2.5.2. Linux 11.02.08.02 Performance Guide

Read This First

All performance numbers provided in this document are gathered using following Evaluation Modules unless otherwise specified.

Name

Description

AM62Lx EVM

AM62Lx Evaluation Module rev E1-1 with ARM running at 1.4GHz, DDR data rate 1600 MT/S

Table: Evaluation Modules

About This Manual

This document provides performance data for each of the device drivers which are part of the Processor SDK Linux package. This document should be used in conjunction with release notes and user guides provided with the Processor SDK Linux package for information on specific issues present with drivers included in a particular release.

If You Need Assistance

For further information or to report any problems, contact https://e2e.ti.com/ or https://support.ti.com/

2.5.2.1. System Benchmarks

2.5.2.1.1. LMBench

LMBench is a collection of microbenchmarks of which the memory bandwidth and latency related ones are typically used to estimate processor memory system performance. More information about lmbench at https://lmbench.sourceforge.net/whatis_lmbench.html and https://lmbench.sourceforge.net/man/lmbench.8.html

Latency: lat_mem_rd-stride128-szN, where N is equal to or smaller than the cache size at given level measures the cache miss penalty. N that is at least double the size of last level cache is the latency to external memory.

Bandwidth: bw_mem_bcopy-N, where N is equal to or smaller than the cache size at a given level measures the achievable memory bandwidth from software doing a memcpy() type operation. Typical use is for external memory bandwidth calculation. The bandwidth is calculated as byte read and written counts as 1 which should be roughly half of STREAM copy result.

Execute the LMBench with the following:

cd /opt/ltp
./runltp -P j721e-idk-gw -f ddt/lmbench -s LMBENCH_L_PERF_0001
Table 2.3 LMBench Benchmarks

Benchmarks

am62lxx_evm-fs: perf

af_unix_sock_stream_latency (microsec)

33.14 (min 32.70, max 34.50)

af_unix_socket_stream_bandwidth (mbs)

458.31 (min 416.22, max 548.49)

bw_file_rd-io-1mb (mb/s)

971.18 (min 906.62, max 1128.85)

bw_file_rd-o2c-1mb (mb/s)

515.41 (min 508.13, max 525.95)

bw_mem-bcopy-16mb (mb/s)

742.11 (min 704.91, max 849.57)

bw_mem-bcopy-1mb (mb/s)

755.61 (min 687.76, max 942.68)

bw_mem-bcopy-2mb (mb/s)

748.93 (min 681.43, max 958.47)

bw_mem-bcopy-4mb (mb/s)

781.73 (min 727.40, max 944.73)

bw_mem-bcopy-8mb (mb/s)

801.94 (min 738.69, max 994.53)

bw_mem-bzero-16mb (mb/s)

2250.16 (min 2104.16, max 2805.54)

bw_mem-bzero-1mb (mb/s)

1502.54 (min 687.76, max 2800.63)

bw_mem-bzero-2mb (mb/s)

1499.05 (min 681.43, max 2807.02)

bw_mem-bzero-4mb (mb/s)

1516.48 (min 727.40, max 2806.03)

bw_mem-bzero-8mb (mb/s)

1526.92 (min 738.69, max 2807.02)

bw_mem-cp-16mb (mb/s)

422.87 (min 397.28, max 517.95)

bw_mem-cp-1mb (mb/s)

1450.86 (min 401.50, max 3062.20)

bw_mem-cp-2mb (mb/s)

1387.71 (min 399.24, max 2898.03)

bw_mem-cp-4mb (mb/s)

1383.65 (min 434.12, max 2868.93)

bw_mem-cp-8mb (mb/s)

1376.95 (min 436.56, max 2827.85)

bw_mem-fcp-16mb (mb/s)

751.75 (min 710.13, max 879.75)

bw_mem-fcp-1mb (mb/s)

1538.99 (min 777.24, max 2800.63)

bw_mem-fcp-2mb (mb/s)

1534.98 (min 762.92, max 2807.02)

bw_mem-fcp-4mb (mb/s)

1547.58 (min 788.72, max 2806.03)

bw_mem-fcp-8mb (mb/s)

1556.02 (min 804.67, max 2807.02)

bw_mem-frd-16mb (mb/s)

1328.40 (min 1227.75, max 1609.50)

bw_mem-frd-1mb (mb/s)

1057.89 (min 777.24, max 1525.94)

bw_mem-frd-2mb (mb/s)

1063.22 (min 762.92, max 1550.87)

bw_mem-frd-4mb (mb/s)

1080.32 (min 788.72, max 1555.81)

bw_mem-frd-8mb (mb/s)

1093.89 (min 804.67, max 1611.60)

bw_mem-fwr-16mb (mb/s)

2270.76 (min 2123.71, max 2819.38)

bw_mem-fwr-1mb (mb/s)

1876.87 (min 1218.45, max 3062.20)

bw_mem-fwr-2mb (mb/s)

1828.62 (min 1203.61, max 2898.03)

bw_mem-fwr-4mb (mb/s)

1809.91 (min 1247.47, max 2868.93)

bw_mem-fwr-8mb (mb/s)

1802.83 (min 1250.00, max 2827.85)

bw_mem-rd-16mb (mb/s)

1345.30 (min 1264.92, max 1616.65)

bw_mem-rd-1mb (mb/s)

953.97 (min 525.76, max 1599.43)

bw_mem-rd-2mb (mb/s)

940.68 (min 511.31, max 1592.61)

bw_mem-rd-4mb (mb/s)

