"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nbtrfs: flush delalloc workers queue before stopping cleaner kthread during unmount\n\nDuring the unmount path, at close_ctree(), we first stop the cleaner\nkthread, using kthread_stop() which frees the associated task_struct, and\nthen stop and destroy all the work queues. However after we stopped the\ncleaner we may still have a worker from the delalloc_workers queue running\ninode.c:submit_compressed_extents(), which calls btrfs_add_delayed_iput(),\nwhich in turn tries to wake up the cleaner kthread - which was already\ndestroyed before, resulting in a use-after-free on the task_struct.\n\nSyzbot reported this with the following stack traces:\n\n BUG: KASAN: slab-use-after-free in __lock_acquire+0x78/0x2100 kernel/locking/lockdep.c:5089\n Read of size 8 at addr ffff8880259d2818 by task kworker/u8:3/52\n\n CPU: 1 UID: 0 PID: 52 Comm: kworker/u8:3 Not tainted 6.13.0-rc1-syzkaller-00002-gcdd30ebb1b9f #0\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024\n Workqueue: btrfs-delalloc btrfs_work_helper\n Call Trace:\n <TASK>\n __dump_stack lib/dump_stack.c:94 [inline]\n dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120\n print_address_description mm/kasan/report.c:378 [inline]\n print_report+0x169/0x550 mm/kasan/report.c:489\n kasan_report+0x143/0x180 mm/kasan/report.c:602\n __lock_acquire+0x78/0x2100 kernel/locking/lockdep.c:5089\n lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5849\n __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]\n _raw_spin_lock_irqsave+0xd5/0x120 kernel/locking/spinlock.c:162\n class_raw_spinlock_irqsave_constructor include/linux/spinlock.h:551 [inline]\n try_to_wake_up+0xc2/0x1470 kernel/sched/core.c:4205\n submit_compressed_extents+0xdf/0x16e0 fs/btrfs/inode.c:1615\n run_ordered_work fs/btrfs/async-thread.c:288 [inline]\n btrfs_work_helper+0x96f/0xc40 fs/btrfs/async-thread.c:324\n process_one_work kernel/workqueue.c:3229 [inline]\n process_scheduled_works+0xa66/0x1840 kernel/workqueue.c:3310\n worker_thread+0x870/0xd30 kernel/workqueue.c:3391\n kthread+0x2f0/0x390 kernel/kthread.c:389\n ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147\n ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244\n </TASK>\n\n Allocated by task 2:\n kasan_save_stack mm/kasan/common.c:47 [inline]\n kasan_save_track+0x3f/0x80 mm/kasan/common.c:68\n unpoison_slab_object mm/kasan/common.c:319 [inline]\n __kasan_slab_alloc+0x66/0x80 mm/kasan/common.c:345\n kasan_slab_alloc include/linux/kasan.h:250 [inline]\n slab_post_alloc_hook mm/slub.c:4104 [inline]\n slab_alloc_node mm/slub.c:4153 [inline]\n kmem_cache_alloc_node_noprof+0x1d9/0x380 mm/slub.c:4205\n alloc_task_struct_node kernel/fork.c:180 [inline]\n dup_task_struct+0x57/0x8c0 kernel/fork.c:1113\n copy_process+0x5d1/0x3d50 kernel/fork.c:2225\n kernel_clone+0x223/0x870 kernel/fork.c:2807\n kernel_thread+0x1bc/0x240 kernel/fork.c:2869\n create_kthread kernel/kthread.c:412 [inline]\n kthreadd+0x60d/0x810 kernel/kthread.c:767\n ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147\n ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244\n\n Freed by task 24:\n kasan_save_stack mm/kasan/common.c:47 [inline]\n kasan_save_track+0x3f/0x80 mm/kasan/common.c:68\n kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:582\n poison_slab_object mm/kasan/common.c:247 [inline]\n __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264\n kasan_slab_free include/linux/kasan.h:233 [inline]\n slab_free_hook mm/slub.c:2338 [inline]\n slab_free mm/slub.c:4598 [inline]\n kmem_cache_free+0x195/0x410 mm/slub.c:4700\n put_task_struct include/linux/sched/task.h:144 [inline]\n delayed_put_task_struct+0x125/0x300 kernel/exit.c:227\n rcu_do_batch kernel/rcu/tree.c:2567 [inline]\n rcu_core+0xaaa/0x17a0 kernel/rcu/tree.c:2823\n handle_softirqs+0x2d4/0x9b0 kernel/softirq.c:554\n run_ksoftirqd+0xca/0x130 kernel/softirq.c:943\n \n---truncated---"
