---
id: "CVE-2026-98643"
kind: "vulnerability"
aliases: ["CVE-2026-98643"]
title: "Linux kernel BPF sockops: out-of-bounds read leaks kernel memory"
severity: "unknown"
cvss: null
epss: null
exploited: false
status: "open"
category: "infrastructure"
products: []
ecosystems: []
tags: ["bpf", "infrastructure", "kernel", "kubernetes", "linux", "self-hosted"]
not_affected_if: ["default-config", "no-sockops-programs"]
published_at: "2026-10-18T16:17:17+00:00"
first_seen_at: "2026-10-23T21:57:16+00:00"
updated_at: "2026-10-25T05:20:53+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-98643", "first_seen_at": "2026-10-23T21:57:16+00:00"}]
---

# Linux kernel BPF sockops: out-of-bounds read leaks kernel memory

A BPF sockops program that reads skops->rtt_min can read past a request_sock during connection setup, because the kernel casts it to a full tcp_sock without a check. The read returns adjacent kernel memory, which can include pointer fragments.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check whether the kernel is in a release that includes the fix: unknown
- List loaded sockops programs with `bpftool prog show` and look for type sock_ops
- Check whether your environment loads BPF programs at all: `sysctl kernel.unprivileged_bpf_disabled`

## Fix

- Update the Linux kernel to a release that includes the upstream commit fixing rtt_min in sock_ops: unknown
- Until patched, unload sockops BPF programs that read rtt_min if you do not need them

## Timeline

- 2026-10-23T21:57:16+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-98643

## From the sources

In the Linux kernel, the following vulnerability has been resolved:

bpf: Fix out-of-bounds read of rtt_min in sock_ops

A sockops prog reading skops->rtt_min never checks the sk type: on the
tcp_conn_request() path sock_ops->sk is a request_sock (non-full), and the
ctx rewrite casts it to a tcp_sock (full) and reads rtt_min past the end of
the request_sock, returning dirty adjacent memory.

	SEC("sockops")
	int prog(struct bpf_sock_ops *skops)
	{
		switch (skops->op) {
		case BPF_SOCK_OPS_RWND_INIT:
			leak = skops->rtt_min;   /* reads the request_sock OOB */
		...
		}
	}

For instance one such read returned rtt_min=0xffff8881, the high half of a
leaked kernel pointer.

Guarding that cast is exactly what SOCK_OPS_GET_FIELD() does -- it checks
is_locked_tcp_sock and returns 0 when sock_ops->sk is not a locked full
socket. Every other tcp_sock field in sock_ops goes through it; rtt_min is
the only one open-coded, so it skips the check.

Read rtt_min through SOCK_OPS_GET_FIELD() too. rtt_min is a bit special:
it is a struct minmax and we only want the current min, so pass
rtt_min.s[0].v. That is equivalent to the old hand-computed offset

	offsetof(struct tcp_sock, rtt_min) + sizeof_field(struct minmax_sample, t)

(s[0] sits at rtt_min + 0 and .v at + sizeof(.t), i.e. what minmax_get()
returns), so the loaded field is unchanged and only the full-sock guard is
added. The two BUILD_BUG_ON()s that protected the hand-computed offset
are no longer needed.

Before patch:

	0: r1 = *(u64 *)(r1 +0)      ; r1 = skops->sk
	1: r1 = *(u32 *)(r1 +2324)   ; ((tcp_sock *)sk)->rtt_min.s[0].v

After patch:

	0: *(u64 *)(r1 +56) = r9
	1: r9 = *(u8 *)(r1 +50)      ; is_locked_tcp_sock
	2: if r9 == 0 goto pc+4      ; not a locked full sock -> 0
	3: r9 = *(u64 *)(r1 +56)
	4: r1 = *(u64 *)(r1 +0)      ; r1 = skops->sk
	5: r1 = *(u32 *)(r1 +2324)   ; rtt_min.s[0].v
	6: goto pc+2
	7: r9 = *(u64 *)(r1 +56)
	8: r1 = 0

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_