---
id: "CVE-2026-99266"
kind: "vulnerability"
aliases: ["CVE-2026-99266"]
title: "Linux kernel: io_uring zero-copy receive returns wrong buffers above 4GB"
severity: "unknown"
cvss: null
epss: null
exploited: false
status: "open"
category: "runtime"
products: []
ecosystems: []
tags: ["io-uring", "linux-kernel", "runtime", "zcrx"]
not_affected_if: ["no-io-uring-zcrx"]
published_at: "2026-11-03T12:17:33+00:00"
first_seen_at: "2026-11-03T12:20:40+00:00"
updated_at: "2026-11-03T12:22:39+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-99266", "first_seen_at": "2026-11-03T12:20:40+00:00"}]
---

# Linux kernel: io_uring zero-copy receive returns wrong buffers above 4GB

In the Linux kernel's io_uring zero-copy receive path, buffer offsets are computed in 32 bits, so for areas above 4GB the completion entries point the application at the wrong memory. An application can read the wrong data and return buffers it does not own while the real buffers are never given back.

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

## Timeline

- 2026-11-03T12:20:40+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-99266

## From the sources

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

io_uring/zcrx: fix buffer offsets in areas above 4GB

io_zcrx_queue_cqe() computes the offset of the buffer in its area from
net_iov_idx(), which is an unsigned int, so the shift is done in 32 bits
and only the result is widened to the u64 that goes into the CQE. An area
can be up to 1TB in size. For every niov beyond the first 4GB the CQE
points the application at a buffer 4GB, or a multiple of that, below the
one which has the data. It reads the wrong data, and then returns a
buffer it doesn't own via the refill queue while the real one is never
given back.

---

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