How to reset the cpu gracefully without powercycle in case of MCE

[gn0001]

Hi,

We are using Intel® Xeon® Processors D-1528 cpu as host cpu in our hypervisor system, we hit a nested exception due to which hypervisor went into unresponsive state with below logs on console, we recovered from this problem only after manual reboot. Could you please help how to recover from this state without human intervention, because we would have deployed many boxes in the field, for these kind of problems we cannot reboot the hypervisor manually. Please help.

[ 0.444243] smpboot: Booting Node 0, Processors #1

[ 10.461018] smpboot: do_boot_cpu failed(-1) to wakeup CPU#1

[ 10.467559] NMI watchdog: enabled on all CPUs, permanently consumes one hw-PMU counter.

[ 10.476574] #2

[ 20.489592] smpboot: do_boot_cpu failed(-1) to wakeup CPU#2

[ 20.496073] #3

[ 30.508927] smpboot: do_boot_cpu failed(-1) to wakeup CPU#3

[ 30.515416] #4

[ 40.528267] smpboot: do_boot_cpu failed(-1) to wakeup CPU#4

[ 40.534748] #5

[ 50.547654] smpboot: do_boot_cpu failed(-1) to wakeup CPU#5

[ 50.554138] #6 #7

[ 60.569089] smpboot: do_boot_cpu failed(-1) to wakeup CPU#7

[ 60.575586] #8

[ 70.588235] smpboot: do_boot_cpu failed(-1) to wakeup CPU#8

[ 70.594718] #9

[ 80.607321] smpboot: do_boot_cpu failed(-1) to wakeup CPU#9

[ 80.613797] #10

[ 90.626512] smpboot: do_boot_cpu failed(-1) to wakeup CPU#10

[ 90.633091] #11 OK

[ 100.646327] smpboot: do_boot_cpu failed(-1) to wakeup CPU#11

[ 100.652652] Brought up 2 CPUs

[ 100.655962] smpboot: Max logical packages: 6

[ 100.660730] smpboot: Total of 2 processors activated (7599.97 BogoMIPS)

[CarlosAM_INTEL]

Hello, @gn0001​:

Thank you for contacting Intel Embedded Community.

Your design should follow the reset requirements stated in section 4.3, on pages 87 through 92 of the Intel Xeon D-1500 Processor Family External Design Specification [EDS] Volume Three: Electrical document # 544042. This document can be found when you are logged into your Resource & Design Center (RDC) privileged account at the following website:

http://www.intel.com/cd/edesign/library/asmo-na/eng/544042.htm

The RDC Account Support form is the channel to process your account update request and any inconveniences associated with the listed websites. It can be found at:

https://www.intel.com/content/www/us/en/forms/support/my-intel-sign-on-support.html

Best regards,

@Mæcenas_INTEL​.