PT-2026-44350 · Linux · Linux
Published
2026-05-28
·
Updated
2026-05-28
·
CVE-2026-46227
None
No severity ratings or metrics are available. When they are, we'll update the corresponding info on the page.
In the Linux kernel, the following vulnerability has been resolved:
sctp: revalidate list cursor after sctp sendmsg to asoc() in SCTP SENDALL
The SCTP SENDALL path in sctp sendmsg() iterates ep->asocs with
list for each entry safe(), which caches the next entry in @tmp before
the loop body runs. The body calls sctp sendmsg to asoc(), which may
drop the socket lock inside sctp wait for sndbuf().
While the lock is dropped, another thread can SCTP SOCKOPT PEELOFF the
association cached in @tmp, migrating it to a new endpoint via
sctp sock migrate() (list del init() + list add tail() to
newep->asocs), and optionally close the new socket which frees the
association via kfree rcu(). The cached @tmp can also be freed by a
network ABORT for that association, processed in softirq while the
lock is dropped.
sctp wait for sndbuf() revalidates @asoc (the current entry) on re-lock
via the "sk != asoc->base.sk" and "asoc->base.dead" checks, but nothing
revalidates @tmp. After a successful return, the iterator advances to
the stale @tmp, yielding either a use-after-free (if the peeled socket
was closed) or a list-walk onto the new endpoint's list head (type
confusion of &newep->asocs as a struct sctp association *).
Both are reachable from CapEff=0; the type-confusion path gives
controlled indirect call via the outqueue.sched->init sid pointer.
Fix by re-deriving @tmp from @asoc after sctp sendmsg to asoc()
returns. @asoc is known to still be on ep->asocs at that point: the
only callers that list del an association from ep->asocs are
sctp association free() (which sets asoc->base.dead) and
sctp assoc migrate() (which changes asoc->base.sk), and
sctp wait for sndbuf() checks both under the lock before any
successful return; a tripped check propagates as err < 0 and the loop
bails before the re-derive.
The SCTP ABORT path in sctp sendmsg check sflags() returns 0 and the
loop hits 'continue' before sctp sendmsg to asoc() is ever called, so
the @tmp cached by list for each entry safe() still covers the
lock-held free that ba59fb027307 ("sctp: walk the list of asoc
safely") was added for.
Found an issue in the description? Have something to add? Feel free to write us 👾
Related Identifiers
Affected Products
Linux