978.01 (min 513.68, max 1601.71)

bw_mem-rd-8mb (mb/s)

1032.26 (min 681.02, max 1605.14)

bw_mem-rdwr-16mb (mb/s)

732.31 (min 663.63, max 858.23)

bw_mem-rdwr-1mb (mb/s)

503.57 (min 401.50, max 680.27)

bw_mem-rdwr-2mb (mb/s)

494.06 (min 399.24, max 692.28)

bw_mem-rdwr-4mb (mb/s)

553.67 (min 434.12, max 778.21)

bw_mem-rdwr-8mb (mb/s)

585.97 (min 436.56, max 857.72)

bw_mem-wr-16mb (mb/s)

763.43 (min 702.83, max 927.75)

bw_mem-wr-1mb (mb/s)

573.83 (min 525.76, max 686.34)

bw_mem-wr-2mb (mb/s)

559.79 (min 507.29, max 692.28)

bw_mem-wr-4mb (mb/s)

633.10 (min 513.68, max 800.24)

bw_mem-wr-8mb (mb/s)

708.78 (min 627.60, max 871.93)

bw_mmap_rd-mo-1mb (mb/s)

1245.57 (min 1156.29, max 1501.22)

bw_mmap_rd-o2c-1mb (mb/s)

509.91 (min 505.22, max 519.21)

bw_pipe (mb/s)

426.35 (min 382.91, max 518.60)

bw_unix (mb/s)

458.31 (min 416.22, max 548.49)

lat_connect (us)

72.81 (min 71.39, max 76.51)

lat_ctx-2-128k (us)

17.46 (min 11.52, max 25.12)

lat_ctx-2-256k (us)

31.42 (min 26.43, max 34.63)

lat_ctx-4-128k (us)

15.70 (min 12.41, max 17.36)

lat_ctx-4-256k (us)

8.13 (min 0.00, max 16.57)

lat_fs-0k (num_files)

198.17 (min 189.00, max 213.00)

lat_fs-10k (num_files)

84.33 (min 76.00, max 90.00)

lat_fs-1k (num_files)

127.00 (min 113.00, max 140.00)

lat_fs-4k (num_files)

128.17 (min 119.00, max 134.00)

lat_mem_rd-stride128-sz1000k (ns)

49.98 (min 41.71, max 52.46)

lat_mem_rd-stride128-sz125k (ns)

6.19 (min 6.16, max 6.22)

lat_mem_rd-stride128-sz250k (ns)

13.46 (min 6.99, max 21.17)

lat_mem_rd-stride128-sz31k (ns)

3.59 (min 2.42, max 4.45)

lat_mem_rd-stride128-sz50 (ns)

2.40 (min 2.40, max 2.41)

lat_mem_rd-stride128-sz500k (ns)

45.07 (min 38.07, max 48.16)

lat_mem_rd-stride128-sz62k (ns)

5.74 (min 5.72, max 5.75)

lat_mmap-1m (us)

59.83 (min 57.00, max 70.00)

lat_ops-double-add (ns)

3.21

lat_ops-double-div (ns)

17.65 (min 17.64, max 17.66)

lat_ops-double-mul (ns)

3.21 (min 3.21, max 3.22)

lat_ops-float-add (ns)

3.21

lat_ops-float-div (ns)

10.43 (min 10.42, max 10.43)

lat_ops-float-mul (ns)

3.21

lat_ops-int-add (ns)

0.80

lat_ops-int-bit (ns)

0.53

lat_ops-int-div (ns)

4.82 (min 4.81, max 4.82)

lat_ops-int-mod (ns)

5.08 (min 5.08, max 5.09)

lat_ops-int-mul (ns)

3.45 (min 3.44, max 3.46)

lat_ops-int64-add (ns)

0.80

lat_ops-int64-bit (ns)

0.53 (min 0.53, max 0.54)

lat_ops-int64-div (ns)

7.62

lat_ops-int64-mod (ns)

5.88 (min 5.88, max 5.89)

lat_ops-int64-mul (ns)

3.99 (min 3.98, max 4.04)

lat_pagefault (us)

0.93 (min 0.81, max 0.98)

lat_pipe (us)

26.36 (min 24.71, max 27.33)

lat_proc-exec (us)

1447.93 (min 1267.60, max 1563.25)

lat_proc-fork (us)

1269.68 (min 1094.20, max 1328.60)

lat_proc-proccall (us)

0.01

lat_select (us)

38.21 (min 38.12, max 38.41)

lat_sem (us)

4.12 (min 3.74, max 4.31)

lat_sig-catch (us)

6.22 (min 6.07, max 6.38)

lat_sig-install (us)

0.74 (min 0.72, max 0.78)

lat_sig-prot (us)

0.69 (min 0.49, max 0.88)

lat_syscall-fstat (us)

2.20 (min 2.12, max 2.31)

lat_syscall-null (us)

0.51 (min 0.51, max 0.52)

lat_syscall-open (us)

403.50 (min 362.00, max 459.33)

lat_syscall-read (us)

0.91 (min 0.90, max 0.91)

lat_syscall-stat (us)

5.41 (min 5.23, max 5.60)

lat_syscall-write (us)

0.86 (min 0.84, max 0.88)

lat_tcp (us)

1.02

lat_unix (us)

33.14 (min 32.70, max 34.50)

latency_for_0.50_mb_block_size (nanosec)

45.07 (min 38.07, max 48.16)

latency_for_1.00_mb_block_size (nanosec)

