"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nrcu/nocb: Fix WARN_ON_ONCE() in the rcu_nocb_bypass_lock()\n\nFor the kernels built with CONFIG_RCU_NOCB_CPU_DEFAULT_ALL=y and\nCONFIG_RCU_LAZY=y, the following scenarios will trigger WARN_ON_ONCE()\nin the rcu_nocb_bypass_lock() and rcu_nocb_wait_contended() functions:\n\n CPU2 CPU11\nkthread\nrcu_nocb_cb_kthread ksys_write\nrcu_do_batch vfs_write\nrcu_torture_timer_cb proc_sys_write\n__kmem_cache_free proc_sys_call_handler\nkmemleak_free drop_caches_sysctl_handler\ndelete_object_full drop_slab\n__delete_object shrink_slab\nput_object lazy_rcu_shrink_scan\ncall_rcu rcu_nocb_flush_bypass\n__call_rcu_commn rcu_nocb_bypass_lock\n raw_spin_trylock(&rdp->nocb_bypass_lock) fail\n atomic_inc(&rdp->nocb_lock_contended);\nrcu_nocb_wait_contended WARN_ON_ONCE(smp_processor_id() != rdp->cpu);\n WARN_ON_ONCE(atomic_read(&rdp->nocb_lock_contended)) |\n |_ _ _ _ _ _ _ _ _ _same rdp and rdp->cpu != 11_ _ _ _ _ _ _ _ _ __|\n\nReproduce this bug with \"echo 3 > /proc/sys/vm/drop_caches\".\n\nThis commit therefore uses rcu_nocb_try_flush_bypass() instead of\nrcu_nocb_flush_bypass() in lazy_rcu_shrink_scan(). If the nocb_bypass\nqueue is being flushed, then rcu_nocb_try_flush_bypass will return\ndirectly."
"value":"En el kernel de Linux, se resolvi\u00f3 la siguiente vulnerabilidad: rcu/nocb: corrija WARN_ON_ONCE() en rcu_nocb_bypass_lock() Para los kernels creados con CONFIG_RCU_NOCB_CPU_DEFAULT_ALL=y y CONFIG_RCU_LAZY=y, los siguientes escenarios activar\u00e1n WARN_ON_ONCE() en rcu_nocb_bypass_lock( ) y funciones rcu_nocb_wait_contended(): CPU2 CPU11 kthread rcu_nocb_cb_kthread ksys_write rcu_do_batch vfs_write rcu_torture_timer_cb proc_sys_write __kmem_cache_free proc_sys_call_handler kmemleak_free drop_caches_sysctl_handler delete_object_full losa __delete_object encogimiento_slab put_object lazy_rcu_shrink_scan call_rcu rcu_nocb_flush_bypass __call_rcu_commn rcu_nocb_bypass_lock raw_spin_trylock(&rdp->nocb_bypass_lock) fail atomic_inc(&rdp->nocb_lock_contended); rcu_nocb_wait_contended WARN_ON_ONCE(smp_processor_id()!= rdp->cpu); WARN_ON_ONCE(atomic_read(&rdp->nocb_lock_contended)) | |_ _ _ _ _ _ _ _ _ _mismo rdp y rdp->cpu != 11_ _ _ _ _ _ _ _ _ __| Reproduzca este error con \"echo 3 > /proc/sys/vm/drop_caches\". Por lo tanto, esta confirmaci\u00f3n utiliza rcu_nocb_try_flush_bypass() en lugar de rcu_nocb_flush_bypass() en lazy_rcu_shrink_scan(). Si se est\u00e1 vaciando la cola nocb_bypass, entonces rcu_nocb_try_flush_bypass regresar\u00e1 directamente."