PT-2026-61297 · Linux · Linux
CVE-2026-63980
·
Publicado
2026-07-19
·
Atualizado
2026-07-19
CVSS v3.1
7.5
Alta
| Vetor | AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H |
In the Linux kernel, the following vulnerability has been resolved:
net/handshake: Use spin lock bh for hn lock
nvmet tcp state change(), a socket callback that runs in BH context,
can reach handshake req cancel() via nvmet tcp schedule release queue()
and tls handshake cancel(). handshake req cancel() acquires
hn->hn lock with plain spin lock(). If a process-context thread on
the same CPU holds hn->hn lock when a softirq invokes the cancel path,
the lock attempt deadlocks. This is the only caller that invokes
tls handshake cancel() from BH context; every other consumer calls it
from process context.
Deferring the cancel to process context in the NVMe target is not
straightforward: nvmet tcp schedule release queue() must call
tls handshake cancel() atomically with its state transition to
DISCONNECTING. If the cancel were deferred, the handshake completion
callback could fire in the window before the cancel runs, observe the
unexpected state, and return without dropping its kref on the queue.
Reworking that interlock is considerably more invasive than hardening
the handshake lock. Convert all hn->hn lock acquisitions from
spin lock/spin unlock to spin lock bh/spin unlock bh so the lock is
never taken with softirqs enabled.
Correção
Encontrou algum problema na descrição? Tem algo a acrescentar? Fique à vontade para nos escrever 👾
Identificadores relacionados
Produtos afetados
Linux