Kernel Restarting via Busy-Wait Cores to Cut Reboot Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing kernel reboot processes, particularly in systems using kexec, are prolonged by the time-consuming phases of powering off and on physical cores during kernel version updates, negatively impacting user experience.
Innovation Solution
A method involving setting physical cores to a busy waiting state where they query for entry address modifications, allowing them to exit this state and perform initialization upon kernel boot, thereby skipping the power-off and power-on phases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the physical core is powered off and then powered on during kernel version update, then the kernel can be restarted, but the reboot time becomes excessively long
Solution Approach 1:
The physical core enters a busy waiting state before the new kernel is fully loaded, maintaining power while waiting for the entry address to be updated. This preliminary action eliminates the need for power cycling, as the core is already prepared and waiting in a low-power but active state, thus reducing reboot time while ensuring reliable kernel transition
Solution Approach 2:
The patent skips the traditional power-off and power-on phases by using a busy waiting state where the physical core remains powered on but does not execute instructions. This allows the system to rush through the kernel transition by only updating the entry address and having the core resume execution, significantly reducing the time loss associated with full power cycles
2Loss of time
If the physical core enters busy waiting state and queries entry address modifications, then the reboot time is reduced, but the system complexity increases
Solution Approach 1:
The physical core performs self-service by autonomously querying whether its entry address has been modified during the busy waiting state. This self-monitoring capability eliminates the need for external control logic to manage the transition, as the core itself detects when it should exit the busy waiting state and resume execution, thereby reducing overall system complexity despite the added waiting mechanism
Solution Approach 2:
The system implements a feedback mechanism where the physical core continuously queries the entry address modification status. This feedback loop allows the core to automatically detect when the new kernel is ready and transition out of the busy waiting state, providing a simple yet effective control mechanism that reduces reboot time without requiring complex external management
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application provides a kernel reboot method, to reserve an internal memory for storing an entry address of a physical code and a code segment for the physical core. A kernel hotswap scenario is used as an example. When a computer system stops executing an old kernel, an interrupt is sent to the physical core, so that the physical core enters a busy waiting state. When the computer system starts a new kernel, the entry address of the physical core is modified. When finding that the entry address is modified, the physical core in the busy waiting state exits the busy waiting state, performs an initialization procedure of the physical core, and receives task scheduling of the new kernel. In an entire process, all physical cores are in a power-on state. Compared with a conventional technology in which a physical core needs to be powered off and then powered on, according to the method provided in embodiments of this application, a time required for rebooting the kernel can be reduced, and user experience of the computer system user can be improved.