SPRADM6 December   2024 AM62D-Q1

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
    1. 1.1 Load Binaries to AM62D
  5. 2Processor Core Benchmarks
    1. 2.1 C7x DSP Benchmark
      1. 2.1.1 Fast Fourier Transform
      2. 2.1.2 Digital Signal Processing
        1. 2.1.2.1 FIR
        2. 2.1.2.2 Cascade Biquad
        3. 2.1.2.3 Dot Product
      3. 2.1.3 Mathematical Operations
    2. 2.2 Dhrystone on A53 cores
  6. 3Memory System Benchmarks
    1. 3.1 Critical Memory Access Latency
    2. 3.2 UDMA: DDR to DDR Data Copy
    3. 3.3 C7x DRU Performance: Block Copy with DMA
  7. 4Application Specific Benchmarks
    1. 4.1 SBL Boot Time
    2. 4.2 IPC Performance
    3. 4.3 Flash
    4. 4.4 Application Specific Latency
  8. 5Summary
  9. 6References

2.1.2.3 Dot Product

The DSPLIB_dotprod kernel calculates the dot products of two vectors. The kernel supports data types including int8, uint8, int16, uint16, int32, uint32, float and double. Table 2-5 shows performance results of executing the dot product kernel on C7x DSP on AM62D for various data types. While the kernel supports any input vector length, results of only 1024 vectors are reported in this document. Performance of other sizes is listed in DSPLIB's user guide.

Table 2-5 Dot Product Performance on C7x DSP

Data Type

Data Size

EVM Cycles

Cycle/Sample

Int8

1024

131

0.13

Int16

1024

132

0.13

Int16

16384

1112

0.07

Int

1024

199

0.19

Float

1024

229

0.22

Float

32768

4211

0.13

Double

1024

352

0.34