"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nmptcp: init: protect sched with rcu_read_lock\n\nEnabling CONFIG_PROVE_RCU_LIST with its dependence CONFIG_RCU_EXPERT\ncreates this splat when an MPTCP socket is created:\n\n =============================\n WARNING: suspicious RCU usage\n 6.12.0-rc2+ #11 Not tainted\n -----------------------------\n net/mptcp/sched.c:44 RCU-list traversed in non-reader section!!\n\n other info that might help us debug this:\n\n rcu_scheduler_active = 2, debug_locks = 1\n no locks held by mptcp_connect/176.\n\n stack backtrace:\n CPU: 0 UID: 0 PID: 176 Comm: mptcp_connect Not tainted 6.12.0-rc2+ #11\n Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011\n Call Trace:\n <TASK>\n dump_stack_lvl (lib/dump_stack.c:123)\n lockdep_rcu_suspicious (kernel/locking/lockdep.c:6822)\n mptcp_sched_find (net/mptcp/sched.c:44 (discriminator 7))\n mptcp_init_sock (net/mptcp/protocol.c:2867 (discriminator 1))\n ? sock_init_data_uid (arch/x86/include/asm/atomic.h:28)\n inet_create.part.0.constprop.0 (net/ipv4/af_inet.c:386)\n ? __sock_create (include/linux/rcupdate.h:347 (discriminator 1))\n __sock_create (net/socket.c:1576)\n __sys_socket (net/socket.c:1671)\n ? __pfx___sys_socket (net/socket.c:1712)\n ? do_user_addr_fault (arch/x86/mm/fault.c:1419 (discriminator 1))\n __x64_sys_socket (net/socket.c:1728)\n do_syscall_64 (arch/x86/entry/common.c:52 (discriminator 1))\n entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)\n\nThat's because when the socket is initialised, rcu_read_lock() is not\nused despite the explicit comment written above the declaration of\nmptcp_sched_find() in sched.c. Adding the missing lock/unlock avoids the\nwarning."
"value":"En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: mptcp: init: proteger sched con rcu_read_lock Habilitar CONFIG_PROVE_RCU_LIST con su dependencia CONFIG_RCU_EXPERT crea este splat cuando se crea un socket MPTCP: ============================= ADVERTENCIA: uso sospechoso de RCU 6.12.0-rc2+ #11 No contaminado ----------------------------- net/mptcp/sched.c:44 \u00a1Lista de RCU recorrida en una secci\u00f3n que no es de lectura! Otra informaci\u00f3n que podr\u00eda ayudarnos a depurar esto: rcu_scheduler_active = 2, debug_locks = 1 no hay bloqueos retenidos por mptcp_connect/176. seguimiento de pila: CPU: 0 UID: 0 PID: 176 Comm: mptcp_connect No contaminado 6.12.0-rc2+ #11 Nombre del hardware: Bochs Bochs, BIOS Bochs 01/01/2011 Seguimiento de llamadas: dump_stack_lvl (lib/dump_stack.c:123) lockdep_rcu_suspicious (kernel/locking/lockdep.c:6822) mptcp_sched_find (net/mptcp/sched.c:44 (discriminador 7)) mptcp_init_sock (net/mptcp/protocol.c:2867 (discriminador 1)) ? __sock_create (include/linux/rcupdate.h:347 (discriminador 1)) __sock_create (net/socket.c:1576) __sys_socket (net/socket.c:1671) ? __pfx___sys_socket (net/socket.c:1712) ? do_user_addr_fault (arch/x86/mm/fault.c:1419 (discriminator 1)) __x64_sys_socket (net/socket.c:1728) do_syscall_64 (arch/x86/entry/common.c:52 (discriminator 1)) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) Esto se debe a que cuando se inicializa el socket, no se utiliza rcu_read_lock() a pesar del comentario expl\u00edcito escrito encima de la declaraci\u00f3n de mptcp_sched_find() en sched.c. Agregar el bloqueo/desbloqueo faltante evita la advertencia."