},
{
"lang":"es",
"value":"En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: btrfs: vaciar la cola de trabajadores delalloc antes de detener el kthread del limpiador durante el desmontaje Durante la ruta de desmontaje, en close_ctree(), primero detenemos el kthread del limpiador, utilizando kthread_stop() que libera el task_struct asociado, y luego detenemos y destruimos todas las colas de trabajo. Sin embargo, despu\u00e9s de detener el limpiador, es posible que a\u00fan tengamos un trabajador de la cola delalloc_workers ejecutando inode.c:submit_compressed_extents(), que llama a btrfs_add_delayed_iput(), que a su vez intenta despertar el kthread del limpiador, que ya se destruy\u00f3 antes, lo que resulta en un uso despu\u00e9s de la liberaci\u00f3n en el task_struct. Syzbot inform\u00f3 esto con los siguientes seguimientos de pila: ERROR: KASAN: slab-use-after-free en __lock_acquire+0x78/0x2100 kernel/locking/lockdep.c:5089 Lectura de tama\u00f1o 8 en la direcci\u00f3n ffff8880259d2818 por la tarea kworker/u8:3/52 CPU: 1 UID: 0 PID: 52 Comm: kworker/u8:3 No contaminado 6.13.0-rc1-syzkaller-00002-gcdd30ebb1b9f #0 Nombre del hardware: Google Google Compute Engine/Google Compute Engine, BIOS Google 13/09/2024 Cola de trabajo: btrfs-delalloc btrfs_work_helper Seguimiento de llamadas: __dump_stack lib/dump_stack.c:94 [en l\u00ednea] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [en l\u00ednea] print_report+0x169/0x550 mm/kasan/report.c:489 kasan_report+0x143/0x180 mm/kasan/report.c:602 __lock_acquire+0x78/0x2100 kernel/locking/lockdep.c:5089 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5849 __raw_spin_lock_irqsave incluir/linux/spinlock_api_smp.h:110 [en l\u00ednea] _raw_spin_lock_irqsave+0xd5/0x120 kernel/locking/spinlock.c:162 class_raw_spinlock_irqsave_constructor include/linux/spinlock.h:551 [en l\u00ednea] try_to_wake_up+0xc2/0x1470 kernel/sched/core.c:4205 submit_compressed_extents+0xdf/0x16e0 fs/btrfs/inode.c:1615 run_ordered_work fs/btrfs/async-thread.c:288 [en l\u00ednea] btrfs_work_helper+0x96f/0xc40 fs/btrfs/async-thread.c:324 process_one_work kernel/workqueue.c:3229 [en l\u00ednea] process_scheduled_works+0xa66/0x1840 kernel/workqueue.c:3310 Asignado por la tarea 2: kasan_save_stack mm/kasan/common.c:47 [en l\u00ednea] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 unpoison_slab_object mm/kasan/common.c:319 [en l\u00ednea] __kasan_slab_alloc+0x66/0x80 mm/kasan/common.c:345 kasan_slab_alloc include/linux/kasan.h:250 [en l\u00ednea] gancho de alloc de publicaci\u00f3n de losa mm/slub.c:4104 [en l\u00ednea] nodo de alloc de losa mm/slub.c:4153 [en l\u00ednea] nodo de alloc de cach\u00e9 kmem_noprof+0x1d9/0x380 mm/slub.c:4205 nodo de estructura de tareas de asignaci\u00f3n kernel/fork.c:180 [en l\u00ednea] estructura de tareas dup+0x57/0x8c0 kernel/fork.c:1113 proceso de copia+0x5d1/0x3d50 kernel/fork.c:2225 clon de kernel+0x223/0x870 kernel/fork.c:2807 kernel_thread+0x1bc/0x240 kernel/fork.c:2869 create_kthread kernel/kthread.c:412 [en l\u00ednea] kthreadd+0x60d/0x810 kernel/kthread.c:767 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Liberado por la tarea 24: kasan_save_stack mm/kasan/common.c:47 [en l\u00ednea] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:582 poison_slab_object mm/kasan/common.c:247 [en l\u00ednea] __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264 kasan_slab_free include/linux/kasan.h:233 [en l\u00ednea] gancho slab_free mm/slub.c:2338 [en l\u00ednea] slab_free mm/slub.c:4598 [en l\u00ednea] kmem_cache_free+0x195/0x410 mm/slub.c:4700 put_task_struct include/linux/sched/task.h:144 [en l\u00ednea] delayed_put_task_struct+0x125/0x300 kernel/exit.c:227 rcu_do_batch kernel/rcu/tree.c:2567 [en l\u00ednea] rcu_core+0xaaa/0x17a0 kernel/rcu/tree.c:2823 handle_softirqs+0x2d4/0x9b0 kernel/softirq.c:554 run_ksoftirqd+0xca/0x130 kernel/softirq.c:943 ---truncado---"