24.99 (min 0.00, max 52.46)

pipe_bandwidth (mbs)

426.35 (min 382.91, max 518.60)

pipe_latency (microsec)

26.36 (min 24.71, max 27.33)

procedure_call (microsec)

0.01

select_on_200_tcp_fds (microsec)

38.21 (min 38.12, max 38.41)

semaphore_latency (microsec)

4.12 (min 3.74, max 4.31)

signal_handler_latency (microsec)

0.74 (min 0.72, max 0.78)

signal_handler_overhead (microsec)

6.22 (min 6.07, max 6.38)

tcp_ip_connection_cost_to_localhost (microsec)

72.81 (min 71.39, max 76.51)

tcp_latency_using_localhost (microsec)

1.02

2.5.2.1.2. Dhrystone

Dhrystone is a core only benchmark that runs from warm L1 caches in all modern processors. It scales linearly with clock speed.

Please take note, different run may produce different slightly results. This is advised to run this test multiple times in order to get maximum performance numbers.

Execute the benchmark with the following:

runDhrystone
Table 2.4 Dhrystone Benchmarks

Benchmarks

am62lxx_evm-fs: perf

cpu_clock (mhz)

400.00

dhrystone_per_mhz (dmips/mhz)

8.90

dhrystone_per_second (dhrystonep)

6250000.00

2.5.2.1.3. Whetstone

Whetstone is a benchmark primarily measuring floating-point arithmetic performance.

Execute the benchmark with the following:

runWhetstone
Table 2.5 Whetstone Benchmarks

Benchmarks

am62lxx_evm-fs: perf

whetstone (mips)

5833.33 (min 5000.00, max 10000.00)

2.5.2.1.4. Linpack

Linpack measures peak double precision (64 bit) floating point performance in solving a dense linear system.

Table 2.6 Linpack Benchmarks

Benchmarks

am62lxx_evm-fs: perf

linpack (kflops)

515149.67 (min 514088.00, max 516563.00)

2.5.2.1.5. NBench

NBench which stands for Native Benchmark is used to measure macro benchmarks for commonly used operations such as sorting and analysis algorithms. More information about NBench at https://en.wikipedia.org/wiki/NBench and https://nbench.io/articles/index.html

Table 2.7 NBench Benchmarks

Benchmarks

am62lxx_evm-fs: perf

assignment (iterations)

12.95 (min 12.90, max 12.97)

fourier (iterations)

20382.00 (min 20379.00, max 20384.00)

fp_emulation (iterations)

192.40 (min 192.33, max 192.47)

huffman (iterations)

1057.02 (min 1056.80, max 1057.20)

idea (iterations)

3075.14 (min 3074.90, max 3075.50)

lu_decomposition (iterations)

473.43 (min 470.10, max 479.32)

neural_net (iterations)

7.73 (min 7.72, max 7.73)

numeric_sort (iterations)

558.36 (min 555.56, max 562.43)

string_sort (iterations)

146.38 (min 146.36, max 146.39)

2.5.2.1.6. Stream

STREAM is a microbenchmark for measuring data memory system performance without any data reuse. It is designed to miss on caches and exercise data prefetcher and speculative accesses. It uses double precision floating point (64bit) but in most modern processors the memory access will be the bottleneck. The four individual scores are copy, scale as in multiply by constant, add two numbers, and triad for multiply accumulate. For bandwidth, a byte read counts as one and a byte written counts as one, resulting in a score that is double the bandwidth LMBench will show.

Execute the benchmark with the following:

stream_c
Table 2.8 Stream Benchmarks

Benchmarks

am62lxx_evm-fs: perf

add (mb/s)

1651.55 (min 1501.10, max 1914.10)

copy (mb/s)

1735.32 (min 1565.10, max 2033.20)

scale (mb/s)

1830.72 (min 1641.00, max 2134.40)

triad (mb/s)

1632.43 (min 1491.50, max 1888.40)

2.5.2.1.7. CoreMarkPro

CoreMark®-Pro is a comprehensive, advanced processor benchmark that works with and enhances the market-proven industry-standard EEMBC CoreMark® benchmark. While CoreMark stresses the CPU pipeline, CoreMark-Pro tests the entire processor, adding comprehensive support for multicore technology, a combination of integer and floating-point workloads, and data sets for utilizing larger memory subsystems.

Table 2.9 CoreMarkPro Benchmarks

Benchmarks

am62lxx_evm-fs: perf

cjpeg-rose7-preset (workloads/)

37.22 (min 37.04, max 37.31)

core (workloads/)

0.27

coremark-pro ()

705.20 (min 686.04, max 730.09)

linear_alg-mid-100x100-sp (workloads/)

13.09 (min 13.06, max 13.11)

loops-all-mid-10k-sp (workloads/)

0.57 (min 0.56, max 0.60)

nnet_test (workloads/)

0.97

parser-125k (workloads/)

6.35 (min 5.62, max 6.99)

radix2-big-64k (workloads/)

19.56 (min 17.06, max 22.07)

sha-test (workloads/)

71.11 (min 69.93, max 72.46)

zip-test (workloads/)

18.99 (min 18.52, max 19.61)
Table 2.10 CoreMarkProTwoCore Benchmarks

Benchmarks

am62lxx_evm-fs: perf

cjpeg-rose7-preset (workloads/)

72.89 (min 71.94, max 74.07)

core (workloads/)

0.54 (min 0.53, max 0.54)

coremark-pro ()

