Benchmarking a DSP processor
This Master thesis describes the benchmarking of a DSP processor. Benchmarking means measuring the performance in some way. In this report, we have focused on the number of instruction cycles needed to execute certain algorithms. The algorithms we have used in the benchmark are all very common in signal processing today. The results we have reached in this thesis have been compared to benchmarks for other processors, performed by Berkeley Design Technology, Inc. The algorithms were programmed in assembly code and then executed on the instruction set simulator. After that, we proposed changes to the instruction set, with the aim to reduce the execution time for the algorithms. The results from the benchmark show that our processor is at the same level as the ones tested by BDTI. Probably would a more experienced programmer be able to reduce the cycle count even more, especially for some of the more complex benchmarks.
Summary
This master's thesis presents a cycle-accurate benchmarking study of a DSP processor using common signal-processing kernels implemented in assembly. It shows how instruction-cycle measurements reveal bottlenecks, compares results to industry BDTI benchmarks, and evaluates proposed ISA changes to speed up real-time DSP algorithms.
Key Takeaways
- Measure instruction-cycle counts for common DSP kernels (FFT, filters, adaptive routines) to quantify processor performance.
- Implement critical algorithms in assembly to expose pipeline and instruction-set inefficiencies.
- Compare measured results against published BDTI benchmarks to position processor performance.
- Propose and evaluate ISA modifications to reduce cycle counts and improve real-time throughput.
- Estimate trade-offs between instruction-set complexity and cycle-level performance for fixed-point DSP workloads.
Who Should Read This
Advanced DSP engineers, embedded architects, and graduate students who implement or optimize DSP algorithms and want to evaluate or improve processor instruction-set and cycle performance.
Still RelevantAdvanced
Related Documents
- A New Approach to Linear Filtering and Prediction Problems TimelessAdvanced
- A Quadrature Signals Tutorial: Complex, But Not Complicated TimelessIntermediate
- Lecture Notes on Elliptic Filter Design TimelessAdvanced
- Computing FFT Twiddle Factors TimelessAdvanced
- Digital Envelope Detection: The Good, the Bad, and the Ugly TimelessIntermediate
