GHSA-FH4P-V6R7-6956
Vulnerability from github – Published: 2026-08-16 09:30 – Updated: 2026-08-17 06:33In the Linux kernel, the following vulnerability has been resolved:
crypto: algif_skcipher - force synchronous processing on trees without ctx->state
The AIO/async path in skcipher_recvmsg() passes the socket-wide ctx->iv directly into the skcipher request. After io_submit() the socket lock is dropped and the request is processed asynchronously, so a concurrent sendmsg(ALG_SET_IV) can overwrite ctx->iv and make the in-flight request run under an attacker-controlled IV. For CTR/stream modes this is IV/keystream reuse and lets an unprivileged user recover the plaintext of a concurrent operation.
Snapshotting ctx->iv into per-request storage for the async path is not sufficient. For ciphers with statesize == 0 - which includes cbc and ctr - the MSG_MORE inter-chunk IV chaining is carried solely by the in-place req->iv writeback, which a snapshot redirects into per-request memory that af_alg_free_resources() releases on completion, silently producing wrong output. Writing the IV back from the completion callback instead is not possible either: that would require lock_sock() there, but the callback can run in softirq/atomic context, so it must not sleep.
Make the operation synchronous instead, which removes both the IV race and any writeback race. This is equivalent to the upstream resolution, commit fcc77d33a34c ("net: Remove support for AIO on sockets"), which removed the AIO socket path across net/ entirely and so produces the same end state for this file. This patch deviates from that commit deliberately: rather than removing AIO socket support tree-wide, which would be far too invasive for stable, it removes only the AIO branch in crypto/algif_skcipher.c. io_submit() now completes synchronously; AF_ALG async is rarely used in practice.
The -EIOCBQUEUED check in skcipher_recvmsg() is now dead but harmless, and is left alone to keep the fix minimal.
Tested on 6.6.y: attacker IV injection dropped from 2296/200000 to 0/200000 after the change; MSG_MORE chunked CTR output bit-identical to single-shot.
{
"affected": [],
"aliases": [
"CVE-2026-74578"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-08-16T09:16:21Z",
"severity": "HIGH"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\ncrypto: algif_skcipher - force synchronous processing on trees without ctx-\u003estate\n\nThe AIO/async path in skcipher_recvmsg() passes the socket-wide ctx-\u003eiv\ndirectly into the skcipher request. After io_submit() the socket lock is\ndropped and the request is processed asynchronously, so a concurrent\nsendmsg(ALG_SET_IV) can overwrite ctx-\u003eiv and make the in-flight request\nrun under an attacker-controlled IV. For CTR/stream modes this is\nIV/keystream reuse and lets an unprivileged user recover the plaintext of\na concurrent operation.\n\nSnapshotting ctx-\u003eiv into per-request storage for the async path is not\nsufficient. For ciphers with statesize == 0 - which includes cbc and ctr -\nthe MSG_MORE inter-chunk IV chaining is carried solely by the in-place\nreq-\u003eiv writeback, which a snapshot redirects into per-request memory that\naf_alg_free_resources() releases on completion, silently producing wrong\noutput. Writing the IV back from the completion callback instead is not\npossible either: that would require lock_sock() there, but the callback can\nrun in softirq/atomic context, so it must not sleep.\n\nMake the operation synchronous instead, which removes both the IV race and\nany writeback race. This is equivalent to the upstream resolution, commit\nfcc77d33a34c (\"net: Remove support for AIO on sockets\"), which removed the\nAIO socket path across net/ entirely and so produces the same end state for\nthis file. This patch deviates from that commit deliberately: rather than\nremoving AIO socket support tree-wide, which would be far too invasive for\nstable, it removes only the AIO branch in crypto/algif_skcipher.c.\nio_submit() now completes synchronously; AF_ALG async is rarely used in\npractice.\n\nThe -EIOCBQUEUED check in skcipher_recvmsg() is now dead but harmless,\nand is left alone to keep the fix minimal.\n\nTested on 6.6.y: attacker IV injection dropped from 2296/200000 to 0/200000\nafter the change; MSG_MORE chunked CTR output bit-identical to single-shot.",
"id": "GHSA-fh4p-v6r7-6956",
"modified": "2026-08-17T06:33:55Z",
"published": "2026-08-16T09:30:22Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-74578"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/60eafc7b08c6689ea3ad39eff8d97aefc8a087c7"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/73dd3bf704ca6c20639de70c08e9a10bee904a95"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/7b91e51d0eb7cbb07f7f086f9176bd93dbbc85dd"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/b05defc41b27c7d0c05c45f67bf5b91c28f93669"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/bf09b0be8e851f050e98da702d247c14d81b591a"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/d7860b682da55433b5da0591b0e4c1982ecd2689"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/f1a87ca0843d74482402a206ea2dfb315ee9acbd"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/fcc77d33a34cf271702e8daafb6c593e4626776d"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N",
"type": "CVSS_V3"
}
]
}
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.
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.