PT-2026-151774

CVE-2026-99563

·

Publicado

2026-11-19

·

Atualizado

2026-11-19

Descrição por

Pontuação
N/A
Não disponível

Não há classificações de severidade ou métricas disponíveis. Quando houver, atualizaremos as informações correspondentes na página.

In the Linux kernel, the following vulnerability has been resolved:
smb: smbdirect: release failed pending sockets of a listener
A child of a listener only left the listen.pending or listen.ready list via smbdirect socket accept() or when the listener was destroyed. So every connection that failed before it was accepted (negotiate timeout, peer disconnect, invalid negotiate request, ...) stayed on the list and counted against the backlog until the listener was destroyed. After about backlog (10 for ksmbd) such connections smbdirect listen connect request() rejected every new connection with -EBUSY, so an unauthenticated peer was able to permanently stop ksmbd from accepting SMB-Direct connections.
Now a failed child moves itself to a new listen.orphaned list at the end of smbdirect socket cleanup work(), after the disconnect was started, and queues listen.purge orphaned work, which releases it with smbdirect socket release(), without holding the listener's rdma lock handler() lock, in the same way smbdirect socket destroy() does for the remaining children. Orphaned sockets still count against the backlog until they are released.
sc->accept.listener is now only changed under listener->listen.lock and whoever clears it is responsible for releasing the socket: smbdirect socket accept(), purge orphaned work or the destroy of the listener.
In order to make that work:
  • a struct smbdirect socket that is a listener is now freed via kfree rcu(), so that a child can dereference sc->accept.listener under rcu read lock(), even if the listener is released concurrently.
  • smbdirect listen connect request() adds nsc to the listen.pending list and sets nsc->accept.listener before it calls smbdirect accept connect request(), and always returns 0 so the rdma cm core keeps the connection id and nsc owns it. On a non-zero return it schedules nsc's teardown with smbdirect socket schedule cleanup(), which orphans it off the listener so that purge orphaned work releases it; nsc destroys its own connection id with rdma destroy id() in that teardown. The teardown has to be deferred because the rdma cm core holds nsc's id priv->handler mutex across the handler, so nsc->rdma.cm id must not be destroyed synchronously from here. That same handler mutex (and the listener's) also guarantees nsc can't go away under us, so setting the listener before the call is safe and lets a failure orphan nsc.
  • smbdirect accept negotiate recv work() no longer moves a socket to the ready list if it already failed or no longer belongs to the listener. Every accepting socket has a listener, so if sc->accept.listener is already cleared, it no longer sends a negotiate response, as the socket is about to be released.
  • smbdirect socket destroy() of the listener disables purge orphaned work and also releases the orphaned sockets.
Fornecedores
Produtos afetados
Identificadores relacionados
Encontrou algum problema na descrição? Tem algo a acrescentar? Fique à vontade para nos escrever 👾