10 results about "Spin lock" patented technology

The application discloses a kind of hardware spinlock logic interval check devices supporting multiprocessor programming, including interface control module and logic control module, interface control module includes bus interface, n input message queue, n return message queue and register control module, logic control module includes spinlock control module, logic interval check control module, logic interval information storage array and spin queue.N input message queue is connected with external n microprocessor respectively, for receiving and storing input message from microprocessor, spinlock control module is connected with n input message queue, register control module, logic interval check control module, spin queue, n return message queue, register control module is connected with bus interface.The application reduces the software overhead in solid state disk controller, including reducing the node queue, inter-core interaction queue in software maintenance on-chip cache resource occupation, improve the efficiency of interval check.

The invention provides a multi-core platform spin lock management system based on priority scheduling and a control method thereof, a spin lock is provided with a linked list corresponding to the spin lock, and corresponding tasks run by the second core are added to different positions of the linked list according to the priority of the tasks run by the second core which initiates a use request for resources. Setting the current state of the task as a waiting state; after the first core releases the spin lock, sequentially selecting the first second core from the linked list, adjusting the current state of the running task of the first core to be an awakening state, confirming that the current second core has the spin lock corresponding to the resource at the moment, and further confirming that the current second core obtains the right to use the resource; no matter how many CPU processing cores are managed in the operating system, the cores with high-priority task running requirements can preferentially acquire the spin lock, so that the tasks running on the cores can preferentially use corresponding resources, and the real-time performance and the certainty during scheduling of the operating system are improved.

The utility model relates to a maintenance platform for an electric stacking machine, and belongs to the technical field of maintenance of electric stacking machines. Comprising a bottom frame and a bottom plate connected with the bottom frame, a first front stand column, a second front stand column, a second rear stand column and a first rear stand column are arranged at the four corners of the bottom frame respectively, a left guardrail is arranged between the first front stand column and the first rear stand column, and a right guardrail is arranged between the second front stand column and the second rear stand column; a front guardrail is arranged between the first front stand column and the second front stand column, a rear guardrail is arranged between the first rear stand column and the second rear stand column, and the top of the first rear stand column and the top of the second rear stand column are each provided with a set of suspended ceiling corner fittings matched with the spin lock device. The utility model has the following beneficial effects: the electric stacking machine can be used for controlling the lifting of the overhauling platform, so that the overhauling range of the whole machine is expanded, the maintenance of the stacking machine is facilitated, and the risk of accidents in the overhauling process is reduced; maintenance personnel can work more intensively, and the maintenance quality and efficiency are improved.

The application provides a priority scheduling-based multi-core platform spin lock management system and a control method thereof. The spin lock is provided with a linked list corresponding to itself. According to the priority of a second core that initiates a use request for a resource, the corresponding task of the second core is added to different positions of the linked list, and the current state of the task is set to a waiting state. After a first core releases the spin lock, the second core with the highest order in the linked list is selected in sequence, and the current state of the running task of the second core is adjusted to a wake-up state. At this time, it is confirmed that the current second core has the spin lock corresponding to the resource, and then it is confirmed that the current second core obtains the use right of the resource. No matter how many CPU processing cores are managed in the operating system, the core with a high-priority task running demand can obtain the spin lock in priority, and then the running task thereon can use the corresponding resource in priority, thereby improving the real-time performance and certainty of the operating system scheduling.

This invention discloses a lock-aware bounded delay preemption method and a computer-readable medium for an electronic device. Between the guest operating system and the virtual machine monitor, a shared memory control block is established for each dynamically scheduled virtual CPU, containing a lock holding count variable, a delay preemption timestamp variable, and a maximum delay threshold. The guest operating system modifies the spinlock release mechanism and initiates a virtual CPU scheduling request based on the aforementioned variables. The virtual machine monitor executes scheduling by combining the variables and the maximum delay threshold. This invention achieves precise cross-layer synchronization of virtual CPU lock holding states, allowing the virtual machine monitor to directly perceive the spinlock holding state and avoid blind preemption. Its bounded delay mechanism constrains the preemption duration without breaking the original scheduling semantics and resource allocation rules. It requires no additional hardware support or framework reconstruction, preserves the system resource scheduling logic, ensures resource utilization and stability in hybrid scheduling scenarios, and adapts to the real-time and non-real-time service deployment requirements of multi-core virtualization platforms.

