Method and apparatus for supporting persistence and computing device

1. A method of supporting persistence of a computing device including a processor and a memory module including a non-volatile memory, the method comprising:

performing a stop procedure upon power failure, wherein the step of performing the stop procedure comprises

scheduling out a running process task,

storing a state of the process task to a process control block of the memory module,

flushing a cache for the processor, and

flushing a pending memory request;

wherein the processor includes a plurality of core tiles including a master tile and a worker tile, and

wherein the scheduling out the running process task comprises:

scheduling out a process task running on the master tile except a kernel process implementing the stop procedure; and

scheduling out a process task running on the worker tile.

2. The method of claim 1 , further comprising performing a go procedure when power is recovered,

wherein the performing the go procedure comprises scheduling process tasks based on the process control block.

3. The method of claim 2 , wherein the performing the stop procedure further comprises storing a register associated with a program for performing the stop procedure and the go procedure to the memory module.

4. The method of claim 3 , wherein the performing the go procedure further comprises restoring the register from the memory module.

5. The method of claim 1 , wherein the performing the stop procedure further comprises:

reading peripheral information from a peripheral related memory region; and

writing the peripheral information to a device context of the memory module.

6. The method of claim 5 , further comprising performing a go procedure when power is recovered,

wherein the performing the go procedure comprises restoring the peripheral information by reading the device context.

7. The method of claim 1 , wherein the performing the stop procedure comprises, after flushing the pending memory request;

reporting offline to the master tile and powering off, at the worker tile; and

switching a context at the master tile when receiving the offline report from the worker tile.

8. The method of claim 7 , wherein the switching the context comprises raising a system-level exception to switch the context from a kernel to bootloader of a protection mode.

9. The method of claim 8 , wherein the performing the stop procedure comprises, in the protection mode, storing a return address for re-executing a system to the memory module.

10. The method of claim 1 , wherein the performing the stop procedure comprises:

raising a system-level exception to switch a context from a kernel to bootloader of a protection mode; and

executing a first flush instruction for flushing the cache and a second flush instruction for flushing the memory request in the protection mode.

11. The method of claim 1 , wherein the flushing the cache comprises flushing the cache in a user-level without checking a memory region of the memory module.

12. A method of supporting persistence of a computing device including a processor and a memory module including a non-volatile memory, the method comprising:

performing a stop procedure upon power failure, wherein the step of performing the stop procedure comprises

scheduling out a running process task,

storing a state of the process task to a process control block of the memory module,

flushing a cache for the processor, and

flushing a pending memory request, wherein the step of scheduling out the running process task comprises replacing the running process task with an idle task; and

wherein the performing the stop procedure comprises clearing a kernel pointer of referring to the idle task.

13. The method of claim 12 , further comprising performing a go procedure when power is recovered,

wherein the performing the go procedure comprises scheduling process tasks based on the process control block after returning the kernel pointer to an idle task.