{
  "affected": [
    {
      "ecosystem_specific": {
        "urgency": "not yet assigned"
      },
      "package": {
        "ecosystem": "Debian:12",
        "name": "linux"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "ecosystem_specific": {
        "urgency": "not yet assigned"
      },
      "package": {
        "ecosystem": "Debian:13",
        "name": "linux"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "ecosystem_specific": {
        "urgency": "not yet assigned"
      },
      "package": {
        "ecosystem": "Debian:14",
        "name": "linux"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "7.2.6-1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "details": "In the Linux kernel, the following vulnerability has been resolved:  tcp: fix corruption of urgent data on multi-segment retransmit  On the normal xmit path, while in urgent mode we refuse to build a multi-segment TSO packet, so every segment gets its own urg_ptr:  \t/* tcp_write_xmit() */ \tlimit = mss_now; \tif (tso_segs \u003e 1 \u0026\u0026 !tcp_urg_mode(tp)) \t\tlimit = tcp_mss_split_point(...);  The retransmit path has no such guard. __tcp_retransmit_skb() builds a segs \u003e 1 skb and hands it to the GSO layer, which only advances th-\u003eseq per segment and copies urg_ptr verbatim:  \t/* __tcp_retransmit_skb() */ \tlen = cur_mss * segs;\t\t/* segs \u003e 1, no urg_mode check */ \t... \t/* tcp_gso_segment(): bumps seq only, urg_ptr is copied */  urg_ptr is an offset from the segment's own seq, so a copied value points at a different place on each segment. The receiver rebuilds the absolute urgent seq as seg.seq + urg_ptr, so it walks a moving urgent point instead of the one OOB byte:  \tseg1  seq 1     urg_ptr 5001 -\u003e urgent @ 5001   (ok) \tseg2  seq 1001  urg_ptr 5001 -\u003e urgent @ 6001   (wrong, +MSS) \tseg3  seq 2001  urg_ptr 5001 -\u003e urgent @ 7001   (wrong, +2*MSS)  The real OOB byte is never pointed at, so the receiver stops splicing it out and delivers it as normal in-band data, corrupting the stream.  Guard the retransmit length like the xmit path: keep segs = 1 while in urgent mode.",
  "id": "DEBIAN-CVE-2026-90054",
  "modified": "2026-09-23T03:47:40.897230643Z",
  "published": "2026-09-17T17:16:53.873Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://security-tracker.debian.org/tracker/CVE-2026-90054"
    }
  ],
  "upstream": [
    "CVE-2026-90054"
  ]
}