1226.57 (min 1185.87, max 1277.14)

linear_alg-mid-100x100-sp (workloads/)

26.03 (min 25.97, max 26.11)

loops-all-mid-10k-sp (workloads/)

1.03 (min 1.00, max 1.07)

nnet_test (workloads/)

1.93 (min 1.93, max 1.94)

parser-125k (workloads/)

6.48 (min 5.81, max 7.33)

radix2-big-64k (workloads/)

28.39 (min 25.03, max 31.84)

sha-test (workloads/)

140.47 (min 138.89, max 142.86)

zip-test (workloads/)

34.17 (min 32.26, max 35.71)

2.5.2.1.8. MultiBench

MultiBench™ is a suite of benchmarks that allows processor and system designers to analyze, test, and improve multicore processors. It uses three forms of concurrency: Data decomposition: multiple threads cooperating on achieving a unified goal and demonstrating a processor’s support for fine grain parallelism. Processing multiple data streams: uses common code running over multiple threads and demonstrating how well a processor scales over scalable data inputs. Multiple workload processing: shows the scalability of general-purpose processing, demonstrating concurrency over both code and data. MultiBench combines a wide variety of application-specific workloads with the EEMBC Multi-Instance-Test Harness (MITH), compatible and portable with most any multicore processors and operating systems. MITH uses a thread-based API (POSIX-compliant) to establish a common programming model that communicates with the benchmark through an abstraction layer and provides a flexible interface to allow a wide variety of thread-enabled workloads to be tested.

Table 2.11 Multibench Benchmarks

Benchmarks

am62lxx_evm-fs: perf

4m-check (workloads/)

288.19 (min 268.24, max 314.90)

4m-check-reassembly (workloads/)

57.34 (min 52.49, max 63.94)

4m-check-reassembly-tcp (workloads/)

36.65 (min 34.01, max 40.39)

4m-check-reassembly-tcp-cmykw2-rotatew2 (workloads/)

16.34 (min 15.28, max 18.05)

4m-check-reassembly-tcp-x264w2 (workloads/)

0.93 (min 0.91, max 0.95)

4m-cmykw2 (workloads/)

109.44 (min 107.88, max 112.05)

4m-cmykw2-rotatew2 (workloads/)

21.20 (min 19.46, max 23.21)

4m-reassembly (workloads/)

52.91 (min 49.41, max 58.24)

4m-rotatew2 (workloads/)

21.26 (min 19.25, max 23.63)

4m-tcp-mixed (workloads/)

95.76 (min 93.02, max 98.16)

4m-x264w2 (workloads/)

0.94 (min 0.93, max 0.96)

empty-wld (workloads/)

1.00

idct-4m (workloads/)

16.97 (min 16.90, max 17.08)

idct-4mw1 (workloads/)

16.97 (min 16.89, max 17.09)

ippktcheck-4m (workloads/)

288.16 (min 267.84, max 314.11)

ippktcheck-4mw1 (workloads/)

287.54 (min 267.58, max 312.77)

ipres-4m (workloads/)

68.41 (min 62.50, max 76.69)

ipres-4mw1 (workloads/)

68.31 (min 62.47, max 76.18)

md5-4m (workloads/)

18.99 (min 17.91, max 20.68)

md5-4mw1 (workloads/)

19.11 (min 18.16, max 20.15)

rgbcmyk-4m (workloads/)

56.70 (min 56.45, max 57.08)

rgbcmyk-4mw1 (workloads/)

56.64 (min 56.32, max 57.01)

rotate-4ms1 (workloads/)

19.37 (min 17.85, max 20.73)

rotate-4ms1w1 (workloads/)

19.04 (min 17.85, max 20.74)

rotate-4ms64 (workloads/)

19.51 (min 18.04, max 22.62)

rotate-4ms64w1 (workloads/)

19.33 (min 18.01, max 21.44)

x264-4mq (workloads/)

0.50 (min 0.50, max 0.51)

x264-4mqw1 (workloads/)

0.50 (min 0.49, max 0.51)

2.5.2.2. Boot-time Measurement

2.5.2.2.1. Boot media: MMCSD

Table 2.12 Linux boot time MMCSD

Boot Configuration

am62lxx_evm-fs: Boot time in seconds: avg(min,max)

Linux boot time from SD with default rootfs (20 boot cycles)

19.01 (min 17.68, max 27.66)

Boot time numbers [avg, min, max] are measured from “Starting kernel” to Linux prompt across 20 boot cycles.


2.5.2.3. ALSA SoC Audio Driver

  1. Access type - RW_INTERLEAVED

  2. Channels - 2

  3. Format - S16_LE

  4. Period size - 64

Table 2.13 Audio Capture

Sampling Rate (Hz)

am62lxx_evm-fs: Throughput (bits/sec)

am62lxx_evm-fs: CPU Load (%)

11025

352795.83 (min 352795.00, max 352797.00)

0.38 (min 0.29, max 0.48)

16000

511995.00 (min 511994.00, max 511996.00)

0.35 (min 0.20, max 0.79)

22050

705589.67 (min 705586.00, max 705594.00)

0.40 (min 0.30, max 0.53)

24000

705591.50 (min 705589.00, max 705595.00)

0.47 (min 0.37, max 0.60)

32000

1023987.83 (min 1023985.00, max 1023991.00)

0.30 (min 0.21, max 0.40)

44100

1411182.17 (min 1411179.00, max 1411186.00)

0.59 (min 0.50, max 0.72)

