"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm: slub: fix flush_cpu_slab()/__free_slab() invocations in task context.\n\nCommit 5a836bf6b09f (\"mm: slub: move flush_cpu_slab() invocations\n__free_slab() invocations out of IRQ context\") moved all flush_cpu_slab()\ninvocations to the global workqueue to avoid a problem related\nwith deactivate_slab()/__free_slab() being called from an IRQ context\non PREEMPT_RT kernels.\n\nWhen the flush_all_cpu_locked() function is called from a task context\nit may happen that a workqueue with WQ_MEM_RECLAIM bit set ends up\nflushing the global workqueue, this will cause a dependency issue.\n\n workqueue: WQ_MEM_RECLAIM nvme-delete-wq:nvme_delete_ctrl_work [nvme_core]\n is flushing !WQ_MEM_RECLAIM events:flush_cpu_slab\n WARNING: CPU: 37 PID: 410 at kernel/workqueue.c:2637\n check_flush_dependency+0x10a/0x120\n Workqueue: nvme-delete-wq nvme_delete_ctrl_work [nvme_core]\n RIP: 0010:check_flush_dependency+0x10a/0x120[ 453.262125] Call Trace:\n __flush_work.isra.0+0xbf/0x220\n ? __queue_work+0x1dc/0x420\n flush_all_cpus_locked+0xfb/0x120\n __kmem_cache_shutdown+0x2b/0x320\n kmem_cache_destroy+0x49/0x100\n bioset_exit+0x143/0x190\n blk_release_queue+0xb9/0x100\n kobject_cleanup+0x37/0x130\n nvme_fc_ctrl_free+0xc6/0x150 [nvme_fc]\n nvme_free_ctrl+0x1ac/0x2b0 [nvme_core]\n\nFix this bug by creating a workqueue for the flush operation with\nthe WQ_MEM_RECLAIM bit set."
"value":"En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: mm: slub: corrige las invocaciones de flu_cpu_slab()/__free_slab() en el contexto de la tarea. Commit 5a836bf6b09f (\"mm: slub: mover invocaciones de flu_cpu_slab() invocaciones de __free_slab() fuera del contexto IRQ\") movi\u00f3 todas las invocaciones de flu_cpu_slab() a la cola de trabajo global para evitar un problema relacionado con la llamada de desactivate_slab()/__free_slab() desde un Contexto IRQ en n\u00facleos PREEMPT_RT. Cuando se llama a la funci\u00f3n Flush_all_cpu_locked() desde un contexto de tarea, puede suceder que una cola de trabajo con el bit WQ_MEM_RECLAIM configurado termine vaciando la cola de trabajo global, esto causar\u00e1 un problema de dependencia. cola de trabajo: WQ_MEM_RECLAIM nvme-delete-wq:nvme_delete_ctrl_work [nvme_core] se est\u00e1 vaciando! Eventos WQ_MEM_RECLAIM:flush_cpu_slab ADVERTENCIA: CPU: 37 PID: 410 en kernel/workqueue.c:2637 check_flush_dependency+0x10a/0x120 Cola de trabajo: vme-delete-wq nvme_delete_ctrl_work [ nvme_core] RIP: 0010:check_flush_dependency+0x10a/0x120[ 453.262125] Seguimiento de llamadas: __flush_work.isra.0+0xbf/0x220? __queue_work+0x1dc/0x420 Flush_all_cpus_locked+0xfb/0x120 __kmem_cache_shutdown+0x2b/0x320 kmem_cache_destroy+0x49/0x100 bioset_exit+0x143/0x190 blk_release_queue+0xb9/0x100 kobject_cleanup+0 x37/0x130 nvme_fc_ctrl_free+0xc6/0x150 [nvme_fc] nvme_free_ctrl+0x1ac/0x2b0 [nvme_core ] Corrija este error creando una cola de trabajo para la operaci\u00f3n de vaciado con el bit WQ_MEM_RECLAIM establecido."