Title: kernel de Linux
Description: En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: workqueue: Mejorar la escalabilidad del watchdog de workqueue touch En un sistema powerpc de ~2000 CPU, se han observado bloqueos duros en el código de workqueue cuando se ejecuta stop_machine (en este caso debido a la conexión en caliente de la CPU). Esto se debe a que muchas CPU giran en multi_cpu_stop, llamando a touch_nmi_watchdog() que termina llamando a wq_watchdog_touch(). wq_watchdog_touch() escribe en la variable global wq_watchdog_touched, y esta puede encontrarse en la misma línea de caché que otros datos importantes de workqueue, lo que ralentiza las operaciones hasta el punto de bloquearse. En el caso del siguiente seguimiento abreviado, worker_pool_idr estaba en la línea directa, lo que hacía que los bloqueos siempre aparecieran en idr_find. watchdog: CPU 1125 autodetectó BLOQUEO duro @ idr_find Seguimiento de llamadas: get_work_pool __queue_work call_timer_fn run_timer_softirq __do_softirq do_softirq_own_stack irq_exit timer_interrupt decrementer_common_virt * interrupción: 900 (temporizador) en multi_cpu_stop multi_cpu_stop cpu_stopper_thread smpboot_thread_fn kthread Solucione esto haciendo que wq_watchdog_touch() solo escriba en la línea si la última vez que se registró un toque excede 1/4 del umbral del watchdog.

In the Linux kernel, the following vulnerability has been resolved:

workqueue: Improve scalability of workqueue watchdog touch

On a ~2000 CPU powerpc system, hard lockups have been observed in the
workqueue code when stop_machine runs (in this case due to CPU hotplug).
This is due to lots of CPUs spinning in multi_cpu_stop, calling
touch_nmi_watchdog() which ends up calling wq_watchdog_touch().
wq_watchdog_touch() writes to the global variable wq_watchdog_touched,
and that can find itself in the same cacheline as other important
workqueue data, which slows down operations to the point of lockups.

In the case of the following abridged trace, worker_pool_idr was in
the hot line, causing the lockups to always appear at idr_find.

  watchdog: CPU 1125 self-detected hard LOCKUP @ idr_find
  Call Trace:
  get_work_pool
  __queue_work
  call_timer_fn
  run_timer_softirq
  __do_softirq
  do_softirq_own_stack
  irq_exit
  timer_interrupt
  decrementer_common_virt
  * interrupt: 900 (timer) at multi_cpu_stop
  multi_cpu_stop
  cpu_stopper_thread
  smpboot_thread_fn
  kthread

Fix this by having wq_watchdog_touch() only write to the line if the
last time a touch was recorded exceeds 1/4 of the watchdog threshold.

Rejected reason: This CVE ID has been rejected or withdrawn by its CVE Numbering Authority.

kernel.org https://git.kernel.org/stable/c/241bce1c757d0587721512296952e6bba69631ed

kernel.org https://git.kernel.org/stable/c/98f887f820c993e05a12e8aa816c80b8661d4c87

kernel.org https://git.kernel.org/stable/c/9d08fce64dd77f42e2361a4818dbc4b50f3c7dad

kernel.org https://git.kernel.org/stable/c/a2abd35e7dc55bf9ed01e2b3481fa78e086d3bf4

kernel.org https://git.kernel.org/stable/c/da5f374103a1e0881bbd35847dc57b04ac155eb0

In the Linux kernel, the following vulnerability has been resolved:

workqueue: Improve scalability of workqueue watchdog touch

On a ~2000 CPU powerpc system, hard lockups have been observed in the
workqueue code when stop_machine runs (in this case due to CPU hotplug).
This is due to lots of CPUs spinning in multi_cpu_stop, calling
touch_nmi_watchdog() which ends up calling wq_watchdog_touch().
wq_watchdog_touch() writes to the global variable wq_watchdog_touched,
and that can find itself in the same cacheline as other important
workqueue data, which slows down operations to the point of lockups.

In the case of the following abridged trace, worker_pool_idr was in
the hot line, causing the lockups to always appear at idr_find.

  watchdog: CPU 1125 self-detected hard LOCKUP @ idr_find
  Call Trace:
  get_work_pool
  __queue_work
  call_timer_fn
  run_timer_softirq
  __do_softirq
  do_softirq_own_stack
  irq_exit
  timer_interrupt
  decrementer_common_virt
  * interrupt: 900 (timer) at multi_cpu_stop
  multi_cpu_stop
  cpu_stopper_thread
  smpboot_thread_fn
  kthread

Fix this by having wq_watchdog_touch() only write to the line if the
last time a touch was recorded exceeds 1/4 of the watchdog threshold.

kernel.org https://git.kernel.org/stable/c/241bce1c757d0587721512296952e6bba69631ed [No types assigned]

kernel.org https://git.kernel.org/stable/c/98f887f820c993e05a12e8aa816c80b8661d4c87 [No types assigned]

kernel.org https://git.kernel.org/stable/c/9d08fce64dd77f42e2361a4818dbc4b50f3c7dad [No types assigned]

kernel.org https://git.kernel.org/stable/c/a2abd35e7dc55bf9ed01e2b3481fa78e086d3bf4 [No types assigned]

kernel.org https://git.kernel.org/stable/c/da5f374103a1e0881bbd35847dc57b04ac155eb0 [No types assigned]