CVE-2026-90051: tcp: reject non zerocopy devmem tx
In the Linux kernel, the following vulnerability has been resolved:
tcp: reject non zerocopy devmem tx
Devmem tcp tx doesn't work without zero-copy, however it's not currently enforced if NETIFFSG isn't present. In this case, tcpsendmsglocked() will try the copy path and try to copy data from an iovec which consists of offsets into the dma-buf and would normally fail. Moreover, d9c56501c72fd ("net: tcp: block mixing readable and unreadable frags") relies on that and assumes that the devmem binding is present IFF we're using the zero-copy path, which can be used to mix net-iov and pages in a single skb, and break invariants. Let's reject devmem tx without zero-copy.
Note, the parameter check the patch is modifying is too loose, we can create an iouring request with dmabufid and all ZC flags, but which won't have the binding. We replace it with stricter validation.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade to a fixed release to a version that resolves this vulnerability.
Patch d9c56501c72fd - Configuration
Apply the kernel fix so that tcp rejects devmem TX unless it uses the zero-copy path (the patch adds stricter validation and rejects non-zerocopy devmem TX).
Linux kernel tcp_sendmsg_locked() / net: tcp devmem TX handling = reject non-zerocopy devmem tx (i.e., enforce zero-copy path)