18386880. Multi-Threaded, Self-Scheduling Processor simplified abstract (Micron Technology, Inc.)
Multi-Threaded, Self-Scheduling Processor
Organization Name
Inventor(s)
Tony M. Brewer of Plano TX (US)
Multi-Threaded, Self-Scheduling Processor - A simplified explanation of the abstract
This abstract first appeared for US patent application 18386880 titled 'Multi-Threaded, Self-Scheduling Processor
Simplified Explanation
The abstract describes a self-scheduling processor that can execute instructions and provide additional functionality, such as scheduling instructions automatically in response to received work descriptor data packets.
- Processor core executes received instructions
- Core control circuit automatically schedules instructions for execution by processor core based on received work descriptor data packets
- Core control circuit schedules fiber create instruction, reserves memory space for return arguments, and generates work descriptor data packets for execution of multiple threads
- Event processing, data path management, system calls, memory requests, and new instructions are also disclosed
---
- Potential Applications
- High-performance computing - Multi-threaded applications - Real-time processing systems
- Problems Solved
- Efficient scheduling of instructions for execution - Improved memory management - Enhanced performance in multi-threaded environments
- Benefits
- Increased processing efficiency - Optimized resource utilization - Enhanced system performance and responsiveness
Original Abstract Submitted
Representative apparatus, method, and system embodiments are disclosed for a self-scheduling processor which also provides additional functionality. Representative embodiments include a self-scheduling processor, comprising: a processor core adapted to execute a received instruction; and a core control circuit adapted to automatically schedule an instruction for execution by the processor core in response to a received work descriptor data packet. In another embodiment, the core control circuit is also adapted to schedule a fiber create instruction for execution by the processor core, to reserve a predetermined amount of memory space in a thread control memory to store return arguments, and to generate one or more work descriptor data packets to another processor or hybrid threading fabric circuit for execution of a corresponding plurality of execution threads. Event processing, data path management, system calls, memory requests, and other new instructions are also disclosed.