The utility model discloses a child restraint system capable of being positioned on a vehicle seat. The child restraint system comprises a supporting base; the child seat is connected to the supporting base, and the child seat can rotate around the axis relative to the supporting base; the spin lock is used for selectively locking rotation of the child seat relative to the supporting base; and an indicator associated with the spin lock, the indicator being movable in response to movement of the spin lock between a locked position and an unlocked position to visually indicate to a user when the spin lock is in the locked position and when the spin lock is in the unlocked position.

The invention provides a mutual exclusion synchronization mechanism optimization method and device based on a FreeRTOS SMP, equipment and a medium, and relates to the field of FreeRTOS kernel design, and the method comprises the steps: in a FreeRTOS SMP environment, integrating an embedded lock in each kernel object; the embedded lock is a special spin lock for each kernel object; the method comprises the steps of determining a first kernel object to be accessed based on a global task, obtaining an embedded lock of the first kernel object to be accessed through a first core, disabling local interruption of the first core, and accessing the first kernel object to be accessed through the first core; the embedded lock of the first kernel object to be accessed is a first embedded lock; and after the access is completed, the first core is allowed to be locally interrupted, and the first embedded lock is released. According to the scheme, the parallel execution efficiency and the real-time response capability of the multi-core system are improved.

The invention provides a spin lock implementation method and device based on hardware semaphore, equipment and a medium, and relates to the field of embedded operating systems, the method comprises the following steps: binding a spin lock with the hardware semaphore, and allocating a counting register to the spin lock; if the first core requests to acquire the spin lock, verifying whether the current holder of the hardware semaphore is the first core; if yes, reading a count value for validity check, and if valid, adding 1 to the count value and returning; if the core is not the first core, hardware semaphore locking is called; after the hardware semaphore is obtained, whether a current holder is a first core is verified, if yes, whether a count value is 0 is verified, and if yes, 1 is added to the count value, and a memory barrier instruction is inserted; if the first core requests to release the spin lock, verifying whether the first core is a holder of the spin lock; if yes, subtracting 1 from the count value, judging whether the count value is 0 or not, and if yes, calling hardware semaphore for unlocking to complete lock release. The scheme can reduce system power consumption.

This invention discloses a method for dynamic resource scheduling in collaboration between virtualization and operating systems, belonging to the field of collaborative resource scheduling technology. This invention achieves real-time communication between the virtualization layer and the operating system by constructing a bidirectional communication channel based on shared memory and event-driven mechanisms. Combined with a three-level dynamic mapping mechanism, it accurately maps operating system process priorities to virtualization layer vCPU scheduling weights. This, along with refined collaborative scheduling of CPU and I / O resources and load peak / valley prediction, effectively avoids resource allocation misalignment issues, significantly improves physical resource utilization, and is suitable for high-density deployment scenarios with multiple virtual machines. Furthermore, targeted designs such as spinlock-driven forced synchronization scheduling, dynamic adjustment of vCPU time slices, I / O priority pass-through, and queue weight allocation reduce response and processing latency for critical services, ensuring service quality.

The utility model discloses a container spreader spin lock maintenance tool and relates to the technical field of container spreader spin lock maintenance, the container spreader spin lock maintenance tool comprises a jacking rod, a locking clamp and a detection scaleplate, the jacking rod comprises a pressing rod, one end of the pressing rod is provided with a supporting leg and a supporting head, the supporting leg is arranged on one side of the pressing rod in a protruding mode, and the supporting head is arranged on the other side of the pressing rod. The supporting head is located on the other side of the pressing rod, and the included angle between the extending direction and the length direction of the pressing rod is an obtuse angle. The locking clamp comprises a clamp head, and the clamp head can be supported between an upper mounting plate of the lifting appliance spin lock and a lifting pin collision block of the lifting pin mechanism; the detection scale comprises a detection block, and the thickness of the detection block is not larger than the sensing distance of an unlocking and locking sensor on the lifting appliance spin lock. The lifting pin rod solves the technical problem that in the prior art, time and labor are wasted in the lifting process of the lifting pin rod, meanwhile, the lifting pin rod can be conveniently maintained in the jacked state, and potential safety hazards are avoided.