{"cve":"CVE-2026-43244","nvd":{"published":"2026-05-06T12:16:44.873","modified":"2026-10-21T13:18:52.643","status":"Modified","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nkcm: fix zero-frag skb in frag_list on partial sendmsg error\n\nSyzkaller reported a warning in kcm_write_msgs() when processing a\nmessage with a zero-fragment skb in the frag_list.\n\nWhen kcm_sendmsg() fills MAX_SKB_FRAGS fragments in the current skb,\nit allocates a new skb (tskb) and links it into the frag_list before\ncopying data. If the copy subsequently fails (e.g. -EFAULT from\nuser memory), tskb remains in the frag_list with zero fragments:\n\n  head skb (msg being assembled, NOT yet in sk_write_queue)\n  +-----------+\n  | frags[17] |  (MAX_SKB_FRAGS, all filled with data)\n  | frag_list-+--> tskb\n  +-----------+    +----------+\n                   | frags[0] |  (empty! copy failed before filling)\n                   +----------+\n\nFor SOCK_SEQPACKET with partial data already copied, the error path\nsaves this message via partial_message for later completion. For\nSOCK_SEQPACKET, sock_write_iter() automatically sets MSG_EOR, so a\nsubsequent zero-length write(fd, NULL, 0) completes the message and\nqueues it to sk_write_queue. kcm_write_msgs() then walks the\nfrag_list and hits:\n\n  WARN_ON(!skb_shinfo(skb)->nr_frags)\n\nTCP has a similar pattern where skbs are enqueued before data copy\nand cleaned up on failure via tcp_remove_empty_skb(). KCM was\nmissing the equivalent cleanup.\n\nFix this by tracking the predecessor skb (frag_prev) when allocating\na new frag_list entry. On error, if the tail skb has zero frags,\nuse frag_prev to unlink and free it in O(1) without walking the\nsingly-linked frag_list. frag_prev is safe to dereference because\nthe entire message chain is only held locally (or in kcm->seq_skb)\nand is not added to sk_write_queue until MSG_EOR, so the send path\ncannot free it underneath us.\n\nAlso change the WARN_ON to WARN_ON_ONCE to avoid flooding the log\nif the condition is somehow hit repeatedly.\n\nThere are currently no KCM selftests in the kernel tree; a simple\nreproducer is available at [1].\n\n[1] https://gist.github.com/mrpre/a94d431c757e8d6f168f4dd1a3749daa","cvss_v4":null,"cvss_v3":{"version":"3.1","vector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","score":5.5},"cwes":["CWE-401"],"references":["https://git.kernel.org/stable/c/6f129a251c2795ba55dd5cc1fdf7e0af3098d848","https://git.kernel.org/stable/c/7af58f76e4b404a74c836881a845e6652db8a09f","https://git.kernel.org/stable/c/9ea3671d70ee07480d80bebe86696397c4e99fb7","https://git.kernel.org/stable/c/b1e3edf688a88c1a3ac41657055d9c136a08cd25","https://git.kernel.org/stable/c/ca220141fa8ebae09765a242076b2b77338106b0"],"products":["linux:linux_kernel"],"source":"nvd","synced_at":"2026-10-22T16:41:11.231219466+00:00"},"epss":{"cve":"CVE-2026-43244","score":0.00155,"percentile":0.04567,"date":"2026-10-25"},"kev":null,"osv":[],"attribution":["This product uses the NVD API but is not endorsed or certified by the NVD.","EPSS scores: FIRST Exploit Prediction Scoring System, https://www.first.org/epss","Known exploited vulnerabilities: CISA Known Exploited Vulnerabilities Catalog, https://www.cisa.gov/known-exploited-vulnerabilities-catalog","Package advisories: OSV (https://osv.dev), from the GitHub Advisory Database, PyPA, Go, RustSec and other ecosystem databases, under CC-BY 4.0 or CC0","CVE records: the CVE Program's CVE List (https://www.cve.org), under its terms of use (https://www.cve.org/Legal/TermsOfUse). CVE is a registered trademark of The MITRE Corporation; this product is not affiliated with the CVE Program."]}