does the Linux kernel incorporate some mechanism to detect a malfunction at some CPU at the SMP implementations? If, for example, some core at multi-core platforms get stuck or blocked due to an endless loop after disabling the interrupts or if some core get power cutoff due to some HW failure (over-temperature, etc.), will the rest of the system notice this and perform some reaction upon this?
Asked
Active
Viewed 45 times
1
-
Yes, it has few mechanisms to do something about it. – 0andriy Jun 08 '20 at 20:36