ACK: [SRU][Q/N/J][PATCH 0/3] CVE-2026-53224

[Bethany Jamison]

On 7/17/26 11:37 AM, Cengiz Can via kernel-team wrote:
> https://ubuntu.com/security/CVE-2026-53224
>
> [ Impact ]
>
> sctp_unpack_cookie() only verified that the embedded INIT chunk length did not
> exceed the remaining cookie payload, but did not ensure the INIT chunk was
> large enough to hold a complete INIT header. A malformed COOKIE_ECHO can carry
> a truncated INIT chunk, and sctp_process_init() then accesses INIT parameters
> unconditionally, leading to out-of-bounds reads. Additionally,
> raw_addr_list_len is not fully validated against the remaining payload, so when
> cookie authentication is disabled an attacker can supply an oversized length
> and cause sctp_raw_to_bind_addrs() and the address parser to read beyond the
> end of the cookie.
>
> [ Fix ]
>
> questing: backported with AI-assisted adaptation
> noble: backported with AI-assisted adaptation
> jammy: backported with AI-assisted adaptation
> focal: backported with AI-assisted adaptation
> bionic: backported with AI-assisted adaptation
> xenial: backported with AI-assisted adaptation
> trusty: backported with AI-assisted adaptation
>
> [ Test Plan ]
>
> Boot tested.
>
> [ Where Problems Could Occur ]
>
> The change adds stricter length validation to the SCTP cookie unpacking path,
> so an incorrect fix could reject legitimate COOKIE_ECHO chunks and break SCTP
> association establishment. Regressions would be confined to the SCTP protocol
> subsystem.
>
Signed-off-by: Bethany Jamison