CVE-2021-46929

En el kernel de Linux, se resolvió la siguiente vulnerabilidad: sctp: use call_rcu para liberar el endpoint. Este parche tiene como objetivo retrasar la liberación del endpoint llamando a call_rcu() para solucionar otro problema de use-after-free en sctp_sock_dump(): ERROR: KASAN: use-after-free en __lock_acquire+0x36d9/0x4c20 Rastreo de llamadas: __lock_acquire+0x36d9/0x4c20 kernel/locking/lockdep.c:3218 lock_acquire+0x1ed/0x520 kernel/locking/lockdep.c:3844 __raw_spin_lock_bh include/linux/spinlock_api_smp.h :135 [en línea] _raw_spin_lock_bh+0x31/0x40 kernel/locking/spinlock.c:168 spin_lock_bh include/linux/spinlock.h:334 [en línea] __lock_sock+0x203/0x350 net/core/sock.c:2253 lock_sock_nested+0xfe/ 0x120 net/core/sock.c:2774 lock_sock include/net/sock.h:1492 [en línea] sctp_sock_dump+0x122/0xb20 net/sctp/diag.c:324 sctp_for_each_transport+0x2b5/0x370 net/sctp/socket.c: 5091 sctp_diag_dump+0x3ac/0x660 net/sctp/diag.c:527 __inet_diag_dump+0xa8/0x140 net/ipv4/inet_diag.c:1049 inet_diag_dump+0x9b/0x110 net/ipv4/inet_diag.c:1065 netlink_dump+0x6 06/0x1080 neto/ netlink/af_netlink.c:2244 __netlink_dump_start+0x59a/0x7c0 net/netlink/af_netlink.c:2352 netlink_dump_start include/linux/netlink.h:216 [en línea] inet_diag_handler_cmd+0x2ce/0x3f0 net/ipv4/inet_diag.c:1170 __sock_diag_cm re neto /core/sock_diag.c:232 [en línea] sock_diag_rcv_msg+0x31d/0x410 net/core/sock_diag.c:263 netlink_rcv_skb+0x172/0x440 net/netlink/af_netlink.c:2477 sock_diag_rcv+0x2a/0x40 net/core/sock_diag. c:274 Este problema ocurre cuando se quita asoc y se libera el sk antiguo después de obtenerlo mediante asoc->base.sk y antes de llamar a lock_sock(sk). Para evitar que sk se libere, como titular de sk, ep debe estar activo al llamar a lock_sock(). Este parche usa call_rcu() y mueve sock_put y ep free a sctp_endpoint_destroy_rcu(), por lo que es seguro intentar mantener el ep bajo rcu_read_lock en sctp_transport_traverse_process(). Si sctp_endpoint_hold() devuelve verdadero, significa que este ep todavía está vivo, lo hemos retenido y podemos continuar descartándolo; Si devuelve falso, significa que este ep está muerto y puede liberarse después de rcu_read_unlock, y debemos omitirlo. En sctp_sock_dump(), después de bloquear el sk, si este ep es diferente de tsp->asoc->ep, significa que durante este volcado, este asoc se eliminó antes de llamar a lock_sock(), y el sk debe omitirse; Si este ep es el mismo con tsp->asoc->ep, significa que no se produce ningún despegue en este asoc y, debido a lock_sock, tampoco se producirá ningún despegue hasta que se libere_sock. Tenga en cuenta que retrasar la liberación del endpoint no retrasará la liberación del puerto, ya que la liberación del puerto ocurre en sctp_endpoint_destroy() antes de llamar a call_rcu(). Además, liberar el endpoint mediante call_rcu() hace que sea seguro acceder a sk mediante asoc->base.sk en sctp_assocs_seq_show() y sctp_rcv(). Gracias Jones por plantear este problema. v1->v2: - mejorar el registro de cambios. - agregue kfree(ep) a sctp_endpoint_destroy_rcu(), como notó Jakub.

In the Linux kernel, the following vulnerability has been resolved: sctp: use call_rcu to free endpoint This patch is to delay the endpoint free by calling call_rcu() to fix another use-after-free issue in sctp_sock_dump(): BUG: KASAN: use-after-free in __lock_acquire+0x36d9/0x4c20 Call Trace: __lock_acquire+0x36d9/0x4c20 kernel/locking/lockdep.c:3218 lock_acquire+0x1ed/0x520 kernel/locking/lockdep.c:3844 __raw_spin_lock_bh include/linux/spinlock_api_smp.h:135 [inline] _raw_spin_lock_bh+0x31/0x40 kernel/locking/spinlock.c:168 spin_lock_bh include/linux/spinlock.h:334 [inline] __lock_sock+0x203/0x350 net/core/sock.c:2253 lock_sock_nested+0xfe/0x120 net/core/sock.c:2774 lock_sock include/net/sock.h:1492 [inline] sctp_sock_dump+0x122/0xb20 net/sctp/diag.c:324 sctp_for_each_transport+0x2b5/0x370 net/sctp/socket.c:5091 sctp_diag_dump+0x3ac/0x660 net/sctp/diag.c:527 __inet_diag_dump+0xa8/0x140 net/ipv4/inet_diag.c:1049 inet_diag_dump+0x9b/0x110 net/ipv4/inet_diag.c:1065 netlink_dump+0x606/0x1080 net/netlink/af_netlink.c:2244 __netlink_dump_start+0x59a/0x7c0 net/netlink/af_netlink.c:2352 netlink_dump_start include/linux/netlink.h:216 [inline] inet_diag_handler_cmd+0x2ce/0x3f0 net/ipv4/inet_diag.c:1170 __sock_diag_cmd net/core/sock_diag.c:232 [inline] sock_diag_rcv_msg+0x31d/0x410 net/core/sock_diag.c:263 netlink_rcv_skb+0x172/0x440 net/netlink/af_netlink.c:2477 sock_diag_rcv+0x2a/0x40 net/core/sock_diag.c:274 This issue occurs when asoc is peeled off and the old sk is freed after getting it by asoc->base.sk and before calling lock_sock(sk). To prevent the sk free, as a holder of the sk, ep should be alive when calling lock_sock(). This patch uses call_rcu() and moves sock_put and ep free into sctp_endpoint_destroy_rcu(), so that it's safe to try to hold the ep under rcu_read_lock in sctp_transport_traverse_process(). If sctp_endpoint_hold() returns true, it means this ep is still alive and we have held it and can continue to dump it; If it returns false, it means this ep is dead and can be freed after rcu_read_unlock, and we should skip it. In sctp_sock_dump(), after locking the sk, if this ep is different from tsp->asoc->ep, it means during this dumping, this asoc was peeled off before calling lock_sock(), and the sk should be skipped; If this ep is the same with tsp->asoc->ep, it means no peeloff happens on this asoc, and due to lock_sock, no peeloff will happen either until release_sock. Note that delaying endpoint free won't delay the port release, as the port release happens in sctp_endpoint_destroy() before calling call_rcu(). Also, freeing endpoint by call_rcu() makes it safe to access the sk by asoc->base.sk in sctp_assocs_seq_show() and sctp_rcv(). Thanks Jones to bring this issue up. v1->v2: - improve the changelog. - add kfree(ep) into sctp_endpoint_destroy_rcu(), as Jakub noticed.