48000

1535980.33 (min 1535976.00, max 1535985.00)

0.91 (min 0.23, max 1.56)

88200

2822355.00 (min 2822349.00, max 2822363.00)

1.09 (min 0.93, max 1.20)

96000

3071938.67 (min 3071934.00, max 3071943.00)

1.22 (min 0.39, max 3.19)
Table 2.14 Audio Playback

Sampling Rate (Hz)

am62lxx_evm-fs: Throughput (bits/sec)

am62lxx_evm-fs: CPU Load (%)

11025

352943.17 (min 352943.00, max 352944.00)

0.30 (min 0.19, max 0.42)

16000

512208.83 (min 512208.00, max 512209.00)

0.26 (min 0.17, max 0.42)

22050

705883.83 (min 705882.00, max 705888.00)

0.30 (min 0.25, max 0.41)

24000

705886.17 (min 705886.00, max 705887.00)

0.33 (min 0.29, max 0.36)

32000

1024415.50 (min 1024414.00, max 1024417.00)

0.30 (min 0.19, max 0.65)

44100

1411772.17 (min 1411772.00, max 1411773.00)

0.41 (min 0.38, max 0.45)

48000

1536622.00 (min 1536621.00, max 1536623.00)

0.50 (min 0.19, max 0.95)

88200

2823534.17 (min 2823531.00, max 2823536.00)

0.77 (min 0.73, max 0.81)

96000

3073226.00 (min 3073218.00, max 3073234.00)

0.56 (min 0.31, max 1.58)

2.5.2.4. Ethernet

Ethernet performance benchmarks were measured using Netperf 2.7.1 https://hewlettpackard.github.io/netperf/doc/netperf.html Test procedures were modeled after those defined in RFC-2544: https://tools.ietf.org/html/rfc2544, where the DUT is the TI device and the “tester” used was a Linux PC. To produce consistent results, it is recommended to carry out performance tests in a private network and to avoid running NFS on the same interface used in the test. In these results, CPU utilization was captured as the total percentage used across all cores on the device, while running the performance test over one external interface.

UDP Throughput (0% loss) was measured by the procedure defined in RFC-2544 section 26.1: Throughput. In this scenario, netperf options burst_size (-b) and wait_time (-w) are used to limit bandwidth during different trials of the test, with the goal of finding the highest rate at which no loss is seen. For example, to limit bandwidth to 500Mbits/sec with 1472B datagram:

burst_size = <bandwidth (bits/sec)> / 8 (bits -> bytes) / <UDP datagram size> / 100 (seconds -> 10 ms)
burst_size = 500000000 / 8 / 1472 / 100 = 425

wait_time = 10 milliseconds (minimum supported by Linux PC used for testing)

UDP Throughput (possible loss) was measured by capturing throughput and packet loss statistics when running the netperf test with no bandwidth limit (remove -b/-w options).

In order to start a netperf client on one device, the other device must have netserver running. To start netserver:

netserver [-p <port_number>] [-4 (IPv4 addressing)] [-6 (IPv6 addressing)]

Running the following shell script from the DUT will trigger netperf clients to measure bidirectional TCP performance for 60 seconds and report CPU utilization. Parameter -k is used in client commands to summarize selected statistics on their own line and -j is used to gain additional timing measurements during the test.

#!/bin/bash
for i in 1
do
   netperf -H <tester ip> -j -c -l 60 -t TCP_STREAM --
      -k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &

   netperf -H <tester ip> -j -c -l 60 -t TCP_MAERTS --
      -k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
done

Running the following commands will trigger netperf clients to measure UDP burst performance for 60 seconds at various burst/datagram sizes and report CPU utilization.

  • For UDP egress tests, run netperf client from DUT and start netserver on tester.

netperf -H <tester ip> -j -c -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
   -k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE

  • For UDP ingress tests, run netperf client from tester and start netserver on DUT.

netperf -H <DUT ip> -j -C -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
   -k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE

2.5.2.4.1. CPSW/CPSW2g/CPSW3g Ethernet Driver

  • CPSW2g: AM65x, J7200, J721e, J721S2, J784S4, J742S2

  • CPSW3g: AM64x, AM62x, AM62ax, AM62px

TCP Bidirectional Throughput

Table 2.15 CPSW2g TCP Bidirectional Throughput

Command Used

am62lxx_evm-fs: THROUGHPUT (Mbits/sec)

am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)

netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS

1250.65 (min 1162.63, max 1320.05)

99.45 (min 98.50, max 99.91)

TCP Bidirectional Throughput Interrupt Pacing

Table 2.16 CPSW2g TCP Bidirectional Throughput Interrupt Pacing

Command Used

am62lxx_evm-fs: THROUGHPUT (Mbits/sec)

am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)

netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS

1276.53 (min 1096.32, max 1414.15)

98.53 (min 97.46, max 99.97)

UDP Throughput

Table 2.17 CPSW2g UDP Egress Throughput 0 loss

Frame Size(bytes)

am62lxx_evm-fs: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

am62lxx_evm-fs: THROUGHPUT (Mbits/sec)

am62lxx_evm-fs: Packets Per Second (kPPS)

am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)

64

18.00

48.02 (min 45.73, max 49.33)

93.57 (min 89.00, max 96.00)

81.68 (min 79.75, max 83.43)

128

82.00

91.45 (min 88.25, max 95.00)

89.29 (min 86.00, max 93.00)

76.38 (min 50.57, max 81.56)

256

210.00

173.95 (min 160.08, max 184.15)

