Behavioral implementation of a double fault stack in a computer system

A method of exception handling in a virtualized computer system, the virtualized computing system including a hypervisor executing therein, the hypervisor supporting a virtual machine (VM), the method comprising:

• creating, by the hypervisor, a first stack and a double fault stack mapped to locations in a system memory of the virtualized computer system and associated with a physical central processing unit (PCPU) of the virtualized computer system;
• handling, by a second exception handler of the hypervisor, a second exception generated by the PCPU in response to execution of an instruction of a first exception handler of the hypervisor that caused a stack overflow condition of the first stack while handling a first exception generated by the PCPU;
• and
• executing, by the second exception handler, instructions to store register states of the PCPU on the double fault stack.

A computer system, comprising:

• a hardware platform comprising a physical central processing unit (PCPU) and a system memory;
• and
• a software platform executing on the hardware platform, the software platform configured to store a first stack, a second stack, and a double fault stack in the system memory in association with the PCPU, the PCPU configured to call a second exception handler of the software platform in response to a stack overflow condition of the first stack caused by a first exception handler of the software platform, the software platform further configured to:
• store, by the second exception handler, an exception frame on the double fault stack in response to a stack overflow condition of the first stack;
• switch, by the second exception handler, a first stack pointer of the PCPU from pointing to the first stack to pointing to the double fault stack;
• set a current stack pointer of the PCPU to the first stack pointer;
• and
• execute software on the PCPU with the current stack pointer pointing to the double fault stack.