Coprocessor Synchronization via Acknowledgment Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing heterogeneous processor systems face inefficiencies in coordinating the operation of coprocessors, leading to suboptimal overall running efficiency due to the need for a control unit to wait for completion signals and the associated time overhead in synchronizing coprocessors.
Innovation Solution
A method where a first coprocessor acquires a relationship instruction, executes computing instructions, and sends acknowledgment information to a second coprocessor, allowing the coprocessors to self-coordinate their running times based on the acknowledgment, thereby improving efficiency without relying on the control unit to manage synchronization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a control unit manages synchronization between coprocessors by waiting for completion signals, then coordination between coprocessors is achieved, but the overall running efficiency deteriorates due to time overhead and control unit idle waiting
Solution Approach 1:
The patent implements self-service by enabling coprocessors to autonomously manage their own synchronization through relationship instructions embedded in their instruction streams. Each coprocessor independently monitors for acknowledgment signals from other coprocessors and automatically adjusts its execution flow without requiring the control unit to intervene or wait for completion signals. This eliminates the control unit's idle waiting time while maintaining proper coordination.
2Reliability
If the control unit waits for completion signals from coprocessors, then synchronization is maintained, but processing speed deteriorates due to control unit idle time
Solution Approach 1:
The patent applies preliminary action by embedding relationship instructions that contain acknowledgment signal identifiers into the coprocessor instruction streams before execution begins. This allows coprocessors to proactively prepare for synchronization events and immediately recognize when other coprocessors have completed their tasks, eliminating the need for the control unit to wait passively for completion signals and thereby increasing processing speed while maintaining synchronization.
3Device complexity
If coprocessors operate independently without self-coordination, then device complexity is reduced, but productivity deteriorates due to control unit intervention overhead
Solution Approach 1:
The patent implements feedback by having coprocessors monitor for acknowledgment signals generated by other coprocessors during execution. Each coprocessor's relationship instructions specify which acknowledgment signals to wait for, creating a feedback loop where completion information automatically flows between coprocessors. This decentralized feedback mechanism eliminates the need for complex control unit intervention while improving productivity through parallel autonomous operation.
Data Source
AI summary
A method for data processing, a processor chip. The method includes: acquiring a first relationship instruction; executing at least one first computing instruction acquired before the first relationship instruction based on the first relationship instruction; and sending acknowledgment information based on the first relationship instruction in response to completing executing the at least one first computing instruction, to cause a second coprocessor receiving the acknowledgment information to revert to a state of acquiring a second computing instruction after the second relationship instruction acquired by a second coprocessor based on the acknowledgment information.