Trace Data Priority Selection for Integrated Circuit Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing systems face challenges with large volumes of trace data, leading to insufficient resources for buffering and transmission, resulting in discarded critical diagnostic information or device stalling, which can mask or introduce faults during diagnostic investigations.
Innovation Solution
Implementing a priority-based system where trace data is dynamically assigned a priority level, allowing critical data to be guaranteed acceptance while less critical data is discarded if resources are unavailable, thereby preventing impediments to processing and maintaining diagnostic integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If trace data is buffered in the device, then diagnostic information is preserved, but buffer resources are insufficient for large volumes of trace data
Solution Approach 1:
The invention segments trace data into different priority levels (first priority for critical diagnostic data, second priority for less critical data). This segmentation allows the system to differentiate between data that must be preserved and data that can be discarded when buffer resources are insufficient, thereby resolving the contradiction between preserving all trace information and limited buffer capacity.
Solution Approach 2:
The invention applies local quality by assigning different quality attributes (priority levels) to different portions of trace data based on their diagnostic importance. Critical data receives first priority with guaranteed buffer allocation, while less critical data receives second priority and is discarded when resources are constrained, optimizing the use of limited buffer resources.
2Loss of information
If trace data is transmitted out of the device, then diagnostic information is available for analysis, but transmission bandwidth is insufficient for large volumes of trace data
Solution Approach 1:
The invention segments trace data transmission by priority level. First priority data is transmitted with guaranteed bandwidth allocation, while second priority data transmission is discarded when bandwidth is insufficient. This segmentation ensures that critical diagnostic information is always transmitted successfully despite limited bandwidth resources.
Solution Approach 2:
The invention applies local quality by differentiating transmission quality based on data priority. Critical trace data receives high-priority transmission treatment with reserved bandwidth, while less critical data receives lower-priority treatment and is discarded when transmission resources are constrained, optimizing bandwidth utilization for diagnostic purposes.
3Loss of information
If the device is stalled to preserve trace data, then data is not lost, but device operation timing is altered which may mask or introduce faults
Solution Approach 1:
The invention segments trace data into priority levels, allowing first priority critical data to be preserved through selective buffering without stalling the entire device. Second priority data is discarded when resources are insufficient. This segmentation enables data preservation without the need for device stalling, thereby maintaining fault detection accuracy.
Solution Approach 2:
The invention applies local quality by assigning different handling qualities to different data priorities. First priority data receives guaranteed acceptance and buffering without affecting device timing, while second priority data is discarded when resources are constrained. This differentiated approach preserves critical diagnostic information without altering device operation timing or masking faults.
4Loss of information
If all trace data is accepted without priority differentiation, then no data is discarded, but buffer resources are quickly exhausted and processing is impeded
Solution Approach 1:
The invention segments trace data into first priority (critical) and second priority (less critical) categories. First priority data is accepted with guaranteed buffer allocation and processing, while second priority data is discarded when buffer resources are insufficient. This segmentation prevents buffer exhaustion from non-critical data while ensuring critical diagnostic information is preserved and processed.
Solution Approach 2:
The invention applies local quality by assigning different acceptance qualities to different data priorities. First priority data receives guaranteed acceptance and processing resources, while second priority data receives best-effort handling and is discarded when resources are constrained. This differentiated quality approach optimizes both data preservation and processing throughput by focusing resources on critical information.
Data Source
AI summary
An integrated circuit 2 is provided with a data source 6 in the form of a processor executing program instructions connected via a bus interconnect 16 to a trace output device 8. The trace output device 8 is memory mapped. Different memory addresses that are mapped to the trace output device 8 are associated with different priority levels. Trace data written to at least one memory address has a first level of priority in which it is either accepted or the transfer is stalled until the data can be processed by the trace output device 8. Another level of priority associated with a different memory address is such that the data is always accepted but is discarded if the trace output device 8 does not have the ability to process, e.g. store that data at that time.