85.00 (min 78.00, max 90.00)

76.66 (min 59.01, max 80.28)

1024

978.00

470.51 (min 73.73, max 704.77)

57.50 (min 9.00, max 86.00)

56.11 (min 7.01, max 79.70)

1518

1472.00

666.11 (min 646.32, max 702.83)

54.71 (min 53.00, max 58.00)

72.76 (min 71.76, max 74.10)
Table 2.18 CPSW2g UDP Ingress Throughput 0 loss

Frame Size(bytes)

am62lxx_evm-fs: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

am62lxx_evm-fs: THROUGHPUT (Mbits/sec)

am62lxx_evm-fs: Packets Per Second (kPPS)

am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)

64

18.00

2.28 (min 2.25, max 2.36)

4.14 (min 4.00, max 5.00)

2.97 (min 1.94, max 6.96)

128

82.00

4.96 (min 4.40, max 5.43)

4.86 (min 4.00, max 5.00)

3.14 (min 2.03, max 4.94)

256

210.00

9.04 (min 1.02, max 10.85)

4.17 (min 0.00, max 5.00)

3.31 (min 0.98, max 6.33)

1024

978.00

38.40 (min 6.55, max 43.42)

4.50 (min 1.00, max 5.00)

4.55 (min 1.23, max 8.23)

1518

1472.00

52.60 (min 4.71, max 62.41)

4.17 (min 0.00, max 5.00)

5.17 (min 0.70, max 9.26)
Table 2.19 CPSW2g UDP Ingress Throughput possible loss

Frame Size(bytes)

am62lxx_evm-fs: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

am62lxx_evm-fs: THROUGHPUT (Mbits/sec)

am62lxx_evm-fs: Packets Per Second (kPPS)

am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)

am62lxx_evm-fs: Packet Loss %

64

18.00

72.96 (min 69.39, max 76.56)

142.57 (min 136.00, max 150.00)

82.64 (min 81.65, max 85.19)

72.14 (min 61.99, max 82.53)

128

82.00

144.83 (min 137.49, max 154.44)

141.43 (min 134.00, max 151.00)

85.13 (min 83.80, max 86.33)

68.86 (min 55.17, max 78.22)

256

210.00

280.59 (min 262.68, max 303.33)

137.00 (min 128.00, max 148.00)

85.01 (min 82.70, max 87.15)

49.21 (min 31.44, max 66.24)

1024

978.00

809.96 (min 576.67, max 890.96)

99.00 (min 70.00, max 109.00)

88.52 (min 84.69, max 92.80)

7.02 (min 4.65, max 10.64)

1518

1472.00

787.28 (min 704.86, max 853.41)

66.83 (min 60.00, max 72.00)

82.25 (min 73.30, max 87.11)

6.83 (min 2.70, max 10.83)

2.5.2.5. Linux OSPI Flash Driver

UBIFS

Table 2.20 OSPI Flash Driver

Buffer size (bytes)

am62lxx_evm-fs: Write UBIFS Throughput (Mbytes/sec)

am62lxx_evm-fs: Write UBIFS CPU Load (%)

am62lxx_evm-fs: Read UBIFS Throughput (Mbytes/sec)

am62lxx_evm-fs: Read UBIFS CPU Load (%)

102400

3.84 (min 3.55, max 4.17)

29.68 (min 24.49, max 34.58)

11.32 (min 11.19, max 11.63)

33.47 (min 31.25, max 35.29)

262144

3.77 (min 3.49, max 4.27)

28.92 (min 24.49, max 33.65)

11.31 (min 11.20, max 11.55)

33.93 (min 31.25, max 37.14)

524288

3.76 (min 3.45, max 4.15)

28.78 (min 25.49, max 33.98)

11.28 (min 11.19, max 11.50)

33.96 (min 31.25, max 35.29)

1048576

3.77 (min 3.44, max 4.13)

28.45 (min 24.72, max 32.20)

11.18 (min 11.05, max 11.31)

37.67 (min 35.29, max 38.89)

RAW

Table 2.21 OSPI Raw Flash Driver

File size (Mbytes)

am62lxx_evm-fs: Raw Read Throughput (Mbytes/sec)

50

17.36 (min 16.95, max 17.67)

2.5.2.6. EMMC Driver

Warning

IMPORTANT: The performance numbers can be severely affected if the media is mounted in sync mode. Hot plug scripts in the filesystem mount removable media in sync mode to ensure data integrity. For performance sensitive applications, umount the auto-mounted filesystem and re-mount in async mode.

2.5.2.6.1. EMMC EXT4 FIO 1G

Table 2.22 EMMC EXT4 FIO 1G

Buffer size (bytes)

am62lxx_evm-fs: Write EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Write EXT4 CPU Load (%)

am62lxx_evm-fs: Read EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Read EXT4 CPU Load (%)

1m

128.17 (min 128.00, max 129.00)

8.56 (min 8.30, max 8.86)

179.33 (min 179.00, max 180.00)

7.98 (min 7.62, max 8.22)

4m

127.50 (min 125.00, max 128.00)

5.46 (min 5.28, max 5.63)

179.50 (min 179.00, max 180.00)

4.95 (min 4.74, max 5.08)

4k

86.62 (min 86.40, max 86.90)

59.76 (min 59.27, max 60.23)

90.67 (min 88.60, max 93.50)

56.33 (min 55.48, max 56.62)

256k

122.83 (min 122.00, max 123.00)

11.83 (min 11.60, max 12.02)

