Managing paging I/O errors during hypervisor page fault processing

A shared memory partition data processing system comprising:

• a physical memory comprising a shared memory pool for at least one shared memory partition of the data processing system, the at least one shared memory partition comprising a plurality of virtual processors;
• a processor supporting the plurality of virtual processors, the plurality of virtual processors sharing at least a portion of the shared memory pool of the physical memory of the data processing system;
• a hypervisor that connects the shared memory partition to at least the shared memory pool and to the paging space;
• a paging service partition that facilitates paging-in of memory pages from the paging space to the shared memory pool;
• wherein the hypervisor memory manager, responsive to the paging service partition encountering an I/O paging error when attempting to read from the paging space, receives a paging failure indication from the paging service partition, and wherein the hypervisor memory manager sends a simulated machine check interrupt instruction to the shared memory partition, and wherein the virtual processor is placed in a wait state subsequent to the virtual processor attempting to access a page that is paged-out;
• and
• a machine check handler associated with the shared memory partition, wherein the virtual processor in the wait state is allowed to execute at the machine check handler.

An article of manufacture comprising:

• at least one non-transitory computer-readable storage medium having computer-readable program code logic to address a hypervisor page fault in a shared memory partition data processing system, the computer-readable program code logic when executing on a processor performs:
• responsive to a paging service partition encountering an I/O paging error when attempting to read from a paging space, sending a paging failure indication to a hypervisor that is connected to a shared memory partition and to a paging service partition:
• sending a simulated machine check interrupt instruction from the hypervisor to the shared memory partition;
• responsive to the virtual processor attempting to access apage that is paged-out, placing the virtual processor in a wait state;
• and
• responsive to a machine check handler associated with the shared memory partition obtaining control, allowing the virtual processor in the wait state to begin executing at the machine check handler.

A method for managing paging I/O errors during hypervisor page fault processing comprising:

• in response to a hypervisor page fault in requesting memory that is not resident in a shared memory pool, sending an I/O paging request to an external storage paging space;
• in response to a paging service partition encountering an I/O paging error, sending a paging failure indication a hypervisor;
• sending a simulated machine check interrupt instruction from the hypervisor to the shared memory partition;
• transferring control to a machine check handler associated with the shared memory partition;
• performing data analysis at the machine check handler utilizing an error log to attempt to isolate the I/O paging error to a process or a set of processes in the shared memory partition;
• terminating the process or set of processes associated with the I/O paging error;
• and
• clearing or initializing the page associated with the I/O paging error.