ops that has a \"struct module *owner\" member,\nit can be extended either by the regular kernel module or\nby the bpf_struct_ops. bpf_try_module_get() will be used\nto do the refcounting and different refcount is done\nbased on the owner pointer. When CONFIG_MODULES=n,\nthe btf_id of the \"struct module\" is missing:\n\nWARN: resolve_btfids: unresolved symbol module\n\nThus, the bpf_try_module_get() cannot do the correct refcounting.\n\nNot all subsystem's struct_ops requires the \"struct module *owner\" member.\ne.g. the recent sched_ext_ops.\n\nThis patch is to disable bpf_struct_ops registration if\nthe struct_ops has the \"struct module *\" member and the\n\"struct module\" btf_id is missing. The btf_type_is_fwd() helper\nis moved to the btf.h header file for this test.\n\nThis has happened since the beginning of bpf_struct_ops which has gone\nthrough many changes. The Fixes tag is set to a recent commit that this\npatch can apply cleanly. Considering CONFIG_MODULES=n is not\ncommon and the age of the issue, targeting for bpf-next also.","publishedAt":"2025-03-06T15:54:03.253Z","updatedAt":"2026-08-05T11:47:30.918Z","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":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"b777b14c2a4a4e2322daf8e8ffd42d2b88831b17","versionType":"git"},{"version":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"2324fb4e92092837ee278fdd8d60c48ee1a619ce","versionType":"git"},{"version":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"96ea081ed52bf077cad6d00153b6fba68e510767","versionType":"git"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/btf.h","kernel/bpf/bpf_struct_ops.c","kernel/bpf/btf.c"]},{"vendor":"Linux","product":"Linux","defaultStatus":"affected","versions":[{"version":"6.9","status":"affected"},{"version":"0","status":"unaffected","lessThan":"6.9","versionType":"semver"},{"version":"6.12.13","status":"unaffected","lessThanOrEqual":"6.12.*","versionType":"semver"},{"version":"6.13.2","status":"unaffected","lessThanOrEqual":"6.13.*","versionType":"semver"},{"version":"6.14","status":"unaffected","lessThanOrEqual":"*","versionType":"original_commit_for_fix"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/btf.h","kernel/bpf/bpf_struct_ops.c","kernel/bpf/btf.c"]}],"references":["https://git.kernel.org/stable/c/b777b14c2a4a4e2322daf8e8ffd42d2b88831b17","https://git.kernel.org/stable/c/2324fb4e92092837ee278fdd8d60c48ee1a619ce","https://git.kernel.org/stable/c/96ea081ed52bf077cad6d00153b6fba68e510767"],"weaknesses":["CWE-416"],"ingestedAt":"2026-10-29T20:08:12.933Z","source":{"repository":"CVEProject/cvelistV5","path":"cves/2024/58xxx/CVE-2024-58060.json","commit":"556c9d3a8cd17e218bb8e452ba7cd619a384ff47"},"raw":{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2024-58060","state":"PUBLISHED","datePublished":"2025-03-06T15:54:03.253Z","dateUpdated":"2026-08-05T11:47:30.918Z","assignerShortName":"Linux","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","dateReserved":"2025-03-06T15:52:09.179Z"},"containers":{"cna":{"title":"bpf: Reject struct_ops registration that uses module ptr and the module btf_id is missing","descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Reject struct_ops registration that uses module ptr and the module btf_id is missing\n\nThere is a UAF report in the bpf_struct_ops when CONFIG_MODULES=n.\nIn particular, the report is on tcp_congestion_ops that has\na \"struct module *owner\" member.\n\nFor struct_ops that has a \"struct module *owner\" member,\nit can be extended either by the regular kernel module or\nby the bpf_struct_ops. bpf_try_module_get() will be used\nto do the refcounting and different refcount is done\nbased on the owner pointer. When CONFIG_MODULES=n,\nthe btf_id of the \"struct module\" is missing:\n\nWARN: resolve_btfids: unresolved symbol module\n\nThus, the bpf_try_module_get() cannot do the correct refcounting.\n\nNot all subsystem's struct_ops requires the \"struct module *owner\" member.\ne.g. the recent sched_ext_ops.\n\nThis patch is to disable bpf_struct_ops registration if\nthe struct_ops has the \"struct module *\" member and the\n\"struct module\" btf_id is missing. The btf_type_is_fwd() helper\nis moved to the btf.h header file for this test.\n\nThis has happened since the beginning of bpf_struct_ops which has gone\nthrough many changes. The Fixes tag is set to a recent commit that this\npatch can apply cleanly. Considering CONFIG_MODULES=n is not\ncommon and the age of the issue, targeting for bpf-next also."