178.33 (min 178.00, max 179.00)

10.08 (min 9.65, max 10.27)

2.5.2.6.2. EMMC RAW FIO 1G

Table 2.23 EMMC RAW FIO 1G

Buffer size (bytes)

am62lxx_evm-fs: Write Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Write Raw CPU Load (%)

am62lxx_evm-fs: Read Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Read Raw CPU Load (%)

1m

127.83 (min 122.00, max 129.00)

7.85 (min 7.48, max 7.99)

180.00

7.63 (min 7.16, max 8.07)

4m

129.00

5.33 (min 5.10, max 5.56)

180.00

4.67 (min 4.51, max 4.96)

4k

95.57 (min 93.50, max 96.40)

47.29 (min 45.85, max 48.11)

94.90 (min 94.00, max 95.80)

53.35 (min 52.55, max 54.08)

256k

122.67 (min 118.00, max 124.00)

9.76 (min 9.54, max 10.07)

178.17 (min 178.00, max 179.00)

9.31 (min 9.00, max 9.65)

2.5.2.6.3. EMMC EXT4

Table 2.24 EMMC EXT4

Buffer size (bytes)

am62lxx_evm-fs: Write EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Write EXT4 CPU Load (%)

am62lxx_evm-fs: Read EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Read EXT4 CPU Load (%)

102400

90.91 (min 83.00, max 97.71)

20.38 (min 17.87, max 25.90)

180.36 (min 177.78, max 181.30)

37.46 (min 36.45, max 39.09)

262144

94.89 (min 83.93, max 100.22)

20.88 (min 18.72, max 25.51)

183.72 (min 177.40, max 185.29)

39.71 (min 36.70, max 50.88)

524288

95.62 (min 84.98, max 101.40)

20.84 (min 18.31, max 26.53)

186.81 (min 186.21, max 187.09)

32.93 (min 31.82, max 34.26)

1048576

95.42 (min 82.97, max 100.97)

21.14 (min 18.45, max 26.53)

186.89 (min 186.56, max 187.07)

32.73 (min 32.11, max 33.33)

5242880

96.51 (min 85.25, max 101.40)

20.64 (min 17.87, max 25.82)

186.37 (min 185.98, max 186.68)

32.98 (min 32.43, max 33.64)

2.5.2.6.4. EMMC VFAT

Table 2.25 EMMC VFAT

Buffer size (bytes)

am62lxx_evm-fs: Write VFAT Throughput (Mbytes/sec)

am62lxx_evm-fs: Write VFAT CPU Load (%)

am62lxx_evm-fs: Read VFAT Throughput (Mbytes/sec)

am62lxx_evm-fs: Read VFAT CPU Load (%)

102400

32.73 (min 16.59, max 37.40)

15.81 (min 14.03, max 18.56)

173.62 (min 172.15, max 174.70)

39.10 (min 37.72, max 40.52)

262144

47.45 (min 20.52, max 63.09)

17.72 (min 13.82, max 23.74)

174.94 (min 164.63, max 177.64)

40.50 (min 36.61, max 55.74)

524288

56.02 (min 22.44, max 71.58)

18.76 (min 12.94, max 22.48)

177.43 (min 177.03, max 177.74)

32.31 (min 30.97, max 34.19)

1048576

61.76 (min 23.47, max 77.61)

19.71 (min 18.09, max 22.20)

177.54 (min 175.95, max 178.35)

34.28 (min 31.62, max 40.71)

5242880

71.24 (min 24.09, max 85.03)

20.62 (min 18.65, max 23.67)

176.90 (min 176.63, max 177.31)

32.18 (min 30.70, max 33.90)

2.5.2.7. UBoot EMMC Driver

Table 2.26 UBOOT EMMC RAW

File size (bytes in hex)

am62lxx_evm-fs: Write Throughput (Kbytes/sec)

am62lxx_evm-fs: Read Throughput (Kbytes/sec)

2000000

112634.05 (min 110702.70, max 116198.58)

174765.56 (min 173375.66, max 175229.95)

4000000

125497.58 (min 122268.66, max 127254.37)

178086.96

2.5.2.8. MMCSD

Warning

IMPORTANT: The performance numbers can be severely affected if the media is mounted in sync mode. Hot plug scripts in the filesystem mount removable media in sync mode to ensure data integrity. For performance sensitive applications, umount the auto-mounted filesystem and re-mount in async mode.

2.5.2.8.1. MMC EXT4 FIO 1G

Table 2.27 MMC EXT4 FIO 1G

Buffer size (bytes)

am62lxx_evm-fs: Write EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Write EXT4 CPU Load (%)

am62lxx_evm-fs: Read EXT4 Throughput (Mbytes/sec)

am62lxx_evm-fs: Read EXT4 CPU Load (%)

1m

35.31 (min 18.70, max 42.30)

3.38 (min 2.18, max 4.24)

87.01 (min 86.40, max 87.50)

5.29 (min 4.42, max 5.60)

4m

34.96 (min 18.70, max 42.40)

2.36 (min 1.75, max 2.97)

86.47 (min 83.20, max 87.40)

3.02 (min 2.66, max 3.38)

4k

3.23 (min 2.74, max 4.29)

7.39 (min 5.78, max 9.52)

13.10 (min 12.90, max 13.50)

12.51 (min 11.54, max 13.83)

256k

31.90 (min 18.00, max 38.20)

5.09 (min 3.67, max 6.32)

83.10 (min 82.00, max 83.90)

