Processor Arbitration Circuit for Time-Sensitive Multi-Request Grants
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional arbitration mechanisms in processor-based devices result in inefficient processing of time-sensitive requests due to the need for multiple arbitration grants, leading to longer lifetimes and slower processing of requests, especially when requests are linked.
Innovation Solution
Implementing a multi-request arbitration grant policy where the arbitration circuit determines the availability of resource elements and issues a single arbitration grant for multiple linked requests when sufficient resources are available, while maintaining a starvation prevention mechanism to ensure fair access for single-request tracker entries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-request arbitration grant policy is used where each request tracker entry has equal priority and each arbitration grant is issued for one request, then the arbitration circuit design is simple, but the processing time for linked requests increases significantly
Solution Approach 1:
The arbitration circuit dynamically adjusts the arbitration grant policy based on the number of linked requests associated with each request tracker entry. When N linked requests are detected, the circuit issues a single arbitration grant that covers all N requests simultaneously, rather than requiring N separate grants. This dynamic adaptation reduces the time required to process linked requests while maintaining simple circuit design.
2Productivity
If a parked arbiter mechanism is used to detect linked commands and park the pointer at the current request tracker entry, then back-to-back arbitration wins are achieved for linked requests, but the design complexity increases and different arbitration grant policies are required
Solution Approach 1:
The arbitration circuit changes the arbitration grant policy parameter based on the detected number of linked requests. Instead of using a fixed single-request policy or a complex parked arbiter mechanism, the circuit simply adjusts its behavior to issue single grants that cover multiple requests when N linked requests are detected. This parameter-based approach achieves high productivity without increasing design complexity.
3Adaptability or versatility
If multiple linked requests wait for separate arbitration grants, then each request can be processed independently, but the total processing time increases and time-sensitive requests suffer from longer lifetimes
Solution Approach 1:
The arbitration circuit merges multiple linked requests into a single arbitration grant when N requests are detected. Instead of processing requests separately and waiting for individual grants, the circuit combines them into one grant that accommodates all N linked requests simultaneously. This merging approach reduces the total processing time and shortens the lifetime of request tracker entries while maintaining the flexibility to handle different request types.
Data Source
AI summary
Providing multi-request arbitration grant policies for time-sensitive arbitration decisions in processor-based devices is disclosed. In this regard, a processor-based device provides an arbitration circuit that is configured to select a request tracker entry of a plurality of request tracker entries of a request tracker circuit to apply a multi-request arbitration grant policy. The arbitration circuit determines a count N of a plurality of requests associated with the request tracker entry, and determines a count R of resource elements that are available of a plurality of resource elements of an arbitrated resource. The arbitration circuit determines whether the count R of resource elements that are available is equal to or greater than N, and if so, issues a single arbitration grant for the plurality of requests associated with the request tracker entry to the request tracker circuit.