OSEC-2026-30
Vulnerability from osv_ocaml – Published: 2026-11-06 12:30 – Updated: 2026-11-06 12:30 – Source websiteSummary
In OCaml before 4.14.5, when the key of an ephemeron is a closure with Infix_tag (a function of a mutually recursive group that is not the first function), the mark phase does not keep the ephemeron's data alive, although the key is alive. The clean phase does treat the key as alive, so it leaves the data field filled. The sweep phase then frees the data block. After a major collection the ephemeron's data field still points to the freed block.
This looks like the part of #7810 that PR #9742 did not cover. PR #9742 fixed the clean phase and the minor collector. It did not change mark_ephe_aux.
Please note that the CVSS score high is under the conditions: - if the application is using ephemerons, - and if the application is available over the network, - and if the attacker can find a way to exploit the memory safety issue in an attack chain.
Reproduction
[@@@alert "-old_ephemeron_api"]
let infix n = let rec f () = n and g () = f () in g
let build key n =
let e = Ephemeron.K1.create () in
let w = Weak.create 1 in
let d = ref n in
Ephemeron.K1.set_key e key;
Ephemeron.K1.set_data e d;
Weak.set w 0 (Some d);
(e, w)
let run key =
let (e, w) = build key 1 in
Gc.full_major ();
Printf.printf "tag=%d weak=%b check_key=%b check_data=%b\n"
(Obj.tag (Obj.repr key)) (Weak.check w 0)
(Ephemeron.K1.check_key e) (Ephemeron.K1.check_data e);
ignore (Sys.opaque_identity key)
let () =
if Sys.argv.(1) = "infix" then run (infix 1) else run (ref 1)
Obj is used only to print the tag of the key. The program does not read the
data after the collection.
Timeline
- 2026-10-04: initial report via email
- 2026-10-09: develop and committed the fix
- 2026-11-03: fix was reviewed and merged
- 2026-11-06: OCaml 4.14.5 released
{
"affected": [
{
"ecosystem_specific": {
"opam_constraint": "ocaml {\u003c \"4.14.5\"}"
},
"package": {
"ecosystem": "opam",
"name": "ocaml",
"purl": "pkg:opam/ocaml"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "4.14.5"
}
],
"type": "ECOSYSTEM"
}
],
"versions": [
"3.07",
"3.07+1",
"3.07+2",
"3.08.0",
"3.08.1",
"3.08.2",
"3.08.3",
"3.08.4",
"3.09.0",
"3.09.1",
"3.09.2",
"3.09.3",
"3.10.0",
"3.10.1",
"3.10.2",
"3.11.0",
"3.11.1",
"3.11.2",
"3.12.0",
"3.12.1",
"4.00.0",
"4.00.1",
"4.01.0",
"4.02.0",
"4.02.1",
"4.02.2",
"4.02.3",
"4.02.4",
"4.03.0",
"4.03.1",
"4.04.0",
"4.04.1",
"4.04.2",
"4.04.3",
"4.05.0",
"4.05.1",
"4.06.0",
"4.06.1",
"4.06.2",
"4.07.0",
"4.07.1",
"4.07.2",
"4.08.0",
"4.08.1",
"4.08.2",
"4.09.0",
"4.09.1",
"4.09.2",
"4.10.0",
"4.10.1",
"4.10.2",
"4.10.3",
"4.11.0",
"4.11.1",
"4.11.2",
"4.11.3",
"4.12.0",
"4.12.1",
"4.12.2",
"4.13.0",
"4.13.1",
"4.13.2",
"4.14.0",
"4.14.1",
"4.14.2",
"4.14.3",
"4.14.4"
]
},
{
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "da301a20833b765e268d677fe693d29fe51793ad"
}
],
"repo": "https://github.com/ocaml/ocaml",
"type": "GIT"
}
]
}
],
"credits": [
{
"name": "Igor Santos-Grueiro",
"type": "REPORTER"
},
{
"name": "Hannes Mehnert",
"type": "COORDINATOR"
},
{
"name": "Florian Angeletti",
"type": "REMEDIATION_REVIEWER"
},
{
"name": "Gabriel Scherer",
"type": "REMEDIATION_DEVELOPER"
}
],
"database_specific": {
"cwe": [
"CWE-416"
],
"human_link": "https://github.com/ocaml/security-advisories/tree/main/advisories/2026/OSEC-2026-30.md",
"osv": "https://github.com/ocaml/security-advisories/tree/generated-osv/2026/OSEC-2026-30.json"
},
"details": "## Summary\n\nIn OCaml before 4.14.5, when the key of an ephemeron is a closure with `Infix_tag` (a function of a mutually recursive group that is not the first function), the mark phase does not keep the ephemeron\u0027s data alive, although the key is alive. The clean phase does treat the key as alive, so it leaves the data field filled. The sweep phase then frees the data block. After a major collection the ephemeron\u0027s data field still points to the freed block.\n\nThis looks like the part of [#7810](https://github.com/ocaml/ocaml/issues/7810) that PR [#9742](https://github.com/ocaml/ocaml/pull/9742) did not cover. PR [#9742](https://github.com/ocaml/ocaml/pull/9742) fixed the clean phase and the minor collector. It did not change `mark_ephe_aux`.\n\nPlease note that the CVSS score high is under the conditions:\n- *if* the application is using ephemerons,\n- and *if* the application is available over the network,\n- and *if* the attacker can find a way to exploit the memory safety issue in an attack chain.\n\n## Reproduction\n\n```ocaml\n[@@@alert \"-old_ephemeron_api\"]\nlet infix n = let rec f () = n and g () = f () in g\n\nlet build key n =\n let e = Ephemeron.K1.create () in\n let w = Weak.create 1 in\n let d = ref n in\n Ephemeron.K1.set_key e key;\n Ephemeron.K1.set_data e d;\n Weak.set w 0 (Some d);\n (e, w)\n\nlet run key =\n let (e, w) = build key 1 in\n Gc.full_major ();\n Printf.printf \"tag=%d weak=%b check_key=%b check_data=%b\\n\"\n (Obj.tag (Obj.repr key)) (Weak.check w 0)\n (Ephemeron.K1.check_key e) (Ephemeron.K1.check_data e);\n ignore (Sys.opaque_identity key)\n\nlet () =\n if Sys.argv.(1) = \"infix\" then run (infix 1) else run (ref 1)\n```\n\n`Obj` is used only to print the tag of the key. The program does not read the\ndata after the collection.\n\n## Timeline\n\n- 2026-10-04: initial report via email\n- 2026-10-09: develop and committed the fix\n- 2026-11-03: fix was reviewed and merged\n- 2026-11-06: OCaml 4.14.5 released",
"id": "OSEC-2026-30",
"modified": "2026-11-06T12:30:00Z",
"published": "2026-11-06T12:30:00Z",
"references": [],
"schema_version": "1.7.4",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:L/VA:H/SC:N/SI:N/SA:N/E:P",
"type": "CVSS_V4"
}
],
"summary": "Ephemeron data is freed while still referenced when the key is an `Infix_tag` closure"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.