6.51 (min 5.61, max 6.78)

2.5.2.8.2. MMC RAW FIO 1G

Table 2.28 MMC RAW FIO 1G

Buffer size (bytes)

am62lxx_evm-fs: Write Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Write Raw CPU Load (%)

am62lxx_evm-fs: Read Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Read Raw CPU Load (%)

1m

31.02 (min 18.20, max 44.20)

2.80 (min 1.82, max 3.70)

87.63 (min 87.10, max 88.00)

4.61 (min 4.27, max 5.11)

4m

30.72 (min 18.20, max 43.60)

2.09 (min 1.32, max 2.68)

87.67 (min 87.00, max 88.10)

3.00 (min 2.61, max 3.48)

4k

3.45 (min 2.80, max 4.15)

5.24 (min 4.23, max 6.10)

13.32 (min 13.00, max 13.60)

11.41 (min 10.56, max 12.38)

256k

27.70 (min 17.60, max 38.60)

3.72 (min 2.58, max 4.57)

83.30 (min 82.10, max 84.10)

5.72 (min 4.84, max 6.15)

2.5.2.8.3. MMC EXT4

Table 2.29 MMC EXT4

Buffer size (bytes)

am62lxx_evm-fs: Write Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Write Raw CPU Load (%)

am62lxx_evm-fs: Read Raw Throughput (Mbytes/sec)

am62lxx_evm-fs: Read Raw CPU Load (%)

102400

9.60 (min 8.52, max 10.57)

2.45 (min 1.99, max 3.24)

10.84 (min 10.64, max 11.16)

2.81 (min 2.45, max 3.08)

262144

9.45 (min 8.73, max 10.75)

2.46 (min 1.98, max 3.38)

10.99 (min 10.74, max 11.20)

2.35 (min 2.06, max 2.67)

524288

9.52 (min 8.75, max 10.56)

2.42 (min 1.92, max 3.28)

11.13 (min 10.90, max 11.49)

2.10 (min 1.94, max 2.18)

1048576

9.46 (min 8.86, max 10.52)

2.50 (min 1.96, max 3.77)

11.20 (min 11.08, max 11.38)

2.23 (min 2.01, max 2.66)

5242880

9.53 (min 8.79, max 10.77)

2.38 (min 1.91, max 3.95)

11.46 (min 11.05, max 12.02)

2.23 (min 1.96, max 2.49)

The performance numbers were captured using the following:

  • SanDisk Max Endurance SD card (SDSQQVR-032G-GN6IA)

  • Partition was mounted with async option

2.5.2.9. UBoot MMCSD

2.5.2.9.1. UBOOT MMCSD FAT

Table 2.30 UBOOT MMCSD FAT

File size (bytes in hex)

am62lxx_evm-fs: Write Throughput (Kbytes/sec)

am62lxx_evm-fs: Read Throughput (Kbytes/sec)

400000

17604.61 (min 12962.03, max 19692.31)

22714.58 (min 22505.49, max 22882.68)

800000

20272.80 (min 18004.40, max 21501.31)

23152.29 (min 23076.06, max 23206.80)

1000000

19450.02 (min 16015.64, max 21250.32)

23311.43 (min 23239.72, max 23339.03)

The performance numbers were captured using the following:

  • SanDisk Max Endurance SD card (SDSQQVR-032G-GN6IA)


2.5.2.10. USB Driver

2.5.2.10.1. USB Device Controller

Table 2.31 USBDEVICE HIGHSPEED SLAVE_READ_THROUGHPUT

Number of Blocks

am62lxx_evm-fs: Throughput (MB/sec)

150

41.14 (min 36.20, max 43.50)
Table 2.32 USBDEVICE HIGHSPEED SLAVE_WRITE_THROUGHPUT

Number of Blocks

am62lxx_evm-fs: Throughput (MB/sec)

150

37.43 (min 31.00, max 41.40)

2.5.2.11. Low Power Performance

2.5.2.11.1. Power Performance

These are power measurements taken while the device is in various low power modes like Deep Sleep and RTC Only + DDR.

Table 2.33 Deep Sleep Power Performance

Rail name

Rail voltage(V)

Power (mW)

vdd_core

0.75

4.06

soc_dvdd_1v8

1.80

1.48

soc_dvdd_3v3

3.29

3.59

vdda_1v8

1.80

1.38

vdd_lpddr4_pmic2

1.10

1.04

vdd_rtc

0.74

0.01

vdd_rtc_1v8

1.80

0.02

Total

11.57
Table 2.34 RTC Only + DDR Power Performance

Rail name

Rail voltage(V)

Power (mW)

vdd_core

0.00

0.00

soc_dvdd_1v8

1.80

1.41

soc_dvdd_3v3

3.29

1.09

vdda_1v8

0.00

0.00

vdd_lpddr4_pmic2

1.10

1.09

vdd_rtc

0.75

0.01

vdd_rtc_1v8

1.80

0.02

Total

3.62

2.5.2.11.2. Resume Latency Performance

Table 2.35 LPM Resume Latency Performance

Low Power Mode

Total Resume Latency (ms)

RTC Only + DDR

456.44

Deep Sleep

109.96

The performance numbers are measured without the Linux printk logs. To remove the Linux printk logs, run the following commands in the terminal:

# Detach kernel serial console
consoles=$(find /sys/bus/platform/devices/*.serial/ -name console)
for console in ${consoles}; do
     echo -n N > ${console}
done

Note

The measurements shown are from using the default SDK with no extra optimizations.