Flexible Bit Field Search in Hardware Section Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hardware section filters are limited by byte boundaries, requiring multiple runs to filter data patterns spanning multiple bytes or packets and applying multiple rules, leading to inefficiencies in modern applications like digital television receivers.
Innovation Solution
A flexible section filter apparatus and method that allows filtering to start and end at any bit within a 128-bit field, enabling range filtering and multiple rule applications through linking multiple section filters, eliminating the need for multiple runs and reducing CPU involvement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the section filter checks for matches byte by byte starting at the beginning of each byte, then the filtering process is simple to implement, but the filter is limited by byte boundaries and cannot efficiently handle data patterns that span multiple bytes or packets
Solution Approach 1:
The 128-bit section filter is divided into 16 independent 8-bit byte lanes, each capable of parallel processing. This segmentation allows the filter to maintain simplicity at the byte level while achieving flexibility through bitwise operations across all bytes simultaneously, resolving the contradiction between adaptability and implementation complexity.
Solution Approach 2:
The patent introduces a bit-level dimension of operation within each byte lane by using bitwise shift and mask operations. This allows the filter to operate on arbitrary bit fields within bytes without changing the underlying byte-oriented architecture, achieving flexibility while maintaining simple byte-by-byte processing logic.
2Adaptability or versatility
If the section filter processes data spanning multiple bytes or packets, then the filter can handle diverse data patterns, but multiple runs of the filtering process are required which reduces efficiency
Solution Approach 1:
The patent combines multiple filtering operations into a single pass by merging the results of bitwise operations across all 16 byte lanes. The accumulator combines match results from each byte, and the final match determination merges these results to identify patterns spanning multiple bytes, achieving both adaptability and efficiency in one filtering run.
Solution Approach 2:
The patent performs preliminary bitwise operations (shifts and masks) on each byte lane before the final match determination. This preliminary processing prepares the data in advance so that the final comparison can efficiently determine whether a pattern matches across byte boundaries without requiring multiple separate filtering passes.
3Reliability
If multiple section filter descriptor rules are applied, then the filter can enforce multiple filtering criteria, but the entire process must be run multiple times increasing processing time
Solution Approach 1:
The patent enables continuous filtering by maintaining the accumulated match results across multiple descriptor rule evaluations. Instead of restarting the filtering process for each rule, the system continuously updates the match status based on new rules while preserving previous results, thereby enforcing multiple criteria without requiring separate filtering runs and reducing processing time.
4Adaptability or versatility
If the filter requires CPU involvement to run the entire filtering process multiple times, then complex filtering scenarios can be handled, but CPU resources are consumed and processing speed decreases
Solution Approach 1:
The patent implements self-service filtering by enabling the hardware section filter to autonomously handle complex filtering scenarios through its accumulated match results and bitwise operation capabilities. The filter independently processes patterns spanning multiple bytes and evaluates multiple descriptor rules without requiring CPU intervention to restart or reconfigure the filtering process, thereby maintaining adaptability while reducing CPU involvement and increasing automation.
Data Source
AI summary
Systems, methods, and other embodiments associated with flexible bit field search are described. According to one embodiment, an apparatus includes a filter configured to receive data packets and a descriptor. The descriptor includes a header and at least one filter descriptor rule. The header identifies a filtering mode. The at least one filter descriptor rule includes instructions that identify a filtering operation in the filtering mode. The filter is also configured to filter the data packets based, at least in part, on the filtering operation identified in the at least one filter descriptor rule.