Dynamic Trace Aggregation with Platform Description Identifiers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Dynamic systems with non-static topologies in the semiconductor and computing industry face challenges in tracing and debugging due to missing information about trace sources, leading to undecodable traces and increased effort in identifying and triaging issues, which hampers product releases.

Innovation Solution

A debug system that uses platform description identifiers (PDIDs) to dynamically describe complex systems, allowing for dynamic changes such as component addition or removal, and provides a hierarchical decoder structure to decode traces effectively, reducing the risk of unusable data and improving trace analysis tools.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a static assignment of trace sources and protocols is used, then the decoding process is simple and reliable, but the system cannot adapt to dynamic topologies where components are added or removed

Engineering Contradiction:
Improveadaptability to dynamic system topologyVSAvoidcomplexity of trace decoding system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic trace source identification by replacing static protocol assignments with runtime-determined identification. The system now dynamically assigns trace sources to protocols based on actual system state, allowing adaptation to changing topologies while maintaining decoding reliability through structured identification mechanisms.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The trace system performs self-identification by embedding source identification information within the trace data stream itself. This self-service mechanism allows the trace system to automatically identify and decode its own sources without external intervention, adapting dynamically to system changes while maintaining autonomous operation.

Inventive Principle:
Principle #25Self-service

2Reliability

If no source identification information is provided in the trace stream, then the trace data transmission is efficient with lower bandwidth requirements, but the receiving tool cannot decode the traces without a priori knowledge of the system

Engineering Contradiction:
Improvereliability of trace decodingVSAvoidamount of data transmitted in trace stream
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts source identification information from the trace data stream in a structured manner, separating identification metadata from the actual trace content. This extraction approach enables reliable decoding by providing necessary identification information while minimizing the overhead added to the trace stream.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The identification mechanism serves multiple functions simultaneously: it identifies trace sources, determines protocol types, and enables dynamic adaptation to system topology changes. This multi-functional approach ensures reliable decoding without requiring separate identification systems, thereby limiting additional data transmission requirements.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If trace sources are dynamically added or removed in the system, then the system flexibility is improved, but the receiving tool cannot effectively leverage available tracing systems due to missing origin information

Engineering Contradiction:
Improvesystem flexibility for dynamic component changesVSAvoidloss of trace origin information
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system performs preliminary registration and identification of trace sources when they are first introduced, creating a baseline knowledge base before dynamic changes occur. This preliminary action ensures that when components are dynamically added or removed, the receiving tool already has the necessary identification information to decode traces from new sources while maintaining awareness of system topology changes.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11513940B2System, apparatus and method for dynamic tracing in a system
Publication Date: 2022.11.29 INTEL CORP
  • US11513940B2 patent drawing
  • US11513940B2 patent drawing
  • US11513940B2 patent drawing

AI summary

In one embodiment, an apparatus includes: a first trace source to generate a plurality of first trace messages and a first local platform description identifier to identify the first trace source; a second trace source to generate a plurality of second trace messages and a second local platform description identifier to identify the second trace source; and a trace aggregator coupled to the first and the second trace sources, the trace aggregator to generate a global platform description identifier for the apparatus and output a trace stream including the global platform destination identifier, the first and second local platform description identifiers, the plurality of first trace messages and the plurality of second trace messages. Other embodiments are described and claimed.