}],"affected":[{"vendor":"Linux","product":"Linux","defaultStatus":"unaffected","versions":[{"version":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"b777b14c2a4a4e2322daf8e8ffd42d2b88831b17","versionType":"git"},{"version":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"2324fb4e92092837ee278fdd8d60c48ee1a619ce","versionType":"git"},{"version":"1611603537a4b88cec7993f32b70c03113801a46","status":"affected","lessThan":"96ea081ed52bf077cad6d00153b6fba68e510767","versionType":"git"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/btf.h","kernel/bpf/bpf_struct_ops.c","kernel/bpf/btf.c"]},{"vendor":"Linux","product":"Linux","defaultStatus":"affected","versions":[{"version":"6.9","status":"affected"},{"version":"0","status":"unaffected","lessThan":"6.9","versionType":"semver"},{"version":"6.12.13","status":"unaffected","lessThanOrEqual":"6.12.*","versionType":"semver"},{"version":"6.13.2","status":"unaffected","lessThanOrEqual":"6.13.*","versionType":"semver"},{"version":"6.14","status":"unaffected","lessThanOrEqual":"*","versionType":"original_commit_for_fix"}],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["include/linux/btf.h","kernel/bpf/bpf_struct_ops.c","kernel/bpf/btf.c"]}],"references":[{"url":"https://git.kernel.org/stable/c/b777b14c2a4a4e2322daf8e8ffd42d2b88831b17"},{"url":"https://git.kernel.org/stable/c/2324fb4e92092837ee278fdd8d60c48ee1a619ce"},{"url":"https://git.kernel.org/stable/c/96ea081ed52bf077cad6d00153b6fba68e510767"}],"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 attack is driven entirely through local syscalls — bpf(BPF_MAP_CREATE/MAP_UPDATE_ELEM/LINK_CREATE) to register the struct_ops congestion control, setsockopt(TCP_CONGESTION) to bind a socket to it, and close() to drop the map. Although the freed object is subsequently dereferenced while processing TCP ACKs, a remote peer cannot perform the registration/free setup.\nAC:L - On an affected kernel every step is deterministic and attacker-driven — register the struct_ops, attach it to a socket the attacker owns, close the fds to free the map, then drive traffic over loopback to hit the dangling icsk_ca_ops; there is no race and no dependence on memory layout the attacker cannot influence. The build-time CONFIG_MODULES=n precondition defines the vulnerable configuration rather than a condition to be won at exploit time, and it is trivially observable in advance (absent /proc/modules).\nPR:L - Creating the struct_ops map goes through bpf_token_capable(token, CAP_BPF), which BPF tokens explicitly delegate into unprivileged container user namespaces — the model container runtimes and BPF agents routinely use — and the socket, setsockopt(TCP_CONGESTION) and loopback-traffic halves of the attack need no capability at all.\nUI:N - The attacker performs the entire sequence with its own process, sockets and file descriptors; no victim action, mount, or configuration change by another user is required.\nS:U - The dangling tcp_congestion_ops pointer is dereferenced by the same kernel the attacker already has local access to; no VM, IOMMU, or sandbox boundary is crossed.\nC:H - The freed st_map allocation holding kvalue.data can be reclaimed and shaped by the attacker while sockets still read through icsk_ca_ops, turning the stale pointer into a controlled kernel-memory read primitive that leaks arbitrary kernel data.\nI:H - The kernel makes indirect calls through function pointers in the freed tcp_congestion_ops and into freed trampoline/bpf_prog_pack executable pages, giving a direct control-flow hijack primitive after heap or exec-page reclaim — full arbitrary write and code execution in kernel context.\nA:H - Even unweaponized, the use-after-free causes an oops or panic on the next congestion-control callback, and installing the bpf CA as the netns default poisons every new TCP socket in that namespace, wedging the network stack."}]}],"providerMetadata":{"shortName":"Linux","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","dateUpdated":"2026-08-05T11:47:30.918Z"},"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.9","versionEndExcluding":"6.12.13"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.9","versionEndExcluding":"6.13.2"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.9","versionEndExcluding":"6.14"}]}]}],"x_generator":{"engine":"bippy-1.2.0"}},"adp":[{"title":"CISA ADP Vulnrichment","metrics":[{"cvssV3_1":{"scope":"UNCHANGED","version":"3.1","baseScore":7.8,"attackVector":"LOCAL","baseSeverity":"HIGH","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","integrityImpact":"HIGH","userInteraction":"NONE","attackComplexity":"LOW","availabilityImpact":"HIGH","privilegesRequired":"LOW","confidentialityImpact":"HIGH"}},{"other":{"type":"ssvc","content":{"id":"CVE-2024-58060","role":"CISA Coordinator","options":[{"Exploitation":"none"},{"Automatable":"no"},{"Technical Impact":"total"}],"version":"2.0.3","timestamp":"2025-03-06T16:04:02.766596Z"}}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-416","lang":"en","type":"CWE","description":"CWE-416 Use After Free"}]}],"providerMetadata":{"shortName":"CISA-ADP","orgId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","dateUpdated":"2025-03-06T16:04:30.923Z"}}]}}}